clinical handbook of psychological...

CLINICAL HANDBOOK OFPSYCHOLOGICAL DISORDERS

CLINICAL HANDBOOK OF

PSYCHOLOGICALDISORDERS

A Step-by-Step Treatment Manual

F O U R T H E D I T I O N

EDITED BY

DAVID H. BARLOW

THE GUILFORD PRESSNew York London

To BeverlyFor love, loyalty, and dedication

© 2008 The Guilford PressA Division of Guilford Publications, Inc.72 Spring Street, New York, NY 10012www.guilford.com

All rights reserved

No part of this book may be reproduced, translated, stored in a retrieval system,or transmitted, in any form or by any means, electronic, mechanical, photocopying,microfilming, recording, or otherwise, without written permission from the Publisher.

Printed in the United States of America

This book is printed on acid-free paper.

Last digit is print number: 9 8 7 6 5 4 3 2 1

The authors have checked with sources believed to be reliable in their efforts to provideinformation that is complete and generally in accord with the standards of practice that areaccepted at the time of publication. However, in view of the possibility of human error orchanges in medical sciences, neither the authors, nor the editor and publisher, nor any otherparty who has been involved in the preparation or publication of this work warrants thatthe information contained herein is in every respect accurate or complete, and they are notresponsible for any errors or omissions or the results obtained from the use of suchinformation. Readers are encouraged to confirm the information contained in this book withother sources.

Library of Congress Cataloging-in-Publication Data

Clinical handbook of psychological disorders : a step-by-step treatment manual / edited byDavid H. Barlow.—4th ed.

p. ; cm.Includes bibliographical references and index.ISBN-13: 978-1-59385-572-7 (hardcover : alk. paper)ISBN-10: 1-59385-572-9 (hardcover : alk. paper)1. Behavior therapy—Handbooks, manuals, etc. 2. Medical protocols—Handbooks,

manuals, etc. I. Barlow, David H.[DNLM: 1. Mental Disorders—therapy. 2. Psychotherapy—methods. WM 420 C6415

2008]RC489.B4C584 2008616.89′1—dc22

2007015429

About the Editor

David H. Barlow, PhD, is Professor of Psychology and Psychiatry and Founder andDirector Emeritus of the Center for Anxiety and Related Disorders at Boston Uni-versity. He has published over 500 articles and chapters and close to 50 books andclinical manuals—some translated in over 20 languages, including Arabic, Chinese,and Russian—mostly in the area of emotional disorders and clinical researchmethodology. Dr. Barlow has been the recipient of numerous awards throughout hiscareer, most recently the American Board of Professional Psychology’s DistinguishedService Award to the Profession of Psychology in 2006. He is past president of theSociety of Clinical Psychology (Division 12) of the American Psychological Associa-tion and the Association for Advancement of Behavior Therapy (now the Associatonfor Behavioral and Cognitive Therapies) and past editor of the journals BehaviorTherapy, the Journal of Applied Behavior Analysis, and Clinical Psychology: Scienceand Practice. Dr. Barlow was chair of the American Psychological Association TaskForce on Psychological Intervention Guidelines, a member of the DSM-IV TaskForce of the American Psychiatric Association, and a cochair of the DSM-IV workgroup for revising the anxiety disorder categories. He is a Diplomate in Clinical Psy-chology of the American Board of Professional Psychology and maintains a privatepractice.

v

Contributors

Michael E. Addis, PhD, Department of Psychology, Clark University, Worcester,Massachusetts

Laura B. Allen, PhD, Center for Anxiety and Related Disorders and Department ofPsychology, Boston University, Boston, Massachusetts

Amy K. Bach, PhD, Department of Psychiatry and Human Behavior, Brown University,Providence, Rhode Island

David H. Barlow, PhD, Center for Anxiety and Related Disorders and Departmentof Psychology, Boston University, Boston, Massachusetts

Aaron T. Beck, MD, Department of Psychiatry, School of Medicine, University ofPennsylvania, Philadelphia, Pennsylvania, and Beck Institute for Cognitive Therapy andResearch, Bala Cynwyd, Pennsylvania

Kathryn L. Bleiberg, PhD, Department of Psychiatry, Weill Cornell Medical College,Cornell University, New York, New York

Andrew Christensen, PhD, Department of Psychology, University of California, LosAngeles, California

Zafra Cooper, DPhil, Department of Psychiatry, University of Oxford, Oxford, UnitedKingdom

Michelle G. Craske, PhD, Department of Psychology, University of California, LosAngeles, California

Elizabeth T. Dexter-Mazza, PsyD, Department of Psychology, University of Washington,Seattle, Washington

vi

Sona Dimidjian, PhD, Department of Psychology, University of Colorado, Boulder,Colorado

Christopher G. Fairburn, MD, Department of Psychiatry, Oxford University, Oxford,United Kingdom

Edna B. Foa, PhD, Department of Psychiatry and Center for the Treatment and Study ofAnxiety, University of Pennsylvania, Philadelphia, Pennsylvania

Martin E. Franklin, PhD, Department of Psychiatry and Center for the Treatment andStudy of Anxiety, University of Pennsylvania, Philadelphia, Pennsylvania

Sarah H. Heil, PhD, Departments of Psychiatry and Psychology, University of Vermont,Burlington, Vermont

Richard G. Heimberg, PhD, Adult Anxiety Clinic and Department of Psychology, TempleUniversity, Philadelphia, Pennsylvania

Ruth Herman-Dunn, PhD, private practice, Seattle, Washington

Stephen T. Higgins, PhD, Departments of Psychiatry and Psychology, University ofVermont, Burlington, Vermont

Neil S. Jacobson, PhD (deceased), Department of Psychology, University of Washington,Seattle, Washington

Marsha M. Linehan, PhD, Departments of Psychology and Psychiatry and BehavioralSciences, University of Washington, Seattle, Washington

Leanne Magee, MA, Adult Anxiety Clinic and Department of Psychology, TempleUniversity, Philadelphia, Pennsylvania

John C. Markowitz, MD, Department of Psychiatry, New York State Psychiatric Institute,New York, New York

Christopher R. Martell, PhD, private practice, Seattle, Washington, and Departmentof Psychiatry and Behavioral Sciences, University of Washington, Seattle, Washington

Barbara S. McCrady, PhD, Center on Alcoholism, Substance Abuse, and Addictions,and Department of Psychology, University of New Mexico, Albuquerque, New Mexico

R. Kathryn McHugh, MA, Center for Anxiety and Related Disorders and Departmentof Psychology, Boston University, Boston, Massachusetts

David J. Miklowitz, PhD, Department of Psychology, University of Colorado,Boulder, Colorado

Candice M. Monson, PhD, National Center for PTSD, VA Boston Healthcare System,Boston, Massachusetts

Patricia A. Resick, PhD, National Center for PTSD, VA Boston Healthcare System,Boston, Massachusetts

Shireen L. Rizvi, PhD, Department of Psychology, New School for Social Research,New York, New York

Jayne L. Rygh, PhD, Cognitive Therapy Center of New York, New York, New York

Roz Shafran, PhD, School of Psychology and Clinical Language Science, ReadingUniversity, Reading, United Kingdom

Contributors vii

Stacey C. Sigmon, PhD, Departments of Psychiatry and Psychology, Universityof Vermont, Burlington, Vermont

Nicholas Tarrier, PhD, Division of Clinical Psychology, School of Psychological Sciences,University of Manchester, Manchester, United Kingdom

Cynthia L. Turk, PhD, Anxiety Clinic and Department of Psychology, WashburnUniversity, Topeka, Kansas

Arthur D. Weinberger, PhD, Cognitive Therapy Center of New York, New York, NewYork

Jennifer G. Wheeler, PhD, private practice, Seattle, Washington

G. Terence Wilson, PhD, Graduate School of Applied and Professional Psychology,Rutgers, The State University of New Jersey, Piscataway, New Jersey

John P. Wincze, PhD, Department of Psychiatry and Human Behavior and Departmentof Psychology, Brown University, Providence, Rhode Island

Jeffrey E. Young, PhD, Department of Psychiatry, Columbia University, and CognitiveTherapy Center of New York, New York, New York

viii Contributors

Preface

Evidence-based practice (EBP) is one of those ideas that comes along occasionallyand takes the world by storm. Although some of the tenets of EBP have been aroundfor decades (as has this Handbook), it is only in the past 10 years that EBP has beenformally identified as a systematic method of delivering clinical care (Institute ofMedicine, 2001; Sackett, Strauss, Richardson, Rosenberg, & Haynes, 2000).

Since that time the “tipping point” (Gladwell, 2000) for EBP has clearly occurred,and health care policymakers and governments as well as professional societiesaround the world have collectively decided that the delivery of health care, includingbehavioral health care, should be based on evidence (APA Task Force on Evidence-Based Practice, 2006). Fulfilling this mandate comprises the goals of EBP, and hasalso been the goal of this book since the first edition was published in 1985.

The fourth edition of this book continues to represent a distinct departure fromany number of similar books reviewing advances in the treatment of psychologicaldisorders from the perspective of EBP. Over the past two decades we have developeda technology of behavior change that necessarily differs from disorder to disorder.

This technology consists of a variety of techniques or procedures with more or lessproven effectiveness for a given disorder (and increasingly for classes of disorders).Naturally, we have more evidence of the effectiveness of these treatments for somedisorders than for others. It also has become more apparent since the earlier editionsthat considerable clinical skill is required to apply this technology most effectively.Therefore, this book, in its fourth edition, is not another review of therapeutic pro-cedures for a given problem with recommendations for further research. Rather, it isa detailed description of actual treatment protocols in which experienced cliniciansimplement the technology of behavior change in the context of the most frequentlyencountered disorders.

ix

In this edition, the originators of some of the best-known psychological treatmentprotocols have revised and updated the descriptions of their interventions to reflectthe latest developments in an increasingly powerful array of psychological treat-ments. Among these revisions to existing chapters, several deserve comment. PatriciaA. Resick, Candice M. Monson, and Shireen L. Rizvi have updated their chapter onposttraumatic stress disorder with a new and tragic case fresh from the battlefieldsof Iraq. Their successful treatment of this individual suffering from the unspeakable(and intolerable) trauma of war is one consequence of today’s headlines that seldommakes it into print. Drug abuse continues as a scourge that ruins individual lives, thefunctioning of families, and the very fabric of society. Stephen T. Higgins, Stacey C.Sigmon, and Sarah H. Heil have expanded their chapter from a more narrow focuson cocaine to a broader focus on all serious drugs of abuse. The body of evidencesupporting interpersonal psychotherapy (IPT) has continued to grow in recent years,and for the first time in this edition two of the leading clinicians and teachers of IPT,Kathryn L. Bleiberg and John C. Markowitz, illustrate their fascinating approach tothe treatment of depression in the context of an illuminating case.

In addition, there is much that appears for the first time in this edition, reflectingmajor developments in the past several years. Four original treatment protocolsmake their appearance—two that focus on specific disorders and two that cut acrossdisorders. Chapter 8, by Sona Dimidjian, Christopher R. Martell, Michael E. Addis,and Ruth Herman-Dunn, describes the application of behavioral activation to de-pression. A variety of new evidence suggests that this approach is one of the moreexciting innovations in recent years, but because it is so new, few are aware of the in-tricacies of its application to depressed patients. Chapter 11, by Nicholas Tarrier,presents the latest developments on psychological approaches to schizophrenia di-rected mostly at the “positive” symptoms of hallucinations and delusions. Much ofthis work has been innovated in the United Kingdom, and is unfamiliar to many inNorth America despite the robust nature of evidence for efficacy.

Finally, there is growing consensus that the future of EBP will be to distill princi-ples of effective change that cut across diagnostic conditions, making them moregenerally applicable. Two of these “unified” or “transdiagnostic” protocols are pre-sented for the first time in this edition. Chapter 5 presents our own unified approachto emotional disorders, and Chapter 14 describes a transdiagnostic approach to eat-ing disorders by the originators, Christopher G. Fairburn, Zafra Cooper, RozShafran, and G. Terence Wilson. As with all chapters, the nuts and bolts of clinicalapplication are emphasized.

As with the previous editions, this book was motivated by countless clinical psy-chology graduate students, psychiatric residents, and other mental health profes-sionals, either in training or in practice, asking, “But how do I do it?” Realizing thatthere is no single source in which to find step-by-step treatment protocols for use asa guide to practice, this book attempts to fill the void. To accomplish this purpose, anumber of specific topics are common to most chapters. Each chapter begins with abrief review of our knowledge of the specific disorder (or class of disorders), fol-lowed by a description of the particular model or mini-theory that guides the tech-nology utilized with the disorder in question. This model, or mini-theory, typicallyanswers the question, What particular facets of the disorder should be assessed and

x Preface

treated? While clinical application always dilutes theoretical models, clinicians willrecognize behavioral and systems approaches with some psychodynamic contribu-tions as the predominant theoretical context.

This model is followed by a description of the typical setting in which the treat-ment is carried out. The setting varies from disorder to disorder, ranging from themore usual office setting to the home environment of the patient. Similar detailed de-scriptions of the social context of treatment (e.g., the importance of the involvementof family or friends) as well as therapist and client variables that are important with-in the context of the particular problem are discussed. For example, therapist vari-ables that may be important in implementing techniques for treatment of agorapho-bia or couple distress are described. In addition, the implications for treatment ofclient variables such as dependency and unassertiveness in individuals with panicdisorder with agoraphobia are discussed.

A detailed description of the actual step-by-step process of assessment and treat-ment follows, liberally sprinkled in many chapters with transcripts of therapy ses-sions. Important components of this process are the specifics of the rationale givento the patient before treatment, as well as typical problems that arise during the im-plementation of the technology. Where data exist, information on clinical predictorsof success or failure is provided.

In accomplishing the rather ambitious goals described above, I was very fortunatein this edition of the book, as in previous editions, to have leading clinicians and re-searchers document in some detail how they actually treat their patients. Onceagain, these authorities reported that the number of details they had to include in or-der to convey how they actually applied their treatment programs went far beyondtheir expectations. My hope is that practicing clinicians and clinical students every-where will benefit from acquaintance with these details.

In closing, I would like to express my deep appreciation to Mara Fleischer, my re-search and administrative assistant during the editing of this book. She worked withme and the authors every step of the way. I am sure this information will come inhandy as she is now pursuing her own doctorate in clinical psychology.

DAVID H. BARLOW, PhD

REFERENCES

APA Task Force on Evidence-Based Practice. (2006). Evidence-based practice in psychology. American Psy-chologist, 61, 271–285.

Gladwell, M. (2000). The tipping point: How little things can make a big difference. Boston: Little, Brown.Institute of Medicine. (2001). Crossing the quality chasm: A new health system for the 21st century. Washing-

ton, DC: National Academies Press.Sackett, D. L., Strauss. S. E., Richardson, W. S., Rosenberg, W., & Haynes, R. B. (2000). Evidence-based medi-

cine: How to practice and teach EBM (2nd ed.). London: Churchill Livingstone.

Preface xi

Contents

CHAPTER 1. Panic Disorder and Agoraphobia 1MICHELLE G. CRASKE and DAVID H. BARLOW

CHAPTER 2. Posttraumatic Stress Disorder 65PATRICIA A. RESICK, CANDICE M. MONSON,and SHIREEN L. RIZVI

CHAPTER 3. Social Anxiety Disorder 123CYNTHIA L. TURK, RICHARD G. HEIMBERG, and LEANNE MAGEE

CHAPTER 4. Obsessive–Compulsive Disorder 164MARTIN E. FRANKLIN and EDNA B. FOA

CHAPTER 5. Emotional Disorders: A Unified Protocol 216LAURA B. ALLEN, R. KATHRYN MCHUGH,and DAVID H. BARLOW

CHAPTER 6. Cognitive Therapy for Depression 250JEFFREY E. YOUNG, JAYNE L. RYGH, ARTHUR D. WEINBERGER,and AARON T. BECK

CHAPTER 7. Interpersonal Psychotherapy for Depression 306KATHRYN L. BLEIBERG and JOHN C. MARKOWITZ

CHAPTER 8. Behavioral Activation for Depression 328SONA DIMIDJIAN, CHRISTOPHER R. MARTELL, MICHAEL E. ADDIS,and RUTH HERMAN-DUNN

xiii

CHAPTER 9. Dialectical Behavior Therapyfor Borderline Personality Disorder

365

MARSHA M. LINEHAN and ELIZABETH T. DEXTER-MAZZA

CHAPTER 10. Bipolar Disorder 421DAVID J. MIKLOWITZ

CHAPTER 11. Schizophrenia and Other Psychotic Disorders 463NICHOLAS TARRIER

CHAPTER 12. Alcohol Use Disorders 492BARBARA S. MCCRADY

CHAPTER 13. Drug Abuse and Dependence 547STEPHEN T. HIGGINS, STACEY C. SIGMON, and SARAH H. HEIL

CHAPTER 14. Eating Disorders: A Transdiagnostic Protocol 578CHRISTOPHER G. FAIRBURN, ZAFRA COOPER, ROZ SHAFRAN,and G. TERENCE WILSON

CHAPTER 15. Sexual Dysfunction 615JOHN P. WINCZE, AMY K. BACH, and DAVID H. BARLOW

CHAPTER 16. Couple Distress 662ANDREW CHRISTENSEN, JENNIFER G. WHEELER,and NEIL S. JACOBSON

Author Index 691

Subject Index 706

xiv Contents

C H A P T E R 1

Panic Disorder and Agoraphobia

MICHELLE G. CRASKEDAVID H. BARLOW

The treatment protocol described in this chapter represents one of the success stories inthe development of empirically supported psychological treatments. Results from numer-ous studies indicate that this approach provides substantial advantages over placebomedication or alternative psychosocial approaches containing “common” factors, such aspositive expectancies and helpful therapeutic alliances. In addition, this treatment formsan important part of every clinical practice guideline in either public health or othersources from countries around the world, describing effective treatments for panic disor-der and agoraphobia. Results from numerous studies evaluating this treatment protocol,both individually and in combination with leading pharmacological approaches, suggestthat this approach is equally effective as the best pharmacological approaches in the shortterm and more durable over the long term. But this treatment protocol has not stood still.For example, we have learned a great deal in the past 5 years about neurobiologicalmechanisms of action in fear reduction, and the best psychological methods for effectingthese changes. In this chapter we present the latest version of this protocol, incorporatingthese changes and additions as illustrated in a comprehensive account of the treatment of“Julie.”—D. H. B.

Advances continue in the development of bio-psychosocial models and cognitive-behavioraltreatments for panic disorder and agoraphobia.The conceptualization of panic disorder as anacquired fear of certain bodily sensations, andagoraphobia as a behavioral response to theanticipation of such bodily sensations or theircrescendo into a full-blown panic attack, con-tinues to be supported by experimental, clini-cal, and longitudinal research. Furthermore,the efficacy of cognitive-behavioral treatmentsthat target fear of bodily sensations and associ-ated agoraphobic situations is well established.In addition to presenting an up-to-date review

of treatment outcome data, this chapter coversrecent theoretical and empirical developmentsin reference to etiological factors, the roleof comorbid diagnoses in treatment, ways ofoptimizing learning during exposure therapy,and the effect of medication on cognitive-behavioral treatments. The chapter concludeswith a detailed, session-by-session outline ofcognitive-behavioral treatment for panic disor-der with agoraphobia (PDA). This protocol hasbeen developed in our clinics; the full proto-col is detailed in available treatment manuals(Barlow & Craske, 2006; Craske & Barlow,2006).

1

NATURE OF PANICAND AGORAPHOBIA

Panic Attacks

“Panic attacks” are discrete episodes of intensedread or fear, accompanied by physical andcognitive symptoms, as listed in the DSM-IV-TR panic attack checklist (American Psychiat-ric Association, 2000). Panic attacks are dis-crete by virtue of their sudden or abrupt onsetand brief duration, as opposed to gradual-ly building anxious arousal. Panic attacks inpanic disorder often have an unexpected qual-ity, meaning that from the patient’s perspective,they appear to happen without an obvious trig-ger or at unexpected times. Indeed, the diagno-sis of panic disorder is given in the case of re-current “unexpected” panic attacks, followedby at least 1 month of persistent concern abouttheir recurrence and their consequences, or bya significant change in behavior consequent tothe attacks (American Psychiatric Association,1994).

As with all basic emotions (Izard, 1992),panic attacks are associated with strong actiontendencies; Most often, these are urges to es-cape, and less often, urges to fight. These fightand flight tendencies usually involve elevatedautonomic nervous system arousal needed tosupport such fight–flight reactivity. Further-more, perceptions of imminent threat or dan-ger, such as death, loss of control, or socialridicule, often accompany such fight–flight re-activity. However, the features of urgency to es-cape, autonomic arousal, and perception ofthreat are not present in every self-reported oc-currence of panic. For example, despite evi-dence for elevated heart rate or other indices ofsympathetic nervous system activation duringpanic attacks on average (e.g., Wilkinson etal., 1998), Margraf, Taylor, Ehlers, Roth, andAgras (1987) found that 40% of self-reportedpanic attacks were not associated with acceler-ated heart rate. Moreover, in general, patientswith panic disorder are more likely than non-anxious controls to report arrhythmic heartrate in the absence of actual arrhythmias(Barsky, Clearly, Sarnie, & Ruskin, 1994).Heightened anxiety about signs of autonomicarousal may lead patients to perceive cardiacevents when none exist (Barlow, Brown, &Craske, 1994; Craske & Tsao, 1999). We be-lieve that self-reported panic in the absence ofheart rate acceleration or other indices of auto-

nomic activation reflects anticipatory anxietyrather than true panic (Barlow et al., 1994), es-pecially because more severe panics are moreconsistently associated with accelerated heartrate (Margraf et al., 1987). Another example ofdiscordance occurs when perceptions of threator danger are refuted despite the report of in-tense fear. This has been termed “noncog-nitive” panic (Rachman, Lopatka, & Levitt,1988). Finally, the urgency to escape is some-times weakened by situational demands forcontinued approach and endurance, suchas performance expectations or job demands,thus creating discordance between behavioralresponses on the one hand, and verbal or physi-ological responses on the other.

A subset of individuals with panic disorderexperience nocturnal panic attacks. “Noctur-nal panic” refers to waking from sleep in astate of panic with symptoms that are very sim-ilar to panic attacks during wakeful states(Craske & Barlow, 1989; Uhde, 1994). Noc-turnal panic does not refer to waking fromsleep and panicking after a lapse of wakingtime, or nighttime arousals induced by night-mares or environmental stimuli (e.g., unex-pected noises). Instead, nocturnal panic is anabrupt waking from sleep in a state of panic,without an obvious trigger. Nocturnal panic at-tacks reportedly most often occur between 1and 3 hours after sleep onset, and only occa-sionally more than once per night (Craske &Barlow, 1989). Surveys of select clinical groupssuggest that nocturnal panic is relatively com-mon among individuals with panic disorder:44–71% report having experienced nocturnalpanic at least once, and 30–45% report re-peated nocturnal panics (Craske & Barlow,1989; Krystal, Woods, Hill, & Charney, 1991;Mellman & Uhde, 1989; Roy-Byrne, Mellman,& Uhde, 1988; Uhde, 1994). Individuals whosuffer frequent nocturnal panic often becomefearful of sleep and attempt to delay sleep on-set. Avoidance of sleep may result in chronicsleep deprivation, which in turn precipitatesmore nocturnal panics (Uhde, 1994).

“Nonclinical” panic attacks occur occasion-ally in approximately 3–5% of people in thegeneral population who do not otherwise meetcriteria for panic disorder (Norton, Cox, &Malan, 1992). Also, panic attacks occur acrossa variety of anxiety and mood disorders(Barlow et al., 1985), and are not limited topanic disorder. As stated earlier, the defining

2 CLINICAL HANDBOOK OF PSYCHOLOGICAL DISORDERS

feature of panic disorder is not the presence ofpanic attacks per se, but involves addition-al anxiety about the recurrence of panic orits consequences, or a significant behavioralchange because of the panic attacks. It is theadditional anxiety about panic combined withcatastrophic cognitions in the face of panic thatdifferentiate between the person with panic dis-order and the occasional nonclinical panicker(e.g., Telch, Lucas, & Nelson, 1989) or the per-son with other anxiety disorders who also hap-pens to panic. The following scenario exempli-fies the latter point.

PATIENT: Sometimes I lay awake at night think-ing about a million different things. I thinkabout what is going to happen to my daugh-ter if I get sick. Who will look after her, orwhat would happen if my husband died andwe didn’t have enough money to give mydaughter a good education? Then I thinkabout where we would live and how wewould cope. Sometimes I can work myselfup so much that my heart starts to race, myhands get sweaty, and I feel dizzy and scared.So I have to stop myself from thinking aboutall those things. I usually get out of bed andturn on the TV—anything to get my mindoff the worries.

THERAPIST: Do you worry about the feelings ofa racing heart, sweating, and dizziness hap-pening again?

PATIENT: No. They’re unpleasant, but they arethe least of my concerns. I am more worriedabout my daughter and our future.

This scenario illustrates the experience ofpanic that is not the central focus of the per-son’s anxiety. More likely, this woman has gen-eralized anxiety disorder, and her uncontrolla-ble worry leads her to panic on occasion. Thenext example is of someone with social phobia,who becomes very concerned about panickingin social situations, because the possibility of apanic attack increases her concerns about beingjudged negatively by others.

PATIENT: I am terrified of having a panic attackin meetings at work. I dread the thought ofothers noticing how anxious I am. Theymust be able to see my hands shaking, thesweat on my forehead, and worst of all, myface turning red.

THERAPIST: What worries you most about oth-ers noticing your physical symptoms?

PATIENT: That they will think that I am weirdor strange.

THERAPIST: Would you be anxious in the meet-ings if the panic attacks were fully prevent-able?

PATIENT: I would still be worried about doingor saying the wrong thing. It is not just thepanic attacks that worry me.

THERAPIST: Are you worried about panic at-tacks in any other situations?

PATIENT: Formal social events and sometimeswhen I meet someone for the first time.

In this case, even though the patient experi-ences panic attacks, the real concern is aboutbeing judged negatively by others consequentto panic attacks, and the panic attacks do notoccur in situations other than social ones.Hence, this presentation is most aptly de-scribed as social phobia.

Agoraphobia

“Agoraphobia” refers to avoidance or endur-ance with dread of situations from which es-cape might be difficult or help unavailable inthe event of a panic attack, or in the event ofdeveloping symptoms that could be incapaci-tating and embarrassing, such as loss of bowelcontrol or vomiting. Typical agoraphobic sit-uations include shopping malls, waiting inline, movie theaters, traveling by car or bus,crowded restaurants, and being alone. “Mild”agoraphobia is exemplified by the person whohesitates about driving long distances alone butmanages to drive to and from work, prefers tosit on the aisle at movie theaters but still goes tomovies, and avoids crowded places. “Moder-ate” agoraphobia is exemplified by the personwhose driving is limited to a 10-mile radiusfrom home and only if accompanied, whoshops at off-peak times and avoids large super-markets, and who avoids flying or traveling bytrain. “Severe” agoraphobia refers to very lim-ited mobility, sometimes even to the point ofbecoming housebound.

Not all persons who panic develop agora-phobia, and the extent of agoraphobia thatemerges is highly variable (Craske & Barlow,1988). Various factors have been investigated

Panic Disorder and Agoraphobia 3

as potential predictors of agoraphobia. Al-though agoraphobia tends to increase as his-tory of panic lengthens, a significant pro-portion of individuals panic for many yearswithout developing agoraphobic limitations.Nor is agoraphobia related to age of onset orfrequency of panic (Cox, Endler, & Swinson,1995; Craske & Barlow, 1988; Kikuchi et al.,2005; Rapee & Murrell, 1988). Some studiesreport more intense physical symptoms duringpanic attacks when there is more agoraphobia(e.g., de Jong & Bouman, 1995; Goisman etal., 1994; Noyes, Clancy, Garvey, & Anderson,1987; Telch, Brouillard, Telch, Agras, & Tay-lor, 1989). Others fail to find such differences(e.g., Cox et al., 1995; Craske, Miller, Ro-tunda, & Barlow, 1990). On the one hand,fears of dying, going crazy, or losing control donot relate to level of agoraphobia (Cox et al.,1995; Craske, Rapee, & Barlow, 1988). Onthe other hand, concerns about social conse-quences of panicking may be stronger whenthere is more agoraphobia (Amering et al.,1997; de Jong & Bouman, 1995; Rapee &Murrell, 1988; Telch, Brouilard, et al., 1989).In addition, in a recent investigation, Kikuchiand colleagues (2005) found that individualswho developed agoraphobia within 6 monthsof the onset of panic disorder had a higherprevalence of generalized anxiety disorder butnot major depression. However, whether thesocial evaluation concerns or comorbidity areprecursors or are secondary to agoraphobia re-mains to be determined. Occupational statusalso predicts agoraphobia, accounting for 18%of the variance in one study (de Jong &Bouman, 1995). Perhaps the strongest predic-tor of agoraphobia is sex; the ratio of males tofemales shifts dramatically in the direction offemale predominance as level of agoraphobiaworsens (e.g., Thyer, Himle, Curtis, Cameron,& Nesse, 1985).

PRESENTING FEATURES

From the latest epidemiological study, the Na-tional Comorbidity Survey Replication (NCS-R; Kessler, Berglund, Demler, Jin, & Walters,2005; Kessler, Chiu, Demler, & Walters, 2005)prevalence estimates for panic disorder with orwithout agoraphobia (PD/PDA) are 2.7% (12month) and 4.7% (lifetime). These rates arehigher than those reported in the original NCS(Kessler et al., 1994) and the older Epide-

miologic Catchment Area (ECA; Myers et al.,1984) study.

Individuals with agoraphobia who seektreatment almost always report that a historyof panic preceded their development of avoid-ance (Goisman et al., 1994; Wittchen, Reed, &Kessler, 1998). In contrast, epidemiologicaldata indicate that a subset of the populationexperiences agoraphobia without a history ofpanic disorder: 0.8% in the last 12 months(Kessler, Chiu, et al., 2005) and 1.4% lifetimeprevalence (Kessler, Berglund, et al., 2005).The discrepancy between clinical and epidemi-ological data has been attributed to misdiag-nosis of generalized anxiety, specific and socialphobias, and reasonable cautiousness aboutcertain situations (e.g., walking alone in un-safe neighborhoods) as agoraphobia in epide-miological samples (Horwath, Lish, Johnson,Hornig, & Weissman, 1993), and to the factthat individuals who panic are more likely toseek help (Boyd, 1986).

Rarely does the diagnosis of PD/PDA occurin isolation. Commonly co-occurring Axis Iconditions include specific phobias, social pho-bia, dysthymia, generalized anxiety disorder,major depressive disorder, and substance abuse(e.g., Brown, Campbell, Lehman, Grishman, &Mancill, 2001; Goisman, Goldenberg, Vasile,& Keller, 1995; Kessler, Chiu, et al., 2005).Also, 25–60% of persons with panic disorderalso meet criteria for a personality disorder,mostly avoidant and dependent personality dis-orders (e.g., Chambless & Renneberg, 1988).However, the nature of the relationship be-tween PD/PDA and personality disorders re-mains unclear. For example, comorbidity ratesare highly dependent on the method used to es-tablish Axis II diagnosis, as well as the co-occurrence of depressed mood (Alneas &Torgersen, 1990; Chambless & Renneberg,1988). Moreover, the fact that abnormal per-sonality traits improve and some “personalitydisorders” even remit after successful treat-ment of PD/PDA (Black, Monahan, Wesner,Gabel, & Bowers, 1996; Mavissakalian &Hamman, 1987; Noyes, Reich, Suelzer, &Christiansen, 1991) raises questions about thevalidity of Axis II diagnoses. The issue ofcomorbidity with personality disorders and itseffect on treatment for PD/PDA is described inmore detail in a later section.

The modal age of onset is late teenage yearsand early adulthood (Kessler, Berglund, et al.,2005). In fact, a substantial proportion of ado-


lescents report panic attacks (e.g., Hayward etal., 1992), and panic disorder in children andadolescents tends to be chronic and comorbidwith other anxiety, mood, and disruptive disor-ders (Biederman, Faraone, Marrs, & Moore,1997). Treatment is usually sought at a muchlater age, around 34 years (e.g., Noyes et al.,1986). The overall ratio of females to males isapproximately 2:1 (Kessler et al., 2006), and,as mentioned already, the ratio shifts dramati-cally in the direction of female predominanceas level of agoraphobia worsens (e.g., Thyer etal., 1985).

Most (approximately 72%) (Craske et al.,1990) report identifiable stressors around thetime of their first panic attack, including inter-personal stressors and stressors related to phys-ical well-being, such as negative drug experi-ences, disease, or death in the family. However,the number of stressors does not differ from thenumber experienced prior to the onset of othertypes of anxiety disorders (Pollard, Pollard, &Corn, 1989; Rapee, Litwin, & Barlow, 1990;Roy-Byrne, Geraci, & Uhde, 1986). Approxi-mately one-half report having experienced pan-icky feelings at some time before their firstpanic, suggesting that onset may be either in-sidious or acute (Craske et al., 1990).

Finally, PD/PDA tend to be chronic condi-tions, with severe financial and interpersonalcosts; that is, only a minority of untreated indi-viduals remit without subsequent relapse with-in a few years (30%), although a similar num-ber experience notable improvement, albeitwith a waxing and waning course (35%)(Katschnig & Amering, 1998; Roy-Byrne &Cowley, 1995). Also, individuals with panicdisorder overutilize medical resources com-pared to the general public and individualswith other “psychiatric” disorders (e.g., Katonet al., 1990; Roy-Byrne et al., 1999).

HISTORY OF PSYCHOLOGICALTREATMENT FOR PANIC DISORDERAND AGORAPHOBIA

It was not until the publication of DSM-III(American Psychiatric Association, 1980) thatPD/PDA was recognized as a distinct anxietyproblem. Until that time, panic attacks wereviewed primarily as a form of free-floating anx-iety. Consequently, psychological treatment ap-proaches were relatively nonspecific. They in-cluded relaxation and cognitive restructuring

for stressful life events in general (e.g., Barlowet al., 1984). Many presumed that pharmaco-therapy was necessary for the control of panic.In contrast, the treatment of agoraphobia wasquite specific from the 1970s onward, with pri-marily exposure-based approaches to targetfear and avoidance of specific situations. How-ever, relatively little consideration was given topanic attacks in either the conceptualization ortreatment of agoraphobia. The development ofspecific panic control treatments in the middleto late 1980s shifted interest away from agor-aphobia. Interest in agoraphobia was sub-sequently renewed, specifically in terms ofwhether panic control treatments are sufficientfor the management of agoraphobia, andwhether their combination with treatmentsthat directly target agoraphobia is superioroverall. We address these questions in more de-tail after describing the conceptualization thatunderlies cognitive-behavioral approaches tothe treatment of panic and agoraphobia.

CONCEPTUALIZATION OFETIOLOGICAL AND MAINTAININGFACTORS FOR PANIC DISORDERAND AGORAPHOBIA

Several independent lines of research (Barlow,1988; Clark, 1986; Ehlers & Margraf, 1989)converged in the 1980s on the same basic con-ceptualization of panic disorder as an acquiredfear of bodily sensations, particularly sensa-tions associated with autonomic arousal. Psy-chological and biological predispositions arebelieved to enhance the vulnerability to acquiresuch fear. These interacting vulnerabilities havebeen organized into an etiological conceptionof anxiety disorders in general, referred to as“triple vulnerability theory” (Barlow, 1988,2002; Suárez, Bennett, Goldstein, & Barlow, inpress). First, genetic contributions to the devel-opment of anxiety and negative affect consti-tute a generalized (heritable) biological vul-nerability. Second, evidence also supports ageneralized psychological vulnerability to ex-perience anxiety and related negative affectivestates, characterized by a diminished senseof control arising from early developmentalexperiences. Although the unfortunate co-occurrence of generalized biological and psy-chological vulnerabilities may be sufficient toproduce anxiety and related states, particularlygeneralized anxiety disorder and depression, a


third vulnerability seems necessary to accountfor the development of at least some specificanxiety disorders, including panic disorder;that is, early learning experiences in some in-stances seem to focus anxiety on particular ar-eas of concern. In panic disorder, the experi-ence of certain somatic sensations becomesassociated with a heightened sense of threatand danger. This specific psychological vulner-ability, when coordinated with the generalizedbiological and psychological vulnerabilitiesmentioned earlier, seems to contribute to thedevelopment of panic disorder. Fear condition-ing, avoidant responding, and information-processing biases are believed to perpetuatesuch fear. It is the perpetuating factors that aretargeted in the cognitive-behavioral treatmentapproach. What follows is a very brief reviewof some contributory factors with practical rel-evance for panic disorder.

Three Vulnerability Factors

Genetics and Temperament

The temperament most associated with anxietydisorders, including panic disorder, is neuroti-cism (Eysenck, 1967; Gray, 1982), or prone-ness to experience negative emotions in re-sponse to stressors. A closely linked construct,“negative affectivity,” is the tendency to experi-ence a variety of negative emotions across a va-riety of situations, even in the absence of ob-jective stressors (Watson & Clark, 1984).Structural analyses confirm that negative affectis a higher-order factor that distinguishes indi-viduals with each anxiety disorder (and depres-sion) from controls with no mental disorder:Lower-order factors discriminate among anxi-ety disorders, with “fear of fear” being the fac-tor that discriminates panic disorder fromother anxiety disorders (Brown, Chorpita, &Barlow, 1998; Zinbarg & Barlow, 1996). Theanxiety disorders load differentially on nega-tive affectivity, with more pervasive anxietydisorders, such as generalized anxiety disorder,loading more heavily, panic disorder loading atan intermediate level, and social anxiety disor-der loading the least (Brown et al., 1998).1

However, these findings derive from cross-sectional data sets.

Longitudinal prospective evidence for therole of neuroticism in predicting the onset ofpanic disorder is relatively limited. Specifically,neuroticism predicted the onset of panic at-

tacks in adolescents (Hayward, Killen,Kraemer, & Taylor, 2000; Schmidt, Lerew, &Jackson, 1997, 1999), and “emotional reactiv-ity” at age 3 was a significant variable in theclassification of panic disorder in 18- to 21-year-old males (Craske, Poulton, Tsao, &Plotkin, 2001). Ongoing studies, such as theNorthwestern/UCLA Youth Emotion Project,are comprehensively evaluating the role of neu-roticism in the prediction of subsequent panicdisorder.

Numerous multivariate genetic analyses ofhuman twin samples consistently attribute ap-proximately 30–50% of variance in neuroti-cism to additive genetic factors (Eley, 2001;Lake, Eaves, Maes, Heath, & Martin, 2000).In addition, anxiety and depression appear tobe variable expressions of the heritable ten-dency toward neuroticism (Kendler, Heath,Martin, & Eaves, 1987). Symptoms of panic(i.e., breathlessness, heart pounding) may beadditionally explained by a unique source ofgenetic variance that is differentiated fromsymptoms of depression and anxiety (Kendleret al., 1987) and neuroticism (Martin, Jardine,Andrews, & Heath, 1988).

Analyses of specific genetic markers remainpreliminary and inconsistent. For example,panic disorder has been linked to a locuson chromosome 13 (Hamilton et al., 2003;Schumacher et al., 2005) and chromosome 9(Thorgeirsson et al., 2003), but the exact genesremain unknown. Findings regarding markersfor the cholecystokinin-B receptor gene havebeen inconsistent (cf. Hamilton et al. [2001]and van Megen, Westenberg, Den Boer, &Kahn [1996]). Also, association and linkagestudies implicate the adenosine receptor gene inpanic disorder (Deckert et al., 1998; Hamiltonet al., 2004). But studies of genes involved inneurotransmitter systems associated with fearand anxiety have produced inconsistent results(see Roy-Byrne, Craske, & Stein, 2006). Thus,there is no evidence at this point for a specificlink between genetic markers and tempera-ment, on the one hand, and panic disorder onthe other. Rather, neurobiological factors seemto comprise a nonspecific biological vulnerabil-ity.

Anxiety Sensitivity

As described earlier, neuroticism is viewed as ahigher-order factor characteristic of all anxietydisorders, with “fear of fear” being more


unique to panic disorder. The construct “fearof fear” overlaps with the construct anxietysensitivity, or the belief that anxiety and its as-sociated symptoms may cause deleterious phys-ical, social, and psychological consequencesthat extend beyond any immediate physicaldiscomfort during an episode of anxiety orpanic (Reiss, 1980). Anxiety sensitivity is ele-vated across most anxiety disorders, but it isparticularly elevated in panic disorder (e.g.,Taylor, Koch, & McNally, 1992; Zinbarg &Barlow, 1996), especially the Physical Con-cerns subscale of the Anxiety Sensitivity Index(Zinbarg & Barlow, 1996; Zinbarg, Barlow, &Brown, 1997). Therefore, beliefs that physicalsymptoms of anxiety are harmful seem to beparticularly relevant to panic disorder and maycomprise a specific psychological vulnerability.

Anxiety sensitivity is presumed to confer arisk factor for panic disorder, because it primesfear reactivity to bodily sensations. In support,anxiety sensitivity predicts subjective distressand reported symptomatology in response toprocedures that induce strong physical sensa-tions, such as CO2 inhalation (Forsyth, Palav,& Duff, 1999), balloon inflation (Messenger &Shean, 1998), and hyperventilation (Sturges,Goetsch, Ridley, & Whittal, 1998) in nonclini-cal samples, even after researchers control forthe effects of trait anxiety (Rapee & Medoro,1994). In addition, several longitudinal studiesindicate that high scores on the Anxiety Sensi-tivity Index predict the onset of panic attacksover 1- to 4-year intervals in adolescents (Hay-ward et al., 2000), college students (Maller& Reiss, 1992), and community sampleswith specific phobias or no anxiety disorders(Ehlers, 1995). The predictive relationship re-mains after controlling for prior depression(Hayward et al., 2000). In addition, AnxietySensitivity Index scores predicted spontaneouspanic attacks and worry about panic (and anxi-ety more generally), during an acute militarystressor (i.e., 5 weeks of basic training), evenafter controlling for history of panic attacksand trait anxiety (Schmidt et al., 1997, 1999).Finally, panic attacks themselves elevate anxi-ety sensitivity over a 5-week period in adults(Schmidt et al., 1999), and over a 1-year periodin adolescents, albeit to a lesser extent (Weems,Hayward, Killen, & Taylor, 2002).

However, we (Bouton, Mineka, & Barlow,2001) have noted that the relationship betweenanxiety sensitivity and panic attacks in thesestudies is relatively small, not exclusive to

panic, and is weaker than the relationship be-tween panic and neuroticism. Furthermore,these studies have evaluated panic attacks andworry about panic, but not the prediction ofdiagnosed panic disorder. Thus, the causal sig-nificance of anxiety sensitivity for panic disor-der remains to be fully understood.

History of Medical Illness and Abuse

Other studies highlight the role of medical ill-nesses as contributing to a specific psychologi-cal vulnerability for panic disorder. For exam-ple, using the Dunedin Multidisciplinary Studydatabase, we found that experience with per-sonal respiratory disturbance (and parentalpoor health) as a youth predicted panic disor-der at age 18 or 21 (Craske et al., 2001). Thisfinding is consistent with reports of more respi-ratory disturbance in the history of patientswith panic disorder compared to other patientswith anxiety disorders (Verburg, Griez, Meijer,& Pols, 1995). Furthermore, in a recent study,first-degree relatives of patients with panic dis-order had a significantly higher prevalence ofchronic obstructive respiratory disease, andasthma in particular, than first-degree relativesof patients with other anxiety disorders (vanBeek, Schruers, & Friez, 2005).

Childhood experiences of sexual and physicalabuse may also prime panic disorder. Retrospec-tive reports of such childhood abuse were associ-ated with panic disorder onset at ages 16–21years in a recent longitudinal analysis of NewZealanders from birth to age 21 (Goodwin,Fergusson, & Horwood, 2005). This finding isconsistent with multiple cross-sectional studiesin both clinical and community samples (e.g.,Bandelow et al., 2002; Kendler et al., 2000;Kessler, Davis, & Kendler, 1997; Moisan &Engels, 1995; Stein et al., 1996). The associationwith childhood abuse is stronger for panic disor-der than for other anxiety disorders, such as so-cial phobia (Safren, Gershuny, Marzol, Otto, &Pollack, 2002; Stein et al., 1996) and obsessive–compulsive disorder (Stein et al., 1996). In addi-tion, some studies reported an association be-tween panic disorder and exposure to violencebetween other family members, generallyinterparental violence (e.g., Bandelow et al.,2002; Moisan & Engels, 1995), whereas themost recent study did not (Goodwin et al.,2005). Retrospective reporting of childhoodabuse and familial violence in all of these studies,however, limits the findings.


Interoceptive Awareness

Patients with panic disorder, as well as non-clinical panickers, appear to have heightenedawareness of, or ability to detect, bodily sensa-tions of arousal (e.g., Ehlers & Breuer, 1992,1996; Ehlers, Breuer, Dohn, & Feigenbaum,1995; Zoellner & Craske, 1999). Discrepantfindings (e.g., Antony et al., 1995; Rapee,1994) exist but have been attributed to meth-odological artifact (Ehlers & Breuer, 1996).Ability to perceive heartbeat, in particular,appears to be a relatively stable individual-difference variable given that it does not dif-fer between untreated and treated patientswith panic disorder (Ehlers & Breuer, 1992),or from before to after successful treatment(Antony, Meadows, Brown, & Barlow, 1994;Ehlers et al., 1995). Thus, interoceptive accu-racy may be a predisposing trait for panic dis-order. Ehlers and Breuer (1996) suggested that“although good interoception is consideredneither a necessary nor a sufficient conditionfor panic disorder, it may enhance the probabil-ity of panic by increasing the probability ofperceiving sensations that may trigger an at-tack if perceived as dangerous” (p. 174).Whether interoceptive awareness is learned,and represents another specific psychologicalvulnerability, or is more dispositional remainsto be determined.

Separate from interoception is the issue ofpropensity for intense autonomic activation.As noted earlier, some evidence points to aunique genetic influence on the reported expe-rience of breathlessness, heart pounding, and asense of terror (Kendler et al., 1987). Con-ceivably, cardiovascular reactivity presents aunique physiological predisposition for panicdisorder. In support of this, cardiac symptomsand shortness of breath predict later develop-ment of panic attacks and panic disorder (Keyl& Eaton, 1990). Unfortunately, these data de-rive from report of symptoms, which is nota good index of actual autonomic state(Pennebaker & Roberts, 1992) and may in-stead reflect interoception.

Initial Panic Attacks

From an evolutionary standpoint, fear is a nat-ural and adaptive response to threatening stim-uli. However, the fear experienced during thefirst unexpected panic attack is often unjusti-fied due to the lack of an identifiable trigger or

antecedent; hence, it represents a “false alarm”(Barlow, 1988, 2002). The large majority ofinitial panic attacks are recalled as occurringoutside of the home, while driving, walking, atwork, or at school (Craske et al., 1990), gener-ally in public (Lelliott, Marks, McNamee, &Tobena, 1989), and on a bus, plane, subway, orin social-evaluative situations (Shulman, Cox,Swinson, Kuch, & Reichman, 1994). We(Barlow, 1988; Craske & Rowe, 1997b) be-lieve situations that set the scene for initialpanic attacks are ones in which bodily sensa-tions are perceived as posing the most threat,because of impairment of functioning (e.g.,driving), entrapment (e.g., air travel, elevators),negative social evaluation (e.g., job, formal so-cial events), or distance from safety (e.g., unfa-miliar locales). Entrapment concerns may beparticularly salient for subsequent devel-opment of agoraphobia (Faravelli, Pallanti,Biondi, Paterniti, & Scarpato, 1992).

Maintenance Factors

Acute “fear of fear” (or, more accurately, anxi-ety focused on somatic sensations) that devel-ops after initial panic attacks in vulnerable in-dividuals refers to anxiety about certain bodilysensations associated with panic attacks (e.g.,racing heart, dizziness, paresthesias) (Barlow,1988; Goldstein & Chambless, 1978), and isattributed to two factors. The first is intero-ceptive conditioning, or conditioned fear of in-ternal cues, such as elevated heart rate, becauseof their association with intense fear, pain, ordistress (Razran, 1961). Specifically, intero-ceptive conditioning refers to low-level somaticsensations of arousal or anxiety becoming con-ditioned stimuli, so that early somatic compo-nents of the anxiety response come to elicit sig-nificant bursts of anxiety or panic (Bouton etal., 2001). An extensive body of experimentalliterature attests to the robustness of inter-oceptive conditioning (e.g., Dworkin &Dworkin, 1999), particularly with regard toearly interoceptive drug-onset cues becomingconditioned stimuli for larger drug effects (e.g.,Sokolowska, Siegel, & Kim, 2002). In addi-tion, interoceptive conditioned responses arenot dependent on conscious awareness of trig-gering cues (Razran, 1961); thus, they havebeen observed in patients under anesthesia(e.g., Block, Ghoneim, Fowles, Kumar, &Pathak, 1987). Within this model, then, slightchanges in relevant bodily functions that are


not consciously recognized may elicit condi-tioned anxiety or fear and panic due to previ-ous pairings with panic (Barlow, 1988; Boutonet al., 2001).

The second factor, offered by Clark (1986)to explain acute fear of panic-related body sen-sations, is catastrophic misappraisals of bodilysensations (misinterpretation of sensations assigns of imminent death, loss of control, etc.).Debate continues as to the significance of cata-strophic misappraisals of bodily sensations ver-sus conditioned (emotional, non-cognitively-mediated) fear responding. We have taken issuewith the purely cognitive model of panic disor-der by stating that it cannot account for panicattacks devoid of conscious cognitive appraisalwithout turning to constructs such as “auto-matic appraisals,” which prove to be untest-able (Bouton et al., 2001). Catastrophic mis-appraisals may accompany panic attacksbecause they are a natural part of the constella-tion of responses that go with panic, or becausethey have been encouraged and reinforcedmuch like sick role behaviors during child-hood. In addition, such thoughts may becomeconditioned stimuli that trigger anxiety andpanic, as demonstrated via panic inductionthrough presentation of pairs of words in-volving sensations and catastrophic outcomes(Clark et al., 1988). In this case, catastrophiccognitions may well be sufficient to elicit con-ditioned panic attacks, but not necessary.

Whether cognitively or noncognitivelybased, excessive anxiety over panic-relatedbodily sensations in panic disorder is well sup-ported. Persons with panic disorder endorsestrong beliefs that bodily sensations associatedwith panic attacks cause physical or mentalharm (e.g., Chambless, Caputo, Bright, &Gallagher, 1984; McNally & Lorenz, 1987).They are more likely to interpret bodily sensa-tions in a catastrophic fashion (Clark et al.,1988), and to allocate more attentional re-sources to words that represent physical threat,such as “disease” and “fatality” (e.g., Ehlers,Margraf, Davies, & Roth, 1988; Hope, Rapee,Heimberg, & Dombeck, 1990); catastrophewords, such as “death” and “insane” (e.g.,Maidenberg, Chen, Craske, Bohn, &Bystritsky, 1996; McNally, Riemann, Louro,Lukach, & Kim, 1992); and heartbeat stimuli(Kroeze & van den Hout, 2000). Also, individ-uals with panic disorder show enhanced brainpotentials in response to panic-related words(Pauli, Amrhein, Muhlberger, Dengler, &

Wiedemann, 2005). In addition, they are morelikely to become anxious in procedures thatelicit bodily sensations similar to the ones expe-rienced during panic attacks, including benigncardiovascular, respiratory, and audioves-tibular exercises (Antony, Ledley, Liss, &Swinson, 2006; Jacob, Furman, Clark, &Durrant, 1992), as well as more invasive proce-dures, such as CO2 inhalations, compared topatients with other anxiety disorders (e.g.,Perna, Bertani, Arancio, Ronchi, & Bellodi,1995; Rapee, 1986; Rapee, Brown, Antony, &Barlow, 1992) or healthy controls (e.g.,Gorman et al., 1994). The findings are not fullyconsistent, however, because patients withpanic disorder did not differ from patients withsocial phobia in response to an epinephrinechallenge (Veltman, van Zijderveld, Tilders, &van Dyck, 1996). Nonetheless, individualswith panic disorder also fear signals thatostensibly reflect heightened arousal and falsephysiological feedback (Craske & Freed,1995; Craske, Lang, et al., 2002;Ehlers, Margraf, Roth, Taylor, & Birnbaumer,1988).

Distress over bodily sensations is likely togenerate ongoing distress for a number of rea-sons. First, in the immediate sense, autonomicarousal generated by fear in turn intensifies thefeared sensations, thus creating a reciprocatingcycle of fear and sensations that is sustaineduntil autonomic arousal abates or the individ-ual perceives safety. Second, because bodilysensations that trigger panic attacks are not al-ways immediately obvious, they may generatethe perception of unexpected or “out of theblue” panic attacks (Barlow, 1988) that causeseven further distress (Craske, Glover, &DeCola, 1995). Third, the perceived uncontrol-lability, or inability to escape or terminatebodily sensations, again, is likely to generateheightened anxiety (e.g., Maier, Laudenslager,& Ryan, 1985; Mineka et al., 1984). Unpre-dictability and uncontrollability, then, are seenas enhancing general levels of anxiety about“When is it going to happen again?” and“What do I do when it happens?”, thereby con-tributing to high levels of chronic anxious ap-prehension (Barlow, 1988, 2002). In turn, anx-ious apprehension increases the likelihood ofpanic by directly increasing the availability ofsensations that have become conditioned cuesfor panic and/or attentional vigilance for thesebodily cues. Thus, a maintaining cycle of panicand anxious apprehension develops. Also, sub-


tle avoidance behaviors are believed to main-tain negative beliefs about feared bodily sensa-tions (Clark & Ehlers, 1993). Examplesinclude holding onto objects or persons for fearof fainting, sitting and remaining still for fearof a heart attack, and moving slowly or search-ing for an escape route because one fears actingfoolish (Salkovskis, Clark, & Gelder, 1996).Finally, anxiety may develop over specific con-texts in which the occurrence of panic wouldbe particularly troubling (i.e., situations associ-ated with impairment, entrapment, negativesocial evaluation, and distance from safety).These anxieties may contribute to agora-phobia, which in turn maintains distress bypreventing disconfirmation of catastrophicmisappraisals and extinction of conditioned re-sponding.

TREATMENT VARIABLES

Setting

There are several different settings for conduct-ing cognitive-behavioral therapy for panicdisorder and agoraphobia. The first, the outpa-tient clinic–office setting, is suited to psych-oeducation, cognitive restructuring, assign-ment and feedback regarding homeworkassignments, and role-play rehearsals. In addi-tion, certain exposures can be conducted in theoffice setting, such as interoceptive exposure tofeared bodily sensations described later. Re-cently, outpatient settings have extended frommental health settings to primary care suites(e.g., Craske, Roy-Byrne, et al., 2002; Roy-Byrne et al., 2005; Sharp, Power, Simpson,Swanson, & Anstee, 1997). This extension isparticularly important because of the higherprevalence of panic disorder in primary caresettings (e.g., Shear & Schulberg, 1995;Tiemens, Ormel, & Simon, 1996). However,whether a mental health or a primary care of-fice is being used, the built-in safety signals ofsuch an office may limit the generalizability oflearning that takes place in that setting. For ex-ample, learning to be less afraid in the presenceof the therapist, or in an office located near amedical center, may not necessarily generalizeto conditions in which the therapist is not pres-ent, or the perceived safety of a medical centeris not close by. For this reason, homework as-signments to practice cognitive-behavioralskills in a variety of different settings are partic-ularly important.

In the second setting, the natural environ-ment, cognitive restructuring and other anxietymanagement skills are put into practice, andthe patient faces feared situations. The latter iscalled in vivo exposure and can be conductedwith the aid of the therapist or alone.Therapist-directed exposure is particularly use-ful for patients who lack a social network tosupport in vivo exposure assignments, andmore valuable than self-directed exposure forpatients with more severe agoraphobia(Holden, O’Brien, Barlow, Stetson, &Infantino, 1983). Therapist-directed exposureis essential to guided mastery exposure, inwhich the therapist gives corrective feedbackabout the way the patient faces feared situa-tions to minimize unnecessary defensive behav-iors. For example, patients are taught to drivein a relaxed position at the wheel and to walkacross a bridge without holding the rail. On theone hand, guided mastery exposure has beenshown to be more effective than “stimulus ex-posure” when patients attempt simply to en-dure the situation alone until fear subsides,without the benefit of ongoing therapist feed-back (Williams & Zane, 1989). On the otherhand, self-directed exposure is very valuablealso, especially to the degree that it encouragesindependence and generalization of the skillslearned in treatment to conditions in which thetherapist is not present. Thus, the most benefi-cial approach in the natural environment is toproceed from therapist-directed to self-directedexposure.

In an interesting variation that combines theoffice and the natural environment, telephone-guided treatment, therapists direct patientswith agoraphobia by phone to conduct in vivoexposure to feared situations (NcNamee,O’Sullivan, Lelliot, & Marks, 1989; Swinson,Fergus, Cox, & Wickwire, 1995) or provide in-struction in panic control skills (Cote,Gauthier, Laberge, Cormier, & Plamondon,1994). In addition, one small study showedthat cognitive-behavioral therapy was as effec-tive when delivered by videoconference as inperson (Bouchard et al., 2004).

Self-directed treatments, with minimal directtherapist contact, take place in the natural en-vironment, and are beneficial for highly moti-vated and educated patients (e.g., Ghosh &Marks, 1987; Gould & Clum, 1995; Gould,Clum, & Shapiro, 1993; Lidren et al., 1994;Schneider, Mataix-Cols, Marks, & Bachofen,2005). On the other hand, self-directed treat-


ments are less effective for more severely af-fected patients (Holden et al., 1983), or thosewith more comorbidity (Hecker, Losee,Roberson-Nay, & Maki, 2004), less motiva-tion, and less education; or for patients whoare referred as opposed to recruited throughadvertisement (Hecker, Losee, Fritzler, & Fink,1996). Self-directed treatments have expandedbeyond workbooks and manuals to computer-ized and Internet versions (e.g., Carlbring,Ekselius, & Andersson, 2003; Richards, Klein,& Austen, 2006; Richards, Klein, & Carlbring,2003). In general, these treatments yield posi-tive results, although not quite as positiveas fully therapist-delivered treatments. Spe-cifically, a four-session computer-assistedcognitive-behavioral therapy for panic disorderwas less effective than 12 sessions of therapist-delivered cognitive-behavioral therapy at post-treatment, although the groups did not differ atfollow-up (Newman, Kenardy, Herman, &Taylor, 1997). More recently, 12 sessions oftherapist-delivered cognitive-behavioral ther-apy was more effective than six sessionsof either therapist-delivered or computer-augmented therapy (Kenardy et al., 2003).Also, findings from computerized programs foremotional disorders in general indicate thatsuch treatments are more acceptable and suc-cessful when combined with therapist involve-ment (e.g., Carlbring et al., 2003).

The third setting, the inpatient facility, ismost appropriate when conducting very inten-sive cognitive-behavioral therapy (e.g., dailytherapist contact), or treating severely disabledpersons who can no longer function at home.In addition, certain medical or drug complica-tions may warrant inpatient treatment. Thegreatest drawback to the inpatient setting ispoor generalization to the home environment.Transition sessions and follow-up booster ses-sions in an outpatient clinic–office or in the pa-tient’s own home facilitate generalization.

Format

Cognitive-behavioral therapy for panic disor-der and agoraphobia may be conducted in indi-vidual or group formats. Several clinical out-come studies have used group treatments (e.g.,Craske, DeCola, Sachs, & Pontillo, 2003; Ev-ans, Holt, & Oei, 1991; Feigenbaum, 1988;Hoffart, 1995; Telch et al., 1993). The fact thattheir outcomes are generally consistent withthe summary statistics obtained from individu-

ally formatted treatment suggests that grouptreatment is as effective as individual therapy.Also, Lidren and colleagues (1994) found thatgroup therapy is as effective as individualbibliotherapy, although they did not include acomparison with individualized cognitive-behavioral therapy. In direct comparisons, aslight advantage is shown for individual for-mats. Specifically, Neron, Lacroix, and Chaput(1995) compared 12–14 weekly sessions of in-dividual or group cognitive-behavioral therapy(N = 20), although the group condition re-ceived two additional 1-hour individual ses-sions. The two conditions were equally effec-tive for measures of panic and agoraphobia atposttreatment and 6-month follow-up. How-ever, the individual format was more successfulin terms of generalized anxiety and depressivesymptoms by the follow-up point. In addition,individual treatments resulted in more clini-cally significant outcomes than group formatsin primary care (Sharp, Power, & Swanson,2004). Furthermore, 95% of individuals as-signed to the waiting-list condition in the latterstudy stated a clear preference for individualtreatment when given the choice at the end ofthe waiting list.

Most studies of cognitive-behavioral therapyfor panic and agoraphobia involve 10–20weekly treatment sessions. Several studiesshow that briefer treatments may be effectiveas well. Evans and colleagues (1991) compareda 2-day group cognitive-behavioral treatmentto a waiting-list condition, although withoutrandom assignment. The 2-day program com-prised lectures (3 hours); teaching skills, suchas breathing, relaxation, and cognitive chal-lenging (3 hours); in vivo exposure (9 hours);and group discussion plus a 2-hour supportgroup for significant others. Eighty-five percentof treated patients were reported to be eithersymptom-free or symptomatically improved,and these results were maintained 1 year later.In contrast, the waiting-list group did not dem-onstrate significant changes. A recent pilotstudy similarly indicated effectiveness with in-tensive cognitive-behavioral therapy over 2days (Deacon & Abramowitz, 2006). Otherstudies have evaluated the effectiveness ofcognitive-behavioral therapy when deliveredover a fewer number of sessions. In a random-ized study, patients with PDA who awaitedpharmacotherapy treatment were assigned tofour weekly sessions of either cognitive-behavioral therapy or supportive nondirective


therapy (Craske, Maidenberg, & Bystritsky,1995). Cognitive-behavioral therapy was moreeffective than supportive therapy, particularlywith less severely affected patients, althoughthe results were not as positive as those typi-cally seen with more sessions. Also, we foundthat up to six sessions (average of three ses-sions) of cognitive-behavioral therapy com-bined with medication recommendationsyielded significantly greater improvements onan array of measures, including quality of life,compared to treatment as usual for individualswith panic disorder in primary care settings(Roy-Byrne et al., 2005). Notably, however, thetreatment effects substantially increased as thenumber of cognitive-behavioral therapy ses-sions (up to six) and follow-up booster phonecall sessions (up to six) increased (Craske et al.,2006). Finally, in a direct comparison, resultswere equally effective whether cognitive-behavioral therapy was delivered across thestandard 12 sessions or across approximately 6sessions (Clark et al., 1999).

Interpersonal Context

Interpersonal context variables have been re-searched in terms of the development, mainte-nance, and treatment of agoraphobia. The rea-son for this research interest is apparent fromthe following vignettes:

“My husband really doesn’t understand. Hethinks it’s all in my head. He gets angry at mefor not being able to cope. He says I’m weakand irresponsible. He resents having to driveme around, and doing things for the kidsthat I used to do. We argue a lot, because hecomes home tired and frustrated from workonly to be frustrated more by the problemsI’m having. But I can’t do anything withouthim. I’m so afraid that I’ll collapse into ahelpless wreck without him, or that I’ll bealone for the rest of my life. As cruel as hecan be, I feel safe around him because he al-ways has everything under control. He al-ways knows what to do.”

This vignette illustrates dependency on thesignificant other for a sense of safety despite anonsympathetic response that may only serveto increase background stress for the patient.The second vignette illustrates inadvertent rein-forcement of fear and avoidance through atten-tion from the significant other.

“My boyfriend really tries hard to help me.He’s always cautious of my feelings anddoesn’t push me to do things that I can’t do.He phones me from work to check on me.He stays with me and holds my hand when Ifeel really scared. He never hesitates to leavework and take me home if I’m having a badtime. Only last week we visited some of hisfriends, and we had to leave. I feel guilty be-cause we don’t do the things we used to en-joy doing together. We don’t go to the mov-ies anymore. We used to love going to ballgames, but now its too much for me. I am sothankful for him. I don’t know what I woulddo without him.”

Perhaps some forms of agoraphobia repre-sent a conflict between desire for autonomyand dependency in interpersonal relationships(Fry, 1962; Goldstein & Chambless, 1978). Inother words, the “preagoraphobic” is trappedin a domineering relationship without the skillsneeded to activate change. However, the con-cept of a distinct marital system that predis-poses toward agoraphobia lacks empirical evi-dence. That is not to say that marital orinterpersonal systems are unimportant to ago-raphobia. For example, interpersonal discord/dissatisfaction may represent one of severalpossible stressors that precipitate panic attacks.Also, interpersonal relations may be negativelyimpacted by the development of agoraphobia(Buglass, Clarke, Henderson, & Presley, 1977),and in turn contribute to its maintenance. Notunlike one of the earlier vignettes, consider thewoman who has developed agoraphobia andnow relies on her husband to do the shoppingand other errands. These new demands uponthe husband lead to resentment and maritaldiscord. The marital distress adds to back-ground stress, making progress and recoveryeven more difficult for the patient.

Aside from whether interpersonal dysregula-tion contributes to the onset or maintenance ofPD/PDA, some studies suggest that poor mari-tal relations adversely impact exposure-basedtreatments (Bland & Hallam, 1981; Dewey &Hunsley, 1989; Milton & Hafner, 1979). How-ever, other studies show no relationship be-tween marital distress and outcome fromcognitive-behavioral therapy (Arrindell &Emmelkamp, 1987; Emmelkamp, 1980;Himadi, Cerny, Barlow, Cohen, & O’Brien,1986). Another line of research suggests thatinvolving significant others in every aspect of


treatment may override potential negative im-pacts of poor marital relations on phobic im-provement (Barlow, O’Brien, & Last, 1984;Cerny, Barlow, Craske, & Himadi, 1987). Fur-thermore, involvement of significant othersresulted in better long-term outcomes fromcognitive-behavioral therapy for agoraphobia(Cerny et al., 1987). Similarly, communicationstraining with significant others, compared torelaxation training, after 4 weeks of in vivo ex-posure therapy, resulted in significantly greaterreductions on measures of agoraphobia byposttreatment (Arnow, Taylor, Agras, & Telch,1985), an effect that was maintained over an 8-month follow-up. Together, these studies sug-gest the value of including significant others inthe treatment for agoraphobia.

Yet another question is the degree to whichtreatment for panic disorder and agorapho-bia influences marital/interpersonal relations.Some have noted that successful treatment canhave deleterious effects (Hafner, 1984; Hand &Lamontagne, 1976). Others note that it has noeffect or a positive effect on marital functioning(Barlow et al., 1983; Himadi et al., 1986). We(Barlow et al., 1983) suggested that when nega-tive effects do occur, it may be because expo-sure therapy is conducted intensively, withoutthe significant other’s involvement, whichcauses major role changes that the significantother perceives as being beyond his or her con-trol. This again speaks to the value of involvingsignificant others in the treatment process.

Therapist Variables

Only a few studies have evaluated therapistvariables in relation to cognitive-behavioraltreatments for anxiety disorders. Williams andChambless (1990) found that patients with ag-oraphobia who rated their therapists as caring/involved, and as modeling self-confidence,achieved better outcomes on behavioral ap-proach tests. However, an important confoundin this study was that patient ratings of thera-pist qualities may have depended on patientresponses to treatment. Keijsers, Schaap,Hoogduin, and Lammers (1995) reviewed find-ings regarding therapist relationship factorsand behavioral outcome. They concluded thatempathy, warmth, positive regard, and genu-ineness assessed early in treatment predict posi-tive outcome; patients who view their thera-pists as understanding and respectful improvethe most; and patient perceptions of therapist

expertness, self-confidence, and directivenessrelate positively to outcome, although not con-sistently. In their own study of junior therapistswho provided cognitive-behavioral treatmentfor PD/PDA, Keijsers and colleagues (1995)found that more empathic statements andquestioning occurred in Session 1 than in latersessions. In Session 3, therapists became moreactive and offered more instructions and expla-nations. In Session 10, therapists employedmore interpretations and confrontations thanpreviously. Most importantly, directive state-ments and explanations in Session 1 predictedpoorer outcome. Empathic listening in Ses-sion 1 related to better behavioral outcome,whereas empathic listening in Session 3 relatedto poorer behavioral outcome. Thus, they dem-onstrated the advantages of different interac-tional styles at different points in therapy.

Most clinicians assume that therapist train-ing and experience improve the chances of suc-cessful outcome. Some believe this to be thecase particularly with respect to the cognitiveaspects of cognitive-behavioral therapy (e.g.,Michelson et al., 1990), and some indirect evi-dence for this supposition exists. Specifically,cognitive-behavioral therapy conducted by“novice” therapists in a medical setting(Welkowitz et al., 1991) was somewhat less ef-fective in comparison to the same therapy con-ducted by inexperienced but highly trainedtherapists in a psychological setting (Barlow,Craske, Czerny, & Klosko, 1989), or by experi-enced and highly trained therapists in a com-munity mental health setting (Wade, Treat, &Stuart, 1998). Huppert and colleagues (2001),who directly evaluated the role of therapist ex-perience, found that, in general, therapist expe-rience positively related to outcome, seeminglybecause these therapists were more flexible inadministering the treatment and better able toadapt it to the individual being treated. Obvi-ously, there is a need for more evaluation of therole of therapist experience and training incognitive-behavioral therapy.

Equally, if not more important is the need toevaluate how much training of either novice orexperienced therapists is necessary to at-tain therapeutic competency in cognitive-behavioral therapy. This is critically importantin the current environment of dissemination ofcognitive-behavioral treatments for anxietydisorders to real-world settings, in which train-ing procedures must be adequate but not socostly that they are prohibitive and therefore


not disseminable. Ongoing research in our set-tings is addressing exactly these issues. Othersare investigating the benefits of training generalpractitioners in cognitive-behavioral therapyfor panic disorder (Heatley, Ricketts, & For-rest, 2005).

Patient Variables

There has been a recent interest in the effect ofcomorbidity upon the outcomes of cognitive-behavioral therapy for PD/PDA. Brown, An-tony, and Barlow (1995) found that comorbidi-ty with other anxiety disorders did not predictresponse to cognitive-behavioral therapy over-all, although social phobia was unexpectedlyassociated with superior outcome for PD/PDA.In contrast, we (Tsao, Lewin, & Craske, 1998)found a trend for comorbidity that comprisedmostly other anxiety disorders to be associatedwith slightly lower rates of overall success. In asubsequent study, however, we replicated thefinding by Brown et al. (1995) of no relation-ship between baseline comorbidity comprisingmostly other anxiety disorders, and either im-mediate or 6-month outcome for PD/PDA(Tsao, Mystkowski, Zucker, & Craske, 2002).

Depressive disorders are highly comorbidwith PD/PDA (e.g., Goisman et al., 1994). Incontrast to expectations and to pharmacologytrials, the available evidence does not consis-tently demonstrate detrimental effects of initialdepression upon outcome from cognitive-behavioral therapy for PD/PDA. On the onehand, several studies found no relationshipwith outcome, regardless of whether depres-sion was the principal diagnosis or secondaryto PD/PDA (Brown et al., 1995; Laberge,Gauthier, Cote, Plamondon, & Cormier, 1993;McLean, Woody, Taylor, & Koch, 1998). Onthe other hand, Mennin and Heimberg’s (2000)review led them to conclude a mixed pattern ofresults given evidence that patients withoutmajor depression showed greater reductions infears of bodily sensations (Laberge et al.,1993), that patients with primary, but not sec-ondary, depression had worse outcomes thanthose without depression (Maddock & Blacker,1991), and that treatment completers were lesslikely than noncompleters to have comorbiddepression (Wade et al., 1998). Some proposethat depression impedes engagement incognitive-behavioral therapy homework exer-cises. However, McLean and colleagues (1998)reported no relationship between depression

and compliance with cognitive-behavioral ther-apy homework. Similarly, Murphy, Michelson,Marchione, Marchione, and Testa (1998)found that depressed persons with PD/PDA en-gaged in as many self-directed exposures asnondepressed persons, although the depressedgroup reported more subjective anxiety duringexposures.

A relatively high co-occurrence exists be-tween PD/PDA and avoidant, dependent, andhistrionic personality disorders (e.g., Reich etal., 1994). Questions of diagnostic reliabilityand validity aside, comorbid personality disor-ders are sometimes associated with poorer re-sponse than usual to cognitive-behavioral ther-apy for PD/PDA (e.g., Hoffart & Hedley, 1997;Marchand, Goyer, Dupuis, & Mainguy, 1998).However, closer examination reveals that al-though individuals with comorbid personalitydisorders have greater severity of PD/PDA atpre- and post–cognitive-behavioral therapy, therate of decrease in PD/PDA symptoms usuallyis not affected by the comorbid personality dis-order. Thus, Dreessen, Arntz, Luttels, andSallaerts (1994) and van den Hout, Brouwers,and Oomen (2006) found that comorbid per-sonality disorders did not affect response tocognitive-behavioral therapy for PD/PDA.Moreover, Hofmann and colleagues (1998)found that scores on questionnaire subscalesreflecting Axis II personality disorders did notpredict panic disorder treatment response to ei-ther cognitive-behavioral therapy or to medica-tion. In fact, some personality traits may asso-ciate positively with outcome, as was reportedby Rathus, Sanderson, Miller, and Wetzler(1995) with respect to compulsive personalityfeatures.

Substance-related disorders also commonlyco-occur with PD/PDA. On the one hand, in aseries of single cases (N = 3), Lehman, Brown,and Barlow (1998) demonstrated successfulcontrol of panic attacks in individuals whowere abusing alcohol. On the other hand, theaddition of anxiety treatment to a relapse pre-vention program for abstinent individuals witha primary diagnosis of alcohol dependence anda comorbid diagnosis of PDA or social phobiadecreased anxiety symptoms relative to a re-lapse prevention program alone (Schade et al.,2005). However, adding the anxiety treatmentdid not affect rates of alcohol relapse in thatstudy.

Another source of comorbidity is medicalconditions, such as cardiac arrhythmias or


asthma, that may slow improvement ratesgiven the additional complications involved indiscriminating between anxiety and diseasesymptomatology, increases in actual medicalrisk, and the stress of physical diseases. Al-though the effect of medical comorbidity onoutcome has not been assessed to date,cognitive-behavioral therapy for panic disorderhas been shown to alleviate self-reported physi-cal health symptoms (Schmidt et al., 2003).

Other patient variables include socioeco-nomic status and general living conditions. Weevaluated perceived barriers to receiving men-tal health treatment in our primary care studyof panic disorder (Craske, Golinelli, et al.,2005). Commonly reported barriers includedinability to find out where to go for help(43%), worry about cost (40%), lack of cover-age by one’s health plan (35%), and inability toget an appointment soon enough (35%). Also,in our multicenter trial, attrition fromcognitive-behavioral and/or medication treat-ment for panic disorder with minimal agora-phobia was predicted by lower education,which in turn was dependent on lower income(Grilo et al., 1998). Similarly, level of educationand motivation were associated with dropoutrates in another sample, although the effectswere small (Keijsers, Kampman, & Hoogduin,2001). Low education–income may reflect lessdiscretionary time to engage in activities suchas weekly treatment. Consider the woman whois a mother of two, a full-time clerk, whosehusband is on disability due to back injury, orthe full-time student who works an extra 25hours a week to pay his way through school.Under these conditions, treatment assignmentsof daily in vivo exposure exercises are muchless likely to be completed. Frustration withlack of treatment progress is likely to result.Therapeutic success requires either a change inlifestyle that allows the cognitive-behavioraltreatment to become a priority or terminationof therapy until a later time, when life circum-stances are less demanding. In fact, these kindsof life-circumstance issues may explain thetrend for African Americans to show less treat-ment benefit in terms of mobility, anxiety, andpanic attacks, than European Americans(Friedman & Paradis, 1991; Williams &Chambless, 1994). Although, in contrast tothese two studies, Friedman, Paradis, andHatch (1994) found equivalent outcomesacross the two racial groups, and the resultsfrom another study yielded outcomes from a

female African American sample that werejudged to be comparable to those of EuropeanAmericans (Carter, Sbrocco, Gore, Marin, &Lewis, 2003). The influence of ethnic and cul-tural differences on treatment outcome and de-livery clearly needs more evaluation.

Finally, patients’ understanding of the natureof their problem may be important to the suc-cess of cognitive-behavioral treatments. Giventhe somatic nature of panic disorder, many pa-tients seek medical help first. Beyond that,however, differences in the way the problem isconceptualized could lead to the perceptionthat pharmacological or analytical treatmentapproaches are more credible than cognitive-behavioral treatment approaches. For example,individuals who strongly believe their condi-tion is due to “a neurochemical imbalance”may be more likely to seek medication and torefute psychological treatments. Similarly, indi-viduals who attribute their condition to “some-thing about my past—it must be unconsciousinfluences” may resist cognitive-behavioral in-terpretations. Also, Grilo and colleagues(1998) found that patients with PD/PDA whoattributed their disorder to specific stressors intheir lives were more likely to drop out ofcognitive-behavioral or medication treatment,perhaps because they saw the offered treatmentas irrelevant.

Concurrent Pharmacological Treatment

Many more patients receive medications thancognitive-behavioral therapy for panic disorderand agoraphobia, partly because primary carephysicians are usually the first line of treat-ment. Thus, one-half or more of patients withpanic disorder who attend psychology researchclinics already are taking anxiolytic medica-tions. The obvious questions, therefore, are theextent to which cognitive-behavioral therapyand medications have a synergistic effect, andhow medications impact cognitive-behavioraltherapy.

Results from large clinical trials, includingour own multisite trial (Barlow, Gorman,Shear, & Woods, 2000), suggest no advantageduring or immediately after the conclusion oftreatment combining cognitive-behavioral andpharmacological approaches. Specifically, bothindividual cognitive-behavioral and drug treat-ment and a combination treatment were imme-diately effective following treatment. Further-more, following medication discontinuation,


the combination of medication and cognitive-behavioral therapy fared worse than cognitive-behavioral therapy alone, suggesting thepossibility that state- (or context-) dependentlearning in the presence of medication mayhave attenuated the new learning that occursduring cognitive-behavioral therapy. On theother hand, in the primary care setting, wefound that the addition of even just one com-ponent of cognitive-behavioral therapy to med-ications for PD/PDA resulted in statisticallyand clinically significant improvements atposttreatment and 12 months later (Craske,Golinelli, et al., 2005).

More recently, our multisite collaborativeteam has been investigating long-term strate-gies in the treatment of panic disorder. We ex-amined sequential combination strategies todetermine whether this approach was moreadvantageous than simultaneously combiningtreatments. In this study, currently in prepara-tion for publication, 256 patients with panicdisorder with all levels of agoraphobia com-pleted 3 months of initial treatment withcognitive-behavioral therapy. Fifty-eight ofthose patients did not reach an optimal level offunctioning (high end-state functioning) andentered a trial in which they received eithercontinued cognitive-behavioral therapy orparoxetine. Paroxetine was administered forup to 1 year, whereas cognitive-behavioraltherapy was delivered twice a month for 3months. At the end of the 1-year period, therewas a strong suggestion, represented as a statis-tical trend, that more of the patients receivingparoxetine achieved responder status com-pared to those receiving continued cognitive-behavioral treatment. Specifically, 60% of thenonresponders receiving paroxetine became re-sponders, compared to 35% receiving contin-ued cognitive-behavioral therapy (p ≤ .083).Further evaluation of effect sizes will help us toevaluate the importance of this difference. Thisstudy also evaluated long-term strategies formaintaining gains in those patients who re-sponded to cognitive-behavioral therapy, as de-scribed below.

In another study with similar results, pa-tients who did not respond to cognitive-behavioral therapy also benefited more fromthe addition of a serotonergic drug (paroxe-tine) to continued cognitive-behavioral therapythan from the addition of a drug placebo, withsubstantially different effect sizes (Kampman,Keijsers, Hoogduin, & Hendriks, 2002). Con-

versely, individuals who are resistant to phar-macotherapy may respond positively tocognitive-behavioral therapy, although thesefindings were part of an open trial without ran-domization (Heldt et al., 2006).

Findings from the combination of fast-actinganxiolytics and, specifically, the high-potencybenzodiazepines with behavioral treatmentsfor agoraphobia are contradictory (e.g., Markset al., 1993; Wardle et al., 1994). Nevertheless,several studies have reliably demonstrated thedetrimental effects of chronic use of high-potency benzodiazepines on short-term andlong-term outcome in cognitive-behavioraltreatments for panic or agoraphobia (e.g.,Otto, Pollack, & Sabatino, 1996; van Balkom,de Beurs, Koele, Lange, & van Dyck, 1996;Wardle et al., 1994). Specifically, there is evi-dence for more attrition, poorer outcome, andmore relapse with chronic use of high-potencybenzodiazepines. In addition, use of benzo-diazepines as needed was associated withpoorer outcome than regular use or no use inone small naturalistic study (Westra, Stewart,& Conrad, 2002).

Finally, the cost-effectiveness of cognitive-behavioral and medication treatments aloneversus in combination requires further evalua-tion; currently, cognitive-behavioral therapy isconsidered to be more cost-effective (e.g., dis-ability costs, work days missed, health careuse) than pharmacotherapy (Heuzenroeder etal., 2004).

Understanding the ways in which psy-chotropic medications influence cognitive-behavioral therapy may prove useful for devel-oping methods that optimize the combinationof these two approaches to treatment. First,medications, particularly fast-acting, potentmedications that cause a noticeable shift instate and are used on an as-needed basis (e.g.,benzodiazepines, beta-blockers), may contrib-ute to relapse, because therapeutic success is at-tributed to them rather than to cognitive-behavioral therapy. Patients’ resultant lack ofperceived self-control may increase relapse po-tential when medication is withdrawn or con-tribute to maintenance of a medication regimenunder the assumption that it is necessary tofunctioning. In support, attribution of thera-peutic gains to alprazolam, and lack of confi-dence in coping without alprazolam, evenwhen given in conjunction with behavioraltherapy, predicted relapse (Basoglu, Marks,Kilic, Brewin, & Swinson, 1994). Second, med-


ications may assume the role of safety signals,or objects to which persons erroneously attrib-ute their safety from painful, aversive out-comes. Safety signals contribute to mainte-nance of fear and avoidance in the long term(Hermans, Craske, Mineka, & Lovibond,2006) and may interfere with corrections ofmisappraisals of bodily symptoms. Third, med-ications may block the capacity to experiencefear, which, at least initially in exposure ther-apy, is a positive predictor of overall outcome(for a review, see Craske & Mystkowski,2006). Fourth, medications may reduce themotivation to engage in practices of cognitive-behavioral skills, especially ones that effec-tively reduce panic and anxiety. Finally,learning that takes place under the influence ofmedications may not necessarily generalizeto the time when medications are removed,thus contributing to relapse (Bouton &Swartzentruber, 1991). Some of these pointsare illustrated in the following vignettes:

“I had been through a program of cognitive-behavioral therapy, but it was really thePaxil that helped. Because I was feeling somuch better, I considered tapering off themedication. At first I was very concernedabout the idea. I had heard horror storiesabout what people go through when with-drawing. However, I thought it would be OKas long as I tapered slowly. So, I graduallyweaned myself off. It really wasn’t that bad.Well, I had been completely off the medica-tion for about a month when the problemstarted all over again. I remember sitting in arestaurant, feeling really good because I wasthinking about how much of a problem res-taurants used to be for me before, and howeasy it seemed now. Then, whammo. I be-came very dizzy and I immediately thought,‘Oh no, here it comes.’ I had a really badpanic attack. All I could think of was whydidn’t I stay on the medication.”

“I started to lower my dose of Xanax. I wasOK for the first couple of days. . . . I felt re-ally good. Then, when I woke up on Fridaymorning, I felt strange. My head felt reallytight and I worried about having the sameold feelings all over again. The last thing Iwant to do is to go through that again. So Itook my usual dose of Xanax and, within afew minutes, I felt pretty good again. I need

the medication. I can’t manage without itright now.”

Continuation of exposure after medication iswithdrawn may offset relapse, because it en-hances attributions of personal mastery and re-duces the safety signal function of medications.In addition, opportunities to practice exposureand cognitive and behavioral strategies withoutthe aid of medication overcome state depend-ency and enhance generalization of therapeuticgains once treatment is over.

CASE STUDY

Julie, a 33-year-old European American,mother of two, lives with Larry, her husband of8 years. For the past 3 years she has beenchronically anxious and panic stricken. She de-scribes her panic attacks as unbearable and in-creasing in frequency. The first time she feltpanicky was just over 3 years ago, when shewas rushing to be by her grandmother’s side inthe last moments before she died. Julie wasdriving alone on the freeway. She remembersfeeling as if everything were moving in slowmotion, as if the cars were standing still, andthings around her seemed unreal. She recalledfeeling short of breath and detached. However,it was so important to reach her destinationthat she did not dwell on how she felt untillater. After the day was over, she reflected uponhow lucky she was not to have had an accident.A few weeks later, the same type of feeling hap-pened again when driving on the freeway. Thistime it occurred without the pressure of gettingto her dying grandmother. It scared Julie be-cause she was unable to explain the feelings.She pulled off to the side of the road and calledher husband, who came to meet her. She fol-lowed him home, feeling anxious all the way.

Now, Julie has these feelings in many situa-tions. She describes her panic attacks as feel-ings of unreality, detachment, shortness ofbreath, a racing heart, and a general fear of theunknown. It is the unreality that scares her themost. Consequently, Julie is sensitive to any-thing that produces “unreal” types of feelings,such as the semiconsciousness that occurs justbefore falling asleep, the period when daylightchanges to night, bright lights, concentratingon the same thing for long periods of time, al-cohol or drugs, and being anxious in general.Even though she has a prescription for


Klonopin (a high-potency benzodiazepine), sherarely, if ever, uses it because of her general fearof being under the influence of a drug, or offeeling an altered state of consciousness. Shewants to be as alert as possible at all times, butshe keeps the Klonopin with her in the eventthat she has no other way of managing herpanic. She does not leave home without theKlonopin. Julie is very sensitive to her body ingeneral; she becomes scared of anything thatfeels a little different than usual. Even coffee,which she used to enjoy, is distressing to hernow because of its agitating and racy effects.She was never a big exerciser, but to think ofexerting herself now is also scary. Julie reportsthat she is constantly waiting for the next panicattack to occur. She avoids freeways, driving onfamiliar surface streets only. She limits herselfto a 10-mile radius from home. She avoidscrowds and large groups as well, partly becauseof the feeling of too much stimulation andpartly because she is afraid to panic in front ofothers. In general, she prefers to be with herhusband or her mother. However, she can domost things as long as she is within her“safety” region.

Julie describes how she differs from the wayshe used to be: how weak and scared she isnow. The only other incident similar to her cur-rent panic attacks occurred in her early 20s,when she had a negative reaction to smokingmarijuana. Julie became very scared of the feel-ing of losing control and feared that she wouldnever return to reality. She has not taken drugssince then. Otherwise, there is no history of se-rious medical conditions, or any previous psy-chological treatment. Julie had some separa-tion anxiety and was shy as a young child andthroughout her teens. However, her social anx-iety improved throughout her 20s to the pointthat until the onset of her panic attacks, shewas mostly very comfortable around people.Since the onset of her panic attacks, Julie hasbecome concerned that others will notice thatshe appears anxious. However, her social anxi-ety is limited to panic attacks and does not re-flect a broader social phobia.

In general, Julie’s appetite is good, but hersleep is restless. At least once a week she wakesabruptly in the middle of the night, feelingshort of breath and scared, and has great diffi-culty going to sleep when her husband travels.In addition to worrying about her panic at-tacks, Julie worries about her husband and herchildren, although these latter worries are sec-

ondary to her worry about panicking and arenot excessive. She has some difficulty concen-trating but is generally able to function at homeand at work, because of the familiarity of herenvironment and the safety she feels in thepresence of her husband. Julie works part-timeas the manager of a business that she and herhusband own. She sometimes becomes de-pressed about her panic and the limitations onhow far she can travel. Occasionally she feelshopeless about the future, doubting whethershe will ever be able to escape the anxiety. Al-though the feelings of hopelessness and theteariness never last than more than a few days,Julie has generally had a low-grade depressedmood since her life became restricted by thepanic attacks.

Julie’s mother and her uncle both had panicattacks when they were younger. Julie is nowworried that her oldest child is showing signsof being overly anxious, because he is hesitantabout trying new things or spending time awayfrom home.

ASSESSMENT

A functional behavioral analysis depends onseveral different modes of assessment, whichwe describe next.

Interviews

An in-depth interview is the first step in estab-lishing diagnostic features and the profile ofsymptomatic and behavioral responses. Severalsemistructured and fully structured interviewsexist. The Anxiety Disorders InterviewSchedule—Fourth Edition (ADIS-IV; Di Nardo,Brown, & Barlow, 1994) primarily assessesanxiety disorders, as well as mood andsomatoform disorders. Psychotic and drug con-ditions are screened by this instrument also.The ADIS-IV facilitates gathering the necessaryinformation to make a differential diagnosisamong anxiety disorders and offers a means todistinguish between clinical and subclinicalpresentations of a disorder. Data on the fre-quency, intensity, and duration of panic at-tacks, as well as details on avoidance behavior,are embedded within the ADIS-IV; this infor-mation is necessary for tailoring treatment toeach individual’s presentation. The value ofstructured interviews is in their contribution toa differential diagnosis and interrater reliabil-


ity. Interrater agreement ranges from satisfac-tory to excellent for the various anxiety disor-ders using the ADIS-IV (Brown, Di Nardo,Lehman, & Campbell, 2001).

Similarly, the Schizophrenia and AffectiveDisorders Schedule—Lifetime Version (modi-fied for the study of anxiety) produces reliablediagnoses for most of the anxiety disorders(generalized anxiety disorder and simplephobia are the exceptions) (Manuzza, Fyer,Liebowitz, & Klein, 1990), as does the Struc-tured Clinical Interview for DSM-IV (SCID),which covers all of the mental disorders (First,Spitzer, Gibbon, & Williams, 1994).

Differential diagnosis is sometimes difficultbecause, as described earlier, panic is a ubiqui-tous phenomenon (Barlow, 1988) that occursacross a wide variety of emotional disorders. Itis not uncommon for persons with specificphobias, social phobia, generalized anxiety dis-order, obsessive–compulsive disorder, and post-traumatic stress disorder to report panic at-tacks. For Julie, there was a differentialdiagnostic question regarding social phobiaand PDA. Shown in Figure 1.1 are the ADIS-IVquestions that addressed this differentiation(Julie’s answers are in italics).

As demonstrated in Figure 1.1, Julie experi-ences panic attacks in social situations and isconcerned about being negatively evaluated byothers if her anxiety becomes visibly apparent.However, despite her history of shyness, Julie’scurrent social discomfort is based primarily onthe possibility of panicking. Because of this,and because she meets the other criteria forPDA (i.e., uncued/nonsocial panic attacks andpervasive apprehension about future panic at-tacks), the social distress is best subsumedunder the domain of PDA. If Julie reported thatshe experiences panic attacks in social situa-tions only, or that she worries about panic at-tacks in social situations only, then a diagnosisof social phobia would be more probable. A re-port of uncued panic attacks, as well as self-consciousness about things that she might door say in social situations regardless of the oc-currence of panic, would be consistent with adual diagnosis of PDA and social phobia. Ingeneral, individuals with PDA may continue tofeel anxious even when playing a passive role ina social setting, whereas a patient with socialphobia is more likely to feel relaxed when he orshe is not the center of attention and does notanticipate being evaluated or judged (Dattilio& Salas-Auvert, 2000).

The same types of diagnostic questioning areuseful for distinguishing between PDA andclaustrophobia. Other differential diagnosticissues can arise with respect to somatoform dis-orders, real medical conditions, and avoidantor dependent personality disorders.

Medical Evaluation

A medical evaluation is generally recom-mended, because several medical conditionsshould be ruled out before assigning the diag-nosis of PD/PDA. These include thyroid condi-tions, caffeine or amphetamine intoxication,drug withdrawal, or pheochromocytoma (arare adrenal gland tumor). Furthermore, cer-tain medical conditions can exacerbate PD/PDA, although it is likely to continue evenwhen the symptoms are under medical control.Mitral valve prolapse, asthma, allergies, andhypoglycemia fall into this latter category. Ac-cording to the model described earlier, thesemedical conditions exacerbate PD/PDA to theextent that they elicit the feared physical sensa-tions. For example, mitral valve prolapse some-times produces the sensation of a heart flutter,asthma produces shortness of breath, andhypoglycemia produces dizziness and weak-ness, all of which overlap with symptoms ofpanic and may therefore become conditionedcues for panic.

Self-Monitoring

Self-monitoring is a very important part of as-sessment and treatment for panic disorder–agoraphobia. Retrospective recall of past epi-sodes of panic and anxiety, especially whenmade under anxious conditions, may inflateestimates of panic frequency and intensity(Margraf et al., 1987; Rapee, Craske, &Barlow, 1990). Moreover, such inflation maycontribute to apprehension about future panic.In contrast, ongoing self-monitoring generallyyields more accurate, less inflated estimates(for a comprehensive review of self-monitoringfor panic and anxiety, see Craske & Tsao,1999). Also, ongoing self-monitoring is be-lieved to contribute to an objective self-awareness. Objective self-monitoring replacesnegative affect-laden self-statements such as “Ifeel horrible. This is the worst its ever been—my whole body is out of control” with “Myanxiety level is 6. My symptoms include tremu-lousness, dizziness, unreal feelings, and short-


ness of breath—and this episode lasted 10 min-utes.” Objective self-awareness usually reducesnegative affect. Finally, self-monitoring pro-vides feedback for judging progress and usefulmaterial for in-session discussions.

Panic attacks are recorded in the Panic At-tack Record, a version of which is shown inFigure 1.2. This record is to be completed assoon as possible after a panic attack occurs;therefore, it is carried on-person (wallet size).Daily levels of anxiety, depression, and worry

about panic are monitored with the DailyMood Record shown in Figure 1.3. This recordis completed at the end of each day. Finally, ac-tivities may be recorded by logging daily excur-sions in a diary, or by checking off activitiescompleted from an agoraphobia checklist.

A common problem with self-monitoring isnoncompliance. Sometimes noncompliance isdue to misunderstanding or lack of perceivedcredibility in self-monitoring. Most often, how-ever, noncompliance is due to anticipation of


FIGURE 1.1. Julie’s responses to ADIS-IV questions.

Parts of ADIS-IV Panic Disorder Section

Do you currently have times when you feel a sudden rush of intense fear or discomfort? Yes.

In what kinds of situations do you have those feelings? Driving, especially on freeways . . . alone athome . . . at parties or in crowds of people.

Did you ever have those feelings come “from out of the blue,” for no apparent reason, or insituations where you did not expect them to occur? Yes.

How long does it usually take for the rush of fear/discomfort to reach its peak level? It varies,sometimes a couple of seconds and at other times it seems to build more slowly.

How long does the fear/discomfort usually last at its peak level? Depends on where I am at thetime. If it happens when I’m alone, sometimes it is over within a few minutes or even seconds. If I’m in acrowd, then it seems to last until I leave.

In the last month, how much have you been worried about, or how fearful have you been abouthaving another panic attack?

0 1 2 3 4 5 6 7 8

No worryno fear

Rarely worried/mildfear

Occasionallyworried/moderate

fear

Frequently worried/severe fear

Constantly worried/extreme fear

Parts of ADIS-IV Social Phobia Section

In social situations, where you might be observed or evaluated by others, or when meeting newpeople, do you feel fearful, anxious, or nervous? Yes.

Are you overly concerned that you might do and/or say something that might embarrass orhumiliate yourself in front of others, or that others may think badly of you? Yes.

What are you concerned will happen in these situations? That others will notice that I am anxious.My face turns white and my eyes look strange when I panic. I am worried that I’ll flip out in front ofthem, and they won’t know what to do.

Are you anxious about these situations because you are afraid that you will have an unexpectedpanic attack? Yes (either a panic or that I’ll feel unreal).

Other than when you are exposed to these situations, have you experienced an unexpected rushof fear/anxiety? Yes.

more anxiety as a result of monitoring. This isparticularly true for individuals whose pre-ferred style of coping is to distract themselvesas much as possible, and to avoid “quiet”times, when thoughts of panic might becomeoverwhelming: “Why should I make myselfworse by asking myself how bad I feel?” InJulie’s case, the self-monitoring task was partic-ularly difficult, because explicit reminders ofher anxiety elicited strong concerns about los-ing touch with reality. Prompting, reassurance

that anxiety about self-monitoring would sub-side with perseverance at self-monitoring, andemphasis on objective versus subjective self-monitoring were helpful for Julie. In addition,cognitive restructuring in the first few sessionshelped Julie to be less afraid of the feelings ofunreality; therefore, she was less afraid to bereminded of those feelings by self-monitoring.Finally, therapist attention to the self-monitoredinformation and corrective feedback about themethod of self-monitoring at the start of each


FIGURE 1.2. Julie’s Panic Attack Record.

Date 2/16/06 Time began 5:20 P.M.

Triggers Home alone and shortness of breath

Expected x Unexpected

Maximum Fear 0——1——2——3——4——5——6——7——8——9——10

None Mild Moderate Strong Extreme

Check all symptoms present to at least a mild degree:

Chest pain or discomfort Sweating x

Heart racing/palpitations/pounding x Nausea/upset stomach

Short of breath x Dizzy/unsteady/lightheaded/faint

Shaking/trembling x Chills/hot flushes

Numbness/tingling Feelings of unreality x

Feelings of choking Fear of dying

Fear of losing control/going crazy x

Thoughts: I am going crazy, I will lose control

Behaviors: Called my mother

FIGURE 1.3. Julie’s Daily Mood Record.

0——1——2——3——4——5——6——7——8——9——10

None Mild Moderate Strong Extreme

DateAverageanxiety

Averagedepression

Average worryabout panic

2/16 7 5 7

2/17 5 4 5

2/18 4 4 5

2/19 4 3 4

2/20 4 4 5

2/21 2 1 1

2/22 2 2 2

treatment session reinforced Julie’s self-monitoring.

Standardized Inventories

Several standardized self-report inventoriesprovide useful information for treatment plan-ning and are sensitive markers of therapeuticchange. The Anxiety Sensitivity Index (Reiss,Peterson, Gursky, & McNally, 1986) has re-ceived wide acceptance as a trait measure ofthreatening beliefs about bodily sensations.It has good psychometric properties andtends to discriminate between panic disorder–agoraphobia and other types of anxiety disor-ders (e.g., Taylor et al., 1992; Telch, Sherman,& Lucas, 1989), especially the Physical Con-cerns subscale (Zinbarg et al., 1997). Morespecific information about which particularbodily sensations are feared the most and whatspecific misappraisals occur most often may beobtained from the Body Sensations and Agora-phobia Cognitions Questionnaire (Chamblesset al., 1984). The Mobility Inventory (Cham-bless, Caputo, Gracely, Jasin, & Williams,1985) lists agoraphobic situations rated interms of degree avoidance when alone andwhen accompanied. This instrument is veryuseful for establishing in vivo exposure hierar-chies. Measures of trait anxiety include theState–Trait Anxiety Inventory (Speilberger,Gorsuch, Lushene, Vagg, & Jacobs, 1983) andthe Beck Anxiety Inventory (Beck, Epstein,Brown, & Steer, 1988).

In addition, we have developed two stan-dardized self-report inventories that are usefulfor panic disorder and agoraphobia. The first,the Albany Panic and Phobia Questionnaire(Rapee, Craske, & Barlow, 1995), is a 32-itemquestionnaire designed to assess fear andavoidance of activities that produce fearedbodily sensations, as well as more typical ago-raphobia and social situations. Factor analysesconfirmed three distinct factors labeled Agora-phobia, Social Phobia, and Interoceptive Fears.The questionnaire has adequate psychometricproperties and is useful in profiling agora-phobic versus interoceptive avoidance. The sec-ond, the Anxiety Control Questionnaire, is a30-item scale that assesses perceived lack ofcontrol over anxiety-related events and occur-rences, such as internal emotional reactions orexternally threatening cues (Rapee, Craske,Brown, & Barlow, 1996). This scale is designedto assess locus of control, but in a more specific

and targeted manner relevant to anxiety andanxiety disorders compared to more generallocus-of-control scales. A revised 15-item ver-sion yields three factors, Emotion Control,Threat Control, and Stress Control, with ahigher-order dimension of perceived control(Brown, White, Forsyth, & Barlow, 2004).Changes in this scale from pre to posttreatmentpredicted reductions in comorbidity at follow-up in one study (Craske et al., 2007). Finally,measures of interpersonal context include theDyadic Adjustment Scale (Spanier, 1976), andthe Marital Happiness Scale (Azrin, Naster, &Jones, 1973).

Behavioral Tests

The behavioral test is a useful measure of de-gree of avoidance of specific interoceptive cuesand external situations. Behavioral approachtests can be standardized or individually tai-lored. The standardized behavioral test foragoraphobic avoidance usually involves walk-ing or driving a particular route, such as a 1-mile loop around the clinic setting. Standard-ized behavioral tests for anxiety about physicalsensations involve exercises that induce panic-like symptoms, such as spinning in a circle,running in place, hyperventilating, and breath-ing through a straw (Barlow & Craske, 2006).Anxiety levels are rated at regular intervalsthroughout the behavioral tests, and actual dis-tance or length of time is measured. The disad-vantage of standardized behavioral tests is thatthe specific task may not be relevant to all pa-tients (e.g., a 1-mile walk or running in placemay be only mildly anxiety provoking); hence,the value of individually tailored tasks. In thecase of agoraphobia, this usually entails at-tempts at three to five individualized situationsthat the patient has identified as ranging fromSomewhat difficult to Extremely difficult, suchas driving two exits on freeway, waiting in abank line, or shopping in a local supermarketfor 15 minutes. For anxiety about physical sen-sations, individually tailored behavioral testsentail exercises designed specifically to inducethe sensations feared most by a given patient,and may include a tongue depressor to inducesensations of gagging, smells to induce sensa-tions of nausea, or nose plugs to induce sensa-tions of difficulty breathing. As with standard-ized tests, ongoing levels of anxiety and degreeof approach behavior are measured in relationto individually tailored behavioral tests.


Individually tailored behavioral tests aremore informative for clinical practice, althoughthey confound between-subject comparisonsfor research purposes. On the one hand, stan-dardized and individually tailored behavioraltests are susceptible to demand biases for bothfear and avoidance prior to treatment, and im-provement after treatment (Borkovec, Weerts,& Bernstein, 1977). On the other hand, behav-ioral tests are an important supplement to self-report of agoraphobic avoidance, because pa-tients tend to underestimate what they canactually achieve (Craske et al., 1988). In addi-tion, behavioral tests often reveal important in-formation for treatment planning of which theindividual is not yet fully aware. For example,the tendency to remain close to supports, suchas railings or walls, may not be apparent untilone observes the patient walk through a shop-ping mall. In Julie’s case, the importance ofchanges from daylight to night was not appar-ent until she was asked to drive on a section ofroad as a behavioral test. Her response wasthat it was too late in the day to drive, becausedusk made her feel as if things were unreal.Similarly, it was not until Julie completed abehavioral test that we recognized the impor-tance of air-conditioning when Julie was driv-ing. Julie believed that the cool air blowing onher face helped her to remain “in touch with re-ality.” Finally, we noticed that her physical pos-ture while driving was a factor that contributedto anxiety: Julie’s shoulders were hunched, sheleaned toward the wheel, and she held thewheel very tightly. All of these were targeted inthe treatment: driving at dusk was included inher hierarchy; air-conditioning was regarded asa safety signal from which she should beweaned; and driving in a more relaxed positionwas part of mastery exposure.

Psychophysiology

Ongoing physiological measures are not verypractical tools for clinicians, but they can pro-vide important information. In particular, thediscrepancy described earlier between reportsof symptoms and actual physiological arousal(i.e., report of heart rate acceleration in ab-sence of actual heart rate acceleration) mayserve as a therapeutic demonstration of the roleof attention and cognition in symptom produc-tion. Similarly, actual recordings provide datato disconfirm misappraisals such as “My heartfeels like its going so fast that it will explode”

or “I’m sure my blood pressure is so high that Icould have a stroke at any minute.” Finally,baseline levels of physiological functioning,which are sometimes dysregulated in anxiousindividuals, may be sensitive measures of treat-ment outcome (e.g., Craske, Golinelli, et al.,2005).

Functional Analysis

The various methods of assessment provide thematerial for a full functional analysis for Julie.Specifically, the topography of her panic attackis as follows: most common symptoms includea feeling of unreality, shortness of breath, andracing heart; average frequency is three perweek; each panic attack on average lasts from afew seconds to 5 minutes, if Julie is not in acrowd; in terms of apprehension, Julie worriesabout panic 75% of the day; and she hasmostly expected panic attacks but some unex-pected ones as well. Julie has both situationaland internal antecedents to her panic attacks.The situational antecedents include driving onfreeways; crowds of people; being alone, espe-cially at night; restaurants; dusk; reading andconcentrating for long periods of time; and aer-obic activity. The internal antecedents includeheart rate fluctuations, lightheaded feelings,hunger feelings, weakness due to lack of food,thoughts of the “big one” happening, thoughtsof not being able to cope with this for muchlonger, and anger. Her misappraisals aboutpanic attack symptoms include beliefs that shewill never return to normality, that she will gocrazy or lose control, and that others will thinkshe is weird. Her behavioral reactions to panicattacks include escape behaviors such as pull-ing off to the side of the road, leaving restau-rants and other crowded places, calling herhusband or mother, and checking for herKlonopin. Her behavioral reactions to the an-ticipation of panic attacks include avoidance ofdriving long distances alone, driving on unfa-miliar roads and freeways or at dusk, crowdedareas, exercise, quiet time with nothing to do,and doing one thing for a long period of time.In addition, she tries not to think about anxietyor feelings of unreality. Her safety signals andsafety-seeking behaviors include having herKlonopin on hand at all times, always knowingthe location of husband, and having the air-conditioning on. The consequences of her PDAaffect her family: Julie’s husband is concernedand supportive, but her mother thinks she


should pull herself together because “it’s all inher head.” In addition, Julie works but has cutback the number of hours, and she travels andsocializes much less. Her general mood in-cludes some difficulty concentrating and sleep-ing, restlessness, headaches, and muscularpains and aches. In addition, she is occasionallytearful, sad, and hopeless, and generally feelsdown.

COMPONENTS OFCOGNITIVE-BEHAVIORAL THERAPY

The components of the cognitive-behavioraltreatment described in this section are integrat-ed into a session-by-session treatment programin the next section.

Education

The treatment begins with education about thenature of panic disorder, the causes of panicand anxiety, and the ways panic and anxietyare perpetuated by feedback loops amongphysical, cognitive, and behavioral responsesystems. In addition, specific descriptions ofthe psychophysiology of the fight–flight re-sponse are provided, as well as an explanationof the adaptive value of the various physiologi-cal changes that occur during panic and anxi-ety. The purpose of this education is to correctthe common myths and misconceptions aboutpanic symptoms (i.e., beliefs about going crazy,dying, or losing control) that contribute topanic and anxiety. The survival value of alarmreactions (panic attacks) is emphasizedthroughout.

Education also distinguishes between thestate of anxiety and the emotion of fear/panic, both conceptually and in terms of itsthree response modes (subjective, physiologi-cal, and behavioral). This distinction is cen-tral to the model of panic disorder and to theremainder of the treatment. Anxiety is viewedas a state of preparation for future threat,whereas panic is the fight–flight emotion elic-ited by imminent threat. Panic/fear is charac-terized by (1) perception or awareness of im-minent threat, (2) sudden autonomicdischarge, and (3) fight–flight behavior. Anxi-ety is characterized by (1) perception orawareness of future threat, (2) chronic ten-sion, and (3) cautiousness, avoidance, anddisruption of performance.

Self-Monitoring

Self-monitoring is considered essential to thepersonal scientist model of cognitive-behavioraltherapy. Self-monitoring is introduced as a wayto enhance objective self-awareness and in-crease accuracy in self-observation. As notedearlier, patients are asked to keep at least twotypes of records. The first, a Panic Attack Re-cord, is completed as soon after each panic at-tack as possible; this record provides a de-scription of cues, maximal distress, symptoms,thoughts, and behaviors. The second, a DailyMood Record, is completed at the end of eachday to record overall or average levels of anxi-ety, depression, and whatever else is consideredimportant to record. Additionally, patients maykeep a daily record of activities or situationscompleted or avoided.

Breathing Retraining

Breathing retraining is a central componentearly on in the development of panic-controltreatments, because many panic patients de-scribe symptoms of hyperventilation as beingvery similar to their panic attack symptoms. Itis noteworthy, however, that hyperventilationsymptom report does not always accuratelyrepresent hyperventilation physiology: only50% or fewer patients show actual reductionsin end-tidal carbon dioxide values during panicattacks (Hibbert & Pilsbury, 1989; Holt & An-drews, 1989; Hornsveld, Garssen, FiedelijDop, & van Spiegel, 1990).

In early conceptualizations, panic attackswere related to stress-induced respiratorychanges that either provoke fear because theyare perceived as threatening or augment fear al-ready elicited by other phobic stimuli (Clark,Salkovskis, & Chalkley, 1985). Several studiesillustrated a positive effect of breathing retrain-ing. Kraft and Hoogduin (1984) found that sixbiweekly sessions of breathing retraining andprogressive relaxation reduced panic attacksfrom 10 to 4 per week, but were no more effec-tive than either repeated hyperventilation pluscontrol of symptoms by breathing into a bag oridentification of life stressors and problemsolving. Other studies were uncontrolled re-ports that combined breathing retrainingand cognitive restructuring, sometimes within vivo exposure (Clark et al., 1985; Rapee,1985; Salkovskis, Warwick, Clark, & Wessels,1986).


More recently, the value of breathing retrain-ing has been questioned. For example, it is un-clear whether breathing retraining alone is ther-apeutic for agoraphobia, and several studiessuggest that the addition of breathing retrainingalone does not improve upon in vivo exposure(e.g., de Beurs, van Balkom, Lange, Koele, & vanDyck, 1995). We found breathing retraining tobe slightly less effective than interoceptive expo-sure when each was added to cognitive restruc-turing and in vivo exposure (Craske, Rowe,Lewin, & Noriega-Dimitri, 1997), and in an-other study, the inclusion of breathing retrainingresulted in poorer outcomes than cognitive-behavioral therapy without breathing retrain-ing, although the findings were not robust(Schmidt et al., 2000). From their review of effi-cacy and mechanisms of action, Garssen, deRuiter, and van Dyck (1992) concluded thatbreathing retraining probably effects change notthrough breathing per se, but through distrac-tion and/or a sense of control. Given the recentrecognition that tolerance of fear and anxietymay be a more critical learning experience thanthe elimination of fear (see Eifert & Forsyth,2005), breathing retraining has beendeemphasized, because it may become a methodof avoidance of physical symptoms or a safetybehavior, and thereby be antitherapeutic. Whenit is included in the treatment, it is essential thatpatients not rely upon breathing retraining as amethod of avoidance or safety seeking.

Applied Relaxation

A form of relaxation known as applied relax-ation has shown good results as a treatment forpanic attacks. Applied relaxation entails train-ing patients in progressive muscle relaxation(PMR) until they are skilled in cue control re-laxation, at which point relaxation is used as acoping skill for practicing exposure to itemsfrom a hierarchy of anxiety-provoking tasks. Atheoretical basis for relaxation as a treatmentfor panic attacks has not been elaborated be-yond the provision of a somatic counter-response to the muscular tension that is likelyto occur during anxiety and panic. However,evidence does not lend support to this notion(Rupert, Dobbins, & Mathew, 1981). An alter-native suggestion is that, as with breathing re-training, fear and anxiety are reduced to the ex-tent that relaxation provides a sense of controlor mastery (Bandura, 1977; Rice & Blanchard,1982). The procedures and mechanisms ac-

countable for therapeutic gains are furtherclouded in the case of applied forms of relax-ation given the involvement of exposure-basedprocedures as anxiety-provoking situations arefaced.

Ost (1988) reported very favorable resultswith applied PMR: 100% of an applied PMRgroup (N = 8) were panic-free after 14 sessionsin comparison to 71.7% of a nonapplied PMRgroup (N = 8). Furthermore, the results of thefirst group were maintained at follow-up (ap-proximately 19 months after treatment com-pletion): All members of the applied PMRgroup were classified as high end state (i.e.,nonsymptomatic) at follow-up, compared to25% of the nonapplied PMR group. Michelsonand colleagues (1990) combined applied PMRwith breathing retraining and cognitive train-ing for 10 panickers. By treatment completion,all subjects were free of “spontaneous” panics,all but one were free of panic attacks alto-gether, and all met criteria for high end-statefunctioning. However, the specific contributionof applied PMR to these results is not known.Two subsequent studies by Ost (Ost &Westling, 1995; Ost, Westling, & Hellstrom,1993) indicate that applied relaxation was aseffective as in vivo exposure and cognitive ther-apy. In contrast, we (Barlow et al., 1989) foundthat applied PMR was relatively ineffective forpanic attacks, although we excluded all formsof interoceptive exposure from the hierarchy oftasks to which PMR was applied, which wasnot necessarily the case in the studies by Ost.Clark and colleagues (1994) found that cogni-tive therapy was superior to applied PMRwhen conducted with equal amounts of in vivoexposure, whereas Beck, Stanley, Baldwin,Deagle, and Averill (1994) found very few dif-ferences between cognitive therapy and PMRwhen each was administered without exposureprocedures.

Cognitive Restructuring

Initially, cognitive therapy for panic disorderand agoraphobia did not directly targetmisappraisals of bodily sensations, but insteadfostered coping self-statements in anxiety-provoking situations. Michelson, Mavissa-kalian, and Marchione (1985) published thefirst of their series of investigations comparingdifferent behavioral treatments to variouscoping-oriented cognitive treatments for agora-phobia. They compared paradoxical intention,


graduated exposure, and progressive deep mus-cle relaxation, although all participants con-ducted self-directed in vivo exposure betweensessions. At posttreatment and 3 months later,paradoxical intention demonstrated equivalentrates of improvement, but significantly moreparticipants remained symptomatic comparedto those treated with graduated exposure andrelaxation. Michelson, Mavissakalian, andMarchione (1988) replicated this design withalmost twice as many participants. Contrary tothe first study, few significant differences weredetected between treatments. Lack of differ-ences was replicated in a third study(Michelson et al., 1990). Thus, coping-orientedcognitive treatments appeared to be as effectiveas behaviorally oriented treatments, althoughthe cognitive treatments were all heavily con-taminated by behavioral self-directed expo-sure. In a slightly different design, Murphy,Michelson, Marchione, Marchione, and Testa(1998) compared cognitive therapy combinedwith therapist- and self-directed exposure, re-laxation combined with therapist- and self-directed exposure, and just therapist and self-directed exposure. Again, overall there werefew significant differences, although the condi-tion that included cognitive therapy yielded themost potent and stable changes. Without theself-directed exposure component, Emmel-kamp and colleagues found that coping-oriented cognitive therapy (rational–emotivetherapy and self-instruction training) was sig-nificantly less effective than prolonged in vivoexposure for agoraphobia on an array ofbehavioral and self-report measures of anxietyand avoidance (Emmelkamp, Brilman, Kuiper,& Mersch, 1986; Emmelkamp, Kuipers, &Eggeraat, 1978; Emmelkamp & Mersch,1982).

Cognitive therapy that targets misappraisalsof bodily sensations is clearly effective withsamples with mild to moderate levels of agora-phobia, producing results that are either as ef-fective as or superior to applied relaxation(Arntz & van den Hout, 1996; Beck et al.,1994; Clark et al., 1994; Ost & Westling,1995; Stanley et al., 1996). Results with moresevere levels of agoraphobia are mixed. Onestudy indicated that cognitive therapy targetingmisappraisals of bodily sensations is as effec-tive as guided mastery exposure delivered in-tensively over 6 weeks for individuals withmoderate to severe agoraphobia (Hoffart,1995), and other studies showed that cognitive

restructuring combined with breathing retrain-ing and/or interoceptive exposure is as effectiveas self-directed in vivo exposure (Craske et al.,2003; de Ruiter, Garssen, Rijken, &Kraaimaat, 1989; Rijken, Kraaimaat, deRuiter, & Garssen, 1992) for individuals withvarying levels of agoraphobia. Other studiesfound that cognitive therapy is slightly less ef-fective than guided mastery and in vivo expo-sure for agoraphobia (Bouchard et al., 1996;Williams & Falbo, 1996). Furthermore, severalstudies found no added benefit when cognitivetherapy that targeted misappraisals of bodilysensations was added to in vivo exposure (Ost,Thulin, & Ramnero, 2004; van den Hout,Arntz, & Hoekstra, 1994).

Behavioral exposure-based strategies areusually included in cognitive therapy as vehi-cles for obtaining data that disconfirmmisappraisals. The importance of exposure-based strategies to the effectiveness of cognitivetherapy is not known, although 2 weeks of fo-cused cognitive therapy with antiexposure in-structions reduced panic attacks in all but oneof a series of seven cases in a single-case, mul-tiple baseline design (Salkovskis, Clark, &Hackmann, 1991).

In terms of implementation, cognitive ther-apy begins to provide a treatment rationalewith discussion of the role of thoughts in gener-ating emotions. Next, thoughts are recognizedas hypotheses rather than fact, and are there-fore open to questioning and challenge. De-tailed self-monitoring of emotions and associ-ated cognitions is instituted to identify specificbeliefs, appraisals, and assumptions. Once rele-vant cognitions are identified, they are catego-rized into types of typical errors that occurduring heightened emotion, such as over-estimations of risk of negative events orcatastrophizing of meaning of events. The pro-cess of categorization, or labeling of thoughts,is consistent with a personal scientist modeland facilitates an objective perspective bywhich the validity of the thoughts can be evalu-ated. Thus, in labeling the type of cognitive dis-tortion, the patient is encouraged to use an em-pirical approach to examine the validity of hisor her thoughts by considering all of the avail-able evidence. Therapists use Socratic question-ing to help patients make guided discoveriesand question their anxious thoughts. Next,more evidence-based alternative hypotheses aregenerated. In addition to surface-level apprais-als (e.g., “That person is frowning at me be-


cause I look foolish”), core-level beliefs orschemas (e.g., “I am not strong enough to with-stand further distress” or “I am unlikable”) arequestioned in the same way. Importantly, cog-nitive restructuring is not intended as a directmeans of minimizing fear, anxiety, or unpleas-ant symptoms. Instead, cognitive restructuringis intended to correct distorted thinking; even-tually fear and anxiety are expected to subside,but their diminution is not the first goal of cog-nitive therapy.

Exposure

Exposure is a critical phase of treatment andonce begun, is a major focus of treatment ses-sions as well as between treatment sessionhomework, since limited exposure practice isof small benefit and may even be detrimental.The exposure is designed to disconfirm mis-appraisals and extinguish conditioned emo-tional responses to external situations and con-texts, through in vivo exposure, as well as tobodily sensations, through interoceptive expo-sure.

In Vivo Exposure

In vivo exposure refers to repeated and system-atic real-life exposure, in this case, to agora-phobic situations. As indicated from the studiesreviewed earlier, a long history of research hasestablished the efficacy of in vivo exposure foragoraphobia.

Most often, in vivo exposure is conducted ina graduated manner, proceeding from the leastto the most anxiety-provoking situations on anavoidance hierarchy. However, there is someevidence to suggest that intensive or ungradu-ated exposure may be effective. In a study byFeigenbaum (1988), treatment sessions wereconducted in a massed format over the courseof 6–10 consecutive days. One group receivedungraded exposure (N = 25), beginning withthe most feared items from avoidance hierar-chies. Another group received graded exposure(N = 23), beginning with the least feared hier-archy items. Approximately one-third of thisseverely agoraphobic sample was houseboundat initial assessment. At posttreatment and 8months later, the conditions proved to beequally effective (although, intriguingly, thegraded group reported the treatment to bemore distressing). However, ungraded expo-sure was clearly superior at the 5-year follow-

up assessment: 76% of the intensive group ver-sus 35% of the graded group reportedthemselves to be completely free of symptoms.When 104 subjects were added to the intensiveexposure format, the same results were ob-tained. Of 129 subjects, 78% were reportedlycompletely symptom-free 5 years later. Thisdramatic set of results suggests that an in-tensive approach, which is likely to producehigher levels of arousal than a graduated ap-proach, can be very beneficial (at least whenconducted in a massed format). Unfortunately,the validity of the outcome measures in thisstudy is somewhat questionable, and replica-tion by independent investigators has yet to bereported.

Critical to in vivo exposure is the removal ofsafety signals and safety behaviors. Examplesof safety signals include other people, water,money (to call for help), empty or full medica-tion bottles, exit signs, and familiar landmarkswhen traveling. Safety behaviors similarly pro-vide a sense of safety, and include seeking reas-surance or checking for exits. Reliance onsafety signals and safety behaviors attenuatedistress in the short term but maintain exces-sive anxiety in the long term. With the thera-pist’s guidance, the patient identifies and findsways gradually to eliminate his or her ownsafety signals and behaviors. In addition, invivo exposure is eventually combined withinteroceptive exposure, by deliberately induc-ing feared sensations in feared situations.

The amount of time devoted to in vivo expo-sure is very dependent on the patient’s agora-phobia profile. Obviously, more time is neededfor patients with more severe agoraphobia.Also, as reviewed earlier, evidence indicatesthat inclusion of significant others in the treat-ment process can improve treatment outcomes(e.g., Cerny et al., 1987). The benefit obtainedfrom involving significant others may dependon the pervasiveness of agoraphobia and theextent to which family roles and interactionshave been affected by or contribute to theagoraphobic pattern.

Interoceptive Exposure

In interoceptive exposure, the goal is to deliber-ately induce feared physical sensations a suffi-cient number of times, and long enough eachtime so that misappraisals about the sensationsare disconfirmed and conditioned anxiety re-sponses are extinguished. A series of studies


have reported on the effects of interoceptive ex-posure independent of other therapeutic strate-gies. Early on, Bonn, Harrison, and Rees(1971) and Haslam (1974) observed successfulreduction in reactivity with repeated infusionsof sodium lactate (a drug that produces panic-type bodily sensations). However, panic wasnot monitored in these investigations. Griezand van den Hout (1986) compared six ses-sions of graduated CO2 inhalations to a treat-ment regimen of propranolol (a beta-blockerchosen because it suppresses symptoms in-duced by CO2 inhalations), both conductedover the course of 2 weeks. CO2 inhalationtreatment resulted in a mean reduction from 12to 4 panic attacks, which was superior to theresults from propranolol. In addition, inhala-tion treatment resulted in significantly greaterreductions in reported fear of sensations. A 6-month follow-up assessment suggested mainte-nance of treatment gains, although panic fre-quency was not reported. Beck and Shipherd(1997) similarly found positive effects from re-peated CO2 inhalations, although it had littleeffect on agoraphobia (Beck, Shipherd, &Zebb, 1997). Broocks and colleagues (1998)tested the effects of exercise (with once-weeklysupportive contact from a therapist) in com-parison to clomipramine or drug placebo over10 weeks. The exercise group was trained torun 4 miles, three times per week. Despite highattrition from exercise (31%), exercise wasmore effective than the drug placebo condition.However, clomipramine was superior to exer-cise.

In the first comparison to other cognitiveand behavioral treatments, we (Barlow et al.,1989) compared applied PMR, interoceptiveexposure plus breathing retraining and cogni-tive restructuring, their combination with ap-plied PMR, and a waiting-list control, in a sam-ple with panic disorder with limitedagoraphobia. The two conditions involvinginteroceptive exposure, breathing retrainingand cognitive restructuring, were significantlysuperior to applied PMR and waiting-list con-ditions. The results were maintained 24months following treatment completion for thegroup receiving interoceptive exposure, breath-ing retraining, and cognitive restructuringwithout PMR, whereas the combined grouptended to deteriorate over follow-up (Craske,Brown, & Barlow, 1991). As already men-tioned, we compared interoceptive exposure,cognitive therapy, and in vivo exposure to

breathing retraining, cognitive therapy, and invivo exposure for individuals with varying lev-els of agoraphobia. The condition that in-cluded interoceptive exposure was slightly su-perior to breathing retraining at posttreatmentand 6 months later (Craske et al., 1997). Simi-larly, Ito, Noshirvani, Basoglu, and Marks(1996) found a trend for those who addedinteroceptive exposure to their self-directed invivo exposure and breathing retraining to bemore likely to achieve at least a 50% improve-ment in phobic fear and avoidance. Recently,an intensive, 8-day treatment with a sensation-focused approach was developed for individu-als with moderate to severe agoraphobia, andinitial results are promising (Morisette, Spiegel,& Heinrichs, 2005). But breathing education,breathing retraining, and repeated intero-ceptive exposure to hyperventilation did not in-crease the effectiveness of in vivo exposure foragoraphobia (de Beurs, Lang, van Dyck, &Koele, 1995).

Interoceptive exposure is now a standardcomponent of cognitive-behavioral therapy forpanic disorder (e.g., Barlow et al., 2000;Craske, Lang, et al., 2005), although differentgroups give different emphases to interoceptiveexposure, with some emphasizing it as a meansfor extinguishing fear responses (Barlow &Craske, 2006) and others, as a vehicle fordisconfirming misappraisals (Clark, 1996).

In terms of implementation, a standard listof exercises, such as hyperventilating and spin-ning, are used to establish a hierarchy ofinteroceptive exposures. With a graduated ap-proach, exposure begins with the less distress-ing physical exercises and continues with themore distressing exercises. It is essential thatthe patient endure the sensations beyond thepoint at which they are first noticed, for at least30 seconds to 1 minute, because early termina-tion of the task may eliminate the opportunityto learn that the sensations are not harmful andthat the anxiety can be tolerated. The copingskills of cognitive restructuring and slow dia-phragmatic breathing are used after each exer-cise, followed by a discussion of what the pa-tient learned during the exercise about bodilysensations, fear, and avoidance. These intero-ceptive exercises are practiced daily outside ofthe therapy session to consolidate the processof learning. Interoceptive exposure extends tonaturalistic activities that inherently induce so-matic sensations (e.g., caffeine consumption,exercise).


Optimizing Learning during Exposure

The ways in which learning during exposuretherapy is optimized are open to continuing in-vestigation. In this section, we highlight the lat-est developments in the research.

LENGTH OF AN EXPOSURE PRACTICE

Expectancies regarding the likelihood of aver-sive events are central to human fear condition-ing. For example, contingency awareness (i.e.,knowledge that a specific conditional stimulus[CS] predicts a specific unconditioned stimulus[US]), although of debatable necessity for con-ditioned responding (cf. Lovibond & Shanks[2002] and Ohman & Mineka [2001]) is astrong correlate of conditioned responding.Differential autonomic conditioning in particu-lar is strongly associated with verbal measuresof contingency knowledge (e.g., Purkis & Lipp,2001). Expectancies also are important for ex-tinction; extinction is posited to follow from amismatch between the expectancy of an aver-sive event and the absence of its occurrence(Rescorla & Wagner, 1972), or from the per-ception of a negative change in the rate atwhich aversive events are associated with theCS (Gallistel & Gibbon, 2000); that is, expec-tancies for the US are violated during extinc-tion. Thus, exposure tasks designed to violateexpectancies for negative outcomes are hypoth-esized to be the most effective form of exposure(Craske & Mystkowski, 2006). Indirect evi-dence derived from several studies of phobicsamples indicates that a single, massed expo-sure is more effective than a series of short ex-posures of the same total duration, such as one60-minute duration versus three 20-minute du-rations of exposure (e.g., Chaplin & Levine,1981; Marshall, 1985). Conceivably, thelengthier (massed) exposure is more effective,because it provides sufficient time to learn thataversive outcomes do not occur (i.e., to dis-confirm negative outcome expectancies)(Craske & Mystkowski, 2006). However, nostudy to date has directly evaluated outcomeexpectancies or manipulated exposure dura-tion in relation to outcome expectancies.

Related, however, is the body of work on therole of distraction during exposure, becausedistraction in essence represents disrupted (i.e.,unmassed) exposure. We (Craske, Street, &Barlow, 1989) administered therapist- and self-directed exposure to patients with agoraphobia

in small groups for 11 sessions. In one condi-tion (N = 16), patients were instructed tomonitor bodily sensations and thoughts objec-tively throughout in vivo exposures, and to usethought stopping and focusing self-statementsto interrupt distraction. In a second condition(N = 14), they were taught to use specific dis-traction tasks during in vivo exposures (wordrhymes, spelling, etc.), and to use thought stop-ping and distracting self-statements to inter-rupt the focus of attention upon feared bodilysensations and images. The treatment groupsdid not differ at posttreatment or at follow-upassessment, but, consistent with previousfindings with obsessive–compulsive disorder(Grayson, Foa, & Steketee, 1982), the focusedexposure group improved significantly fromposttreatment to follow-up, in contrast to aslight deterioration in the distracted exposuregroup. However, the degree to which partici-pants were actually distracted versus focusedwas not ascertainable. Also, other results re-garding the detrimental effects of distractionduring exposure therapy have been contradic-tory (e.g., Kamphuis & Telch, 2000; Oliver &Page, 2003; Rodriguez & Craske, 1995; Rose& McGlynn, 1997). The equivocal nature ofthe findings may derive from lack of an opera-tional definition of “distraction,” from con-founds with the affective quality of thedistractor, and from the unknown amount ofdistraction that actually takes place.

Nonetheless, given the recent advances in re-search, showing that neither physiological ha-bituation nor the amount of fear reductionwithin an exposure trial is predictive of overalloutcome (see Craske & Mystkowski, 2006),and given that self-efficacy through perfor-mance accomplishment is predictive of overallphobia reductions (e.g., Williams, 1992), andthat toleration of fear and anxiety may be amore critical learning experience than the elim-ination of fear and anxiety (see Eifert &Forsyth, 2005), the focus now is on staying inthe phobic situation until the specified time,when patients learn that what they are mostworried about never or rarely happens, and/orthat they can cope with the phobic stimulusand tolerate the anxiety. Thus, the length of agiven exposure trial is based not on fear reduc-tion but on the conditions necessary for newlearning, in which fear and anxiety eventuallysubside across trials of exposure. Essentially,the level of fear or fear reduction within a giventrial of exposure is no longer considered an in-


dex of learning, but a reflection of perfor-mance; learning is best measured by the level ofanxiety experienced the next time the patientencounters the phobic situation or at somelater time. Therefore, we have moved awayfrom the model of “Stay in the situation untilfear has declined” to “Stay in the situation un-til you have learned what you need to learn,and sometimes that means learning that youcan tolerate fear.” Exposure tasks, therefore,are to be defined clearly in advance, indepen-dent of level of fear reduction in a given day ofpractice. For example, patients are encouragedto practice inducing sensations of shortness ofbreath for a predetermined amount of time,and driving on the freeway for a predetermineddistance to gain experience that disconfirmswhat they fear most. If patients are most wor-ried about their fear remaining elevatedthroughout the entire exposure, then the goalof exposure is reframed as learning to be ableto tolerate a sustained level of fear. Neverthe-less, there may be occasions when the therapistjudges that the most effective learning comesfrom enduring an exposure task until fear hasdeclined, such as would be the case for patientswho maintain that their fear will decrease onlywhen they exit from the situation.

SCHEDULE OF EXPOSURE PRACTICES

A second way of potentially optimizing expo-sure is through the scheduling of exposure ses-sions. Spacing between exposure days (as op-posed to the duration of a given exposurepractice) pertains to consolidation of learning.Unfortunately, research in human samples hasfailed simultaneously to address both massingwithin exposure trials and spacing between ex-posure trials; that is, studies of spacing betweenexposure days have been conducted withoutensuring necessarily that exposure is suffi-ciently lengthy within each exposure day to vi-olate negative expectancies effectively; hence,the results have been mixed. Foa, Jameson,Turner, and Payne (1980) found greater decre-ments in anxiety and avoidance behavior inthose receiving massed rather than spaced ex-posure sessions for agoraphobia, whereasRamsay, Barends, Brueker, and Kruseman(1966) found spaced schedules to be superiorto massed schedules for desensitization for spe-cific phobias. Chambless (1990) found no dif-ferences between weekly versus daily sessionsof graduated in vivo exposure and training in

respiratory control, distraction techniques, andparadoxical intention. However, some subjectswere unwilling to accept massed exposure, cre-ating a sample selection bias. In addition,Chambless pointed out that her results maylack generalization, because spaced exposure isusually interspersed with homework assign-ments, which may increase outcome efficacy.Nevertheless, she concluded by suggesting thatthe choice for massed versus spaced exposure isthe decision of the therapist and patient. Someof the contradiction arises from inconsistentoperationalization of massed and spacedscheduling across studies. Studies have com-pared arbitrarily chosen fixed durations andschedules of exposure, and sometimes what islabeled as “massed” in one study is labeled as“spaced” in another.

Nonetheless, given the strength of the exper-imental data on spacing of learning trials fornonemotional learning (Bjork & Bjork, 1992),the evidence for superior outcomes from aschedule of progressively increasing durationsbetween exposure trials in circumscribed pho-bias (e.g., Rowe & Craske, 1998), and the evi-dence for substantially improved outcomeswith monthly follow-up phone calls afterweekly cognitive-behavioral therapy for panicdisorder in primary care settings (Craske et al.,2006), a schedule of weekly sessions followedby progressively longer intervals between ses-sions may be advisable.

LEVEL OF AROUSAL WITHINAN EXPOSURE PRACTICE

Clinically, on the one hand, there is wide sub-scription to the theory that corrective learningis maximal when physiological arousal is ini-tially activated, then allowed to subside withinand between exposure sessions (i.e., emotionalprocessing theory) (Foa & McNally, 1996).However, recent post hoc analyses indicate thatthe degree to which physiological respondingdeclines from the beginning to the end of an ex-posure trial is not predictive of overall outcome(see Craske & Mystkowski, 2006). In addition,empirical and theoretical developments suggestthat a certain level of sustained excitation dur-ing extinction training may yield even more ef-fective results upon retesting. Specifically, Cain,Blouin, and Barad (2004) have found thatanxiogenic drugs such as yohimbine facilitateextinction in mice, and in general suggest thatdrugs or conditions that enhance adrenergic


transmission overcome a natural inhibitoryconstraint upon extinction. However, extantdata in humans are limited to post hoc observa-tions of a positive relationship between sus-tained excitation (i.e., heart rate) during expo-sure and overall outcome with circumscribedphobias (e.g., Rowe & Craske, 1998).

On the other hand, there is evidence for det-rimental effects of safety signals and safety be-haviors, which presumably lower anxiety andarousal during exposures. As mentioned ear-lier, common safety signals for patients withpanic disorder are the presence of another per-son, therapists, medications, and food or drink(Barlow, 1988). Although they alleviate distressin the short term, safety signals are assumed tosustain anxiety in the long term (Siddle &Bond, 1988). These effects have been explainedby associative and attributional mechanisms.The associative model assumes that the nega-tive associative strength of the inhibitory stim-ulus cancels out the positive associativestrength of the excitatory stimulus, so thatthere is no change from what is predicted by allcues (Lovibond, Davis, & O’Flaherty, 2000).The attributional model implies that if subjectsattribute the absence of an expected outcometo the inhibitory stimulus, then there is no rea-son to change the causal status of the excit-atory stimulus (Lovibond et al., 2000).

In terms of treatment, Sloan and Telch(2002) reported that claustrophobic partici-pants who received an exposure treatment inwhich they were encouraged to use safety sig-nals, reported more fear at posttest and follow-up than those encouraged to focus on their fearduring exposure. In a subsequent study,Powers, Smits, and Telch (2004) found that theperception of safety (i.e., availability of safetybehaviors regardless of whether they wereused) rather than use of safety was detrimentalto treatment outcome, because level of fear re-duction was unaffected by actual use of safetybehaviors. However, in both studies, the effectsof safety signal encouragement may have beenattributable to distraction, and the results werelimited to circumscribed phobias. In anotherstudy, Salkovskis (1991) showed that “within-situation safety behaviours” interfered with thebenefits of exposure therapy for panic and anx-iety, and that teaching anxious patients to re-frain from these behaviors leads to greater fearreduction after an exposure session. Clearlymuch more direct investigation is needed onthe effects of safety signals and avoidance re-

sponses during exposure therapy, especiallygiven the very direct implications for clinicalpractice.

Such research may be directed at medica-tions that can become safety signals, becausetheir availability reassures patients that thedangers of extreme fear are controllable. Attri-bution of safety to medications impedes correc-tion of misperceived danger (e.g., “It is safe forme to drive on the freeway even when unmedi-cated”), and attribution of therapeutic gains toa medication (alprazolam) in patients withpanic disorder and agoraphobia predicted sub-sequent withdrawal symptoms and relapse(Basoglu et al., 1994). Thus, the greater relapsefollowing exposure combined with anxiolytics(especially high-potency, short-acting drugs)compared to exposure alone (e.g., Marks et al.,1993) may be attributable to medications func-tioning as safety signals.

THE EFFECT OF CONTEXT ON RETURN OF FEAR

A fourth consideration to optimize learningduring exposure therapy derives from condi-tioning models in which extinction involveslearning new, inhibitory CS–no US associationsas opposed to unlearning original CS–US asso-ciations. Thus, Bouton (1993) proposed thatthe original excitatory meaning of the CS is noterased during extinction; rather, an additionalinhibitory meaning is learned. The resultingdual meaning of the CS creates an ambiguitythat is resolved only by the current context ofthe CS. Bouton uses the analogy of an ambigu-ous word; that is, reaction to the word “fire”depends largely on the context in which it oc-curs; “fire” may elicit a panic reaction in acrowded theater and elicit very little reaction ina carnival shooting gallery. Thus, the contextdetermines which meaning is expressed at anygiven time. In terms of anxiety treatments,bodily sensations may mean “sudden death”when experienced in a context that reminds theperson of intense panic attacks before treat-ment, whereas the same sensations may mean“unpleasant but harmless” when experiencedin a context that reminds a person of his or hersuccess with treatment. The effects of contextshifts have been tested in circumscribed pho-bias, and indeed, return of fear is greater whenparticipants are subsequently assessed in a con-text distinctly different rather than the same asthat in which they were treated (for reviews,see Craske & Mystkowski, 2006; Hermans et


al., 2006). Hence, what is learned in the con-text of exposure therapy may not be retrievedat reencounters with the previously feared pho-bic object or situation after therapy is over.

Conceivably, conducting exposure therapyin multiple contexts minimizes the context re-newal effect after therapy is over. Unfortu-nately, extant research with humans is limitedto one study of circumscribed phobias(Vansteenwegen et al., 2007). Because it is notalways feasible to conduct exposures in origi-nal fear-acquisition or multiple contexts, we(Mystkowski, Craske, Echiverri, & Labus,2006) sought to investigate whether a contex-tually based return of fear could be counter-acted via mental rehearsal. Phobic participantswho were instructed to recall the exposurelearning environment just prior to being re-tested with a spider in a novel context showedless return of fear than those who were in-structed to recall unrelated events. Althoughthese findings were based on circumscribedphobias, they raise the possibility that simplyreminding patients to recall their treatment ex-periences may offset return of fear when theyreencounter their previously feared situationsafter treatment is over.

OVERALL EFFICACY OFCOGNITIVE-BEHAVIORAL THERAPY

Cognitive-behavioral therapy, involving mostor all of the components just listed, yieldspanic-free rates in the range of 70–80% andhigh end-state rates (i.e., within normativeranges of functioning) in the range of 50–70%,for panic disorder with minimal agoraphobia(e.g., Barlow et al., 1989; Clark et al., 1994).Two meta-analyses reported very large effectsizes of 1.55 and 0.90 for cognitive-behavioraltherapy for panic disorder (Mitte, 2005;Westen & Morrison, 2001). Also, results gen-erally maintain over follow-up intervals for aslong as 2 years (Craske et al., 1991). One anal-ysis of individual profiles over time suggested aless optimistic picture in that one-third of pa-tients who were panic-free 24 months aftercognitive-behavioral therapy had experienced apanic attack in the preceding year, and 27%had received additional treatment for panicover that same interval of time (Brown &Barlow, 1995). Nevertheless, this approach toanalysis did not take into account the generaltrend toward continuing improvement over

time. Thus, rates of eventual therapeutic suc-cess may be underestimated when success isdefined by continuous panic-free status sincethe end of active treatment.

The effectiveness extends to patients who ex-perience nocturnal panic attacks (Craske,Lang, Aikins, & Mystkowski, 2005). Also,cognitive-behavioral therapy is effective evenwhen there is comorbidity, and some studies in-dicate that comorbidity does not reduce the ef-fectiveness of cognitive-behavioral therapy forpanic disorder (e.g., Allen & Barlow, 2006;Brown, Antony, & Barlow, 1995; McLean etal., 1998). Furthermore, cognitive-behavioraltherapy results in improvements in comorbidanxiety and mood disorders (Brown et al.,1995; Tsao et al., 1998; Tsao, Mystkowski,Zucker, & Craske, 2002, 2005), although re-sults in one study indicated that the benefits forcomorbid conditions may lessen over time,when assessed 2 years later (Brown et al.,1995). Nonetheless, the general finding of im-provement in comorbidity is significant giventhat it suggests the value of remaining focusedon the treatment for panic disorder even whencomorbidity is present, because the comorbidi-ty will be benefited as well, at least up to 1 year.Finally, applications of cognitive-behavioraltherapy have proven very helpful in loweringrelapse rates upon discontinuation of high-potency benzodiazepines (e.g., Otto et al.,1993; Spiegel, Bruce, Gregg, & Nuzzarello,1994).

Results in samples with moderate to severeagoraphobia are generally slightly less positivethan those in samples with no or mild agora-phobia (e.g., Williams & Falbo, 1996). How-ever, data typically show patterns of continuingimprovement over time. Furthermore, Fava,Zielezny, Savron, and Grandi (1995) foundthat only 18.5% of their panic-free patients re-lapsed over a period of 5–7 years afterexposure-based treatment for agoraphobia. Asmentioned, some research suggests that thetrend for improvement after acute treatment isfacilitated by involvement of significant othersin every aspect of treatment for agoraphobia(e.g., Cerny et al., 1987).

As noted earlier, recently, our multicentergroup evaluated strategies for maintaining re-sponse in those who are considerably improvedafter cognitive-behavioral treatment. Spe-cifically, 157 patients who had responded wellto initial treatment were randomized to receiveeither no further cognitive-behavioral treat-


ment or one maintenance session a month for 9months. At that point all treatment was discon-tinued for 1 year. At the end of that year,97.3% of the patients receiving the booster ses-sions continued to maintain their response,whereas 81.9% maintained their responsewithout the booster sessions; that is, 18.1%showed some loss of response compared toonly 2.7% of those receiving the booster ses-sions, a significant difference. In this largestudy, the value of occasional continuedbooster sessions was demonstrated.

Most of the outcome studies to date are con-ducted in university or research settings, withselect samples (although fewer exclusionarycriteria are used in more recent studies). Conse-quently, of major concern is the degree towhich these treatment methods and outcomesare transportable to nonresearch settings, withmore severe or otherwise different populations,and with less experienced or trainedclinicians—a topic that is just now receiving at-tention. Wade and colleagues (1998) used abenchmarking strategy to compare their resultsfrom a community mental health center withresults from research sites. One hundred tenindividuals underwent cognitive-behavioraltherapy for PD/PDA, concomitant with psy-chopharmocotherapy where appropriate.Therapists were trained extensively. As in priorstudies, treatment completion correlated posi-tively with years of education. Overall, the per-cent of panic-free individuals and the percentachieving normative levels of functioning on avariety of measures were comparable topercents obtained from research sites. As men-tioned, we are now evaluating the degree towhich these treatment results can be obtainedin other settings (e.g., primary care) and withless-well-trained therapists. In our first study ofpanic disorder in primary care, we found thatoffering a treatment combination of cognitive-behavioral therapy (up to six sessions) andpharmacotherapy yielded highly significantoutcomes relative to treatment as usual (TAU)in primary care settings, with relatively novicetherapists (Roy-Byrne et al., 2005).

TREATMENT DESCRIPTION:PROTOCOL

What follows is a description of a 12-sessioncognitive-behavioral therapy for PDA tailoredto Julie’s presentation. Of course, the degree to

which the various components of treatment areemphasized vary by the functional assessmentconducted for each patient.

Overview

The basic aim of the treatment protocol is toinfluence directly the catastrophic misap-praisals and avoidance of bodily sensations andagoraphobic situations. This is done firstthrough the provision of accurate informationas to the nature of the fight–flight response. Byprovision of such information, patients aretaught that they experience “sensations” andnot “panics,” and that these sensations are nor-mal and harmless. Second, treatment aims toteach a set of skills for developing evidence-based appraisals regarding bodily sensationsand agoraphobic situations. At the same time,specific information concerning the effects ofhyperventilation and its role in panic attacks isprovided, with extensive practice of breathingretraining. Then, the crux of the treatment in-volves repeated exposure to feared internalcues and agoraphobic situations.

Session 1

The goals of Session 1 are to describe fear andanxiety; to help patients understand the cycli-cal influences among behavioral, physiological,and cognitive responses; to understand thatpanic attack symptoms are not harmful; and tobegin self-monitoring, if it was not already be-gun with the initial assessment. Therapy beginswith identifying anxiety patterns and the situa-tions in which anxiety and panic attacks arelikely to occur. Many patients have difficultyidentifying specific antecedents, reporting thatpanic can occur at almost any time. Therapistshelp patients to identify internal triggers,specifically, negative verbal cognitions, cata-strophic imagery, and physical sensations. Thefollowing interchange took place for Julie:

THERAPIST: In what situations are you mostlikely to panic?

JULIE: Crowded restaurants and when I’m driv-ing on the freeway. But sometimes I am driv-ing along, feeling OK, when all of a sudden ithits. And other times I can be sitting at homefeeling quite relaxed and it just hits. That’swhen I really get scared, because I can’t ex-plain it.


THERAPIST: So, when you are driving on thefreeway, what is the very first thing you no-tice that tells you you’re about to panic?

JULIE: Well, the other cars on the road look as ifthey are moving really slowly.

THERAPIST: And what is the first thing you no-tice when you’re at home?

JULIE: An unreal feeling, like I’m floating.

THERAPIST: So, it sounds like the panic attacksthat seem to occur for no reason are actuallytied in with the sensations of unreality orwhen things look as if they are moving inslow motion.

JULIE: I guess so. I always thought the physicalfeelings were the panic attack, but maybethey start the panic attack.

Next, the three-response system model fordescribing and understanding anxiety andpanic is introduced. This model contributes toan objective self-awareness—to becoming apersonal scientist—and provides the ground-work for an alternative conceptual frameworkfor explaining panic and anxiety that replacesthe patient’s own misassumptions. Patients areasked to describe cognitive, physiological, andbehavioral aspects to their responding: to iden-tify the things that they feel, think, and dowhen they are anxious and panicky. As de-scribed earlier, differences between the re-sponse profiles of anxiety and panic are high-lighted. After grasping the notion of threeresponses that are partially independent, inter-actions among the response systems are de-scribed. The patient is asked to describe thethree-response system components in a recentpanic attack and to identify ways in which theyinteracted to produce heightened distress. Forexample,

THERAPIST: How would you describe the threeparts to the panic attack you had at homelast week?

JULIE: Well, physically, my head felt really light,and my hands were clammy. I thought that Iwould either pass out or that I would some-how dissolve into nothingness. My behaviorwas to lie down and call my husband, whowas at work.

THERAPIST: What was the very first thing younoticed?

JULIE: When I stood up, my head started to feelreally weird, as if it was spinning inside.

THERAPIST: What was your very next reactionto that feeling?

JULIE: I held onto the chair. I thought some-thing was wrong. I thought it could getworse and that I’d collapse.

THERAPIST: So it began with a physical sensa-tion, and then you had some very specificthoughts about those sensations. What hap-pened next?

JULIE: I felt very anxious.

THERAPIST: And what happened next?

JULIE: Well, the dizziness seemed to be gettingworse and worse. I became really concernedthat it was different from any other experi-ence I had ever had. I was convinced that thiswas “it.”

THERAPIST: So, as you became more anxious,the physical feelings and the thoughts thatsomething bad was going to happen intensi-fied. What did you do next?

JULIE: I called my husband and lay on the beduntil he came home. It was horrible.

THERAPIST: Can you see how one thing fed offanother, creating a cycle? That it began witha sensation, then some anxious thoughts,then feeling anxious, then more sensationsand more thoughts, and more fear, and soon?

Reasons why panic attacks first began areaddressed briefly. Patients are informed that itis not necessary to understand the reasons whythey began to panic to benefit from the treat-ment, because factors involved in onset are notnecessarily the same as the factors involved inthe maintenance of a problem. Nevertheless,the initial panic attack is described as a mani-festation of anxiety/stress. The stressors sur-rounding the time of the first panic attack areexplored with the patient, particularly in termsof how stressors may have increased levels ofphysical arousal and primed certain danger-laden cognitive schemas.

Next, the therapist briefly describes thephysiology underlying anxiety and panic, andthe myths about what the physical sensationsmight mean. The main concepts covered in thiseducational phase are (1) the survival value orprotective function of anxiety and panic; (2)the physiological basis to the various sensa-tions experienced during panic and anxiety,and the survival function of the underlying


physiology; and (3) the role of specific learnedand cognitively mediated fears of certain bodilysensations. The model of panic we describedearlier in this chapter is explained. In par-ticular, the concepts of misappraisals andinteroceptive conditioning are explained as ac-counting for panic attacks that seem to occurfrom out of the blue—that are triggered by verysubtle internal cues or physical sensations thatmay occur at any time. Not only does this in-formation reduce anxiety by decreasing uncer-tainty about panic attacks but it also enhancesthe credibility of the subsequent treatment pro-cedures. This information is detailed in a hand-out given to the patient to read over the nextweek (for the handout, see Barlow & Craske,2006).

This information was very important forJulie, because the inability to explain her panicattacks was a major source of distress. Here aresome of the questions she asked in her attemptto understand more fully:

JULIE: So, if I understand you correctly, you’resaying that my panic attacks are the same asthe fear I experienced the time we found aburglar in our house. It doesn’t feel the sameat all.

THERAPIST: Yes, those two emotional states—an unexpected panic attack and fear whenconfronted with a burglar—are essentiallythe same. However, in the case of theburglar, where were you focusing yourattention—on the burglar or on the way youwere feeling?

JULIE: The burglar, of course, although I did no-tice my heart was going a mile a minute.

THERAPIST: And when you have a panic attack,where are you focusing your attention—onthe people around you or on the way you arefeeling?

JULIE: Well, mostly on the way I’m feeling, al-though it depends on where I am at the time.

THERAPIST: Being most concerned about what’sgoing on inside can lead to a very differenttype of experience than being concernedabout the burglar, even though basically thesame physiological response is occurring.For example, remember our description ofthe way fear of sensations can intensify thesensations.

JULIE: But what about the feelings of unreality?How can they be protective or how can feel-

ing unreal help me deal with a danger situa-tion?

THERAPIST: OK, remember that it’s the physio-logical events that are protective—not thesensations. The sensations are just the endresult of those events. Now, feelings of unre-ality can be caused by changes in your bloodflow to your brain (although not danger-ously so), or from overbreathing, or fromconcentrating too intensely on what’s goingon inside you. So the unreality sensation maynot be protective, but the changes in bloodflow and overbreathing are.

JULIE: I understand how I can create a panic at-tack by being afraid of my physical feelings,like my heart racing or feeling unreal. Butsometimes it happens so quickly that I don’thave time to think.

THERAPIST: Yes, these reactions can occur veryquickly, at times automatically. But remem-ber, we are tuned to react instantaneously tothings (including our own bodies) that wethink mean danger. Imagine yourself walk-ing through a dark alley, and you have rea-son to believe that somewhere in the dark-ness lurks a killer. Under those conditions,you would be extremely attentive to anysign, any sound, or any sight of another per-son. If you were walking through the samealley and were sure there were no killers, youmight not hear or detect the same signals youpicked up on in the first case. Now let’stranslate this to panic; the killer in the darkalley is the panic attack, and the signs,sounds, and smells are the physical sensa-tions you think signal the possibility of apanic attack. Given the acute degree of sensi-tivity to physical symptoms that signal apanic attack, it is likely that you are noticingnormal “noises” in your body that youwould otherwise not notice, and on occa-sion, immediately become fearful because ofthose “noises.” In other words, the sensa-tions are often noticeable because you attendto them.

Next, the method of self-monitoring was de-scribed and demonstrated with in-session prac-tice of completing a Panic Attack Record. Juliewas concerned that self-monitoring would onlyelevate her distress, by reminding of the verything she was afraid of (panic and unreality).The therapist clarified the difference betweenobjective and subjective self-monitoring, and


explained that distress would subside as Juliepersevered with self-monitoring.

The homework for this session was to self-monitor panic attacks, daily anxiety, and moodand to read the handout. In fact, we encouragepatients to reread the handout several times,and to actively engage in the material by cir-cling or marking the most personally relevantsections or areas in need of clarification, be-cause effort enhances long-term retention ofthe material learned. Of course, for some pa-tients, reading the material draws their atten-tion to things they fear (just as with self-monitoring). In this case, therapists can discussthe role of avoidance versus that of exposure,and how, with repeated readings, distress levelswill most likely subside.

At the end of the session, Julie suddenly be-came highly anxious. She felt unable to tolerateeither the treatment procedures or her anticipa-tion of them. She became very agitated in theoffice and reported feelings of unreality. Sheopened the office door to find her husband,who was waiting outside. The therapist helpedJulie understand how the cycle of panic hademerged in the current situation: (1) The trig-ger was the treatment description—having toeventually face feared sensations and situa-tions; (2) this was anxiety producing, becauseJulie believed that she could not cope with thetreatment demands, that the treatment wouldcause her so much anxiety that she would “flipout” and lose touch with reality permanently,or that she would never improve because shecould not tolerate the treatment; (3) the currentanxiety in the office elicited sensations of unre-ality and a racing heart; (4) Julie began toworry that she might panic and lose touch withreality permanently within the next few min-utes; (5) the more anxious Julie felt, and thestronger her attempts to escape and find safety,the stronger the physical sensations became;and (6) she felt some relief upon finding herhusband, because his presence reassured herthat she would be safe. Julie was reassured thattreatment would progress at a pace with whichshe was comfortable, but at the same time shewas helped to understand that her acute dis-tress about the feeling of unreality would be theprecise target of this type of treatment, there-fore attesting to the relevance of this treatmentfor her. She was also calmed by preliminarycognitive restructuring of the probability ofpermanently losing touch with reality. After alengthy discussion, Julie became more receptive

to treatment. A team approach to treatmentplanning and progress was agreed upon, sothat Julie did not feel that she would be forcedto do things she did not think she could do.

Session 2

The goals of this session are to begin the devel-opment of a hierarchy of agoraphobic situa-tions and coping skills of breathing retrainingand cognitive restructuring. The individualizedhierarchy comprises situations that range frommild to moderate anxiety, all the way up to ex-treme anxiety. These situations become the ba-sis of graduated in vivo exposure. Although invivo exposure exercises are not scheduled totake place until Session 4, the hierarchy is in-troduced now, so that cognitive restructuringskills can be practiced in relation to each situa-tion on the hierarchy before in vivo exposurebegins. Moreover, the hierarchy will be refinedas a result of the cognitive restructuring prac-tice, because the latter highlights specific fea-tures of agoraphobic situations that are mostanxiety provoking.

Julie was asked to develop a hierarchy overthe following week. She expressed some doubtthat she would ever be able to accomplish any,let alone all, of the items on her hierarchy. Thetherapist helped Julie by asking her to think ofany situation in her lifetime that used to be dif-ficult but became easier with practice. Julie re-membered how anxious she used to be whenshe first started working with customers at herhusband’s office—and how that discomfortsubsided over time. This was used to help Julierealize that the same might happen with the sit-uations listed on her hierarchy. Julie’s final hi-erarchy comprised the following situations:driving home from work alone; sitting in acrowded movie theater; spending 2 hours aloneat home during the day; alone at home as dayturned to night; driving on surface streets toher brother’s house (10 miles) alone; drivingtwo exits on freeway 444, with her husbandfollowing in the car behind; driving two exitson freeway 444, alone; driving four exits onfreeway 444; and driving on the freeway to herbrother’s house alone. Then, Julie was to repeatall of these tasks without taking Klonopin, andwithout knowing the location of her husband.

Breathing retraining also is begun in this ses-sion. Patients are asked to hyperventilate vol-untarily by standing and breathing fast anddeep, as if blowing up a balloon, for 1½ min-


utes. With prompting and encouragement fromthe therapist, patients can often complete thefull 1½ minutes, after which time they areasked to sit, close their eyes, and breathe veryslowly, pausing at the end of each breath, untilthe symptoms have abated. The experience isthen discussed in terms of the degree to whichit produced symptoms similar to those that oc-cur naturally during anxiety or panic. Approxi-mately 50–60% of patients report close sim-ilarity of the symptoms. Often, however,similarity of the symptoms is confused withsimilarity of the anxiety. Because the exercise isconducted in a safe environment and the symp-toms have an obvious cause, most patients ratethe experience as less anxiety provoking than ifthe same symptoms had occurred naturally.This distinction is important to make, becauseit demonstrates the significance of perceivedsafety for the degree of anxiety experienced.Julie rated the hyperventilation exercise as veryanxiety provoking (8 on a 0- to 10-point scale),and rated the symptoms as being quite similarto her panic symptoms (6 on a 0- to 10-pointscale). She terminated the task after approxi-mately 40 seconds, in anticipation of experi-encing a full-blown panic attack. The therapistand Julie discussed this experience in terms ofthe three response systems, and the role ofmisappraisals and interoceptive conditioningdescribed during the previous session.

Then, Julie was briefly educated about thephysiological basis to hyperventilation (seeBarlow & Craske, 2006). As before, the goal ofthe didactic presentation was to allay misinter-pretations of the dangers of overbreathing, andto provide a factual information base on whichto draw when actively challenging misinterpre-tations. The educational content is tailored tothe patient’s own educational level and coveredonly to the degree that it is relevant to the pa-tient.

In the next step, the therapist teaches breath-ing retraining, which begins by teaching pa-tients to rely more on the diaphragm (abdo-men) than chest muscles. In addition, patientsare instructed to concentrate on their breath-ing, by counting on their inhalations, andthinking the word “relax” on exhalations.(Slow breathing is introduced in Session 3.)Therapists model the suggested breathing pat-terns, then provide corrective feedback to pa-tients while they practice in the office setting.

Initial reactions to the breathing exercisemay be negative for patients who are afraid of

respiratory sensations, because the exercise di-rects their attention to breathing. It also can bedifficult for patients who are chronic over-breathers, and patients for whom any interrup-tion of habitual breathing patterns initially in-creases respiratory symptomatology. In bothcases, continued practice is advisable, with re-assurance that sensations such as shortness ofbreath are not harmful. The goal is to usebreathing skills training to encourage contin-ued approach toward anxiety and anxiety-producing situations. On occasion, patientsmistakenly view breathing retraining as a wayof relieving themselves of terrifying symptoms,thus falling into the trap of fearing dire conse-quences should they not succeed in correctingtheir breathing. This is what happened forJulie:

JULIE: So, all I have to do is to slow down mybreathing, then everything will be OK.

THERAPIST: Certainly, slowing down yourbreathing will help to decrease the physicalsymptoms that you feel, but I am not surewhat you mean when you ask whether ev-erything will be OK.

JULIE: That proper breathing will prevent mefrom losing touch with reality—that I won’tdisappear.

THERAPIST: Remember, whether you breatheslowly or quickly, from your chest or fromyour abdomen, you will not disappear. Inother words, it is a misinterpretation tothink that the sense of unreality means thatyou are permanently losing touch with real-ity or that you will disappear. Breathing re-training will help you to feel more relaxedand, therefore, less likely to feel the sense ofunreality, but the sense of unreality is not asign of actual loss of touch with reality anddisappearance.

The homework is to practice diaphragmaticbreathing for at least 10 minutes, two times aday in relaxing environments.

Therapists introduce in this session cognitiverestructuring by explaining that errors in think-ing occur for everyone when anxious, thushelping the patient to expect his or her thinkingto be distorted. Patients are informed that thesedistortions have an adaptive function: Chancesof survival are greater if we perceive danger asprobable and worthy of attention than if weminimize danger. Therefore, anxiety leads us to


judge threatening events as being more likelyand more threatening than they really are.However, the cognitive distortions are unneces-sary, because there is no real threat in the caseof panic disorder.

Then, patients are taught to treat theirthoughts as hypotheses or guesses rather thanas facts. The notions of automatic thinking anddiscrete predictions are also explained, to em-phasize the need of becoming an astute ob-server of one’s own habitual self-statements ineach situation. This leads to a “downward ar-row technique” to identify specific predictionsmade at any given moment, as shown withJulie.

THERAPIST: What is it that scared you aboutfeeling detached in the movie theater lastnight?

JULIE: It is just such a horrible feeling.

THERAPIST: What makes it so horrible?

JULIE: I can’t tolerate it.

THERAPIST: What makes you think you cannottolerate it? What is the feeling of detachmentgoing to do to you that makes you think it ishorrible and intolerable?

JULIE: It might get to be so intense that it over-whelms me.

THERAPIST: And if it overwhelms you, whatwould happen?

JULIE: I could become so distressed that I losetouch with reality.

THERAPIST: What would it mean if you losttouch with reality?

JULIE: That I would be in a different mind stateforever—I would never come back to reality.That I would be so crazy that I would haveto be carted out of the movie theater to amental hospital and locked away forever.

Overly general self-statements, such as “Ifeel terrible—something bad could happen,”are insufficient, nontherapeutic, and may serveto intensify anxiety by virtue of their globaland nondirective nature. Instead, detail inthought content, such as “I am afraid that if Iget too anxious while driving, then I’ll lose con-trol of the wheel and drive off the side of theroad and die,” permits subsequent cognitive re-structuring.

Analysis of anxious thought content yieldstwo broad factors that are labeled as “risk”

and “valence.” These two main types of cogni-tive errors are described to patients. Risktranslates to overestimation, or jumping toconclusions by viewing negative events as beingprobable events, when in fact they are unlikelyto occur. The patient is asked to identifyoverestimations from the anxiety and panic in-cidents over the past couple of weeks: “Canyou think of events that you felt sure were go-ing to happen when you panicked, only to findout in the end that they did not happen at all?”Usually, patients can identify such events easily,but with protestations. For example,

JULIE: Well, several times I thought that I reallywas going to lose it this time . . . that I wouldflip out and never return to reality. It neveractually happened, but it could still happen.

THERAPIST: Why do you think “it” could stillhappen?

JULIE: Part of me feels like I’ve always managedto escape it just in time, by either removingmyself from the situation or by having myhusband help me, or by holding on longenough for the feeling to pass. But what ifnext time I can’t hold on?

THERAPIST: Knowing what we know about ourthoughts when we are anxious, can you clas-sify any of the ideas you just expressed, of“just holding on” or “just escaping in time,”as overestimations?

JULIE: I suppose you’re saying that I can holdon or I can always escape in time.

THERAPIST: More that you feel the need to holdon and the need to escape, because you areoverestimating the likelihood of flipping outand never returning to reality.

JULIE: But it really feels like I will.

THERAPIST: The confusion between what youthink will happen and what actually happensis the very problem that we are addressing inthis session.

The reasons why overestimations persist de-spite repeated disconfirmation are explored.Typically, patients misattribute the absence ofdanger to external safety signals or safety be-haviors (e.g., “I only made it because I man-aged to find help in time,” “If I had not takenXanax last week when I panicked in the store,I’m sure I would have passed out” or “I would-n’t have made it if I hadn’t pulled off the roadin time”), or to “luck,” instead of realizing the


inaccuracy of the original prediction. Similarly,patients may assume that the only reason theyare still alive, sane, and safe, is because the “bigone” has not happened. In this case, patientserr by assuming that intensity of panic attacksincreases the risk of catastrophic outcomes.

The method for countering overestimationerrors is to question the evidence for probabil-ity judgments. The general format is to treatthoughts as hypotheses or guesses rather thanas facts and to examine the evidence and gener-ate alternative, more realistic predictions. Thisis best done by the therapist using a Socraticstyle, so that patients learn the skill of examin-ing the content of their statements and arrive atalternative statements or predictions after theyhave considered all of the evidence. Question-ing of the logic (e.g., “How does a racing heartlead to heart attack?”), or the bases fromwhich judgments are made (e.g., misinforma-tion from others, unusual sensations) is usefulin this regard. Continuing with the previous ex-ample from Julie, the questioning took the fol-lowing course:

THERAPIST: One of the specific thoughts youhave identified is that you will flip out andnever return to reality. What specificallyleads you to think that that is likely to hap-pen?

JULIE: Well, I guess it really feels like that.

THERAPIST: Describe the feelings?

JULIE: Well, I feel spacey and unreal, like thingsaround me are different and that I’m notconnected.

THERAPIST: And why do you think those feel-ings mean that you have actually lost touchwith reality?

JULIE: I don’t know—it feels as if I have.

THERAPIST: So, let’s examine that assumption.What is your behavior like when you feel un-real? For example, do you respond if some-one asks you a question during those epi-sodes?

JULIE: Well, I respond to you even though I feelthat way sometimes in here.

THERAPIST: OK, and can you walk or write ordrive when you feel that way?

JULIE: Yes, but it feels different.

THERAPIST: But you do perform those functionsdespite feeling detached. So, what does thattell you?

JULIE: Well, maybe I haven’t lost completetouch with reality. But what if I do?

THERAPIST: How many times have you felt de-tached?

JULIE: Hundreds and hundreds of times.

THERAPIST: And how many times have you losttouch with reality permanently?

JULIE: Never. But what if the feelings don’t goaway? Maybe I’ll lose it then?

THERAPIST: So what else tells you that this is apossibility?

JULIE: Well, what about my second cousin? Helost it when he was about 25, and now he’sjust a mess. He can hardly function at all,and he is constantly in and out of psychiatrichospitals. They have him on a bunch ofheavy-duty drugs. I’ll never forget the time Isaw him totally out of it. He was talking tohimself in jibberish.

THERAPIST: So, do you make a connection be-tween him and yourself?

JULIE: Yes.

THERAPIST: What are the similarities betweenthe two of you?

JULIE: There are none really. It’s just that he iswhat I think I will become.

THERAPIST: Did he ever feel the way you feelnow?

JULIE: I don’t know.

THERAPIST: And if another one of your cousinshad severe back problems, would you beconcerned that you would end up with se-vere back problems?

JULIE: No.

THERAPIST: Why not?

JULIE: Because it never crosses my mind. It isnot something that I worry about.

THERAPIST: So, it sounds like you think you willend up like your cousin because you areafraid of ending up like him.

JULIE: I suppose so.

THERAPIST: So, let’s look at all of the evidenceand consider some alternatives. You have feltunreal hundreds of times, and you’ve neverlost touch with reality, because you’ve con-tinued to function in the midst of those feel-ings, and they have never lasted. You areafraid of becoming like your cousin, butthere are no data to show that you and he


have the same problem. In fact, the data sug-gest otherwise, because you function and hedoes not. So what is the realistic probabilitythat you will lose touch with reality perma-nently? Use a scale of 0 to 100, where 0 = Nochance at all and 100 = Definitely will hap-pen.

JULIE: Well, maybe it is lower than I thought.Maybe 20%.

THERAPIST: So that would mean that you haveactually lost touch with reality in a perma-nent way once every five times you have feltunreal.

JULIE: When it’s put like that, I guess not.Maybe it’s a very small possibility.

THERAPIST: Yes, so what is an alternative expla-nation?

JULIE: Perhaps the feelings of unreality arecaused by feeling anxious or overbreathing,and having those feelings does not mean thatI am actually losing touch with reality, andthat I am not like my cousin at all.

For homework, in addition to continuationof self-monitoring and practice of diaphrag-matic breathing, Julie was asked to identify heranxious thoughts in relation to every item onher agoraphobia hierarchy, and to use the in-session steps of examining the evidence andgenerating alternative evidence based interpre-tations for errors of overestimating the risk.She was to do the same for every panic attackthat occurred over the next week.

Session 3

The goals of this session are to develop breath-ing retraining and to continue active cognitiverestructuring. The therapist reviews the pa-tient’s week of diaphragmatic breathing prac-tice. Julie was disappointed with her attemptsto practice.

JULIE: I just didn’t seem to be able to do it theright way. Sometimes I would start off OKand then the more I tried, the more it felt likeI was running out of air, and I’d have to takea big gulp between breaths. At other times, Ifelt dizzy and the unreal feelings would start,at which point I would stop and do “busywork” to keep my mind occupied.

THERAPIST: It sounds like quite a few thingswere going on. First of all, remember that

this is a skill, just like learning to ride a bike,and you cannot expect it to be easy from theget-go. Second, it sounds like you experi-enced some uncomfortable physical symp-toms that worried you. You said it felt likeyou were running out of air. Based on whatwe talked about last week, what do youthink might have caused that feeling?

JULIE: Well, maybe I wasn’t getting enough airinto my lungs, because it’s really hard for meto use my diaphragm muscle. I felt like I wassuffocating myself.

THERAPIST: Possibly it’s just a matter of learn-ing to use the diaphragm muscle, but wereyou really suffocating or was it an interpre-tation that you might be suffocating?

JULIE: I don’t know. I’ve had the feeling of suf-focating before, especially when I’m trappedin a crowded room.

THERAPIST: So, how do you know you weresuffocating?

JULIE: I don’t know. It just felt that way.

THERAPIST: So, let’s put the evidence together.You’ve had the feelings before and never suf-focated. As we discussed last time, anxietycan sometimes create a sensation of short-ness of breath even though you are gettingplenty of air. Can you think of an alternativeexplanation?

JULIE: Well, maybe I wasn’t suffocating. Maybeit just felt like that.

Julie’s complaints represent typical concernsthat should be addressed. The next step is toslow the rate of breathing until the patient cancomfortably span a full inhalation and exhala-tion cycle of 6 seconds. Again, the therapistmodels slowed breathing, then provides correc-tive feedback on practice in the session. The pa-tient is instructed to continue to practice slowbreathing in “safe” or relaxing environments,and is discouraged from applying slow breath-ing when anxious or panicking, until fullyskilled in its application.

Also, cognitive restructuring is continued byaddressing the second cognitive error, which in-volves viewing an event as “dangerous,” “in-sufferable,” or “catastrophic,” when in actual-ity it is not. Typical examples of catastrophicerrors are “If I faint, people will think that I’mweak and that would be unbearable” or “Panicattacks are the worst thing I can imagine,” and“The whole evening is ruined if I start to feel


anxious.” “Decatastrophizing” means to facethe worst, to realize that the occurrences arenot as “catastrophic” as stated, and to thinkabout actual ways to cope with negative eventsrather than how “bad” they are. A key princi-ple underlying decatastrophizing is that eventscan be endured even though they are uncom-fortable. Recognition of the time-limited na-ture of discomfort contributes to the develop-ment of a sense of being able to cope. Thecritical distinction here is that although pa-tients might prefer that these events not occur,they can tolerate the discomfort, if necessary.Thus, for the person who states that negativejudgments from others are unbearable, it is im-portant to discuss what he or she would do tocope should someone else make a direct nega-tive judgment. Similarly, for the person whostates that the physical symptoms of panic areintolerably embarrassing, the following type ofquestioning is helpful:

JULIE: I am really worried that I might lose con-trol and do something crazy, like yell andscream.

THERAPIST: Aside from the low likelihood ofthat happening (as we discussed before), letsface the worst and find out what is so badabout it. What would be so horrible aboutyelling and screaming?

JULIE: I could never live it down.

THERAPIST: Well, lets think it through. Whatare the various things you could do in the sit-uation? You have just yelled and screamed—now what?

JULIE: Well, I guess the yelling and screamingwould eventually stop.

THERAPIST: That’s right—at the very least youwould eventually exhaust yourself. Whatelse?

JULIE: Well, maybe I would explain to the peo-ple around me that I was having a really badday but that I would be OK. In other words,reassure them.

THERAPIST: Good. What else?

JULIE: Maybe I would just get away—findsomeplace to calm down and reassure myselfthat the worst is over.

THERAPIST: Good.

JULIE: But what if the police came and took meaway, locked me up in a mental ward?

THERAPIST: Again, lets face the worst. What if

the police did come when you were yellingand screaming, and what if the police didtake you away? As scary as that may soundto you, lets consider what actually wouldhappen.

JULIE: I have this image of myself not being ableto tell them what is really going on—that Iam so out of it I don’t have the ability to letthem know I am just anxious.

THERAPIST: If you were so distraught that youcould not clearly communicate, how longwould that last?

JULIE: You’re right. I would eventually exhaustmyself and then I could speak more clearly.But what if they didn’t believe me?

THERAPIST: What if they did not believe you atfirst? How long would it take before theywould realize that you were not crazy?

JULIE: I guess that after a while they would seethat I was OK, and maybe I could call afriend or my doctor to explain what was go-ing on.

THERAPIST: That’s right. Now remember, all ofthis is about events that are unlikely to hap-pen. At the same time, it is helpful to faceworst-case scenarios (even though unlikely)and realize that they are not as bad as youfirst thought.

The homework for this session, in additionto continued self-monitoring, is to practiceslow and diaphragmatic breathing in relaxingenvironments, and to identify errors of catas-trophizing in relation to each item on the ago-raphobia hierarchy, followed by practice ofdecatastrophizing and generation of ways tocope. In addition, Julie was to use the skill ofdecatastrophizing for panic attacks that oc-curred over the following week.

Session 4

The main goal of this session is to use breathingretraining skills as a coping tool, to review cog-nitive restructuring skills, and to begin in vivoexposure to the first item on the agoraphobiahierarchy.

Now that patients have practiced slow anddiaphragmatic breathing sufficiently in relax-ing environments, they are ready to use thesemethods in distracting environments and inanxious situations. Patients are encouraged touse breathing skills as a coping technique as


they face fear, anxiety, and anxiety-provokingsituations. Some patients use breathing skills asa safety signal or a safety behavior; in otherwords, they believe that they will be at risk forsome mental, physical, or social calamity ifthey do not breathe correctly. This issue cameup with Julie, as shown below.

JULIE: When I panicked during the week, I triedto use the breathing. It didn’t work. It mademe feel worse.

THERAPIST: It sounds as if you might have at-tempted to use the breathing exercise as adesperate attempt to control the feelings youwere experiencing.

JULIE: Yes, that’s right.

THERAPIST: What did you think would havehappened if you had not been able to controlthe feelings?

JULIE: I was really worried that I might not beable to handle the feelings.

THERAPIST: And if you weren’t able to handlethe feelings, what would happen?

JULIE: It just feels like I will lose it, permanently.

THERAPIST: So this is one of those thoughts thatwe were talking about last time. What doesyour evidence tell you about the likelihoodof losing touch with reality permanently?

JULIE: So you mean even if I don’t control mybreathing, then I will be OK?

THERAPIST: Well, you had not lost touch withreality permanently before you learned thebreathing exercise, so what does that tellyou?

JULIE: OK, I get it.

THERAPIST: The breathing exercise is bestthought of as a tool to help you face what-ever is provoking anxiety. So, as you face sit-uations and your anxiety increases, use thebreathing exercise first, then use your cogni-tive skills, so that you can continue to facerather than run away from anxiety.

Patients who consistently use the breathingskills as a safety behavior might be discouragedfrom using the breathing skills, so that theylearn that what they are most worried about ei-ther does not happen or it can be managedwithout using the breathing skills.

In terms of the cognitive restructuring, thera-pists give corrective feedback to patients on the

methods of questioning the evidence to gener-ate realistic probabilities, facing the worst, andgenerating ways of coping with each item onthe agoraphobia hierarchy and any panic at-tacks that occurred over the past week. Par-ticular “corrective” feedback is given whenpatients lack specificity in their cognitive re-structuring (e.g., patients who record that theyare most worried about panicking should beencouraged to detail what it is about panickingthat worries them) or rely on blanket reassur-ance (e.g., patients who record that “Every-thing will be OK” as their evidence and/orways of coping should be encouraged to list theevidence and/or generate actual coping steps).

Next, attention is given to how to practicethe first item on the agoraphobia hierarchy. Ifappropriate, reasons why previous attempts atin vivo exposure may have failed are reviewed.Typical reasons for patients’ past failures at invivo exposure include attempts that are toohaphazard and/or brief, or spaced too farapart, and attempts conducted without a senseof mastery, or while maintaining beliefs that ca-tastrophe is very possible. Julie had tried toface agoraphobic situations in the past, buteach time she had escaped, feeling over-whelmed by panic and terrified of losing touchwith reality permanently. The therapist helpedJulie realize how to approach the agoraphobicsituations differently to benefit from the expo-sure. Julie’s typical safety signals were the pres-ence of her husband, or at least knowing hiswhereabouts, and Klonopin (which she carriedbut rarely used). The therapist discussed theimportance of eventual weaning from thosesafety signals.

As mentioned earlier, the goal of exposuretherapy is not immediate reduction in fear andanxiety; rather the goal is for the patient tolearn something new as a result of exposure.Clarification of what patients are most worriedabout as they face their feared situations andthe conditions that best help patients to learnthat what they are most worried about never orrarely happens, and/or that they can cope withthe situation and tolerate anxiety is essentialfor effective exposure. If a patient is most wor-ried that fear and anxiety will remain elevatedfor the entire duration of the practice, then cor-rective learning involves toleration of sustainedanxiety. For Julie, the first situation on her hier-archy was to drive home from work, alone. Shestated that what most worried her in that situa-


tion was that she would panic and lose touchwith reality, therefore losing control of the carand dying in an accident. She also stated that todrive at dusk was the condition under whichshe was most convinced of these eventualities.Thus, the task that the therapist consideredmost effective in teaching Julie that she wouldnot lose touch with reality and have an acci-dent, or that she could cope with the sensationsof unreality and panic, was to drive home fromwork at dusk.

Delineation of the exposure task as con-cretely as possible, so that patients clearly un-derstand exactly what the practice entails (e.g.,“Walk around inside of mall for 10 minutes bymyself”), reduces uncertainty about whetherthe practice was conducted correctly. Withoutsuch concrete details, patients might decidethat they “failed.” Importantly, the practiceshould not be ended because of anxiety (e.g.,“Continue driving on the freeway until I feelanxious”) because the exposure practice wouldthen reinforce avoidance of anxiety.

Julie was reminded to use her coping skillsshould she panic as she practiced the task; thatis, in moments of fear, patients are encouragedto use their breathing and thinking skills tocomplete the assigned task; the coping skills arenot intended as means to reduce fear and anxi-ety, but to tolerate it.

Patients are encouraged to maintain a regu-lar schedule of repeated in vivo exposure prac-tices at least three times per week, and to con-duct these practices regardless of internal (e.g.,having a “bad day,” feeling ill) or external (e.g.,inclement weather, busy schedules) factors thatmay prompt postponement of practices. Julieexpressed some concerns about being able topractice at least three times over the followingweek:

JULIE: I don’t know if I can practice three times,because more days than not I feel prettyworn down; maybe I can practice on justMonday and Tuesday, because they are thedays I typically feel better.

THERAPIST: What is it you are worried abouthappening if you practice on a day when youalready feel worn down?

JULIE: I feel more fragile on those days.

THERAPIST: And if you feel more fragile, whatmight happen?

JULIE: I just don’t think I could do it. It wouldbe too hard. I might really freak out and losetouch with reality for ever.

THERAPIST: OK, so let’s think about thatthought. What does your experience tellyou? How many times have you perma-nently lost touch with reality, including dayswhen you were worn down?

JULIE: Well, never.

THERAPIST: So, what does that tell you?

JULIE: OK, but it still feels difficult to drive onthose days.

THERAPIST: How about you start with Mondayor Tuesday, but quickly move to the otherdays of the week when you are feeling worndown, so that you get a really good opportu-nity to learn whether you permanently losetouch with reality or not?

The homework for this session involves con-tinued self-monitoring, continued use of cogni-tive restructuring and breathing retraining inthe event of elevated anxiety or panic, andpracticing the first item on the agoraphobia hi-erarchy at least three times, with at least one ofthose times being without her husband Larry.

Session 5

The goals of this session are to review the prac-tice of in vivo exposure, to design another ex-posure task to be practiced over the next week,and to begin interoceptive exposure. Note thatin vivo and interoceptive exposure can be donesimultaneously or sequentially. For Julie, invivo exposure was begun in Session 4, whereasinteroceptive exposure was begun in this ses-sion, but they could easily have been done inthe opposite order.

It is essential to review the week’s practice ofin vivo exposure. An objective evaluation ofperformance is considered necessary to offsetsubjective and damaging self-evaluations. Asdemonstrated in experimental literature onlearning and conditioning, appraisals of aver-sive events after they have occurred can influ-ence anxiety about future encounters with thesame types of aversive events. Any practice thatis terminated prematurely is to be reviewedcarefully for contributing factors, which canthen be incorporated into subsequent trials ofin vivo exposure. Recognition of the precipi-


tant to escape is very important, because theurge to escape is usually based on the predic-tion that continued endurance would result insome kind of danger. For example, patientsmay predict that the sensations will become in-tense and lead to an out-of-control reaction.This prediction can be discussed in terms ofjumping to conclusions and blowing things outof proportion. At the same time, escape itselfneed not be viewed catastrophically (i.e., asembarrassing, or as a sign of failure). In addi-tion, therapists reinforce the use of breathingand cognitive skills to help patients remain inthe situation until the specified duration or taskhas been completed, despite uncomfortablesensations.

Again, it is important for patients to recog-nize that the goal is to repeatedly face situa-tions despite anxiety, not to achieve a total ab-sence of anxiety. Toleration of fear rather thanimmediate fear reduction is the goal for eachexposure practice; this approach leads to aneventual fear reduction. Anxiety that does notdecline over repeated days of in vivo exposuremay result from too much emphasis on imme-diate fear and anxiety reduction; that is, tryingtoo hard or wishing too much for anxiety todecline typically maintains anxiety.

Julie had success with her first in vivo expo-sure practice; she managed to drive home fromwork at dusk, alone, four different times. Shenoted that the first time was easier than she hadexpected; the second was harder, and the onetime she pulled off to the side of the road. Thetherapist helped Julie identify the thoughts andsensations that led her to “escape” from the sit-uation: the sensations of unreality and fears oflosing touch with reality. Julie had waited for afew minutes, then continued driving home—anaction that was highly reinforced by the thera-pist. The third and fourth times were easier.

Julie’s husband Larry attended Session 5, sothat he could learn how to help Julie overcomeher PDA. He was supportive and eager to helpin any way possible, expressing frustration athaving had no idea how to help in the past.

The general principles for involvement ofsignificant others in treatment are as follows.First, a treatment conceptualization is providedto the significant other to reduce his or herfrustration and/or negative attributions aboutthe patient’s emotional functioning (e.g., “Oh,she’s just making it up. There’s nothing reallywrong with her” or “He has been like this sincebefore we were married, and he’ll never

change”). The way in which the agoraphobicproblem has disrupted daily routines and dis-tribution of home responsibilities is exploredand discussed also. Examples might include so-cial activities, leisure activities, and householdchores. The therapist explains that family ac-tivities may be structured around the agora-phobic fear and avoidance to help the patientfunction without intense anxiety. At the sametime, reassignment of the patient’s tasks to thesignificant other may actually reinforce theagoraphobic pattern of behavior. Conse-quently, the importance of complying with invivo exposure homework instructions, eventhough the patient may experience some dis-tress initially, is emphasized.

The significant other is encouraged to be-come an active participant by providing his orher perception of the patient’s behavior andfearfulness, and the impact on the home envi-ronment. Sometimes significant others haveprovided information of which the patient wasnot fully aware, or did not report, particularlyin relation to how the patient’s behavior affectsthe significant other’s own daily functioning.Larry, for example, described how he felt re-stricted at home in the evenings; whereas, be-fore, he occasionally played basketball with hisfriends at the local gym, he now stays at home,because he feels guilty if he leaves Julie alone.

The next step is to describe the role of thesignificant other regarding in vivo exposuretasks. The significant other is viewed as acoach, and the couple is encouraged to ap-proach the tasks as a problem-solving team.This includes deciding exactly where and whento practice in vivo exposure. In preparation forpractices, the patient identifies his or hermisappraisals about the task and generatescognitive alternatives. The significant other isencouraged to help the patient question his orher own “anxious” thoughts. Role plays of thistype of questioning of the patient by the signifi-cant other may be conducted in the session, sothat the therapist can provide corrective feed-back to each partner. Throughout in vivo expo-sure, the significant other reminds the patientto apply cognitive challenges and/or breathingskills. Because the significant other is usually asafety signal, tasks are less anxiety provoking.However, the patient must be weaned from thesafety signal eventually. Therefore, initial at-tempts at facing agoraphobic situations areconducted with the significant other, and latertrials are conducted alone. Weaning from the


significant other may be graduated, as in thecase of (1) Julie driving first with Larry in thecar, (2) with him in a car behind, (3) meetingthe significant other at a destination point, and(4) driving alone.

Very important to the success of this collabo-ration is style of communication. On the onehand, significant others are discouraged frommagnifying the experience of panic and are en-couraged to help the patient apply coping state-ments when anxious. On the other hand, sig-nificant others are encouraged to be patientgiven the fact that progress for the patient maybe erratic. The patient and significant other areinstructed to use a 0- to 10-point rating scale tocommunicate with each other about the pa-tient’s current level of anxiety or distress, as away of diminishing the awkwardness associ-ated with discussion of anxiety, especially inpublic situations. The patient is warned aboutthe potential motivation to avoid discussing hisor her feelings with the significant other, due toembarrassment or an attempt to avoid the anx-iety for fear that such discussion and concen-tration on anxiety may intensify his or her dis-tress level. Avoidance of feelings is discouraged,because distraction is viewed as less beneficialin the long term than is objectively facing what-ever is distressing and learning that predictedcatastrophes do not occur. The patient is reas-sured that the initial discomfort and embar-rassment will most likely diminish as the cou-ple becomes more familiar with discussinganxiety levels and their management. Further-more, the patient’s concerns about the signifi-cant other being insensitive or too pushy areaddressed. For example, a significant othermay presume to know the patient’s level ofanxiety and anxious thoughts without confir-mation from the patient, or the significantother may become angry toward the patient foravoiding or escaping from situations, or beingfearful. All of these issues are described as rela-tively common and understandable patterns ofcommunication that are nevertheless in need ofcorrection. In-session role-playing of moreadaptive communication styles during episodesof heightened anxiety is a useful learning tech-nique. On occasion, more specific communica-tions training may be beneficial, especially ifthe partners frequently argue in their attemptsto generate items or methods for conducting invivo exposure.

The next in vivo exposure task for Julie wasto sit in a crowded movie theater, gradually

moving away from the aisle, toward the middleof the row, because that was the condition inwhich she was most concerned that she wouldlose control and draw attention to herself. Julieand Larry rehearsed their approach to the invivo exposure task in session, while the thera-pist provided corrective feedback using theprinciples of communication and coping de-scribed earlier. They were instructed to practicethis task at least three times over the next week.On at least one occasion, Julie was to practicethe task alone.

Next, interoceptive exposure was intro-duced. As with in vivo exposure, through re-peated exposures to feared sensations, patientslearn that they are not harmed by the sensa-tions, and they achieve increased confidence intheir ability to tolerate symptoms of anxiety.The procedure begins with assessment of thepatient’s response to a series of standardizedexercises. The therapist models each exercisefirst. Then, after the patient has completed theexercise, the therapist records the sensations,anxiety level (0 to 10), sensation intensity (0 to10), and similarity to naturally occurring panicsensations (0 to 10). The exercises includeshaking the head from side to side for 30 sec-onds; placing the head between the legs for 30seconds and lifting the head to an upright posi-tion quickly; running in place or using steps forl minute; holding one’s breath for as long aspossible; complete body muscle tension for 1minute or holding a push-up position for aslong as possible; spinning in a swivel chair for 1minute; hyperventilating for 1 minute; breath-ing through a narrow straw (with closed nasalpassages) or breathing as slowly as possible for2 minutes; and staring at a spot on the wall orat one’s mirror image for 90 seconds. If none ofthese exercises produce sensations at leastmoderately similar to those that occur natu-rally, other, individually tailored exercises aregenerated. For example, tightness around thechest may be induced by a deep breath beforehyperventilating; heat may be induced by wear-ing heavy clothing in a heated room; chokingsensations may be induced by a tongue depres-sor, high-collared sweater, or a necktie; andstartle may be induced by an abrupt, loud noisein the midst of relaxation. For Julie, the sensa-tions produces by hyperventilating, spinning,and staring at a spot on the wall were mostanxiety provoking.

Patients who report little or no fear becausethey feel safe in the presence of the therapist are


asked to attempt each exercise alone, eitherwith the therapist out of the office or at home.At the same time, discussing the influence ofperceived safety as a moderating factor in theamount of fear experienced reinforces the valueof cognitive restructuring. For a minority of pa-tients, the known cause and course of the sen-sations override the fear response; that is, be-cause the sensations are predictably related to aclear cause (the interoceptive exercise), and be-cause the sensations can be relatively easilycontrolled by simply terminating the intero-ceptive exercise, fear is minimal. Under theseconditions, discussion can productively centeron the misassumptions that render naturallyoccurring sensations more frightening than theones produced by the interoceptive exercises.Typically, these misassumptions are that natu-rally occurring sensations are unpredictable,that unpredictable sensations are more harm-ful, and that if naturally occurring sensationsare not controlled, then they pose a potentialthreat. The majority of patients fear at leastseveral of the interoceptive exercises despiteknowing the cause of the sensations and theircontrollability.

Interoceptive exercises rated as producing atleast somewhat similar sensations to naturallyoccurring panic (at least 3 on the 0- to 10-pointscale) are selected for repeated exposure. Agraduated approach is used for interoceptiveexposure, beginning with the lowest item onthe hierarchy established in Session 4. For eachtrial of exposure, the patient is asked to beginthe induction, to indicate when the sensationsare first experienced (e.g., by raising a hand),and to continue the induction for at least 30seconds longer to permit corrective learning.After terminating the induction, anxiety israted, and the patient is given time to applycognitive and breathing coping skills. Finally,the therapist reviews the induction experienceand the application of management strategieswith the patient. During this review, the thera-pist emphasizes the importance of experiencingthe sensations fully during the induction, ofconcentrating objectively on the sensations ver-sus distracting from them, and the importanceof identifying specific cognitions and challeng-ing them by considering all of the evidence. Inaddition, the therapist asks key questions tohelp the patient realize his or her safety (e.g.,“What would have happened if you had con-tinued spinning for another 60 seconds?”), andto generalize to naturally occurring experiences

(e.g., “How is this different from when you feeldizzy at work?”). In other words, cognitive re-structuring extends the cognitive reprocessingalready taking place implicitly as a result of re-peated interoceptive exposure.

Specific, previously unrecognized cognitionssometimes become apparent during repeatedexposure. For example, when Julie began toconduct repeated exposures to hyperventila-tion and spinning, she became more aware ofher implicit assumption that sensations ofspaciness or lightheadedness would lead her tolose control of her limbs. This related to herconcern about causing an accident when driv-ing. During repeated hyperventilation exer-cises, and with prompting of “what ifs” fromthe therapist, Julie discovered her fear of notbeing able to move her arms or legs. The thera-pist then behaviorally challenged this assump-tion by having Julie overbreathe for longerperiods of time, followed immediately by walk-ing, picking up objects, and so on.

Homework practice is very important, be-cause safety signals present in the clinic settingor that derive from the therapist per se may,again, prevent generalizability to the naturalsetting. Patients are instructed to practice theinteroceptive items conducted in session on adaily basis, three times each day. Julie was topractice hyperventilation over the followingweek. She expressed some concern at doing theexercises alone, so the therapist helped Julie touse her cognitive restructuring skills in relationto being alone. In addition, more graduation ofhomework was suggested, so that Julie wouldpractice hyperventilating when her husbandwas at home the first couple of days, then whenhe was not at home the rest of the time.

Sessions 6 and 7

The primary goal of these sessions is to reviewthe past week of in vivo exposure practices, de-sign new exposures, review between-sessionpractices of interoceptive exposure, conduct re-peated interoceptive exposure in session, andassign those as homework for the next week.

The in vivo exposure is reviewed, as in theprevious session. In this case, Julie and Larryhad done well with the movie theater practice.Julie even practiced going to the movies on herown. On that occasion she reported higheranxiety than when she was with Larry for fearof having to get up and leave the theater andworries about bothering others in the audience.


The therapist helped Julie to identify whatworry led her to think about leaving in the firstplace; in other words, what did she think mighthappen if she could not leave? Julie indicatedthat she had thoughts of losing control andcausing a scene, to which she was thenprompted to apply her cognitive restructuringskills of evidence-based analyses and decatas-trophizing. She was ready to move to the nextitems on her hierarchy: to spend 2 hours aloneat home during the day and to stay alone athome as day turned to night. As with every invivo exposure task, Julie identified what shemost feared happening in those situations, andthe best conditions under which to practice tolearn that either those eventualities would nothappen and/or that she could cope with theworst.

The past week of interoceptive exposurepractice is reviewed in session with a mind to-ward avoidance: either overt failure to practice,or covert avoidance by minimizing the intensityor duration of the sensations induced, or limit-ing practice to the presence of a safety signal(e.g., a significant other) or times when back-ground anxiety is minimal. Reasons for avoid-ance may include continued misinterpretationof the dangers of bodily sensations (i.e.,“I don’t want to hyperventilate, because I’mafraid that I won’t be able to stop over-breathing and no one will be there to help me”)or the belief that anxiety will not reduce withrepetition of the task.

For the first week, Julie practiced intero-ceptive exposure exercises about half of thedays between sessions. The therapist used a“downward arrow” method to explore Julie’sreasons for not practicing every day.

JULIE: I tried hyperventilating on my own.However, I wasn’t very successful, because Ifelt too scared and I stopped it as soon as Inoticed the strange feelings.

THERAPIST: What did you think would happenif the sensations became more intense?

JULIE: I thought the feelings would get worseand worse and worse, and just overwhelmme. I didn’t want to have that feeling ofpanic again.

THERAPIST: If you did become overwhelmed,then what would happen to you?

JULIE: Then I’d feel really terrible.

THERAPIST: And if you felt really terrible?

JULIE: Well, nothing. I’d just feel terrible.

THERAPIST: The word “terrible” carries a lot ofmeaning. Let’s see if we can pin down youranxious thoughts that make the feelings soterrible.

JULIE: I just can’t tolerate the feeling.

THERAPIST: What tells you that you cannot tol-erate it? How do you know you can’t toler-ate it?

And the discussion continued, so that Julierealized what was most important for her tolearn by the repeated hyperventilation: Shecould tolerate the sensations and anxiety.

However, after the subsequent week of re-peated practice, Julie remained cautious forfear that the exercises would cause her to revertto her state of several weeks earlier; that is, shewas concerned that the inductions would leaveher in a persistent symptomatic state. Further-more, she was particularly reluctant to practiceinteroceptive exposure at the end of the day,when she was more likely to feel unreal, or on aday when an important social event was sched-uled. Again, these avoidance patterns were re-lated to fears that the symptoms would becometoo intense or result in some type of mental orsocial catastrophe. These types of avoidancepatterns are addressed in the following vi-gnette:

THERAPIST: When did you practice deliberatelyspinning and hyperventilating?

JULIE: Usually in the mornings. One day I left ituntil the end of the day, and that turned outto be a bad idea. I felt terrible.

THERAPIST: Let’s think about that a bit more.What made it terrible when you practiced atthe end of the day?

JULIE: Well, I was already feeling prettyunreal—I usually do around that time of theday. So I was much more anxious about thesymptoms.

THERAPIST: Being more anxious implies thatyou thought the symptoms were more harm-ful. Is that what happened on the day thatyou practiced interoceptive exposure whenyou were already feeling unreal?

JULIE: Yes, I felt that because I was already feel-ing unreal, I was on the edge, and that Imight push myself over the edge if I tried toincrease the feelings of unreality.


THERAPIST: What do you mean by “push my-self over the edge”?

JULIE: That I would make the feelings so intensethat I really would lose it—go crazy.

THERAPIST: So there is one of those hypotheses:to feel more intense unreality means to becloser to going crazy. Let’s examine the evi-dence. Is it necessarily the case that more in-tense unreality means you are closer to crazi-ness?

In sessions, the therapist continued practiceof interoceptive exposure with the next item onJulie’s hierarchy, which was to stare at a spoton the wall and to spin around.

The homework from this session is to con-tinue self-monitoring, in vivo exposure to anitem from the agoraphobia hierarchy at leastthree times, and daily practice of interoceptiveexposure.

Sessions 8 and 9

The primary goals of these sessions are to con-tinue in vivo exposure, as described in the priorsessions, and to extend interoceptive exposureto natural activities. Julie had practiced stayingat home for 2 hours alone during the day andas daylight turned to dusk, with good results.In particular, she experienced a couple of panicattacks during these in vivo exposure practicesbut continued with the assigned practice re-gardless. This was critical for Julie, as it al-lowed her to learn that she could survive thefeeling of panic; it was the first time she had re-mained in a situation despite panicking.

In reviewing the week’s practice of intero-ceptive exposure, it became apparent that Juliewas separating the practices from real-life ex-periences of bodily sensations in a way thatwould limit generalization. This was addressedas follows:

JULIE: After spinning and hyperventilating sev-eral times, I really do feel much less anxious.I was terrified at the start, but now I am onlymildly anxious, if at all. But this is differentthan what happens to me when I’m on thefreeway or at home.

THERAPIST: How is it different?

JULIE: I don’t know when the feelings of dizzi-ness and unreality are going to hit.

THERAPIST: From our previous discussions, let’sthink of potential reasons why you mightfeel dizzy or unreal at a particular time?

JULIE: I know. I have to keep remembering thatit could be my breathing, or just feeling anx-ious, or tired, or a bunch of different things.

THERAPIST: OK. And why is it so important toknow when those feelings will occur?

JULIE: Because I don’t want them to be there atall.

THERAPIST: And why not . . . what are youafraid of?

JULIE: I guess it’s the same old thing . . . that I’lllose it somehow?

THERAPIST: So let’s go back to the cognitive re-structuring that you have been doing. Whatspecifically are you afraid of? How likely is itto happen? What are the alternatives?

JULIE: I understand.

THERAPIST: So, now you see that whether thesensations of dizziness or unreality are pro-duced by anxiety, overbreathing, diet, or theexercises we do here, they’re all the same—they are just uncomfortable physical sensa-tions. The only reason they perturb youmore when you are driving or at home is be-cause of the meaning you still give to them inthose situations.

“Naturalistic” interoceptive exposure refersto exposure to daily tasks or activities that havebeen avoided or endured with dread because ofthe associated sensations. Typical examples in-clude aerobic exercise or vigorous physical ac-tivity, running up flights of stairs, eating foodsthat create a sensation of fullness or are associ-ated with sensations of choking, saunas orsteamy showers, driving with the windowsrolled up and the heater on, caffeine consump-tion, and so on. (Of course, these exercises maybe modified in the event of actual medical com-plications, such as asthma or high blood pres-sure.) From a list of typically feared activitiesand generation of items specific to the individ-ual’s own experience, a hierarchy is established.Each item is ranked in terms of anxiety ratings(0–10). Julie’s hierarchy was as follows: look-ing out through venetian blinds (anxiety = 3);watching One Flew over the Cuckoo’s Nest(anxiety = 4); playing tennis (anxiety = 4); scan-ning labels on a supermarket shelf (anxiety =


5); concentrating on needlework for an hour(anxiety = 6); driving with windows closed andheater on (anxiety = 7); a nightclub with strobelights (anxiety = 8); and rides at Disneyland(anxiety = 10).

Like the symptom exercises, the activity exer-cises are designed to be systematically graduatedand repetitive. Patients may apply the breathingand cognitive skills while the activity is ongoing.This is in contrast to the symptom induction ex-ercises, in which coping skills are used only aftercompletion of the symptom exercise, becausethe activities often are considerably longer thanthe symptom induction exercises. Nevertheless,patients are encouraged to focus on the sensa-tions and experience them fully throughout theactivity, and not use the coping skills to preventor remove the sensations.

Patients are instructed to identify maladap-tive cognitions and rehearse cognitive restruc-turing before beginning each activity. In-sessionrehearsal of the cognitive preparation allowstherapists to provide corrective feedback. Juliedid this with her therapist for her first two nat-uralistic activities, which were to look atvenetian blinds and to watch One Flew overthe Cuckoo’s Nest. Julie realized that she wasmost worried about sensations of unreality andfears of going crazy, although, as a result of hervarious exposure exercises up to this point, shequickly was able to recognize that such sensa-tions were harmless and that she could toleratethem, and that such fears were unrealisticbased on the evidence.

As with all exposures, it is important to iden-tify and remove (gradually, if necessary) safetysignals or protective behaviors, such as porta-ble phones, lucky charms, walking slowly,standing slowly, and staying in close proximityto medical facilities. These safety signals andbehaviors reinforce catastrophic misappraisalsabout bodily sensations. Julie’s safety behaviorswere identified as checking the time on theclock (as a reassurance that she was in touchwith reality) and pinching herself (again, to feelreality). She was asked to practice the two nat-uralistic interoceptive exposures at least threetimes each before the next treatment session,without the safety behaviors.

Sessions 10 and 11

The primary goals of these sessions are to re-view the in vivo and naturalistic exposure exer-

cises over the past week, and to combine exposure to feared and avoided agoraphobic situ-ations with deliberate induction of feared sen-sations into those situations. As with earlierinteroceptive exposure homework assign-ments, it is important to evaluate and correcttendencies to avoid naturalistic interoceptiveexposure tasks, mainly by considering the un-derlying misassumptions that are leading toavoidance. Remember also that a form ofavoidance is to rely on safety signals or safetybehaviors, so careful questioning of the way inwhich the naturalistic exposure was conducted,and under what conditions, may help to iden-tify inadvertent reliance on these unnecessaryprecautions. Julie reported that she was suc-cessful in looking at the venetian blinds, eventhough she experienced sensations of unreality.She had more difficulty watching One Flewover the Cuckoo’s Nest, because it tapped di-rectly into her worst fears of losing touch withreality permanently; she tried but terminatedthe film early. The second time she watched itwith Larry, who prompted Julie to rememberher cognitive and breathing skills, and she wasable to watch the entire film. She watched thefilm one more time on her own. Two new natu-ralistic exposure items were selected for thecoming week, with special attention to wean-ing or removing safety signals and safety be-haviors, and rehearsal of cognitive restructur-ing in session. For Julie, these were playingtennis (something she had avoided for years)and scanning items on supermarket shelves.

The notion of deliberately inducing fearedbodily symptoms within the context of fearedagoraphobic situations derives from the evi-dence that compound relationships betweenexternal and internal cues can be the most po-tent anxiogenic agent; that is, it is neither justthe situation nor just the bodily sensation thattriggers distress, but the combination of thebodily sensation and the situation that is mostdistressing. Thus, effective exposure targetsboth types of cues. Otherwise, patients run therisk of later return of fear. For example, re-peated practice walking through a shoppingmall without feeling dizzy does not adequatelyprepare patients for occasions on which theyfeel dizzy walking through a shopping mall,and without such preparation, patients may belikely to panic and escape should they feel dizzyin this or similar situations in the future.Wearing heavy clothing in a restaurant helps


patients to learn to be less afraid of not only therestaurant but also of feeling hot in a restau-rant. Other examples include drinking coffeebefore any of the agoraphobic tasks, turningoff the air-conditioning or turning on theheater while driving, breathing very slowly in acrowded area, and so on.

Patients choose an item from their hierarchyof agoraphobia situations, either one alreadycompleted or a new item, and also choosewhich symptom to induce and ways of induc-ing that symptom in that situation. Julie’s taskwas to drink coffee as she went to a movie. Sheexpressed the following concerns:

JULIE: Do you really think I am ready to drinkcoffee and go to the movie?

THERAPIST: What worries you about the combi-nation of coffee and the movie theater?

JULIE: Well, I’ve practiced in the movie theatersa lot, so that feels pretty good, but the coffeeis going to make me feel very anxious.

THERAPIST: And if you feel very anxious in themovie theater, then what?

JULIE: Then, I don’t know what. Maybe I willget those old feelings again, like I have to getout.

THERAPIST: Based on everything you havelearned, how can you manage those feelings?

JULIE: Well, I guess my number one rule is neverto leave a situation because I am feeling anx-ious. I will stick it out, no matter what.

THERAPIST: That sounds great. It means you areaccepting the anxiety and taking the oppor-tunity to learn that you can tolerate it. Whatelse?

JULIE: I can ask myself what is the worst thatcan happen. I know I am not going to die orgo crazy. I will probably feel my heart rategoing pretty fast because of the coffee.

THERAPIST: And if your heart rate goes fast,what does that mean?

JULIE: I guess it just means that my heart ratewill go fast.

THERAPIST: This will be a really good way foryou to learn that you can tolerate the anxietyand the symptoms of a racing heart.

The homework for this session is to continueself-monitoring, to practice in vivo exposurecombined with interoceptive exposure, and tocontinue naturalistic interoceptive exposure.

Session 12

The last treatment session reviews the princi-ples and skills learned and provides the patientwith a template of coping techniques for poten-tial, high-risk situations in the future. Julie fin-ished the program after 12 sessions, by whichtime she had not panicked in 8 weeks, rarelyexperienced dizziness or feelings of unreality,and was driving further distances. There weresome situations still in need of exposure prac-tices (e.g., driving very long distances awayfrom home and on the freeway at dusk). How-ever, Julie and Larry agreed to continue in vivoexposure practices over the next few months toconsolidate her learning and to continue herimprovement.

CONCLUSION

As noted earlier in this chapter, cognitive-behavioral treatments for panic disorder andagoraphobia are highly effective and representone of the success stories of psychotherapy. Be-tween 80 and 100% of patients undergoingthese treatments will be panic free at the end oftreatment and maintain these gains for up to 2years. These results reflect substantially moredurability than medication treatments. Fur-thermore, between 50 and 80% of these pa-tients reach a point of “high end state,” mean-ing within normative realms of symptoms andfunctioning, and many of the remainder haveonly residual symptomatology. Nevertheless,major difficulties remain.

First, these treatments are not foolproof. Asmany as 50% of patients retain substantialsymptomatology despite improvement frombaseline, and this is particularly likely for thosewith more severe agoraphobia. Further re-search must determine how treatments can beimproved or better individualized to alleviatecontinued suffering. For example, one of us(D. H. B.) saw a patient several years ago whohad completed an initial course of treatmentbut required continued periodic visits for over4 years. This patient was essentially improvedfor approximately 9 months but found himselfrelapsing during a particularly stressful time atwork. A few booster sessions restored his func-tioning, but he was back in the office 6 monthslater with reemerging symptomatology. Thispattern essentially continued for 4 years andwas characterized by symptom-free periods fol-


lowed by (seemingly) stress-related relapses.Furthermore, the reemerging panic disorderwould sometimes last from 3 to 6 months be-fore disappearing again, perhaps with the helpof a booster session.

Although this case was somewhat unusual inour experience, there was no easy explanationfor this pattern of relapses and remissions. Thepatient, who has a graduate degree, understoodand accepted the treatment model and fully im-plemented the treatment program. There wasalso no question that he fully comprehendedthe nature of anxiety and panic, and the intri-cacies of the therapeutic strategies. While in theoffice, he could recite chapter and verse on thenature of these emotional states, as well as thedetailed process of his own reaction while inthese states. Nevertheless, away from the of-fice, the patient found himself repeatedly hop-ing that he would not “go over the brink” dur-ing a panic, despite verbalizing very clearly theirrationality of this concept while in the office.In addition, he continued to attempt to reduceminor physiological symptoms associated withanxiety and panic, despite a full rational under-standing of the nature of these symptoms (in-cluding the fact that they are the same symp-toms that he experienced during a state ofexcitement, which he enjoyed). His limited tol-erance of these physical sensations was alsopuzzling in view of his tremendous capacity toendure pain.

Any number of factors might account forwhat seemed to be “overvalued ideation” orvery strongly held irrational ideas during peri-ods of anxiety, including the fact that the pa-tient has several relatives who have repeatedlybeen hospitalized for emotional disorders(seemingly mood disorders or schizoaffectivedisorder). Nevertheless, the fact remains thatwe do not know why this patient did not re-spond as quickly as most people. Eventually hemade a full recovery, received several promo-tions at work, and considered treatment to bethe turning point in his life. But it took 5 years.

Other patients, as noted earlier, seem unin-terested in engaging in treatment, preferring toconceptualize their problems as chemical im-balances. Still others have difficulty graspingsome of the cognitive strategies, and further at-tempts are necessary to make these treatmentsmore “user-friendly.”

It also may seem that this structured,protocol-driven treatment is applied in a verystandard fashion across individuals. Nothing

could be further from the truth. The clinical artinvolved in this, and in all treatments describedin this book, requires a careful adaptation tothese treatment strategies to the individualcase. Many of Julie’s symptoms revolvedaround feelings of unreality (derealization anddepersonalization). Emphasizing rational ex-planations for the production of such feelings,as well as adapting cognitive and exposure ex-ercises to maximize these sensations, is an im-portant part of this treatment program. Al-though standard interoceptive provocationexercises seemed sufficient to produce relevantsymptomatology in Julie’s case, we have had todevelop new procedures to deal with peoplewith more idiosyncratic symptoms and fears,particularly those involving feelings of unreal-ity or dissociation. Other innovations in bothcognitive and behavioral procedures will be re-quired by individual therapists as they applythese procedures.

Although these new treatments seem highlysuccessful when applied by trained therapists,treatment is not readily available to individualswith these disorders. In fact, these treatments,although brief and structured, are far more dif-ficult to deliver than, for example, pharmaco-logical treatments (which are also often misap-plied). Furthermore, few people are currentlyskilled in the application of these treatments.What seems to be needed for these and othersuccessful psychosocial treatments is a newmethod of disseminating them, so that theyreach the maximum number of patients. Modi-fication of these treatment protocols into moreuser-friendly formats, as well as brief periods oftraining for qualified therapists to a point ofcertification, would be important steps in suc-cessfully delivering these treatments. This maybe difficult to accomplish.

NOTE

1. Specific phobias were not assessed, but by beingmost circumscribed, they would be hypothesized toload the least on negative affectivity.

REFERENCES

Allen, L. B., & Barlow, D. H. (2006). The treatment ofpanic disorder: Outcomes and basic processes. In B.O. Rothbaum (Ed.), Pathological anxiety: Emotionalprocessing in etiology and treatment (pp. 166–180).New York: Guilford Press.


Alneas, R., & Torgersen, S. (1990). DSM-III personalitydisorders among patients with major depression,anxiety disorders, and mixed conditions. Journal ofNervous and Mental Disease, 178, 693–698.

American Psychiatric Association. (1980). Diagnosticand statistical manual of mental disorders (3rd ed.).Washington, DC: Author.

American Psychiatric Association. (1994). Diagnosticand statistical manual of mental disorders (4th ed.).Washington, DC: Author.

American Psychiatric Association. (2000). Diagnosticand statistical manual of mental disorders (4th ed.,text rev.). Washington, DC: Author.

Amering, M., Katschnig, H., Berger, P., Windhaber, J.,Baischer, W., & Dantendorfer, K. (1997). Embarrass-ment about the first panic attack predicts agorapho-bia in disorder patients. Behaviour Research andTherapy, 35, 517–521.

Antony, M. M., Brown, T. A., Craske, M. G., Barlow, D.H., Mitchell, W. B., & Meadows, E. A. (1995). Accu-racy of heartbeat perception in panic disorder, socialphobia, and nonanxious subjects. Journal of AnxietyDisorders, 9, 355–371.

Antony, M. M., Ledley, D. B., Liss, A., & Swinson, R. P.(2006). Responses to symptom induction exercises inpanic disorder. Behaviour Research and Therapy, 44,85–98.

Antony, M. M., Meadows, E. A., Brown, T. A., &Barlow, D. H. (1994). Cardiac awareness before andafter cognitive-behavioral treatment for panic disor-der. Journal of Anxiety Disorders, 8, 341–350.

Arnow, B. A., Taylor, C. B., Agras, W. S., & Telch, M. J.(1985). Enhancing agoraphobia treatment outcomeby changing couple communication patterns. Behav-ior Therapy, 16, 452–467.

Arntz, A., & van den Hout, M. (1996). Psychologicaltreatments of panic disorder without agoraphobia:Cognitive therapy versus applied relaxation. Behav-iour Research and Therapy, 34, 113–121.

Arrindell, W., & Emmelkamp, P. (1987). Psychologicalstates and traits in female agoraphobics: A controlledstudy. Journal of Psychopathology and BehavioralAssessment, 9, 237–253.

Azrin, N., Naster, B., & Jones, R. (1973). Reciprocitycounselling: A rapid learning-based procedure formarital counselling. Behaviour Research and Ther-apy, 11, 365–382.

Bandelow, B., Spath, C., Tichaner, G. A., Brooks, A.,Hajak, G., & Ruther, E. (2002). Early traumatic lifeevents, parental attitudes, family history, and birthrisk factors in patients with panic disorder. Compre-hensive Psychiatry, 43, 269–278.

Bandura, A. (1977). Self-efficacy: Toward a unifyingtheory of behavioral change. Psychological Review,84, 191–215.

Barlow, D. H. (1988). Anxiety and its disorders: The na-ture and treatment of anxiety and panic. New York:Guilford Press.

Barlow, D. H. (2002). Anxiety and its disorders: The na-

ture and treatment of anxiety and panic (2nd ed.).New York: Guilford Press.

Barlow, D. H., Brown, T. A., & Craske, M. G. (1994).Definitions of panic attacks and panic disorder in theDSM-IV: Implications for research. Journal of Ab-normal Psychology, 103, 553–564.

Barlow, D. H., Cohen, A., Waddell, M., Vermilyea, J.,Klosko, J., Blanchard, E., et al. (1984). Panic andgeneralized anxiety disorders: Nature and treatment.Behavior Therapy, 15, 431–449.

Barlow, D. H., & Craske, M. G. (2006). Mastery ofyour anxiety and panic: Patient workbook (4th ed.).New York: Oxford University Press.

Barlow, D. H., Craske, M. G., Cerny, J. A., & Klosko, J.S. (1989). Behavioral treatment of panic disorder.Behavior Therapy, 20, 261–282.

Barlow, D. H., Gorman, J. M., Shear, M. K., & Woods,S. W. (2000). Cognitive-behavioral therapy, imipra-mine, or their combination for panic disorder: A ran-domized controlled trial. Journal of the AmericanMedical Association, 283(19), 2529–2536.

Barlow, D. H., O’Brien, G. T., & Last, C. G. (1984).Couples treatment of agoraphobia. Behavior Ther-apy, 15(1), 41–58.

Barlow, D. H., O’Brien, G. T., Last, C. G., & Holden, A.E. (1983). Couples treatment of agoraphobia. In K.D. Craig & R. J. McMahon (Eds.), Advances in clini-cal behavior therapy (pp. 99–127). New York: Brun-ner/Mazel.

Barlow, D. H., Vermilyea, J., Blanchard, E., Vermilyea,B., Di Nardo, P., & Cerny, J. (1985). Phenomenon ofpanic. Journal of Abnormal Psychology, 94, 320–328.

Barsky, A. J., Cleary, P. D., Sarnie, M. K., & Ruskin, J.N. (1994). Panic disorder, palpitations, and theawareness of cardiac activity. Journal of Nervous andMental Disease, 182, 63–71.

Basoglu, M., Marks, I. M., Kilic, C., Brewin, C. R., &Swinson, R. P. (1994). Alprazolam and exposure forpanic disorder with agoraphobia: Attribution of im-provement to medication predicts subsequent re-lapse. British Journal of Psychiatry, 164, 652–659.

Beck, A. T., Epstein, N., Brown, G., & Steer, R. A.(1988). An inventory for measuring clinical anxiety:Psychometric properties. Journal of Consulting andClinical Psychology, 56, 893–897.

Beck, J. G., & Shipherd, J. C. (1997). Repeated expo-sure to interoceptive cues: Does habituation of fearoccur in panic disorder patients?: A preliminary re-port. Behaviour Research and Therapy, 35, 551–557.

Beck, J. G., Shipherd, J. C., & Zebb, B. J. (1997). Howdoes interoceptive exposure for panic disorderwork?: An uncontrolled case study. Journal of Anxi-ety Disorders, 11, 541–556.

Beck, J. G., Stanley, M. A., Baldwin, L. E., Deagle, E. A.,& Averill, P. M. (1994). Comparison of cognitivetherapy and relaxation training for panic disorder.Journal of Consulting and Clinical Psychology, 62,818–826.


Biederman, J., Faraone, S. V., Marrs, A., & Moore, P.(1997). Panic disorder and agoraphobia in consecu-tively referred children and adolescents. Journal ofthe American Academy of Child and Adolescent Psy-chiatry, 36(12), 214–223.

Bjork, R. A., & Bjork, E. L. (1992). A new theory of dis-use and an old theory of stimulus fluctuation. In A.Healy, S. Kosslyn, & R. Shiffrin (Eds.), From learningprocesses to cognitive processes: Essays in honor ofWilliam K. Estes (Vol. 2, pp. 35–67). Hillsdale, NJ:Erlbaum.

Black, D. W., Monahan, P., Wesner, R., Gabel, J., &Bowers, W. (1996). The effect of fluvoxamine, cogni-tive therapy, and placebo on abnormal personalitytraits in 44 patients with panic disorder. Journal ofPersonality Disorders, 10, 185–194.

Bland, K., & Hallam, R. (1981). Relationship betweenresponse to graded exposure and marital satisfactionin agoraphobics. Behaviour Research and Therapy,19, 335–338.

Bonn, J. A., Harrison, J., & Rees, W. (1971). Lactate-induced anxiety: Therapeutic application. BritishJournal of Psychiatry, 119, 468–470.

Borkovec, T., Weerts, T., & Bernstein, D. (1977). As-sessment of anxiety. In A. Ciminero, K. Calhoun, &H. Adams (Eds.), Handbook of behavioral assess-ment (pp. 367–428). New York: Wiley.

Bouchard, S., Gauthier, J., Laberge, B., French, D.,Pelletier, M., & Godbout, D. (1996). Exposure ver-sus cognitive restructuring in the treatment of panicdisorder with agoraphobia. Behaviour Research andTherapy, 34, 213–224.

Bouchard, S., Paquin, B., Payeur, R., Allard, M., Rivard,V., Gournier, T., et al. (2004). Delivering cognitive-behavior therapy for panic disorder with agorapho-bia in videoconference [Special issue: Telemedicine inCanada]. Telemedicine Journal and E-Health, 10(1),13–24.

Bouton, M. E. (1993). Context, time and memory re-trieval in the interference paradigms of Pavlovianlearning. Psychological Bulletin, 114, 90–99.

Bouton, M. E., Mineka, S., & Barlow, D. H. (2001). Amodern learning-theory perspective on the etiologyof panic disorder. Psychological Review, 108(1), 4–32.

Bouton, M. E., & Swartzentruber, D. (1991). Sources ofrelapse after extinction in Pavlovian conditioningand instrumental conditioning. Behavioral Neurosci-ence, 104, 44–55.

Boyd, J. H. (1986). Use of mental health services for thetreatment of panic disorder. American Journal ofPsychiatry, 143, 1569–1574.

Broocks, A., Bandelow, B., Pekrun, G., George, A.,Meyer, T., Bartmann, U., et al. (1998). Comparisonof aerobic exercise, clomipramine, and placebo in thetreatment of panic disorder. American Journal ofPsychiatry, 155, 603–609.

Brown, T. A., Antony, M. M., & Barlow, D. H. (1995).Diagnostic comorbidity in panic disorder: Effect on

treatment outcome and course of comorbid diagno-ses following treatment. Journal of Consulting andClinical Psychology, 63, 408–418.

Brown, T. A., & Barlow, D. H. (1995). Long-term out-come in cognitive-behavioral treatment of panic dis-order: Clinical predictors and alternative strategiesfor assessment. Journal of Consulting and ClinicalPsychology, 63, 754–765.

Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham,J. R., & Mancill, R. B. (2001). Current and lifetimecomorbidity of the DSM-IV anxiety and mood disor-ders in a large clinical sample. Journal of AbnormalPsychology, 110(4), 585–599.

Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998).Structural relationships among dimensions of theDSM-IV anxiety and mood disorders and dimensionsof negative affect, positive affect, and autonomicarousal. Journal of Abnormal Psychology, 107(2),179–192.

Brown, T. A., Di Nardo, P. A., Lehman, C. L., & Camp-bell, L. A. (2001). Reliability of DSM-IV anxiety andmood disorders: Implications for the classification ofemotional disorders. Journal of Abnormal Psychol-ogy, 110(1), 49–58.

Brown, T. A., White, K. S., Forsyth, J. P., & Barlow, D.H. (2004). The structure of perceived emotional con-trol: Psychometric properties of a revised AnxietyControl Questionnaire. Behavior Therapy, 35(1),75–99.

Buglass, P., Clarke, J., Henderson, A., & Presley, A.(1977). A study of agoraphobic housewives. Psycho-logical Medicine, 7, 73–86.

Cain, C. K., Blouin, A. M., & Barad, M. (2004).Adrenergic transmission facilitates extinction of con-ditional fear in mice. Learning and Memory, 11(2),179–187.

Carlbring, P., Ekselius, L., & Andersson, G. (2003).Treatment of panic disorder via the Internet: A ran-domized trial of CBT vs. applied relaxation. Journalof Behavior Therapy and Experimental Psychiatry,34, 129–140.

Carter, M. M., Sbrocco, T., Gore, K. L., Marin, N. W.,& Lewis, E. L. (2003). Cognitive-behavioral grouptherapy versus a wait-list control in the treatment ofAfrican American women with panic disorder. Cog-nitive Therapy and Research, 27(5), 505–518.

Cerny, J. A., Barlow, D. H., Craske, M. G., & Himadi,W. G. (1987). Couples treatment of agoraphobia: Atwo-year follow-up. Behavior Therapy, 18, 401–415.

Chambless, D. L. (1990). Spacing of exposure sessionsin treatment of agoraphobia and simple phobia.Behavior Therapy, 21, 217–229.

Chambless, D. L., Caputo, G., Bright, P., & Gallagher,R. (1984). Assessment of fear in agoraphobics: TheBody Sensations Questionnaire and the AgoraphobicCognitions Questionnaire. Journal of Consulting andClinical Psychology, 52, 1090–1097.

Chambless, D. L., Caputo, G., Gracely, S., Jasin, E., &Williams, C. (1985). The Mobility Inventory for Ag-


oraphobia. Behaviour Research and Therapy, 23,35–44.

Chambless, D. L., & Renneberg, B. (1988, September).Personality disorders of agoraphobics. Paper pre-sented at World Congress of Behavior Therapy, Edin-burgh, Scotland.

Chaplin, E. W., & Levine, B. A. (1981). The effects oftotal exposure duration and interrupted versus con-tinuous exposure in flooding therapy. Behavior Ther-apy, 12(3), 360–368.

Clark D. A. (1996). Panic disorder: From theory to ther-apy. In P. M. Salkovskis (Ed.), From frontiers of cog-nitive therapy: The state of the art and beyond (pp.318–344). New York: Guilford Press.

Clark, D. M. (1986). A cognitive approach to panic. Be-haviour Research and Therapy, 24, 461–470.

Clark, D. M., & Ehlers, A. (1993). An overview of thecognitive theory and treatment of panic disorder. Ap-plied and Preventive Psychology, 2, 131–139.

Clark, D. M., Salkovskis, P., & Chalkley, A. (1985). Re-spiratory control as a treatment for panic attacks.Journal of Behavior Therapy and Experimental Psy-chiatry, 16, 23–30.

Clark, D. M., Salkovskis, P., Gelder, M., Koehler, C.,Martin, M., Anastasiades, P., et al. (1988). Tests of acognitive theory of panic. In I. Hand & H. Wittchen(Eds.), Panic and phobias II (pp. 71–90). Berlin:Springer-Verlag.

Clark, D. M., Salkovskis, P. M., Hackmann, A., Middle-ton, H., Anastasiades, P., & Gelder, M. (1994). Acomparison of cognitive therapy, applied relaxationand imipramine in the treatment of panic disorder.British Journal of Psychiatry, 164, 759–769.

Clark, D. M., Salkovskis, P. M., Hackmann, A., Wells,A., Ludgate, J., & Gelder, M. (1999). Brief cognitivetherapy for panic disorder: A randomized controlledtrial. Journal of Consulting and Clinical Psychology,67, 583–589.

Cote, G., Gauthier, J. G., Laberge, B., Cormier, H. J., &Plamondon, J. (1994). Reduced therapist contact inthe cognitive behavioral treatment of panic disorder.Behavior Therapy, 25, 123–145.

Cox, B. J., Endler, N. S., & Swinson, R. P. (1995). Anexamination of levels of agoraphobic severity inpanic disorder. Behaviour Research and Therapy, 33,57–62.

Craske, M. G., & Barlow, D. H. (1988). A review of therelationship between panic and avoidance. ClinicalPsychology Review, 8, 667–685.

Craske, M. G., & Barlow, D. H. (1989). Nocturnalpanic. Journal of Nervous and Mental Disease,177(3), 160–167.

Craske, M. G., & Barlow, D. H. (2006). Mastery ofyour anxiety and panic: Therapist guide (3rd ed.).New York: Oxford University Press.

Craske, M. G., Brown, T. A., & Barlow, D. H. (1991).Behavioral treatment of panic disorder: A two-yearfollow-up. Behavior Therapy, 22, 289–304.

Craske, M. G., DeCola, J. P., Sachs, A. D., & Pontillo,

D. C. (2003). Panic control treatment of agorapho-bia. Journal of Anxiety Disorders, 17(3), 321–333.

Craske, M. G., Farchione, T., Allen, L., Barrios, V.,Stoyanova, M., & Rose, D. (2007). Cognitive behav-ioral therapy for panic disorder and comorbidity:More of the same or less of more. Behaviour Re-search and Therapy, 45(6), 1095–1109.

Craske, M. G., & Freed, S. (1995). Expectations aboutarousal and nocturnal panic. Journal of AbnormalPsychology, 104, 567–575.

Craske, M. G., Glover, D., & DeCola, J. (1995). Pre-dicted versus unpredicted panic attacks: Acute versusgeneral distress. Journal of Abnormal Psychology,104, 214–223.

Craske, M. G., Golinelli, D., Stein, M. B., Roy-Byrne, P.,Bystritsky, A., & Sherbourne, C. (2005). Does the ad-dition of cognitive behavioral therapy improve panicdisorder treatment outcome relative to medicationalone in the primary-care setting? PsychologicalMedicine, 35(11), 1645–1654.

Craske, M. G., Lang, A. J., Aikins, D., & Mystkowski,J. L. (2005). Cognitive behavioral therapy for noctur-nal panic. Behavior Therapy, 36, 43–54.

Craske, M. G., Lang, A. J., Rowe, M., DeCola, J. P.,Simmons, J., Mann, C., et al. (2002). Presleep attri-butions about arousal during sleep: Nocturnal panic.Journal of Abnormal Psychology, 111, 53–62.

Craske, M. G., Maidenberg, E., & Bystritsky, A. (1995).Brief cognitive-behavioral versus non directive ther-apy for panic disorder. Journal of Behavior Therapyand Experimental Psychiatry, 26, 113–120.

Craske, M. G., Miller, P. P., Rotunda, R., & Barlow, D.H. (1990). A descriptive report of features of initialunexpected panic attacks in minimal and extensiveavoiders. Behaviour Research and Therapy, 28, 395–400.

Craske, M. G., & Mystkowski, J. L. (2006). Exposuretherapy and extinction: Clinical studies. In M. G.Craske, D. Hermans, & D. Vansteenwegen (Eds.),Fear and learning: From basic processes to clinicalimplications (pp. 217–233). Washington, DC: Amer-ican Psychological Association.

Craske, M. G., Poulton, R., Tsao, J. C. I., & Plotkin, D.(2001). Paths to panic–agoraphobia: An exploratoryanalysis from age 3 to 21 in an unselected birth co-hort. American Journal of Child and Adolescent Psy-chiatry, 40, 556–563.

Craske, M. G., Rapee, R. M., & Barlow, D. H. (1988).The significance of panic–expectancy for individualpatterns of avoidance. Behavior Therapy, 19, 577–592.

Craske, M. G., & Rowe, M. K. (1997a). A comparisonof behavioral and cognitive treatments of phobias. InG. C. L. Davey (Ed.), Phobias—a handbook of the-ory, research and treatment (pp. 247–280). West Sus-sex, UK: Wiley.

Craske, M. G., & Rowe, M. K. (1997b). Nocturnalpanic. Clinical Psychology: Science and Practice, 4,153–174.


Craske, M. G., Rowe, M., Lewin, M., & Noriega-Dimitri, R. (1997). Interoceptive exposure versusbreathing retraining within cognitive-behaviouraltherapy for panic disorder with agoraphobia. BritishJournal of Clinical Psychology, 36, 85–99.

Craske, M. G., Roy-Byrne, P., Stein, M. B., Donald-Sherbourne, C., Bystritsky, A., Katon, W., et al.(2002). Treating panic disorder in primary care: Acollaborative care intervention. General HospitalPsychiatry, 24(3), 148–155.

Craske, M. G., Roy-Byrne, P., Stein, M. B., Sullivan, G.,Hazlett-Stevens, H., Bystritsky, A., et al. (2006). CBTintensity and outcome for panic disorder in a primarycare setting. Behavior Therapy, 37, 112–119.

Craske, M. G., Street, L., & Barlow, D. H. (1989). In-structions to focus upon or distract from internalcues during exposure treatment for agoraphobicavoidance. Behaviour Research and Therapy, 27,663–672.

Craske, M. G., & Tsao, J. I. C. (1999). Self-monitoringwith panic and anxiety disorders. Psychological As-sessment, 11, 466–479.

Dattilio, F. M., & Salas-Auvert, J. A. (2000). Panic dis-order: Assessment and treatment through a wide-an-gle lens. Phoenix, AZ: Zeig, Tucker.

Deacon, B., & Abramowitz, J. (2006). A pilot study oftwo-day cognitive-behavioral therapy for panic dis-order. Behaviour Research and Therapy, 44, 807–817.

de Beurs, E., Lange, A., van Dyck, R., & Koele, P.(1995). Respiratory training prior to exposure invivo in the treatment of panic disorder with agora-phobia: Efficacy and predictors of outcome. Austra-lian and New Zealand Journal of Psychiatry, 29,104–113.

de Beurs, E., van Balkom, A. J., Lange, A., Koele, P.,& van Dyck, R. (1995). Treatment of panic dis-order with agoraphobia: Comparison of fluvoxa-mine, placebo, and psychological panic managementcombined with exposure and of exposure in vivoalone. American Journal of Psychiatry, 152, 683–691.

Deckert, J., Nothen, M. M., Franke, P., Delmo, C.,Fritze, J., Knapp, M., et al. (1998). Systematic muta-tion screening and association study of the A1 andA2a adenosine receptor genes in panic disorder sug-gest a contribution of the A2a gene to the develop-ment of disease. Molecular Psychiatry, 3, 81–85.

de Jong, M. G., & Bouman, T. K. (1995). Panic disor-der: A baseline period: Predictability of agoraphobicavoidance behavior. Journal of Anxiety Disorders, 9,185–199.

De Ruiter, C., Garssen, B., Rijken, H., & Kraaimaat,F. (1989). The hyperventilation syndrome in panicdisorder, agoraphobia, and generalized anxiety dis-order. Behaviour Research and Therapy, 27(4),447–452.

Dewey, D., & Hunsley, J. (1990). The effects of maritaladjustment and spouse involvement on the behavior-

al treatment of agoraphobia: A meta-analytic review.Anxiety Research, 2(2), 69–83.

Di Nardo, P., Brown, T. A., & Barlow, D. H. (1994).Anxiety Disorders Interview Schedule—Fourth Edi-tion (ADIS-IV). New York: Oxford University Press.

Dreessen, L., Arntz, A., Luttels, C., & Sallaerts, S.(1994). Personality disorders do not influence the re-sults of cognitive behavior therapies for anxiety dis-orders. Comprehensive Psychiatry, 35(4), 265–274.

Dworkin, B. R., & Dworkin, S. (1999). Heterotopic andhomotopic classical conditioning of the baroreflex.Integrative Physiological and Behavioral Science,34(3), 158–176.

Ehlers, A. (1995). A 1-year prospective study of panicattacks: Clinical course and factors associated withmaintenance. Journal of Abnormal Psychology, 104,164–172.

Ehlers, A., & Breuer, P. (1992). Increased cardiac aware-ness in panic disorder. Journal of Abnormal Psychol-ogy, 101, 371–382.

Ehlers, A., & Breuer, P. (1996). How good are patientswith panic disorder at perceiving their heartbeats?Biological Psychology, 42, 165–182.

Ehlers, A., Breuer, P., Dohn, D., & Fiegenbaum, W.(1995). Heartbeat perception and panic disorder:Possible explanations for discrepant findings. Behav-iour Research and Therapy, 33, 69–76.

Ehlers, A., & Margraf, J. (1989). The psychophysio-logical model of panic attacks. In P. M. G.Emmelkamp (Ed.), Anxiety disorders: Annual seriesof European research in behavior therapy (Vol. 4,pp. 1–29). Amsterdam: Swets.

Ehlers, A., Margraf, J., Davies, S., & Roth, W. T.(1988). Selective processing of threat cues in subjectswith panic attacks. Cognition and Emotion, 2, 201–219.

Ehlers, A., Margraf, J., Roth, W. T., Taylor, C. B., &Birbaumer, N. (1988). Anxiety induced by false heartrate feedback in patients with panic disorder. Behav-iour Research and Therapy, 26(1), 1–11.

Eifert, G. H., & Forsyth, J. P. (2005). Acceptance andcommitment therapy for anxiety disorders: A practi-tioner’s treatment guide to using mindfulness, accep-tance, and value-based behavior change strategies.Oakland, CA: New Harbinger.

Eley, T. C. (2001). Contributions of behavioral geneticsresearch: Quantifying genetic, shared environmentaland nonshared environmental influences. In M. W.Vasey & M. R. Dadds (Eds.), The developmentalpsychopathology of anxiety (pp. 45–59). New York:Oxford University Press.

Emmelkamp, P. (1980). Agoraphobic’s interpersonalproblems. Archives of General Psychiatry, 37, 1303–1306.

Emmelkamp, P. M., Brilman, E., Kuiper, H., & Mersch,P. (1986). The treatment of agoraphobia: A compari-son of self-instructional training, rational emotivetherapy, and exposure in vivo. Behavior Modifica-tion, 10, 37–53.


Emmelkamp, P. M., Kuipers, A. C., & Eggeraat, J. B.(1978). Cognitive modification versus prolonged ex-posure in vivo: A comparison with agoraphobics assubjects. Behaviour Research and Therapy, 16, 33–41.

Emmelkamp, P. M., & Mersch, P. P. (1982). Cognitionand exposure in vivo in the treatment of agorapho-bia: Short-term and delayed effects. Cognitive Ther-apy and Research, 6, 77–90.

Evans, L., Holt, C., & Oei, T. P. S. (1991). Long termfollow-up of agoraphobics treated by brief intensivegroup cognitive behaviour therapy. Australian andNew Zealand Journal of Psychiatry, 25, 343–349.

Eysenck, H. J. (1967). The biological basis of personal-ity. Springfield, IL: Charles C. Thomas.

Faravelli, C., Pallanti, S., Biondi, F., Paterniti, S., &Scarpato, M. A. (1992). Onset of panic disorder.American Journal of Psychiatry, 149, 827–828.

Fava, G. A., Zielezny, M., Savron, G., & Grandi, S.(1995). Long-term effects of behavioural treatmentfor panic disorder with agoraphobia. British Journalof Psychiatry, 166, 87–92.

Feigenbaum, W. (1988). Long-term efficacy of ungradedversus graded massed exposure in agoraphobics. In I.Hand & H. Wittchen (Eds.), Panic and phobias:Treatments and variables affecting course and out-come. Berlin: Springer-Verlag.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J.B. W. (1994). Structured Clinical Interview for Axis IDSM-IV Disorders. New York: Biometric ResearchDepartment, New York State Psychiatric Institute.

Foa, E. B., Jameson, J. S., Turner, R. M., & Payne, L. L.(1980). Massed vs. spaced exposure sessions in thetreatment of agoraphobia. Behaviour Research andTherapy, 18, 333–338.

Foa, E. B., & McNally, R. J. (1996). Mechanisms ofchange in exposure therapy. In R. M. Rapee (Ed.),Current controversies in the anxiety disorders(pp. 329–343). New York: Guilford Press.

Forsyth, J. P., Palav, A., & Duff, K. (1999). The absenceof relation between anxiety sensitivity and fear con-ditioning using 20% versus 13% CO2-enriched air asunconditioned stimuli. Behaviour Research andTherapy, 37(2), 143–153.

Friedman, S., & Paradis, C. (1991). African-Americanpatients with panic disorder and agoraphobia. Jour-nal of Anxiety Disorders, 5, 35–41.

Friedman, S., Paradis, C. M., & Hatch, M. (1994).Characteristics of African-American and white pa-tients with panic disorder and agoraphobia. Hospitalande Community Psychiatry, 45, 798–803.

Fry, W. (1962). The marital context of an anxiety syn-drome. Family Process, 1, 245–252.

Gallistel, C. R., & Gibbon, J. (2000). Time, rate, andconditioning. Psychological Review, 107(2), 289–344.

Garssen, B., de Ruiter, C., & van Dyck, R. (1992).Breathing retraining: A rational placebo? ClinicalPsychology Review, 12, 141–153.

Ghosh, A., & Marks, I. M. (1987). Self-treatment of ag-oraphobia by exposure. Behavior Therapy, 18, 3–16.

Goisman, R. M., Goldenberg, I., Vasile, R. G., & Keller,M. B. (1995). Comorbidity of anxiety disorders in amulticenter anxiety study. Comprehensive Psychia-try, 36, 303–311.

Goisman, R. M., Warshaw, M. G., Peterson, L. G., Rog-ers, M. P., Cuneo, P., Hunt, M. F., et al. (1994). Panic,agoraphobia, and panic disorder with agoraphobia:Data from a multicenter anxiety disorders study.Journal of Nervous and Mental Disease, 182, 72–79.

Goldstein, A. J., & Chambless, D. L. (1978). A re-analysis of agoraphobia. Behavior Therapy, 9, 47–59.

Goodwin, R. D., Fergusson, D. M., & Horwood, L. J.(2005). Childhood abuse and familial violence andthe risk of panic attacks and panic disorder in youngadulthood. Psychological Medicine, 35, 881–890.

Gorman, J. M., Papp, L. A., Coplan, J. D., Martinez, J.M., Lennon, S., Goetz, R. R., et al. (1994).Anxiogenic effects of CO2 and hyperventilation inpatients with panic disorder. American Journal ofPsychiatry, 151(4), 547–553.

Gould, R. A., & Clum, G. A. (1995). Self-help plus min-imal therapist contact in the treatment of panic disor-der: A replication and extension. Behavior Therapy,26, 533–546.

Gould, R. A., Clum, G. A., & Shapiro, D. (1993). Theuse of bibliotherapy in the treatment of panic: A pre-liminary investigation. Behavior Therapy, 24, 241–252.

Gray, J. A. (1982). The neuropsychology of anxiety: Anenquiry into the functions of the septo-hippocampalsystem. New York: Oxford University Press.

Grayson, J. B., Foa, E. B., & Steketee, G. (1982). Habit-uation during exposure treatment: Distraction versusattention-focusing. Behaviour Research and Therapy,20, 323–328.

Griez, E., & van den Hout, M. A. (1986). CO2 inhala-tion in the treatment of panic attacks. Behaviour Re-search and Therapy, 24, 145–150.

Grilo, C. M., Money, R., Barlow, D. H., Goddard, A.W., Gorman, J. M., Hofmann, S. G., et al. (1998).Pretreatment patient factors predicting attrition froma multicenter randomized controlled treatment studyfor panic disorder. Comprehensive Psychiatry, 39,323–332.

Hafner, R. J. (1984). Predicting the effects on husbandsof behavior therapy for agoraphobia. Behaviour Re-search and Therapy, 22, 217–226.

Hamilton, S. P., Fyer, A. J., Durner, M., Heiman, G. A.,Baisre de Leon, A., Hodge, S. E., et al. (2003). Fur-ther genetic evidence for a panic disorder syndromemapping to chromosome 13q. Proceedings of Na-tional Academy of Science USA, 100, 2550–2555.

Hamilton, S. P., Slager, S. L., De Leon, A. B., Heiman, G.A., Klein, D. F., Hodge, S. E., et al. (2004). Evidencefor genetic linkage between a polymorphism in the


adenosine 2A receptor and panic disorder.Neuropsychopharmacology, 29, 558–65.

Hamilton, S. P., Slager, S. L., Helleby, L., Heiman, G. A.,Klein, D. F., Hodge, S. E., et al. (2001). No associa-tion or linkage between polymorphisms in the genesencoding cholecystokinin and the cholecystokinin Breceptor and panic disorder. Molecular Psychiatry, 6,59–65.

Hand, I., & Lamontagne, Y. (1976). The exacerbationof interpersonal problems after rapid phobia re-moval. Psychotherapy: Theory, Research and Prac-tice, 13, 405–411.

Haslam, M. T. (1974). The relationship between the ef-fect of lactate infusion on anxiety states and theiramelioration by carbon dioxide inhalation. BritishJournal of Psychiatry, 125, 88–90.

Hayward, C., Killen, J. D., Hammer, L. D., Litt, I. F.,Wilson, D. M., Simmonds, B., et al. (1992). Pubertalstage and panic attack history in sixth- and seventh-grade girls. American Journal of Psychiatry, 149,1239–1243.

Hayward, C., Killen, J. D., Kraemer, H. C., & Taylor, C.B. (2000). Predictors of panic attacks in adolescents.Journal of the American Academy of Child and Ado-lescent Psychiatry, 39(2), 1–8.

Heatley, C., Ricketts, T., & Forrest, J. (2005). Traininggeneral practitioners in cognitive behavioural ther-apy for panic disorder: Randomized-controlled trial.Journal of Mental Health, 14(1), 73–82.

Hecker, J. E., Losee, M. C., Fritzler, B. K., & Fink, C. M.(1996). Self-directed versus therapist-directed cogni-tive behavioral treatment for panic disorder. Journalof Anxiety Disorders, 10, 253–265.

Hecker, J. E., Losee, M. C., Roberson-Nay, R., & Maki,K. (2004). Mastery of your anxiety and panic andbrief therapist contact in the treatment of panic dis-order. Journal of Anxiety Disorders, 18(2), 111–126.

Heldt, E., Manfro, G. G., Kipper, L., Blaya, C., Isolan,L., & Otto, M. W. (2006). One-year follow-up ofpharmacotherapy-resistant patients with panic disor-der treated with cognitive-behavior therapy: Out-come and predictors of remission. Behaviour Re-search and Therapy, 44(5), 657–665.

Hermans, D., Craske, M. G., Mineka, S., & Lovibond,P. F. (2006). Extinction in human fear conditioning.Biological Psychiatry, 60, 361–368.

Heuzenroeder, L., Donnelly, M., Haby, M. M.,Mihalopoulos, C., Rossell, R., Carter, R., et al.(2004). Cost-effectiveness of psychological and phar-macological interventions for generalized anxiety dis-order and panic disorder. Australian and New Zea-land Journal of Psychiatry, 38(8), 602–612.

Hibbert, G., & Pilsbury, D. (1989). Hyperventilation: Isit a cause of panic attacks? British Journal of Psychi-atry, 155, 805–809.

Himadi, W., Cerny, J., Barlow, D., Cohen, S., &O’Brien, G. (1986). The relationship of marital ad-justment to agoraphobia treatment outcome. Behav-iour Research and Therapy, 24, 107–115.

Hoffart, A. (1995). A comparison of cognitive andguided mastery therapy of agoraphobia. BehaviourResearch and Therapy, 33, 423–434.

Hoffart, A., & Hedley, L. M. (1997). Personality traitsamong panic disorder with agoraphobia patients be-fore and after symptom-focused treatment. Journalof Anxiety Disorders, 11, 77–87.

Hofmann, S. G., Shear, M. K., Barlow, D. H., Gorman,J. M., Hershberger, D., Patterson, M., et al. (1988).Effects of panic disorder treatments on personalitydisorder characteristics. Depression and Anxiety,8(1), 14–20.

Holden, A. E. O., O’Brien, G. T., Barlow, D. H., Stetson,D., & Infantino, A. (1983). Self-help manual for ago-raphobia: A preliminary report of effectiveness.Behavior Therapy, 14, 545–556.

Holt, P., & Andrews, G. (1989). Hyperventilation andanxiety in panic disorder, agoraphobia, and general-ized anxiety disorder. Behaviour Research and Ther-apy, 27, 453–460.

Hope, D. A., Rapee, R. M., Heimberg, R. G., &Dombeck, M. J. (1990). Representations of the self insocial phobia: Vulnerability to social threat. Cogni-tive Therapy and Research, 14, 177–189.

Hornsveld, H., Garssen, B., Fiedeldij Dop, M., & vanSpiegel, P. (1990). Symptom reporting during volun-tary hyperventilation and mental load: Implicationsfor diagnosing hyperventilation syndrome. Journal ofPsychosomatic Research, 34, 687–697.

Horwath, E., Lish, J. D., Johnson, J., Hornig, C. D., &Weissman, M. M. (1993). Agoraphobia withoutpanic: Clinical reappraisal of an epidemiologic find-ing. American Journal of Psychiatry, 150, 1496–1501.

Huppert, J. D., Bufka, L. F., Barlow, D. H., Gorman, J.M., Shear, M. K., & Woods, S. W. (2001). Therapist,therapist variables, and cognitive-behavioral therapyoutcome in a multicenter trial for panic disorder.Journal of Consulting and Clinical Psychology,69(5), 747–755.

Ito, L. M., Noshirvani, H., Basoglu, M., & Marks, I. M.(1996). Does exposure to internal cues enhance ex-posure to external exposure to external cues in ago-raphobia with panic. Psychotherapy and Psycho-somatics, 65, 24–28.

Izard, C. E. (1992). Basic emotions, relations amongemotions, and emotion cognition relations. Psycho-logical Review, 99, 561–565.

Jacob, R. G., Furman, J. M., Clark, D. B., & Durrant, J.D. (1992). Vestibular symptoms, panic, and phobia:Overlap and possible relationships. Annals of Clini-cal Psychiatry, 4(3), 163–174.

Kamphuis, J. H., & Telch, M. J. (2000). Effects of dis-traction and guided threat reappraisal on fear reduc-tion during exposure-based treatments for specificfears. Behaviour Research and Therapy, 38(12),1163–1181.

Kampman, M., Keijsers, G. P. J., Hoogduin, C. A. L., &Hendriks, G.-J. (2002). A randomized, double-blind,


placebo-controlled study of the effects of adjunctiveparoxetine in panic disorder patients unsuccessfullytreated with cognitive-behavioral therapy alone.Journal of Clinical Psychiatry, 63(9), 772–777.

Katon, W., Von Korff, M., Lin, E., Lipscomb, P.,Russo, J., Wagner, E., et al. (1990). Distressed highutilizers of medical care: DSM-III-R diagnoses andtreatment needs. General Hospital Psychiatry,12(6), 355–362.

Katschnig, H., & Amering, M. (1998). The long-termcourse of panic disorder and its predictors. Journal ofClinical Psychopharmacology, 18(6, Suppl. 2), 6S–11S.

Keijsers, G. P., Kampman, M., & Hoogduin, C. A.(2001). Dropout prediction in cognitive behaviortherapy for panic disorder. Behavior Therapy, 32(4),739–749.

Keijsers, G. P., Schaap, C. P., Hoogduin, C. A., &Lammers, M. W. (1995). Patient–therapist interac-tion in the behavioral treatment of panic disorderwith agoraphobia. Behavior Modification, 19, 491–517.

Kenardy, J. A., Dow, M. G., Johnston, D. W., Newman,M. G., Thomson, A., & Taylor, C. B. (2003). A com-parison of delivery methods of cognitive-behavioraltherapy for panic disorder: An internationalmulticenter trial. Journal of Consulting and ClinicalPsychology, 71(6), 1068–1075.

Kendler, K. S., Bulik, C. M., Silberg, J., Hettema, J. M.,Myers, J., & Prescott, C. A. (2000). Childhood sex-ual abuse and adult psychiatric and substance usedisorders in women: An epidemiological and co-twinanalysis. Archives of General Psychiatry, 57, 953–959.

Kendler, K. S., Heath, A. C., Martin, N. G., & Eaves, L.J. (1987). Symptoms of anxiety and symptoms of de-pression: Same genes, different environments? Ar-chives of General Psychiatry, 44, 451–457.

Kessler, R. C., Berglund, P., Demler, O., Jin, R., &Walters, E. E. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the Na-tional Comorbidity Survey Replication. Archives ofGeneral Psychiatry, 62, 593–602.

Kessler, R. C., Chiu, W. T., Demler, O., & Walters, E. E.(2005). Lifetime prevalence and age-of-onset distri-butions of DSM-IV disorders in the National Comor-bidity Survey Replication. Archives of General Psy-chiatry, 62(6), 593–602.

Kessler, R. C., Chiu, W. T., Jin, R., Ruscio, A. M., Shear,K., & Walters, E. E. (2006). The epidemiology ofpanic attacks, panic disorder, and agoraphobia in theNational Comorbidity Survey Replication. Archivesof General Psychiatry, 63, 415–424.

Kessler, R. C., Davis, C. G., & Kendler, K. S. (1997).Childhood adversity and adult psychiatric disorder inthe U.S. National Comorbidity Survey. PsychologicalMedicine, 27, 1101–1119.

Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C.B., Hughes, M., Eshkeman, S., et al. (1994). Lifetimeand 12 month prevalence of DSM-III-R psychiatric

disorders in the United States: Results from the Na-tional Comorbidity Study. Archives of GeneralPsychiatry, 51, 8–19.

Keyl, P. M., & Eaton, W. W. (1990). Risk factors for theonset of panic disorder and other panic attacks in aprospective, population-based study. American Jour-nal of Epidemiology, 131, 301–311.

Kikuchi, M., Komuro, R., Hiroshi, O., Kidani, T.,Hanaoka, A., & Koshino, Y. (2005). Panic disorderwith and without agoraphobia: Comorbidity withina half-year of the onset of panic disorder. Psychiatryand Clinical Neurosciences, 58, 639–643.

Kraft, A. R., & Hoogduin, C. A. (1984). The hyper-ventiliation syndrome: A pilot study of the effective-ness of treatment. British Journal of Psychiatry, 145,538–542.

Kroeze, S., & van den Hout, M. A. (2000). Selective at-tention for cardiac information in panic patients. Be-haviour Research and Therapy, 38, 63–72.

Krystal, J. H., Woods, S. W., Hill, C. L., & Charney, D.S. (1991). Characteristics of panic attack subtypes:Assessment of spontaneous panic, situational panic,sleep panic, and limited symptom attacks. Compre-hensive Psychiatry, 32(6), 474–480.

Laberge, B., Gauthier, J. G., Cote, G., Plamondon, J., &Cormier, H. J. (1993). Cognitive-behavioral therapyof panic disorder with secondary major depression: Apreliminary investigation. Journal of Consulting andClinical Psychology, 61, 1028–1037.

Lake, R. I., Eaves, L. J., Maes, H. H., Heath, A. C., &Martin, N. G. (2000). Further evidence against theenvironmental transmission of individual differencesin neuroticism from a collaborative study of 45,850twins and relatives of two continents. Behavior Ge-netics, 30(3), 223–233.

Lehman, C. L., Brown, T. A., & Barlow, D. H. (1998).Effects of cognitive-behavioral treatment for panicdisorder with agoraphobia on concurrent alcoholabuse. Behavior Therapy, 29, 423–433.

Lelliott, P., Marks, I., McNamee, G., & Tobena, A.(1989). Onset of panic disorder with agoraphobia:Toward an integrated model. Archives of GeneralPsychiatry, 46, 1000–1004.

Lidren, D. M., Watkins, P., Gould, R. A., Clum, G. A.,Asterino, M., & Tulloch, H. L. (1994). A comparisonof bibliotherapy and group therapy in the treatmentof panic disorder. Journal of Consulting and ClinicalPsychology, 62, 865–869.

Lovibond, P. F., Davis, N. R., & O’Flaherty, A. S.(2000). Protection from extinction in human fearconditioning. Behaviour Research and Therapy, 38,967–983.

Lovibond, P. F., & Shanks, D. R. (2002). The role ofawareness in Pavlovian conditioning: Empirical evi-dence and theoretical implications. Journal of Exper-imental Psychology: Animal Behavior Processes, 28,3–26.

Maddock, R. J., & Blacker, K. H. (1991). Response totreatment in panic disorder with associated depres-sion. Psychopathology, 24(1), 1–6.


Maidenberg, E., Chen, E., Craske, M., Bohn, P., &Bystritsky, A. (1996). Specificity of attentional bias inpanic disorder and social phobia. Journal of AnxietyDisorders, 10, 529–541.

Maier, S. F., Laudenslager, M. L., & Ryan, S. M. (1985).Stressor controllability, immune function and endog-enous opiates. In F. R. Brush & J. B. Overmeier(Eds.), Affect, conditioning and cognition: Essays onthe determinants of behavior (pp. 183–201).Hillsdale, NJ: Erlbaum.

Maller, R. G., & Reiss, S. (1992). Anxiety sensitivity in1984 and panic attacks in 1987. Journal of AnxietyDisorders, 6(3), 241–247.

Mannuzza, S., Fyer, A. J., Liebowitz, M. R., & Klein,D. F. (1990). Delineating the boundaries of socialphobia: Its relationship to panic disorder and ago-raphobia. Journal of Anxiety Disorders, 4(1), 41–59.

Marchand, A., Goyer, L. R., Dupuis, G., & Mainguy, N.(1998). Personality disorders and the outcome ofcognitive-behavioural treatment of panic disorderwith agoraphobia. Canadian Journal of BehaviouralScience, 30(1), 14–23.

Margraf, J., Taylor, C. B., Ehlers, A., Roth, W. T., &Agras, W. S. (1987). Panic attacks in the natural envi-ronment. Journal of Nervous and Mental Disease,175, 558–565.

Marks, I. M., Swinson, R. P., Basoglu, M., Kuck, K.,Noshirvani, H., O’Sullivan, G., et al. (1993).Alprazolam and exposure alone and combined inpanic disorder with agoraphobia: A controlled studyin London and Toronto. British Journal of Psychia-try, 162, 776–787.

Marshall, W. L. (1985). The effects of variable exposurein flooding therapy. Behavior Therapy, 16, 117–135.

Martin, N. G., Jardine, R., Andrews, G., & Heath, A.C. (1988). Anxiety disorders and neuroticism: Arethere genetic factors specific to panic? ActaPsychiatrica Scandinavica, 77, 698–706.

Mavissakalian, M., & Hamman, M. (1987). DSM-IIIpersonality disorder in agoraphobia: II. Changeswith treatment. Comprehensive Psychiatry, 28, 356–361.

McLean, P. D., Woody, S., Taylor, S., & Koch, W. J.(1998). Comorbid panic disorder and major depres-sion: Implications for cognitive-behavioral therapy.Journal of Consulting and Clinical Psychology, 66,240–247.

McNally, R. J., & Lorenz, M. (1987). Anxiety sensitiv-ity in agoraphobics. Journal of Behavior Therapyand Experimental Psychiatry, 18(1), 3–11.

McNally, R. J., Riemann, B. C., Louro, C. E., Lukach,B. M., & Kim, E. (1992). Cognitive processing ofemotional information in panic disorder. BehaviourResearch and Therapy, 30, 143–149.

McNamee, G., O’Sullivan, G., Lelliott, P., & Marks, I.M. (1989). Telephone-guided treatment for house-bound agoraphobics with panic disorder: Exposurevs. relaxation. Behavior Therapy, 20, 491–497.

Mellman, T. A., & Uhde, T. W. (1989). Sleep panic at-

tacks: New clinical findings and theoretical implica-tions. American Journal of Psychiatry, 146, 1204–1207.

Mennin, D. S., & Heimberg, R. G. (2000). The impactof comorbid mood and personality disorders in thecognitive-behavioral treatment of panic disorder.Clinical Psychology Review, 20(3), 339–357.

Messenger, C., & Shean, G. (1998). The effects of anxi-ety sensitivity and history of panic on reactions tostressors in a non-clinical sample. Journal of Behav-ior Therapy, 29, 279–288.

Michelson, L., Mavissakalian, M., & Marchione, K.(1985). Cognitive and behavioral treatments of ago-raphobia: Clinical, behavioral, and psychophysio-logical outcomes. Journal of Consulting and ClinicalPsychology, 53, 913–925.

Michelson, L., Mavissakalian, M., & Marchione, K.(1988). Cognitive, behavioral, and psychophysio-logical treatments of agoraphobia: A comparativeoutcome investigation. Behavior Therapy, 19, 97–120.

Michelson, L., Mavissakalian, M., Marchione, K.,Ulrich, R., Marchione, N., & Testa, S. (1990).Psychophysiological outcome of cognitive, behavior-al, and psychophysiologically-based treatments ofagoraphobia. Behaviour Research and Therapy, 28,127–139.

Milton, F., & Hafner, J. (1979). The outcome of behav-ior therapy for agoraphobia in relation to marital ad-justment. Archives of General Psychiatry, 36, 807–811.

Mineka, S., Cook, M., & Miller, S. (1984). Fear condi-tioned with escapable and inescapable shock: The ef-fects of a feedback stimulus. Journal of ExperimentalPsychology: Animal Behavior Processes, 10, 307–323.

Mitte, K. A. (2005). Meta-analysis of the efficacy ofpsycho- and pharmacotherapy in panic disorder withand without agoraphobia. Journal of Affective Dis-orders, 88, 27–45.

Moisan, D., & Engels, M. L. (1995). Childhood traumaand personality disorder in 43 women with panic dis-order. Psychological Reports, 76, 1133–1134.

Morisette, S. B., Spiegel, D. A., & Heinrichs, N. (2005).Sensation-focused intensive treatment for panic dis-order with moderate to severe agoraphobia. Cogni-tive and Behavioral Practice, 12(1), 17–29.

Murphy, M. T., Michelson, L. K., Marchione, K.,Marchione, N., & Testa, S. (1998). The role of self-directed in vivo exposure in combination with cogni-tive therapy, relaxation training, or therapist-assistedexposure in the treatment of panic disorder with ago-raphobia. Behaviour Research and Therapy, 12,117–138.

Myers, J., Weissman, M., Tischler, C., Holzer, C.,Orvaschel, H., Anthony, J., et al. (1984). Six-monthprevalence of psychiatric disorders in three commu-nities. Archives of General Psychiatry, 41, 959–967.

Mystkowski, J. L., Craske, M. G., Echiverri, A. M., &Labus, J. S. (2006). Mental reinstatement of context


and return of fear in spider-fearful participants.Behavior Therapy, 37(1), 49–60.

Neron, S., Lacroix, D., & Chaput, Y. (1995). Group vsindividual cognitive behaviour therapy in panic dis-order: An open clinical trial with a six month follow-up. Canadian Journal of Behavioural Science, 27,379–392.

Newman, M. G., Kenardy, J., Herman, S., & Taylor, C.B. (1997). Comparison of palmtop-computer-assisted brief cognitive-behavioral treatment tocognitive-behavioral treatment for panic disorder.Journal of Consulting and Clinical Psychology, 65,178–183.

Norton, G. R., Cox, B. J., & Malan, J. (1992). Nonclin-ical panickers: A critical review. Clinical PsychologyReview, 12, 121–139.

Noyes, R., Clancy, J., Garvey, M. J., & Anderson, D. J.(1987). Is agoraphobia a variant of panic disorder ora separate illness? Journal of Anxiety Disorders, 1,3–13.

Noyes, R., Crowe, R. R., Harris, E. L., Hamra, B. J.,McChesney, C.M., & Chaudhry, D. R. (1986). Rela-tionship between panic disorder and agoraphobia: Afamily study. Archives of General Psychiatry, 43,227–232.

Noyes, R., Reich, J., Suelzer, M., & Christiansen, J.(1991). Personality traits associated with panic disor-der: Change associated with treatment. Comprehen-sive Psychiatry, 32, 282–294.

Ohman, A., & Mineka, S. (2001). Fears, phobias, andpreparedness: Toward an evolved module of fear andfear learning. Psychological Review, 108, 483–522.

Oliver, N. S., & Page, A. C. (2003). Fear reduction dur-ing in vivo exposure to blood–injection stimuli: Dis-traction vs. attentional focus. British Journal of Clin-ical Psychology, 42(1), 13–25.

Ost, L.-G. (1988). Applied relaxation vs. progressive re-laxation in the treatment of panic disorder. Behav-iour Research and Therapy, 26, 13–22.

Ost, L.-G., Thulin, U., & Ramnero, J. (2004). Cognitivebehavior therapy vs exposure in vivo in the treatmentof panic disorder with agoraphobia. Behaviour Re-search and Therapy, 42(1), 1105–1127.

Ost, L. G., & Westling, B. E. (1995). Applied relaxationvs cognitive behavior therapy in the treatment ofpanic disorder. Behaviour Research and Therapy, 33,145–158.

Ost, L. G., Westling, B. E., & Hellstrom, K. (1993). Ap-plied relaxation, exposure in vivo, and cognitivemethods in the treatment of panic disorder with ago-raphobia. Behaviour Research and Therapy, 31,383–394.

Otto, M. W., Pollack, M. H., & Sabatino, S. A. (1996).Maintenance of remission following cognitive behav-ior therapy for panic disorder: Possible deleterious ef-fects of concurrent medication treatment. BehaviorTherapy, 27, 473–482.

Otto, M. W., Pollack, M. H., Sachs, G. S., Reiter, S. R.,Meltzer-Brody, S., & Rosenbaum, J. F. (1993). Dis-

continuation of benzodiazepine treatment: Efficacyof cognitive-behavioral therapy for patients withpanic disorder. American Journal of Psychiatry,150(10), 1485–1490.

Pauli, P., Amrhein, C., Muhlberger, A., Dengler, W., &Wiedemann, G. (2005). Electrocortical evidence foran early abnormal processing of panic-related wordsin panic disorder patients. International Journal ofPsychophysiology, 57, 33–41.

Pennebaker, J. W., & Roberts, T. (1992). Toward a hisand hers theory of emotion: Gender differences invisceral perception. Journal of Social and ClinicalPsychology, 11(30), 199–212.

Perna, G., Bertani, A., Arancio, C., Ronchi, P., &Bellodi, L. (1995). Laboratory response of patientswith panic and obsessive–compulsive disorders to35% CO2 challenges. American Journal of Psychia-try, 152, 85–89.

Pollard, C. A., Pollard, H. J., & Corn, K. J. (1989).Panic onset and major events in the lives of agora-phobics: A test of contiguity. Journal of AbnormalPsychology, 98, 318–321.

Powers, M. B., Smits, J. A. J., & Telch, M. J. (2004).Disentangling the effects of safety behavior utiliza-tion and safety-behavior availability during expo-sure based treatments: A placebo- controlled trial.Journal of Consulting and Clinical Psychology, 72,448–454.

Purkis, H. M., & Lipp, O. V. (2001). Does affectivelearning exist in the absence of contingency aware-ness? Learning and Motivation, 32, 840–899.

Rachman, S., Lopatka, C., & Levitt, K. (1988). Experi-mental analyses of panic: II. Panic patients. Behav-iour Research and Therapy, 26, 33–40.

Ramsay, R. W., Barends, J., Breuker, J., & Kruseman, A.(1966). Massed versus spaced desensitization of fear.Behaviour Research and Therapy, 4(3), 205–207.

Rapee, R. (1986). Differential response to hyperventila-tion in panic disorder and generalized anxiety disor-der. Journal of Abnormal Psychology, 95, 24–28.

Rapee, R. M. (1985). A case of panic disorder treatedwith breathing retraining. Behavior Therapy and Ex-perimental Psychiatry, 16, 63–65.

Rapee, R. M. (1994). Detection of somatic sensations inpanic disorder. Behaviour Research and Therapy, 32,825–831.

Rapee, R. M., Brown, T. A., Antony, M. M., & Barlow,D. H. (1992). Response to hyperventilation and inha-lation of 5.5% carbon dioxide-enriched air across theDSM-III-R anxiety disorders. Journal of AbnormalPsychology, 101, 538–552.

Rapee, R. M., Craske, M. G., & Barlow, D. H. (1990).Subject described features of panic attacks using anew self-monitoring form. Journal of Anxiety Disor-ders, 4, 171–181.

Rapee, R. M., Craske, M. G., & Barlow, D. H. (1995).Assessment instrument for panic disorder that in-cludes fear of sensation-producing activities: The Al-


bany Panic and Phobia Questionnaire. Anxiety, 1,114–122.

Rapee, R. M., Craske, M. G., Brown, T. A., & Barlow,D. H. (1996). Measurement of perceived control overanxiety-related events. Behavior Therapy, 27(2),279–293.

Rapee, R. M., Litwin, E. M., & Barlow, D. H. (1990).Impact of life events on subjects with panic disorderand on comparison subjects. American Journal ofPsychiatry, 147, 640–644.

Rapee, R. M., & Medoro, L. (1994). Fear of physicalsensations and trait anxiety as mediators of the re-sponse to hyperventilation in nonclinical sub-jects. Journal of Abnormal Psychology, 103(4),693–699.

Rapee, R. M., & Murrell, E. (1988). Predictors ofagoraphobic avoidance. Journal of Anxiety Disor-ders, 2, 203–217.

Rathus, J. H., Sanderson, W. C., Miller, A. L., &Wetzler, S. (1995). Impact of personality functioningon cognitive behavioral treatment of panic disorder:A preliminary report. Journal of Personality Disor-ders, 9, 160–168.

Razran, G. (1961). The observable unconscious and theinferable conscious in current soviet psychophysi-ology: Interoceptive conditioning, semantic condi-tioning, and the orienting reflex. Psychological Re-view, 68, 81–147.

Reich, J., Perry, J. C., Shera, D., Dyck, I., Vasile, R.,Goisman, R. M., et al. (1994). Comparison of per-sonality disorders in different anxiety disorder diag-noses: Panic, agoraphobia, generalized anxiety, andsocial phobia. Annals of Clinical Psychiatry, 6(2),125–134.

Reiss, S. (1980). Pavlovian conditioning and humanfear: An expectancy model. Behavior Therapy, 11,380–396.

Reiss, S., Peterson, R., Gursky, D., & McNally, R.(1986). Anxiety sensitivity, anxiety frequency, andthe prediction of fearfulness. Behaviour Researchand Therapy, 24, 1–8.

Rescorla, R. A., & Wagner, A. R. (1972). A theory ofPavolvian conditioning: Variations in the effective-ness of reinforcement and nonreinforcement. In A.H. Black & W. F. Prokasy (Eds.), Classical condition-ing II: Current research and theory (pp. 64–99). NewYork: Appleton-Century-Crofts.

Rice, K. M., & Blanchard, E. B. (1982). Biofeedback inthe treatment of anxiety disorders. Clinical Psychol-ogy Review, 2, 557–577.

Richards, J., Klein, B., & Carlbring, P. (2003). Internet-based treatment for panic disorder. Cognitive Behav-iour Therapy, 32, 125–135.

Richards, J. C., Klein, B., & Austin, D. W. (2006).Internet cognitive behavioural therapy for panic dis-order: Does the inclusion of stress management infor-mation improve end-state functioning? Clinical Psy-chologist, 10(1), 2–15.

Rijken, H., Kraaimaat, F., de Ruiter, C., & Garssen, B.

(1992). A follow-up study on short-term treatment ofagoraphobia. Behaviour Research and Therapy, 30,63–66.

Rodriguez, B. I., & Craske, M. G. (1995). Does distrac-tion interfere with fear reduction during exposure? Atest among animal-fearful subjects [Special issue: Ex-perimental pain as a model for the study of clinicalpain]. Behavior Therapy, 26(2), 337–349.

Rose, M. P., & McGlynn, F. D. (1997). Toward a stan-dard experiment for studying post-treatment returnof fear. Journal of Anxiety Disorders, 11(3), 263–277.

Rowe, M. K., & Craske, M. G. (1998). Effects of anexpanding-spaced vs massed exposure schedule onfear reduction and return of fear. Behaviour Researchand Therapy, 36, 701–717.

Roy-Byrne, P., Craske, M. G., Stein, M. B., Sullivan, G.,Bystritsky, A., Katon, W., et al. (2005). A randomizedeffectiveness trial of cognitive-behavioral therapyand medication for primary care panic disorder. Ar-chives of General Psychiatry, 62, 290–298.

Roy-Byrne, P. P., & Cowley, D. S. (1995). Course andoutcome in panic disorder: A review of recent follow-up studies. Anxiety, 1, 151–160.

Roy-Byrne, P. P., Craske, M. G., & Stein, M. B. (2006).Panic disorder. Lancet, 368, 1023–1032.

Roy-Byrne, P. P., Geraci, M., & Uhde, T. W. (1986). Lifeevents and the onset of panic disorder. AmericanJournal of Psychiatry, 143, 1424–1427.

Roy-Byrne, P. P., Mellman, T. A., & Uhde, T. W. (1988).Biologic findings in panic disorder: Neuroendocrineand sleep-related abnormalities [Special issue: Per-spectives on panic-related disorders]. Journal of Anx-iety Disorders, 2, 17–29.

Roy-Byrne, P. P., Stein, M. B., Russo, J., Mercier, E.,Thomas, R., McQuaid, J., et al. (1999). Panic disor-der in the primary care setting: Comorbidity, disabil-ity, service utilization, and treatment. Journal of Clin-ical Psychiatry, 60(7), 492–499.

Rupert, P. A., Dobbins, K., & Mathew, R. J. (1981).EMG biofeedback and relaxation instructions in thetreatment of chronic anxiety. American Journal ofClinical Biofeedback, 4, 52–61.

Safren, S. A., Gershuny, B. S., Marzol, P., Otto, M. W.,& Pollack, M. H. (2002). History of childhood abusein panic disorder, social phobia, and generalized anx-iety disorder. Journal of Nervous and Mental Dis-ease, 190(7), 453–456.

Salkovskis, P., Clark, D., & Hackmann, A. (1991).Treatment of panic attacks using cognitive therapywithout exposure or breathing retraining. BehaviourResearch and Therapy, 29, 161–166.

Salkovskis, P., Warwick, H., Clark, D., & Wessels, D.(1986). A demonstration of acute hyperventilationduring naturally occurring panic attacks. BehaviourResearch and Therapy, 24, 91–94.

Salkovskis, P. M. (1991). The importance of behaviourin the maintenance of anxiety and panic: A cognitiveaccount [Special issue: The changing face of behav-


ioural psychotherapy]. Behavioural Psychotherapy,19(1), 6–19.

Salkovskis, P. M., Clark, D. M., & Gelder, M. G.(1996). Cognition–behaviour links in the persistenceof panic. Behaviour Research and Therapy, 34, 453–458.

Schade, A., Marquenie, L. A., van Balkom, A. J., Koeter,M. W., de Beurs, E., van den Brink, W., et al. (2005).The effectiveness of anxiety treatment on alcohol-dependent patients with a comorbid phobic disorder:A randomized controlled trial. Alcoholism: Clinicaland Experimental Research, 29(5), 794–800.

Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1997).The role of anxiety sensitivity in the pathogenesis ofpanic: Prospective evaluation of spontaneous panicattacks during acute stress. Journal of Abnormal Psy-chology, 106, 355–364.

Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1999).Prospective evaluation of anxiety sensitivity in thepathogenesis of panic: Replication and extension.Journal of Abnormal Psychology, 108, 532–537.

Schmidt, N. B., McCreary, B. T., Trakowski, J. J., Santi-ago, H. T., Woolaway-Bickel, K., & Ialong, N.(2003). Effects of cognitive behavioral treatment onphysical health status in patients with panic disorder.Behavior Therapy, 34(1), 49–63.

Schmidt, N. B., Woolaway-Bickel, K., Trakowski, J.,Santiago, H., Storey, J., Koselka, M., et al. (2000).Dismantling cognitive-behavioral treatment for panicdisorder: Questioning the utility of breathing retrain-ing. Journal of Consulting and Clinical Psychology,68(3), 417–424.

Schneider, A. J., Mataix-Cois, D., Marks, I. M., &Bachofen, M. (2005). Internet-guided self-help withor without exposure therapy for phobic and panicdisorders. Psychotherapy and Psychosomatics, 74(3),154–164.

Schumacher, J., Jamra, R. A., Becker, T., Klopp, N.,Franke, P., Jacob, C., et al. (2005). Investigation ofthe DAOA/G30 locus in panic disorder. MolecularPsychiatry, 10, 428–429.

Sharp, D. M., Power, K. G., Simpson, R. J., Swanson, V.,& Anstee, J. A. (1997). Global measures of outcomein a controlled comparison of pharmacological andpsychological treatment of panic disorder and agora-phobia in primary care. British Journal of GeneralPractice, 47, 150–155.

Sharp, D. M., Power, K. G., & Swanson, V. (2004). Acomparison of the efficacy and acceptability of groupversus individual cognitive behaviour therapy in thetreatment of panic disorder and agoraphobia in pri-mary care. Clinical Psychology and Psychotherapy,11(2), 73–82.

Shear, M. K., & Schulberg, H. C. (1995). Anxiety disor-ders in primary care. Bulletin of the MenningerClinic, 59(2, Suppl. A), A73–A85.

Shulman, I. D., Cox, B. J., Swinson, R. P., Kuch, K., &Reichman, J. T. (1994). Precipitating events, loca-tions and reactions associated with initial unexpected

panic attacks. Behaviour Research and Therapy, 32,17–20.

Siddle, D. A., & Bond, N. W. (1988). Avoidance learn-ing, Pavlovian conditioning, and the development ofphobias. Biological Psychology, 27, 167–183.

Sloan, T., & Telch, M. J. (2002). The effects of safety-seeking behavior and guided threat reappraisal onfear reduction during exposure: An experimental in-vestigation. Behaviour Research and Therapy, 40(3),235–251.

Sokolowska, M., Siegel, S., & Kim, J. A. (2002).Intraadministration associations: Conditionalhyperalgesia elicited by morphine onset cues. Journalof Experimental Psychology: Animal Behavior Pro-cesses, 28(3), 309–320.

Spanier, G. (1976). Measuring dyadic adjustment: Newscales for assessing the quality of marriage and simi-lar dyads. Journal of Marriage and the Family, 38,15–38.

Spiegel, D. A., Bruce, T. J., Gregg, S. F., & Nuzzarello,A. (1994). Does cognitive behavior therapy assistslow-taper alprazolam discontinuation in panic dis-order? American Journal of Psychiatry, 151(6), 876–881.

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P.R., & Jacobs, G. A. (1983). Manual for the State–Trait Anxiety Inventory (STAI, Form Y). Palo Alto,CA: Consulting Psychologists Press.

Stanley, M. A., Beck, J. G., Averill, P. M., Baldwin, L. E.,Deagle, E. A., & Stadler, J. G. (1996). Patterns ofchange during cognitive behavioral treatment forpanic disorder. Journal of Nervous and Mental Dis-ease, 184, 567–572.

Stein, M. B., Walker, J. R., Anderson, G., Hazen, A. L.,Ross, C. A., Eldridge, G., et al. (1996). Childhoodphysical and sexual abuse in patients with anxietydisorders and a community sample. American Jour-nal of Psychiatry, 153, 275–277.

Sturges, L. V., Goetsch, V. L., Ridley, J., & Whittal, M.(1998). Anxiety sensitivity and response to hyperven-tilation challenge: Physiologic arousal, interoceptiveacuity, and subjective distress. Journal of AnxietyDisorders, 12(2), 103–115.

Suárez, L., Bennett, S., Goldstein, C., & Barlow, D. H.(in press). Understanding anxiety disorders from a“triple vulnerabilities” framework. In M. M. An-thony & M. B. Stein (Eds.). Oxford handbook ofanxiety and related disorders. New York: OxfordUniversity Press.

Swinson, R. P., Fergus, K. D., Cox, B. J., & Wickwire,K. (1995). Efficacy of telephone-administered behav-ioral therapy for panic disorder with agoraphobia.Behaviour Research and Therapy, 33, 465–469.

Taylor, S., Koch, W. J., & McNally, R. J. (1992). Howdoes anxiety sensitivity vary across the anxiety disor-ders? Journal of Anxiety Disorders, 6, 249–259.

Telch, M. J., Brouillard, M., Telch, C. F., Agras, W. S., &Taylor, C. B. (1989). Role of cognitive appraisal in


panic-related avoidance. Behaviour Research andTherapy, 27, 373–383.

Telch, M. J., Lucas, J. A., & Nelson, P. (1989). Nonclin-ical panic in college students: An investigation ofprevalence and symptomatology. Journal of Abnor-mal Psychology, 98, 300–306.

Telch, M. J., Lucas, J. A., Schmidt, N. B., Hanna, H. H.,LaNae, Jaimez, T., et al. (1993). Group cognitive-behavioral treatment of panic disorder. BehaviourResearch and Therapy, 31, 279–287.

Telch, M. J., Sherman, M., & Lucas, J. (1989). Anxietysensitivity: Unitary personality trait or domain spe-cific appraisals? Journal of Anxiety Disorders, 3, 25–32.

Thorgeirsson, T. E., Oskarsson, H., Desnica, N., Kostic,J. P., Stefansson, J. G., Kolbeinsson, H., et al. (2003).Anxiety with panic disorder linked to chromosome9q in Iceland. American Journal of Human Genetics,72, 1221–1230.

Thyer, B. A., Himle, J., Curtis, G. C., Cameron, O. G.,& Nesse, R. M. (1985). A comparison of panic disor-der and agoraphobia with panic attacks. Compre-hensive Psychiatry, 26, 208–214.

Tiemens, B. G., Ormel, J., & Simon, G. E. (1996). Oc-currence, recognition, and outcome of psychologicaldisorders in primary care. American Journal of Psy-chiatry, 153, 636–644.

Tsao, J. C. I., Lewin, M. R., & Craske, M. G. (1998).The effects of cognitive-behavior therapy for panicdisorder on comorbid conditions. Journal of AnxietyDisorders, 12, 357–371.

Tsao, J. C. I., Mystkowski, J. L., Zucker, B. G., &Craske, M. G. (2002). Effects of cognitive-behavioraltherapy for panic disorder on comorbid conditions:replication and extension. Behavior Therapy, 33,493–509.

Tsao, J. C. I., Mystkowski, J. L., Zucker, B. G., &Craske, M. G. (2005). Impact of cognitive-behavioral therapy for panic disorder on comorbidi-ty: A controlled investigation. Behaviour Researchand Therapy, 43, 959–970.

Uhde, T. W. (1994). The anxiety disorders: Phenomenol-ogy and treatment of core symptoms and associatedsleep disturbance. In M. Kryger, T. Roth, & W. De-ment (Eds.), Principles and practice of sleep medicine(pp. 871–898). Philadelphia: Saunders.

van Balkom, A. J., de Beurs, E., Koele, P., Lange, A., &van Dyck, R. (1996). Long-term benzodiazepine useis associated with smaller treatment gain in panic dis-order with agoraphobia. Journal of Nervous andMental Disease, 184, 133–135.

van Beek, N., Schruers, K. R. J., & Friez, E. J. L. (2005).Prevalence of respiratory disorders in first-degree rel-atives of panic disorder patients. Journal of AffectiveDisorders, 87, 337–340.

van den Hout, M., Arntz, A., & Hoekstra, R. (1994).Exposure reduced agoraphobia but not panic, andcognitive therapy reduced panic but not agorapho-bia. Behaviour Research and Therapy, 32, 447–451.

van den Hout, M., Brouwers, C., & Oomen, J. (2006).Clinically diagnosed Axis II co-morbidity and theshort term outcome of CBT for Axis I disorders.Clinical Psychology and Psychotherapy, 13(1), 56–63.

van Megen, H. J., Westenberg, H. G., Den Boer, J. A., &Kahn, R. S. (1996). The panic-inducing properties ofthe cholecystokinin tetrapeptide CCK4 in patientswith panic disorder. European Neuropsychophar-macology, 6, 187–94.

Vansteenwegen, D., Vervliet, B., Iberico, C., Baeyens, F.,van den Bergh, O., & Hermans, D. (2007). The re-peated confrontation with videotapes of spiders inmultiple contexts attenuates renewal of fear inspider-anxious students. Behavior Research andTherapy, 45(6), 1169–1179.

Veltman, D. J., van Zijderveld, G., Tilders, F. J., & vanDyck, R. (1996). Epinephrine and fear of bodily sen-sations in panic disorder and social phobia. Journalof Psychopharmacology, 10(4), 259–265.

Verburg, K., Griez, E., Meijer, J., & Pols, H. (1995). Re-spiratory disorders as a possible predisposing factorfor panic disorder. Journal of Affective Disorders, 33,129–134.

Wade, W. A., Treat, T. A., & Stuart, G. L. (1998). Trans-porting an empirically supported treatment for panicdisorder to a service clinic setting: A benchmarkingstrategy. Journal of Consulting and Clinical Psychol-ogy, 66, 231–239.

Wardle, J., Hayward, P., Higgitt, A., Stabl, M., Blizard,R., & Gray, J. (1994). Effects of concurrent diazepamtreatment on the outcome of exposure therapy in ag-oraphobia. Behaviour Research and Therapy, 32,203–215.

Watson, D., & Clark, L. A. (1984). Negative affectivity:The disposition to experience aversive emotionalstates. Psychological Bulletin, 96(3), 465–490.

Weems, C. F., Hayward, C., Killen, J., & Taylor, C. B.(2002). A longitudinal investigation of anxiety sensi-tivity in adolescence. Journal of Abnormal Psychol-ogy, 111(3), 471–477.

Welkowitz, L., Papp, L., Cloitre, M., Liebowitz, M.,Martin, L., & Gorman, J. (1991). Cognitive-behavior therapy for panic disorder delivered by psy-chopharmacologically oriented clinicians. Journal ofNervous and Mental Disease, 179, 473–477.

Westen, D., & Morrison, K. (2001). A multidimensionalmeta-analysis of treatments for depression, panic,and generalized anxiety disorder: An empirical exam-ination of the status of empirically supported thera-pies. Journal of Consulting and Clinical Psychology,69(6), 875–899.

Westra, H. A., Stewart, S. H., & Conrad, B. E. (2002).Naturalistic manner of benzodiazepine use and cog-nitive behavioral therapy outcome in panic disorderand agoraphobia. Journal of Anxiety Disorders,16(3), 223–246.

Wilkinson, D. J., Thompson, J. M., Lambert, G. W.,Jennings, G. L., Schwarz, R. G., Jefferys, D., et al.


(1998). Sympathetic activity in patients with panicdisorder at rest, under laboratory mental stress, andduring panic attacks. Archives of General Psychiatry,55(6), 511–520.

Williams, K. E., & Chambless, D. (1990). The relation-ship between therapist characteristics and outcomeof in vivo exposure treatment for agoraphobia.Behavior Therapy, 21, 111–116.

Williams, K. E., & Chambless, D. L. (1994). The resultsof exposure-based treatment in agoraphobia. In S.Friedman (Ed.), Anxiety disorders in African Ameri-cans (pp. 149–165). New York: Springer.

Williams, S. L. (1992). Perceived self-efficacy and pho-bic disability. In R. Schwarzer (Ed.), Self-efficacy:Thought control of action (pp. 149–176). Washing-ton, DC: Hemisphere.

Williams, S. L., & Falbo, J. (1996). Cognitive andperformance-based treatments for panic attacks inpeople with varying degrees of agoraphobic dis-ability. Behaviour Research and Therapy, 34, 253–264.

Williams, S. L., & Zane, G. (1989). Guided mastery andstimulus exposure treatments for severe performanceanxiety in agoraphobics. Behaviour Research andTherapy, 27, 237–245.

Wittchen, H.-U., Reed, V., & Kessler, R. C. (1998).The relationship of agoraphobia and panic in acommunity sample of adolescents and youngadults. Archives of General Psychiatry, 55(11),1017–1024.

Zinbarg, R. E., & Barlow, D. H. (1996). Structure ofanxiety and the anxiety disorders: A hierarchicalmodel. Journal of Abnormal Psychology, 105(2),184–193.

Zinbarg, R. E., Barlow, D. H., & Brown, T. A.(1997). Hierarchical structure and general factorsaturation of the Anxiety Sensitivity Index: Evi-dence and implication. Psychological Assessment,9, 277–284.

Zoellner, L. A., & Craske, M. G. (1999). Interoceptiveaccuracy and panic. Behaviour Research and Ther-apy, 37, 1141–1158.


C H A P T E R 2

Posttraumatic Stress Disorder

PATRICIA A. RESICKCANDICE M. MONSON

SHIREEN L. RIZVI

Severe, unexpected trauma may occur in less than a minute but have lifelong conse-quences. The tragedy that is post traumatic stress disorder (PTSD) is brought into starkrelief when the origins of the trauma occur in the context of man’s inhumanity to man. Inthis chapter, the case of “Tom” illustrates the psychopathology associated with PTSD in allits nuances and provides a very personal account of its impact. In one of any number ofevents summarized dryly every day in the middle pages of the newspaper, Tom, in the fogof war in Iraq, shoots and kills a pregnant woman and her young child in the presence ofthe husband and father. The impact of this event devastates him. The sensitive and skilledtherapeutic intervention described in this chapter is a model for new therapists, and beliesthe notion that, in these severe cases, manualized therapy can be rote and automated. Inaddition, the next generation of treatment for PTSD, termed “cognitive processing therapy”by the authors, is sufficiently detailed to allow knowledgeable practitioners to incorporatethis treatment program into their practice. This comprehensive treatment program takesadvantage of the latest developments in our knowledge of the psychopathology of traumaimpact by incorporating treatment strategies specifically tailored to overcome trauma-related psychopathology.—D. H. B.

DIAGNOSIS

Unlike most other psychiatric diagnoses, post-traumatic stress disorder (PTSD) requires aspecific type of event to occur from which theperson affected does not recover. First, to qual-ify for a diagnosis of PTSD according to theDiagnostic and Statistical Manual of MentalDisorders, fourth edition (DSM-IV; AmericanPsychiatric Association, 1994), the individualmust have experienced, witnessed, or otherwisebeen confronted with an event that involvedactual or threatened death, serious injury, orthreat to physical integrity. Second, the individ-

ual’s response to the event must include intensefear, helplessness, or horror. Thus, an event isdefined as traumatic when it has involveddeath or serious injury, or the threat of death orinjury, and the individual experiences strongnegative affect in response to the event (crite-rion A). Symptom criteria fall into three broadcategories: reexperiencing symptoms (criterionB), avoidance and numbing symptoms (crite-rion C), and physiological hyperarousal (crite-rion D). According to criterion B, the re-experiencing symptoms must be experienced inone of the following ways: Memories of thetrauma may intrude into consciousness repeti-

65

tively, without warning, seemingly “out of theblue,” without triggers or reminders to elicitthem. The person with PTSD may experienceintensely vivid reenactment experiences, orflashbacks. Intrusive memories may also occurduring the sleeping state in the form of themati-cally related nightmares. Additionally, whenfaced with cues associated with the traumaticevent, whether actual or symbolic, the individ-ual may exhibit intense psychological reactions(e.g., terror, disgust, depression) and/or physio-logical responses (e.g., increased heart rate,perspiration, and rapid breathing).

These reexperiencing symptoms are gener-ally experienced as distressing and intrusive,because the individual has no control overwhen or how they occur, and they elicit strongnegative emotions associated with the ini-tial trauma (Janoff-Bulman, 1992; Resick &Schnicke, 1992). Fear stimuli (cues) are some-times obvious, such as the combat veteran whoducks in fear when a car backfires because itsounds like gunfire. However, sometimes therelationship between the trauma and the cue isnot immediately clear. For example, one survi-vor of rape was fearful of taking showers, eventhough the rape had occurred away from herhome. However, as she began to deal with therape in treatment, she realized that every timeshe took a shower, she felt very vulnerable be-cause she was alone, naked, had no escaperoutes, and had diminished vision andhearing—all stimuli that reminded her of therape.

Avoidance and numbing symptoms (crite-rion C) reflect the individual’s attempt to gainpsychological and emotional distance from thetrauma. Some have suggested that avoidancesymptoms are a response to reexperiencingsymptoms (Creamer, Burgess, & Pattison,1992), although some factor-analytic studies ofPTSD symptoms suggest that effortful avoid-ance is more strongly associated with reexperi-encing symptoms, whereas numbing is associ-ated with high arousal symptoms (Buckley,Blanchard, & Hickling, 1998; Taylor, Kuch,Koch, Crockett, & Passey, 1998). As traumaticmemories intrude into consciousness, so do thepainful negative emotions associated with theoriginal trauma. Thus, the individual mayavoid thoughts and feelings about the trauma,avoid situations and events reminiscent of thetrauma, or may actually forget significant as-pects of the trauma. Avoidance of the traumamemory leads to a temporary decrease in pain-

ful emotions, which increases avoidancebehavior. Similarly, detachment or numbingsymptoms are an attempt to cut off the aversivefeelings associated with intrusive memories(Astin, Layne, Camilleri, & Foy, 1994; Resick& Schnicke, 1992). This detachment may thengeneralize to all emotions, both positive andnegative. Trauma survivors commonly statethat they no longer have any strong feelings, orthat they feel numb a great deal of the time.This sort of pervasive detachment may inter-fere profoundly with the individual’s ability torelate to others, enjoy daily life, remain pro-ductive, and plan for the future. Trauma sur-vivors have frequently reported highly con-stricted lifestyles after the traumatic experiencedue to the need to avoid reminders of the trau-matic memory and associated emotions. Atleast three types of avoidance behavior are re-quired before diagnosis can be made.

The trauma survivor may also experiencesymptoms of increased physiological arousal(criterion D). This suggests that the individualis in a constant state of “fight or flight,” whichis similar to how the individual’s body re-sponded during the actual traumatic event. Inthis state of alert, the individual is primed to re-act to new threats of danger, even in relatively“safe” situations. During a crisis, this is adap-tive because it facilitates survival. However, asa steady state, hyperarousal interferes withdaily functioning and leads to exhaustion. Inthis state, the individual spends a great deal ofenergy scanning the environment for dangercues (hypervigilance). The individual is likelyto experience sleep disturbance, decreased con-centration, irritability, and an overreactivity tostimuli (exaggerated startle response). There isevidence to suggest that this constant state oftension has deleterious effects on overall physi-cal health (e.g., Kulka et al., 1990). At leasttwo of the criterion D behaviors must be pres-ent for a diagnosis of PTSD.

The previously described symptom criteriamust be met concurrently for at least 1 monthto receive a diagnosis of PTSD, and the symp-toms must be perceived as distressing or causefunctional impairment. A substantial propor-tion of trauma survivors exhibit symptomsconsistent with a PTSD diagnosis immediatelyafter the traumatic event. However, these ratesdrop almost in half within 3 months posttrau-ma, then tend to stabilize. For example, rapetrauma survivors assessed at 2 weeks, 1 month,3 months, 6 months, and 9 months exhibited


PTSD diagnostic rates of 94, 65, 47, 42, and42%, respectively (Rothbaum & Foa, 1993).Thus, after 3 months, PTSD rates did not dropsubstantially. Another study that assessed sur-vivors of rape trauma at approximately 2-weeks and at 3-months postcrime found verysimilar rates of PTSD (Gutner, Rizvi, Monson,& Resick, 2006). At the first time point, 81%of the rape trauma survivors met symptom cri-teria for PTSD (minus the time criterion), andat 3-months postrape, 53% continued to meetthe criteria. Other, more heterogeneous events(greater variability in the severity of the event),such as combat, disasters, or assaults, are asso-ciated with lower rates of PTSD than rape(Kessler, Sonnega, Bromet, Hughes, & Nelson,1995). Delayed onset of PTSD is rare and mayreflect earlier subthreshold symptoms (perhapsdue to dissociation, amnesia, or extensiveavoidance) or a change in the meaning of theevent at a later time (the perpetrator kills alater victim, thereby changing the meaning ofthe event for the survivor).

In the dozen years that have passed sincethese criteria were last reviewed, an expandingbody of research has questioned this diagnosticstructure, as well as specific items included inthe current nosology. For example, at leastseven factor-analytic studies have been con-ducted (e.g., Amdur & Liberzon, 2001;Asmundson et al., 2000; Buckley et al., 1998;King, Leskin, King, & Weathers, 1998; Simms,Watson, & Doebbeling, 2002; Smith, Redd,DuHamel, Vicksberg, & Ricketts, 1999; Tayloret al., 1998), and none have found the three-factor structure that comprises the current di-agnosis. Most typically, researchers have foundeither four or two factors. The four-factor solu-tion separates effortful avoidance from numb-ing. The two-factor solutions pair numbingwith arousal and intrusions with avoidance. Atthe item level, intrusions have been found to belargely sensory, particularly visual images, anddistinct from cognitions (Ehlers et al., 2002).The merger of images and thoughts into onesymptom item is probably a misrepresentationof both the clinical phenomenology, and brainstructure and functioning (Shin, Rauch, &Pitman, 2005). An increasing body of literaturehas implicated guilt and shame in the develop-ment and maintenance of PTSD (Andrews,Brewin, Rose, & Kirk, 2000; Beckham,Feldman, & Kirby, 1998; Glover, Pelesky,Bruno, & Sette, 1990; Henning & Frueh,1997; Kubany et al., 1995; Kubany, Haynes,

Abueg, & Manke, 1996; Leskela, Dieperink, &Thuras, 2002; Wong & Cook, 1992), yet thediagnostic criteria do not include anythingabout guilt or shame.

PREVALENCE

Epidemiological studies have demonstratedhigh rates of trauma exposure and PTSD in thepopulation (Kessler et al., 1995; Kilpatrick,Saunders, Veronen, Best, & Von, 1987; Kulkaet al., 1990). In a national (U.S.) random prob-ability sample of 4,008 women, Resnick, Kil-patrick, Dansky, Saunders, and Best (1993)found a very high rate of trauma experiences(69%). When they extrapolated their results tothe U.S. population based on census statisticsfor 1989, they estimated that 66 millionwomen in the U.S. had experienced at least onemajor traumatic event. Of those who had expe-rienced a criterion A stressor, Resnick and col-leagues found the following lifetime PTSDrates: completed rape, 32%; other sexual as-sault, 31%; physical assault, 39%; homicide offamily or friend, 22%; any crime victimization,26%; noncrime trauma (e.g., natural and man-made disasters, accidents, injuries), 9%.

In the first large national civilian prevalencestudy of the psychological effects of trauma,Kessler and colleagues (1995) surveyed a repre-sentative U.S. national sample of 5,877 persons(2,812 men and 3,065 women). This study,which included both men and women, assessed12 categories of traumatic stressors. Theyfound that a majority of people had experi-enced at least one major traumatic event. Theyfound that, whereas 20.4% women and 8.2%of men were likely to develop PTSD followingexposure to trauma, the rates for specific trau-mas were often much higher. For example, rapewas identified as the trauma most likely to leadto PTSD among men, as well as women, and65% of men and 46% of women who identi-fied rape as their worst trauma developedPTSD. Among men who identified other worsttraumas, the probability of developing PTSDwas 39% of those with combat exposure, 24%with childhood neglect, and 22% who had ex-perienced childhood physical abuse. Amongwomen, aside from rape PTSD was likely to beassociated with physical abuse in childhood(49%), threat with a weapon (33%), sexualmolestation (27%), and physical attack (21%).As with the Resnick and colleagues (1993)

Posttraumatic Stress Disorder 67

study, accidents and natural disasters weremuch less likely to precipitate PTSD amongmen and women. On the other hand, Norris(1992) has pointed out although motor vehicleaccidents (MVAs) are less frequent than sometraumas (e.g., tragic death or robbery), and lesstraumatic than some events (sexual and physi-cal assault), when both frequency and impactare considered together, MVAs may be the sin-gle most significant event. The lifetime fre-quency of MVAs is 23%, and the PTSD rate is12%, which results in a rate of 28 seriously dis-tressed people for every 1,000 adults in theUnited States, just from one type of event.

More recently, Kessler and his colleagues(Kessler, Berglund, et al., 2005; Kessler, Chiu,et al., 2005), reported on another large Na-tional Comorbidity Survey with over 9,200 re-spondents. The overall prevalence of PTSD was6.8% in this study that included people whohad not experienced trauma. This compares to7.8% population prevalence reported in the1995 study (Kessler et al., 1995).

The largest study of combat veterans, theNational Vietnam Veterans ReadjustmentStudy (NVVRS; Kulka et al., 1990), was man-dated by the U.S. Congress in 1983 to assessPTSD and other psychological problems fol-lowing the Vietnam war. During the years ofthe war, over 8 million people served in the U.S.military. Of those, 3.1 million served in Viet-nam (theater veterans) and the remainderserved in other areas abroad or in the UnitedStates (era veterans). Women comprised 7,200of those serving in Vietnam, and over 255,000of those serving elsewhere during the Vietnamera. The NVVRS conducted in-depth inter-views and assessments with three groups:1,632 Vietnam theater veterans, 716 Vietnamera veterans, and 668 nonveterans/civiliancounterparts, for a total of 3,016 participants.

The results of the NVVRS indicated that themajority of Vietnam theater veterans made asuccessful readjustment to civilian life and didnot suffer from PTSD or other problems. How-ever, the researchers also found that 31% ofmale and 27% of female veterans had a full di-agnosis of PTSD at some time during theirlives. Furthermore, 15% of male and 9% of fe-male veterans had PTSD at the time of thestudy, over a decade after the end of the war.These rates translated to 479,000 Vietnam vet-erans with current PTSD. In addition, 11% ofmale and 8% of female veterans were found tohave significant symptoms and distress but did

not meet the full criteria for PTSD. This trans-lated to an additional 350,000 men andwomen in the United States alone who werestill suffering in the aftermath of the Vietnamwar.

Recently the data from the NVVRS were re-evaluated using very strict criteria that only in-cluded those incidents that could be verifiedthrough historical records. Dohrenwend andcolleagues (2006) found very little falsificationof events and a strong relationship between theamount of trauma exposure and rates of PTSD(i.e., the dose–response relationship). They did,however, find lower rates of PTSD after con-trolling for people who developed PTSD beforeor after their deployment to Vietnam and elimi-nating those people with unverifiable events.Using these stricter criteria, they found that18.7% of the veterans met the criteria for war-related PTSD at some point, and 9.1% still hadPTSD when assessed 11–12 years later. Theserates should be considered minimum likelihoodrates given that people can be traumatized byevents that may not be verifiable (e.g., rape, ac-cidents) in historical accounts of the war.

Also recent, with the wars in Iraq and Af-ghanistan, are the first attempts to assess PTSDduring a war (Hoge et al., 2004; Hoge,Auchterlonie, & Milliken, 2006). In the firststudy, Hoge and his colleagues (2004) studied2,530 Army soldiers and Marines before and3,671 after deployment to Iraq or Afghanistan.They found that mental heath problems weresignificantly greater among those who had re-turned from deployment than those who hadnot yet deployed, and that mental health prob-lems were greater in those who deployed toIraq compared with Afghanistan. Prior to de-ployment, 9% of the service personnel ex-ceeded the cutoff used for likely PTSD, whereas11.5% of those deployed to Afghanistan and18–20% of those deployed to Iraq exceededthe cutoff. There was a linear relationship be-tween the number of firefights reported and theseverity of PTSD. Being wounded or otherwisephysically injured was also associated withgreater PTSD symptomatology.

Because the military began screening all mili-tary personnel for PTSD following deploy-ment, the second study was actually apopulation-based study for a 1-year period(May 2003 to April 2004) of 303,905 Armysoldiers and Marines who deployed to Afghan-istan, Iraq, or other locations (Hoge et al.,2006). As with the previous report, service men


and women were more likely to report mentalhealth problems (19.1%) after serving in Iraqthan in Afghanistan (11.3%) or other locations(8.5%). In this study 32,500 women were alsoassessed, comprising 10.7% of the total sam-ple. There was an overall gender difference inmental health concerns, with 23.6% of womenreporting a mental health concern comparedwith 18.6% of men. However, this gender com-parison did not take into account preexistingtraumas or PTSD, exposure to combat traumasor sexual assault, or other variables that mightbe used to explain these differences. Also, thePTSD screen was a four-item questionnaire forwhich two yes answers indicated possiblePTSD. On this screen 9.8% of those who hadserved in Iraq indicated possible PTSD, andthere was again an association between theamount of combat exposure and PTSD.

THEORETICAL MODELS OF PTSD

As researchers and behavioral therapists beganto study and treat survivors of rape trauma andVietnam veterans in the 1970s, they began todraw upon learning theory as an explanationfor the symptoms they were observing.Mowrer’s two-factor theory (1947) of classicaland operant conditioning was first proposed toaccount for posttrauma symptoms (Becker,Skinner, Abel, Axelrod, & Cichon, 1984;Holmes & St. Lawrence, 1983; Keane,Zimering, & Caddell, 1985; Kilpatrick,Veronen, & Best, 1985; Kilpatrick, Veronen, &Resick, 1982). Classical conditioning was usedto explain the high levels of distress and fearthat were observed in trauma victims in reac-tion to trauma-related stimuli. Operant condi-tioning explained the development of PTSDavoidance symptoms and maintenance of fearover time, despite the fact that the uncondi-tioned stimulus, the traumatic stressor, doesnot recur. Because the trauma memory andother cues (conditioned stimuli) elicit fear andanxiety (conditioned emotional responses),these cues are avoided (or escaped from), andthe result is a reduction in fear and anxiety. Inthis manner, avoidance of the conditioned stim-uli is negatively reinforced, which prevents ex-tinction of the link between the trauma cuesand anxiety, which would normally be ex-pected without repetition of the trauma itself.

Although learning theory accounts for muchof the development and maintenance of the

fear and avoidance in PTSD, it does not fullyexplain intrusion symptoms (i.e., the repetitivememories of the trauma that intrude into thesurvivors’ thoughts in both conscious and un-conscious states such as nightmares). Based onLang’s (1977) information processing theory ofanxiety development, Foa, Steketee, and Roth-baum (1989) suggested that PTSD emerges dueto the development of a fear network in mem-ory that elicits escape and avoidance behavior.Mental fear structures include stimuli, re-sponse, and meaning elements. Anything asso-ciated with the trauma may elicit the fear struc-ture or schema and subsequent avoidancebehavior. The fear network in people withPTSD is thought to be stable and broadly gen-eralized, so that it is easily accessed. Chemtob,Roitblat, Hamada, Carlson, and Twentyman(1988) proposed that these structures are al-ways at least weakly activated in individualswith PTSD and guide their interpretation ofevents as potentially dangerous. When the fearnetwork is activated by reminders of the trau-ma, the information in the network enters con-sciousness (intrusive symptoms). Attempts toavoid this activation result in the avoidancesymptoms of PTSD. According to informationprocessing theory, repetitive exposure to thetraumatic memory in a safe environment re-sults in habituation of the fear and subsequentchange in the fear structure. As emotion de-creases, clients with PTSD begin to modifytheir meaning elements spontaneously, andchange their self-statements and reduce theirgeneralization.

Social-cognitive theories are also concernedwith information processing, but they focus onthe impact of trauma on a person’s belief sys-tem and the adjustments that are necessary toreconcile a traumatic event with prior beliefsand expectations. The first and most influentialsocial-cognitive theorist was Horowitz, whomoved from a more psychodynamic to a cogni-tive processing theory. Horowitz (1986) pro-posed that processing is driven by a “comple-tion tendency,” the psychological need for new,incompatible information to be integrated withexisting beliefs. The completion tendency keepsthe trauma information in active memory untilthe processing is complete and the event is re-solved. Horowitz also theorized that there is abasic conflict between the need to resolve andreconcile the event into the person’s history,and the desire to avoid emotional pain. Whenthe images of the event (flashbacks, night-


mares, intrusive recollections), thoughts aboutthe meanings of the trauma, and emotions as-sociated with the trauma become overwhelm-ing, psychological defense mechanisms takeover, and the person exhibits numbing oravoidance. Horowitz suggested that a personwith PTSD oscillates between phases of intru-sion and avoidance, and that if successfullyprocessed, the oscillations become less frequentand less intense. Chronic PTSD would meanthat the event stays in active memory withoutbecoming fully integrated; therefore, it is stillable to stimulate intrusive and avoidant reac-tions.

Several other social-cognitive researchersand theorists have focused more on the actualcontent of the cognitions in PTSD, and proposethat basic assumptions about the world andoneself are “shattered.” Constructivist theoriesare based on the idea that people actively createtheir own internal representations of the world(and themselves). New experiences are as-signed meaning based on a person’s model ofthe world (Janoff-Bulman, 1985, 1992;Mahoney & Lyddon, 1988; McCann &Pearlman, 1990). The task for recovery is to re-construct fundamental beliefs and establishequilibrium. Janoff-Bulman (1985) suggestedthat this process is accomplished by reinter-preting the event to reduce the distance be-tween the prior beliefs and the new beliefs.Other theorists have proposed that if one’s pre-existing beliefs are particularly positive or par-ticularly negative, then more severe PTSDsymptoms result (Foa, 1996; McCann &Pearlman, 1990; Resick & Schnicke, 1992).Foa (1996) focused particularly on beliefs re-garding the predictability and controllability ofthe trauma, whereas McCann and Pearlman(1990) proposed that several areas of cognitionmight be either disrupted or seemingly con-firmed, that is, beliefs regarding safety, trust,control/power, esteem, and intimacy.

Resick and Schnicke (1992, 1993) have ar-gued that posttrauma affect is not limited tofear, and that individuals with PTSD may bejust as likely to experience a range of otherstrong emotions, such as shame, anger, or sad-ness. Emotions such as fear, anger, or sadnessmay emanate directly from the trauma (pri-mary emotions), because the event is inter-preted as dangerous and/or abusive, and resultsin losses. It is possible that secondary, or manu-factured, emotions can also result from faultyinterpretations made by the trauma survivor.

For example, if someone is intentionally at-tacked by another person, the danger of thesituation would lead to a fight–flight response,and the attending emotions might be anger orfear (primary). However, if in the aftermath, aperson blamed him- or herself for the attack,that person might experience shame or embar-rassment. These manufactured emotions wouldhave resulted from thoughts and interpreta-tions about the event rather than from theevent itself.

In a social-cognitive model, affective expres-sion is needed, not for habituation, but for thetrauma memory to be processed fully. It is as-sumed that the natural affect, once accessed,dissipates rather quickly, and that the work ofaccommodating the memory with beliefs canbegin. Once faulty beliefs regarding the event(self-blame, guilt) and overgeneralized beliefsabout oneself and the world (e.g., safety, trust,control, esteem, intimacy) are challenged, thenthe secondary emotions also vanish, along withthe intrusive reminders. The fact that bothstress inoculation training without trauma ex-posure exercises (Foa, Rothbaum, Riggs, &Murdock, 1991; Foa et al., 1999) and cognitivetherapy (e.g., Ehlers et al., 2003; Tarrier et al.,1999) are effective treatments for PTSD under-mines the assumption that habituation or ex-tinction is the sole mechanism of change.

Ehlers and Clark (2000) proposed a cogni-tive model of PTSD that focuses on threat andmemory. Although the event has occurred inthe past, Ehlers and Clark propose that thepeople with PTSD are unable to see the event astime-limited and assume that it has larger im-plications for the future. Individuals withPTSD appraise the event such that they believethemselves to be currently at risk. There areseveral ways in which this misappraisal hap-pens. One is to overgeneralize based on theevent and begin to assume that normal activi-ties are more dangerous than they actually are.They may overestimate the probability that theevent will recur. After the trauma happens,people may misconstrue the meaning of theirPTSD symptoms such that they perceive them-selves to be in greater danger (false alarms areassumed to be true alarms) or interpret theirsymptoms to mean that they cannot cope withevents in the future.

Ehlers and Clark’s (2000) cognitive theoryalso considers the apparent memory distur-bance that occurs with PTSD such that personswith PTSD may have trouble intentionally ac-


cessing their memory of the event but have in-voluntary intrusions of parts of the event. Theypropose that because memory encoded at thetime of the trauma is poorly elaborated andintegrated with other memories with regard todetails, context of time, sequence, and so forth,this might explain why people with PTSD mayhave poor autobiographical memory, yet maybe triggered to have memory fragments thathave a here-and-now quality (no time context)or lack appropriate posttrauma appraisals(e.g., “I did not die”). Like the emotional pro-cessing models, Ehlers and Clark also proposethat strong associative learning is paired withfear responses and may become generalized. Inresponse to the perceptions of threat, peoplewith PTSD adopt various maladaptive copingstrategies, depending on their appraisals. Forexample, people who believe they will go crazyif they think about the traumatic event try toavoid thoughts about the trauma and keeptheir minds occupied all of the time. Someonewho believes that he or she must figure out whythe traumatic event happened to keep it fromhappening again will ruminate about how itcould have been prevented. Those who thinkthat they were being punished for their actionsmay become immobilized and unable to makedecisions. These maladaptive strategies, mostoften avoidance behaviors, may (1) increasesymptoms, (2) prevent change in negative ap-praisals, or (3) prevent change in the traumamemory.

In an attempt to reconcile the theories ofPTSD, Brewin, Dalgleish, and Joseph (1996)proposed a dual-representation theory that in-corporates both information processing andsocial-cognitive theories, and introduces re-search and theory from cognitive science withregard to memory. Brewin and colleagues havesuggested that the concept of a single emo-tional memory is too narrow to describe thefull range of memory that has been evident inresearch and clinical observations. Based onprior research, they proposed that sensory in-put is subject to both conscious and non-conscious processing. The memories that areconscious can be deliberately retrieved andare termed “verbally accessible memories”(VAMs). VAMs contain some sensory informa-tion, information about emotional and physi-cal reactions, and the personal meaning of theevent. Although VAMs might be reasonably de-tailed, they may also be very selective, becauseattention is narrowed under conditions of

stress, and short-term memory capacity may bedecreased.

The other type of memory is theorized to benonconsious and are termed “situationally ac-cessed memories” (SAMs). This type of infor-mation, which is probably much more exten-sive than the autobiographical memories of theevent, cannot be accessed deliberately and isnot as easily altered or edited as the more ex-plicitly accessed VAMs. SAMs comprise sen-sory (e.g., auditory, visual, tactile), physiologi-cal, and motoric information that may beaccessed automatically when a person is ex-posed to a stimulus situation similar in somefashion to the trauma, or when that personconsciously thinks about the trauma. The SAMis then experienced as an intrusive sensory im-age or flashback accompanied by physiologicalarousal.

Dual representation theory posits two typesof emotional reactions: One is conditioned dur-ing the event (e.g., fear, anger), recorded in theSAMs, and activated along with reexperiencedsensory and physiological information. Theother type, secondary emotions, result from theconsequences and implications (meaning) ofthe trauma. These secondary emotions may in-clude not only fear and anger but also guilt,shame, and sadness.

Brewin and colleagues (1996) proposed thatemotional processing of the trauma has two el-ements. One element of the processing is theactivation of SAMS (as suggested by informa-tion processing theories), the purpose of whichis to aid in cognitive readjustment by supplyingthe detailed sensory and physiological informa-tion concerning the trauma. The activation ofSAMs may eventually diminish in frequencywhen they are blocked by the creation of newSAMs, or when they are altered by the incorpo-ration of new information. When the SAMs arebrought into consciousness, they can be alteredby being paired with different bodily states(e.g., relaxation or habituation) or differentconscious thoughts. Eventually, if the SAMs arereplaced or altered sufficiently, there is a reduc-tion in negative emotions and a subsequent re-duction in attentional bias and accessibility ofthe memory.

The second element (as proposed by thesocial-cognitive theorists) is the conscious at-tempt to search for meaning, to ascribe causeor blame, and to resolve conflicts between theevent and prior expectations and beliefs. Thegoal of this process is to reduce the negative


emotions and to restore a sense of relativesafety and control in one’s environment. To ob-tain this second goal, the traumatized personmay have to edit his or her autobiographicalmemory (VAMs) to reconcile conflicts betweenthe event and the person’s belief system. Thetraumatized person may either alter the mem-ory of the event in some way to reestablish thepreexisting belief system or alter preexistingbeliefs and expectations to accommodate thisnew information.

Brewin and colleagues (1996) suggested thatfor cases in which the emotions are primaryand driven by SAMs, exposure therapy may beall that is needed. However, when secondaryemotions are present, when clients are report-ing self-blame, guilt, or shame and consequentdepression, cognitive therapy may be needed.Although both exposure and cognitive thera-pies have been found to be effective in treatingPTSD, no research thus far has matched typesof therapy to client profiles.

Another multirepresentational cognitivemodel called SPAARS (Dalgleish, 2004) wasoriginally proposed to explain everyday emo-tional experience and was then applied toPTSD. This model also endeavors to encom-pass the previous theories. The model pro-poses four types or levels of mental represen-tation systems: the Schematic, Propositional,Analogue, and Associative RepresentationalSystems. The schematic level representsabstract generic information, or schemas.Propositional-level information is verbally ac-cessible meanings, similar to VAMs, whereasinformation at the analogue level is stored as“images” across all types of sensory systems,similar to SAMs. Associative representationsare similar to the fear structures hypothesizedin emotional processing theory as represent-ing the connections between other types ofrepresentations. In the SPAARS model, emo-tions are generated through two routes. One,similar to the Ehlers and Clark (2000) cogni-tive model, is through appraisals at the sche-matic level, in which events are comparedagainst important goals. A person appraisesan event to be threatening if it blocks an im-portant goal, then experiences fear. Becausetraumatic events are threats to survival, theyare appraised as threatening and elicit fear.The second route to emotion is through asso-ciative learning, which is automatic and simi-lar to the fear activation described by Foaand her colleagues (1989).

Within this SPAARS model, a traumaticevent triggers intense appraisal-driven fear,helplessness, or horror, as well as a range ofother emotions. Information about the trau-matic event is encoded in the schematic, propo-sitional, and analogue levels simultaneously.Because the memory of the traumatic eventrepresents an ongoing threat to goals, the per-son is left with low-level fear activation, cogni-tive bias to attend to threat appraisals, and in-trusive sensory images and appraisals. Thetrauma memory exists across different levels ofmental representation but is unincorporatedinto the person’s larger mental representations;the memory may be elicited as flashbacks ornightmares. Such strong memory and emo-tional intrusions result in efforts to copethrough avoidance.

ASSESSMENT

Any comprehensive assessment of PTSD mustcapture whether or not a life event meets the se-riousness and subjective response requirementsof a traumatic stressor (criterion A), as well asthe presence and severity of 17 associatedsymptoms (criteria B–D). Although interview-based measures are considered the “gold stan-dard” for assessing PTSD, a number of self-report measures have been developed in recentyears to provide a quicker, less resource-heavymethod for assessing PTSD.

Assessment of Traumatic Events

The first essential step in the assessment ofPTSD is to identify major traumas in the pa-tient’s history. Often this is difficult to achieve,because many trauma survivors, especially rapeand child sexual abuse trauma survivors, donot spontaneously disclose their trauma his-tory. This is consistent with general patterns ofavoidance of trauma-related reminders andmay reflect shame, embarrassment, and self-blame regarding the incidents. Even when seek-ing treatment for mental health problems, trau-ma survivors often fail to recognize that theirpsychological difficulties may be associatedwith their trauma history. Kilpatrick (1983)suggested several other reasons survivors mightnot be forthcoming with this information, in-cluding fear of a negative reaction to disclo-sure, especially if previous disclosure has re-sulted in disbelief or blame. Additionally, many


trauma survivors do not recognize or labeltheir experience as “trauma,” “rape,” or“abuse,” especially if the assailant was an ac-quaintance or a relative, or if the trauma wasexperienced by many people, as in combat.Finally, in the absence of a strong alliance withthe therapist, many people choose not to dis-close such deeply personal information. It istherefore important for the clinician to forge apositive alliance as early as possible and beforthcoming about the purpose of the question-ing, any limits of confidentiality, and how theobtained information may be used (i.e., diag-nosis, treatment planning, research purposes).

In terms of questions regarding the presenceof traumatic experiences, a behavioral, descrip-tive prompt such as “Has anyone ever madeyou have unwanted sexual contact by physicalforce or threat of force?” is more detailed andis preferable to asking, “Have you ever beenraped?” In the latter case, someone who is mar-ried (or dating) and who has been sexually as-saulted may say “no” because “rape” mightnot be a term that is associated with forced sexby one’s partner. The same problem may existwith child abuse. A patient may indicate thathe was not abused as a child but readily admits,when asked, that a parent whipped him with abelt until he had welts. In general, it is recom-mended that the clinician always begin withbroad questions about experiences, then moveto more specific behaviorally-anchored ques-tions.

In recent years, some structured interviewshave been developed with the primary purposeof assessing traumas in more detail. The Poten-tial Stressful Events Interview (Kilpatrick,Resnick, & Freedy, 1991) has behaviorally an-chored questions that are particularly good forassessing interpersonal victimizations, as wellas a range of other traumatic stressors. TheDSM-IV version of the Clinician-AdministeredPTSD Scale (CAPS; Blake et al., 1995) (re-viewed in more detail later), includes a self-report screening scale (Life Events Checklist),followed by interviewer prompts to establishwhether a trauma meets criterion A.

Many clinicians may choose to circumventsome of these extensive questions by using self-report measures, such as checklists, to acquiresome initial information. Although cliniciansshould not rely on them exclusively, a numberof checklists can be used as a springboard forfurther inquiry. In addition to the Life EventsChecklist, the Traumatic Stress Schedule

(Norris, 1990), the Trauma History Question-naire (THQ; Green, 1996), the Traumatic LifeEvents Questionnaire (TLEQ; Kubany,Haynes, et al., 2000), and the Traumatic EventsScale (Vrana & Lauterbach, 1994) all assess anumber of different types of trauma, includingaccidents, natural disaster, sexual assault, andthreats of, or actual, physical harm. The Post-traumatic Stress Diagnostic Scale (PDS; Foa,1995) has two sections prior to assessment ofsymptoms. The first section assesses 13 poten-tially traumatic events, whereas the second hasquestions to determine whether an event meetsthe definition of criterion A. With regard tocombat in particular, the Combat ExposureScale (Keane, Fairbank, Caddell, & Zimering,1989) has been used widely to assess the degreeof combat exposure.

Structured Diagnostic Interviews

The CAPS, developed by Blake and others(1995), has become, perhaps, the “gold stan-dard” assessment of PTSD and is now the mostwidely used diagnostic interview (Weathers,Keane, & Davidson, 2001; Weathers, Ruscio,& Keane, 1999). The CAPS has several attrac-tive features. In addition to a detailed assess-ment of individual trauma experiences, itassesses both severity and frequency of symp-toms, using specific criteria. Furthermore, theCAPS includes questions on associated featuresof PTSD, including dissociation, survivor guilt,and social and occupational impairment. Addi-tionally, it gives clear guidelines for assessingchanges in behavior following exposure totrauma. The CAPS, the best and most thoroughassessment of PTSD, has a large body of re-search demonstrating its reliability and validityacross a wide variety of trauma populations.One disadvantage is its length of administra-tion, which is, on average, about 1 hour, andthe need for administration by a mental healthclinician. The length of administration may bedecreased slighted by assessing only the 17 coresymptoms.

The Structured Clinical Interview for DSM-IV (SCID; First, Spitzer, Williams, & Gibbon,1995) is one of the most widely used diagnosticscales. The SCID includes assessment of PTSDsymptomatology and was developed for use byexperienced clinicians. Although it assesses allof the symptoms of PTSD and can provide in-formation about whether an individual meetscriteria for the diagnosis, it is important to note


that the interview does not assess for frequencyor severity of individual symptoms. Further-more, by using the SCID, one can only deter-mine a count of the number of positive symp-toms, thereby limiting its utility in research orclinical settings in which a continuous measureof severity may be desirable. Resnick, Kilpat-rick, and Lipovsky (1991) recommended cer-tain modifications of the SCID for use withrape victims, including more sensitive screeningquestions for history of rape and other majortraumatic events.

Another structured interview, the DiagnosticInterview Schedule (DIS; Robins, Helzer,Croughan, & Ratcliff, 1981), is highly struc-tured and has the advantage of requiring lesstraining and experience to administer than theCAPS and SCID. Like the SCID, the DIS resultsin diagnosis but does not have continuous se-verity scores. One potential problem is that thePTSD section assesses exposure to civilian trau-ma, including sexual assault, but uses the term“rape” without any further specification. Thus,the modifications suggested by Resnick andcolleagues (1991) relative to rape may be ap-propriate for this instrument, as well as whenassessing interpersonal traumas. Kessler andcolleagues (1995) also have modified the DISfor better diagnosis of PTSD in large studieswith lay interviewers.

The PTSD Symptom Scale—Interview (PSS-I; Foa, Riggs, Dancu, & Rothbaum, 1993) hasa particular advantage in its ease of administra-tion and brevity, with 17 items and promptsmatching the 17-symptom PTSD criteria. ThePSS-I can result in continuous scores reflectingfrequency of symptoms, or to determine PTSDdiagnosis. Another advantage is its companionself-report measure (PSS-SR), to which scoreson the interview can be compared. Thus, afterconducting an initial interview, one could ad-minister the PSS-SR on a more regular basis(e.g., biweekly) to monitor symptom change,without having to readminister the interviewfrequently. A disadvantage of the interview isthat symptoms are assessed only over a current,2-week period instead of 1 month, so it is pos-sible that some diagnoses may be incorrect.The time frame should be modified if carefuldiagnosis is required.

Self-Report Instruments

There are now a number of self-report scales ofPTSD that have good psychometric properties.

Among them are the PSS-SR (Falsetti, Resnick,Resick, & Kilpatrick, 1993; Foa et al., 1993),the Purdue PTSD Scale—Revised (Lauterbach& Vrana, 1996), the PTSD Checklist (PCL;Weathers, Litz, Herman, Huska, & Keane,1993), the Distressing Event Questionnaire(DEQ; Kubany, Leisen, Kaplan, & Kelly,2000), the Mississippi Scale for Combat-Related PTSD (Keane, Caddell, & Taylor,1988), and the Posttraumatic Stress DiagnosticScale (PDS; Foa, 1995). Most of these scaleswere developed with specific populations, suchas rape trauma survivors (e.g., PSS) or combatveterans (e.g., Mississippi Scale, PCL), and re-search on their validity with other populationsin some cases is minimal. Thus, it is importantfor clinicians to think about their target popu-lation before adopting a measure. Moreover, aswith any self-report measure, there are limita-tions to relying exclusively on questionnairesfor diagnosis or symptom severity. Used in con-junction with structured interviews, however,they can be useful for screening purposes or fordemonstrating changes over time as a result ofa particular intervention.

The Impact of Event Scale (IES; Horowitz,Wilner, & Alvarez, 1979; Weiss & Marmar,1997) and the Mississippi Scale (Keane et al.,1988, 1989) are two of the oldest self-reportmeasures. The IES is useful for measuring trau-ma impact and was revised from its originalversion, which included only intrusion andavoidance symptoms, to include arousal symp-toms as well; thus, it maps onto the DSM-IVcriteria (IES-R; Weiss & Marmar, 1997). Theoriginal 35-item Mississippi Scale, which as-sesses both diagnostic criteria and associatedfeatures of PTSD in combat veterans, a newerversion has been created for use with civilians.The PDS (Foa, 1995), a 49-item scale designedto assess all five PTSD criteria, has strong psy-chometrics. Griffin, Uhlmansiek, Resick, andMechanic (2004) found a strong correlationbetween the PDS and the CAPS.

Two measures of PTSD have been empiri-cally derived from other scales. The KeanePTSD Scale (PK) of the Minnesota MultiphasicPersonality Inventory (MMPI) and MMPI-2has been used successfully to discriminate be-tween Vietnam combat veterans with and with-out PTSD (Keane, Malloy, & Fairbank, 1984;Weathers & Keane, 1999). The SymptomChecklist 90—Revised (SCL-90-R; Derogatis,1983) has also been examined by Saunders,Arata, and Kilpatrick (1990) and by Weathers


and colleagues (1999), who developed PTSDsubscales derived from different sets of itemsfor female crime victims and combat veterans,respectively.

The PCL is widely used in Department ofVeterans Affairs (VA) and military settings witha version (PCL-M) that specifically refers tomilitary-related traumas. The civilian versionof the scale (PCL-C) assesses civilian traumas,and the PCL-S allows the assessor to identifythe specific trauma of reference. Like the PDS,the PCL reflects DSM-IV symptoms of PTSD.

In response to a need to screen large num-bers of people for PTSD after combat or disas-ters, or in medical settings when time is limited,a brief PTSD screen has been developed for usein primary care settings or for large-group ad-ministrations, such as those with military per-sonnel following deployment (Prins et al.,2004). The Primary Care PTSD Screen (PC-PTSD) was developed for such a purpose and isnow being used routinely in the United Stateswith anyone returning from military deploy-ment or receiving any kind of treatment in theVA medical system (Hoge et al., 2006). Thisscale has four yes–no items that represent thefour major symptom clusters found in mostPTSD factor-analytic studies that separateeffortful avoidance from numbing. These fouritems were found to be highly associated withPTSD as measured by the CAPS. In fact, thePC-PTSD outperformed the PCL with regardto sensitivity and specificity, as well as effi-ciency. A cutoff of 3 was recommended as anoptimally efficient score for both men andwomen, and a cutoff of 2 was recommendedfor maximum sensitivity.

Finally, it should be noted that only onetrauma-related scale, the Trauma Symptom In-ventory (TSI; Briere, 1995), includes scales toassess response bias. For forensic purposes, inwhich response bias may be of particular con-cern, the assessor may wish to include the TSIor administer the MMPI-2, which containsboth the PK scale and validity subscales. In ad-dition to clinical scales, the TSI also includessubscales assessing tendencies to overendorseunusual or bizarre symptoms, to respond in aninconsistent or random manner, and to denysymptoms that others commonly endorse. Inaddition to PTSD-related subscales, such as In-trusive Experiences, Defensive Avoidance, andAnxious Arousal, it also includes subscales thatmeasure frequently observed problems: De-pression, Anger, Dissociation, Tension-

Reduction Behaviors, and Disruptions in Self-Perception and Sexual Functioning.

Psychophysiological Assessment

The ideal assessment includes measurement inmultiple response channels, including physio-logical responses. This is especially true inPTSD assessment, because physiological reac-tivity to trauma cues is one of the criteria of thedisorder. However, a psychophysiological testmight not be feasible in clinical settings, be-cause the required technology and expertise arenot always available. Despite this limitation, itis important to be aware of the research in thisarea and to be alert to obvious physiologicalsymptoms in patients when talking about theirtrauma experiences (e.g., signs of agitation,sweating, flushing). Research has demon-strated consistent group differences in physio-logical reactivity between individuals with andwithout PTSD when exposed to trauma-relatedstimuli, such as through the use of individual-ized trauma scripts (for a systematic review ofthis body of research, see Orr, Metzger, Miller,& Kaloupek, 2004). Vietnam veterans withPTSD have been found to be consistently morereactive to combat imagery than combat veter-ans without PTSD, even when the comparisonsamples had other anxiety disorders or otherpsychological problems (Keane, Kolb, et al.,1998; Pitman, Orr, Forgue, & Altman, 1990;Pitman, Orr, Forgue, de Jong, & Claiborn,1987). Similar results have been found in peo-ple with PTSD as a result of MVAs and childsexual abuse (Blanchard, Hickling, Buckley, &Taylor, 1996; Orr et al., 1998).

The largest investigation of physiological re-activity was a multisite study of over 1,300 vet-erans (Keane, Kaloupek, & Kolb, 1998). Usingfour psychophysiological measures, Keane andcolleagues were able to classify correctly two-thirds of those who had PTSD. This indicatesthat whereas psychophysiological reactivitycan help to distinguish between many membersof PTSD and non-PTSD groups, it should notbe used as a sole measure of diagnostic assess-ment. In fact, a number of factors may affectphysiological reactivity, and these must betaken into account when assessing the validityof psychophysiological findings. For example,the presence of psychotropic drugs (i.e.,benzodiazepines, beta-adrenergic blockers) canaffect an individual’s response. Furthermore, ithas been demonstrated that antisocial charac-


teristics can suppress levels of psychophysio-logical responding (Miller, Kaloupek, &Keane, 1999). In addition to people who donot respond physiologically, some people ap-pear to have an alternative response to arousal.

Griffin, Resick, and Mechanic (1997) stud-ied psychophysiological reactivity in recentrape trauma survivors using a methodologythat differed in two significant ways from priorstudies. First, rather than listening to generatedscripts, participants were asked to talk for 5minutes on both a neutral recall topic and theirrapes. These neutral and trauma phases wereinterspersed with baseline conditions. Second,instead of looking at the PTSD group as awhole, the researchers examined physiologicalreactivity by degree of “peritraumatic dissocia-tion” (PD), which refers to the extent to whichsomeone dissociated during the traumaticevent. Griffin and colleagues found that a smallgroup of women with high PD responded in avery different manner than other women withPTSD. Whereas skin conductance and heartrate of women with low PD scores increased asexpected while they were talking about therape, those with high PD scores showed a de-crease in the physiological measures. When theparticipants’ subjective distress during each ofthe phases was examined, the high-PD groupreported the same level of distress as the low-PD group. Therefore, despite experiencingdistress, the physiological responses of thehigh-PD group were suppressed. Griffin andcolleagues speculated that there may be a disso-ciative subtype of subjects with PTSD whophysiologically respond quite differently thanthose with the more phobic type of PTSD.

In general, there are two major aims in as-sessment in clinical practice: diagnosis andtreatment planning. Whether the primary pur-pose of assessment is diagnosis or treatmentplanning, a multidimensional, multiaxial ap-proach is desirable. Because a cross-sectionalview taken at a single point in time may fail tocapture the full range and pattern of symp-toms, a longitudinal approach to assessmenthas been advocated by Denny, Robinowitz, andPenk (1987) and Sutker, Uddo-Crane, andAllain (1991). Certainly for purposes of treat-ment, ongoing assessment of symptom patternsand treatment effectiveness is essential. Evenwhen measuring PTSD cross-sectionally, it hasbeen suggested that multiple measures andmethods be used, depending on the purpose of

the assessment (Keane, Brief, Pratt, & Miller,2007; Weathers & Keane, 1999).

Some final notes regarding assessment are inorder. First, given the empirical evidence link-ing PTSD with increased risk of suicide, suiciderisk should always be carefully assessed andmonitored. The National Women’s Study (Kil-patrick, Edmunds, & Seymour, 1992) foundthat 13% of rape trauma survivors had madea suicide attempt compared to 1% of non-victims. Additionally, 33% of the rape traumasurvivors compared to 8% of nonvictims statedthat they had seriously considered suicide atsome point. Furthermore, the presence ofcomorbid PTSD has been associated with agreater number of suicide attempts among in-dividuals with major depressive disorder(Oquendo et al., 2003). These data highlightthe need for careful monitoring of suicidal ide-ation and behavior among individuals being as-sessed or treated for PTSD.

Second, a growing body of research suggeststhat individuals with PTSD are at increasedrisk of perpetrating physical aggression againstothers. McFall, Fontana, Raskind, andRosenheck (1999) found that male Vietnamveteran inpatients with PTSD were more likelythan inpatients without PTSD or a communitysample of Vietnam veterans to perpetrate actsof violence toward objects or others. Resultsfrom the NVVRS indicated that 33% of Viet-nam veterans with PTSD had assaulted theirpartner within the previous year (Jordan et al.,1992). Unfortunately, risk of violence does notappear to be limited to Vietnam veterans.There is evidence of increased risk of violencein other traumatized populations, includingwomen (e.g., Miller, Kaloupek, Dillon, &Keane, 2004). Given also that an outburst ofanger is one of the DSM-IV symptoms ofPTSD, it is important that history of aggressiveacts, as well as current impulses toward aggres-sion, are carefully assessed and addressed.

TREATMENT

Types of Therapy for PTSD

There have been four predominant forms oftherapy for PTSD: coping, skills-focused treat-ments; exposure-based treatments; cognitivetherapy; and combination treatments and eyemovement desensitization and reprocessing(EMDR) (which may be a combination treat-


ment). Before reviewing the research on treat-ment outcome for PTSD, we describe some ofthe treatment protocols.

Stress Inoculation Training

The earliest comprehensive approach describedspecifically for use with rape trauma survivorswas stress inoculation training (SIT; Kilpatrick& Amick, 1985; Kilpatrick et al., 1982). Basedon Meichenbaum’s (1985) approach to anxiety,its aim is to give clients a sense of mastery overtheir fears by teaching a variety of coping skills.The approach is tailored to the individualproblems and needs of each client, so it is flexi-ble and can be used in individual or group set-tings. SIT is approached in phases. The firstphase, preparation for treatment, includes aneducational element to provide an explanatoryor conceptual framework from which the clientcan understand the nature and origin of her orhis fear and anxiety and make sense of the trau-ma and its aftermath. In SIT, a social learningtheory explanation is used. Along with this,fear and anxiety reactions are explained as oc-curring along three channels (Lang, 1968): (1)the physical or autonomic channel, (2) thebehavioral or motoric channel, and (3) the cog-nitive channel. Specific examples are given foreach, and the patient identifies her or his ownreactions within each channel. Interrelation-ships among the three channels are explainedand discussed. The second phase of SIT is thetraining of coping skills directed at each ofthese channels of response. It includes, in se-quence, a definition of the coping skill, a ratio-nale, an explanation of the mechanism bywhich the skill works, a demonstration of theskill, application by the client of the skill to aproblem area unrelated to the target behaviors,a review of how well the skill worked, and, fi-nally, application and practice of the skill withone of the target fears. Skills taught most oftenfor coping with fear in the physical channel aremuscle relaxation and breathing control.

For the behavioral channel, covert modelingand role playing are the coping skills usuallytaught. The client is taught to visualize a fear oranxiety-provoking situation and to imagineher- or himself confronting it successfully. Forthe cognitive channel, the client is taughtguided self-dialogue. The client is taught to fo-cus on her or his internal dialogue and trainedto label negative, irrational, and maladaptive

self-statements. She or he is then taught to sub-stitute more adaptive self-verbalizations. Self-dialogue is taught in four categories: prepara-tion, confrontation and management, copingwith feelings of being overwhelmed, and rein-forcement. For each of these categories, a seriesof questions and/or statements is generatedthat encourage the client to assess the actualprobability that the negative event will occur,to manage the overwhelming fear and avoid-ance behavior, to control self-criticism and self-devaluation, to engage in the feared behavior,and finally to reinforce her- or himself for mak-ing the attempt and following the steps.

Exposure Techniques

Beginning in the early 1980s, forms of expo-sure therapy were investigated as a treatmentfor PTSD. Although systematic desensitization(SD) has been demonstrated to be effective fortreating PTSD in a number of case study re-ports and controlled studies, it has not beenwidely adopted as a preferred treatment(Bowen & Lambert, 1986; Brom, Kleber, &Defares, 1989; Frank et al., 1988; Frank &Stewart, 1983, 1984; Schindler, 1980; Shalev,Orr, & Pitman, 1992; Turner, 1979). Becausepeople with PTSD may fear and avoid a widerange of trauma-related stimuli, SD may re-quire a number of hierarchies that can be quiteinefficient.

Extended exposure to feared cues or to thetrauma memory itself is a more efficient treat-ment and has been employed more widely.Known variously as direct therapeutic expo-sure (DTE), flooding, or prolonged exposure,these exposure techniques require clients toconfront feared situations in vivo, to imaginethemselves in a fear-producing situation, or torecall their particular trauma for extended peri-ods of time. Rothbaum (1998) experimentedwith the use of virtual reality for treating Viet-nam veterans. The veteran with PTSD can takea virtual helicopter trip in Vietnam, completewith gunfire and other stimuli that may evokememories of traumatic events.

Foa and colleagues (1991; Rothbaum &Foa, 1992) were the first to focus extensivelyon the specific trauma memory rather thanfear-producing stimuli. Prolonged exposure(PE) is conducted individually in 9–12 weeklyor biweekly 90-minute sessions. The first twosessions are for information gathering, treat-


ment planning, and explanation of the treat-ment rationale. Clients are also taught breath-ing retraining. A hierarchical list is generated ofmajor stimuli that are feared and avoided. Cli-ents are instructed to confront feared cues forat least 45 minutes a day, starting with a mod-erately anxiety-provoking stimulus on the hier-archy. Beginning with Session 3, the traumascene is relived in imagination, and the client isasked to describe it aloud in the present tense.The level of detail is left to the client for thefirst two exposures, but thereafter she or he isencouraged to include more and more detailabout external cues and internal cues, such asthoughts, physiological responses, and fearedconsequences. Descriptions are repeated sev-eral times each session (for 60 minutes) andtape-recorded. Clients are assigned homework,to listen to the tape and to engage in in vivotasks. Care is taken in sessions to ensure thatthe client’s anxiety decreases before the sessionis terminated, aided by the therapist, if neces-sary (Foa & Rothbaum, 1998).

Marks, Lovell, Noshirvani, Livanou, andThrasher (1998) have conducted exposuretherapy somewhat differently. Their version ofthe therapy includes five sessions of imaginalexposure, then five sessions of live exposure.During the imaginal exposure, clients are askedto relive the experience aloud in the first per-son, present tense about the details of their ex-perience, and then to imagine and describe crit-ical aspects of the event (rewind and hold).Clients listen to their therapy tapes daily be-tween sessions. During the live exposure por-tion of therapy, clients (most often therapist-accompanied) progress through a hierarchy offeared, avoided, and disabling trauma-relatedstimuli. They are asked to practice the live ex-posure for an hour a day between sessions.

Cognitive Therapy/Cognitive Restructuring

Cognitive therapy for PTSD has generallytaken two forms. One form is more present-focused and typically uses daily diaries or mon-itoring forms to elicit thoughts that the clienthas recorded during the week. These home-work sheets form the basis of the cognitive re-structuring training that occurs during treat-ment through the use of teaching and Socraticquestioning. Clients are taught to identify andto dispute their unrealistic or exaggeratedthoughts about themselves, the world, andtheir futures with more probabilistic reasoning

and evidence-based argument. Examples ofstudies that have used this model of cognitiverestructuring are Blanchard and colleagues(2003) and Foa and colleagues (2005).

The other form of cognitive therapy is moretrauma-focused and constructivist, focusing onthe particular meanings that the event has forthe client and how those interpretations of theevent contradict or seemingly confirm previ-ously held beliefs about self and others. Thesedistorted assumptions about the event (e.g., “Ishould have been able to stop the event, so it ismy fault that it happened”) may maintain a be-lief in a just world or a sense of controllability,but typically at the cost of reduced self-esteem,shame, or guilt. The focus of treatment is onhow clients may have distorted the event itselfto maintain prior beliefs about justice or therole of others (assimilation), or conversely, howthey may have changed their beliefs aboutthemselves and the world too much (over-accommodation) in an attempt to regain asense of control or safety in the present or thefuture (“I cannot trust other people at all anymore”). Treatment includes Socratic question-ing and teaching clients to challenge theirthinking about their traumatic events and theimplications they have constructed through theuse of progressive worksheets. Examples oftrauma-focused cognitive therapy are studiesby Resick, Nishith, Weaver, Astin, and Feuer(2002) or Tarrier and colleagues (1999).

Combination Treatments/Additive Studies

Some studies that we review below refer totheir protocols as cognitive-behavioral therapy(CBT; e.g., Blanchard et al., 2003; Bryant,Harvey, Dang, Sackville, & Basten, 1998).These therapy packages are typically combina-tions of various forms of exposure (imaginal, invivo, written) and cognitive restructuring orcognitive therapy, but they may also include re-laxation or other coping skills. These protocolsmay have been developed as a combinationtreatment to begin with, or may reflect additivestudies in which a new component is added toan existing treatment to determine whether theextra component adds value to the existingtherapy. An important distinction betweenthese two types of protocols is that the combi-nation treatments were designed to be a spe-cific length to accommodate the components ofthe treatment. In other words, the therapypackage was designed to achieve the optimal


goals of the exposure or cognitive components.Additive studies must accommodate to theoriginal protocol length; therefore, they maynot be designed with the optimal amount oftherapy needed to achieve the goals of the com-ponents. In additive studies, the amount of ex-posure, stress management, or cognitive ther-apy in combination is shortened to maintainthe length of the original protocol being com-pared.

STAIR/MPE

Cloitre, Koenen, Cohen, and Han (2002) pro-posed that survivors of child sexual abuse trau-ma have problems of affect regulation and in-terpersonal effectiveness, in addition to theirPTSD, that compromise their ability to profitfrom trauma-focused interventions. Thus, theydeveloped a protocol called STAIR (i.e., skillstraining in affective and interpersonal regula-tion) that included treatment for these prob-lems prior to implementing a modification ofprolonged exposure (MPE). This combinationtreatment first trained patients in emotionmanagement and interpersonal skills for 8weeks, followed by a second phase of treat-ment with imaginal exposure. The imaginal ex-posure phase also included postexposure emo-tional management and cognitive therapy.

COGNITIVE PROCESSING THERAPY

Unlike the additive studies, cognitive process-ing therapy (CPT) was initially developed spe-cifically as a combination treatment to treat thespecific symptoms of PTSD in sexual assaulttrauma survivors (Resick & Schnicke, 1992,1993), but it has since been tested with otherpopulations. CPT, which can be delivered ineither individual or group formats, is a 12-session, structured therapy program that in-cludes a written form of exposure to the trau-matic memory but is predominantly a cognitivetherapy. After an introduction to PTSD symp-toms and the therapy, clients are asked to writean impact statement, a statement of how theirworst traumatic event has affected them. Cli-ents are asked to focus on any self-blame theyhave regarding the trauma and the effects ofthe event on their beliefs about self and others.This statement is used to understand how theymay have distorted the cause of the event orovergeneralized its meaning, such that theirfunctioning has been compromised. For exam-

ple, if someone thought that she or he shouldhave been able to stop the event, then the indi-vidual might feel guilt afterward. If a client de-cided that the event means that no one is to betrusted, then she or he will behave as thoughthat were true.

Before examining the trauma in depth, theclient is taught to label emotions and recognizethe connection among events, thoughts, andfeelings, then is asked to write a detailed ac-count of the worst traumatic event and read itto her- or himself every day. In sessions she orhe reads it to the therapist and is encouraged tofeel whatever emotions emerge. The therapistbegins to challenge faulty thinking about thetraumatic event with Socratic questioning. Af-ter writing and reading the account for a sec-ond time, the therapist focuses on teaching theclient to learn the skill of challenging thoughtsand assumptions with Socratic questions forher- or himself through a series of worksheets.The client is first taught to question a singlethought, then look for patterns of problematicthinking, and, finally, to generate alternative,more balanced thoughts about the event itselfand then overgeneralized assumptions aboutself and world. In the last five sessions, clientsare provided modules to assist them in thinkingabout specific themes that are commonly dis-rupted following traumatic events: safety, trust,power and control, esteem, and intimacy.

Eye Movement Desensitization and Reprocessing(EMDR)

Eye movement desensitization and reprocess-ing therapy (EMDR) is a controversial therapythat evolved not from theory or application ofeffective techniques for other disorders, butfrom a personal observation. As originally de-veloped by Shapiro (1989, 1995), EMDR wasbased on a chance observation that troublingthoughts were resolved when her eyes followedthe waving of leaves during a walk in the park.Shapiro developed EMDR on the basis of thisobservation and argued that lateral eye move-ments facilitate cognitive processing of thetrauma. Subsequently, EMDR was conceptual-ized as a cognitive-behavioral treatment aimedat facilitating information processing of trau-matic events and cognitive restructuring of neg-ative, trauma-related cognitions. In the earlypresentations of EMDR, it was touted as a one-session cure for a range of disorders. However,more recent studies are typically of trauma-


related symptoms, with a course more similarto the other trauma therapies. EMDR is nowdescribed as an eight-phase treatment that in-cludes history taking, client preparation, targetassessment, desensitization, installation, bodyscan, closure, and reevaluation of treatment ef-fects. EMDR includes exposure and cognitivecomponents, as well as the lateral eye move-ments.

In the basic EMDR protocol, clients areasked to identify and focus on a traumatic im-age or memory (target assessment phase).Next, the therapist elicits negative cognitionsor belief statements about the memory. Clientsare asked to assign a rating to the memory andnegative cognitions on an 11-point scale of dis-tress and to identify the physical location of theanxiety. The therapist helps clients generatepositive cognitions that would be preferable toassociate with the memory. These are rated ona 7-point scale of how much the clients believethe statement. Once the therapist has in-structed a client in the basic EMDR procedure,she or he is asked to do four things simulta-neously (desensitization phase): (1) Visualizethe memory; (2) rehearse the negativecognitions; (3) concentrate on the physical sen-sations of the anxiety; and (4) visually track thetherapist’s index finger. While the client doesthis, the therapist rapidly moves his or her in-dex finger back and forth from right to left, 30to 35 cm from the client’s face, with two back-and-forth movements per second. These are re-peated 24 times. Then the client is asked toblank out the memory and take a deep breath.Subsequently, she or he brings back the mem-ory and cognitions and rates the level of dis-tress. Sets of eye movements (saccades) are re-peated until the distress rating equals 0 or 1. Atthis point, the client is asked about how she orhe feels about the positive cognition and gives arating for it (installation phase).

Evidence for Treatment Efficacy

Because so many of the studies have comparedmore than one type of treatment or combina-tions of components, this section is presentedgenerally in chronological order. Only con-trolled trials are presented here. Foa and col-leagues (1991) compared a modified SIT to PE,supportive counseling, and a waiting-list con-trol group, seeing clients individually. Becausethey did not want overlapping techniques intheir comparison, Foa and colleagues elimi-

nated the in vivo exposure component, con-fronting feared cues from SIT. In this study, SITwas most effective immediately posttreatmentin reducing PTSD symptoms, anxiety, and de-pression. However, at 3.5-month follow-up,there was a trend for the exposure approach toshow the greatest efficacy.

Marks and colleagues (1998) compared fourgroups: exposure therapy, cognitive restructur-ing, exposure combined with cognitive restruc-turing, and relaxation training. The partici-pants were 87 men and women who had PTSDfrom a range of traumatic stressors. Seventy-seven participants completed treatment and 52completed the 36-week follow-up. The authorsfound that overall, the cognitive, exposure, andcombined treatments were more effective thanrelaxation, but there were no major differencesamong any of the three treatments. The treat-ment gains were maintained through the 6-month follow-up. Marks and colleagues, in re-sponding to these findings, modified the theory.They suggested that emotions may be viewedas response syndromes that comprise looselylinked reactions of many physiological, behav-ioral, and cognitive components. The intensityof an emotion may be reduced by acting on anyof several components. A change in one com-ponent may affect the others. Several compo-nents can be acted on at the same time by cer-tain treatment combinations.

In an additive study, Foa and colleagues(1999) compared modified SIT, PE, and a com-bination of SIT and PE. They found that thecombination did not improve the results overSIT or PE alone. However, as discussed earlier,it should be pointed out that because of the na-ture of the research, the participants only re-ceived half as much SIT or PE as the other par-ticipants, because they had the same length ofsessions but half SIT and half PE at each ses-sion. It should also be pointed out that neitherof the Foa and colleagues studies were poweredto detect anything but large treatment differ-ences. The small sample sizes may not havebeen able to detect differences that might haveemerged with samples large enough to findsmall to medium effect sizes.

Tarrier and colleagues (1999) comparedimaginal exposure and cognitive therapy. Atposttreatment and follow-up, the investigatorsdid not find a difference between the two thera-pies. However at a 5-year follow-up with 52%of the original sample, no patients who re-ceived cognitive therapy were diagnosed with


PTSD compared to 29% of those receiving ex-posure treatment (Tarrier & Sommerfield,2004). Those receiving exposure treatment re-ported significantly more PTSD and depressivesymptoms than the cognitive therapy group.

Beginning in the early to mid-1990s, a seriesof studies compared EMDR to other treat-ments and examined the necessity of the eyemovements for the treatment. In a sample of 61combat veterans, all of whom met diagnosticcriteria for PTSD, Boudewyns, Hyer, Peralme,Touze, and Kiel (1995) found EMDR to beequally as effective as PE in reducing symptomsof PTSD, depression, anxiety, and acceleratedheart rate. In this study, subjects were ran-domly assigned to one of three groups. Allthree groups received the standard treatment attheir facility, which comprised eight grouptherapy sessions and a few spontaneous indi-vidual sessions. One group received only thisform of treatment; the second and third groupsalso received either five to eight sessions ofEMDR or five to eight sessions of PE. All threegroups improved significantly on symptoms ofPTSD. The therapy-only group did not im-prove on depression and showed increases inanxiety and heart rate, whereas the two othergroups improved on all measures.

Two studies had mixed or negative resultsfor EMDR. The Pitman and colleagues (1993)study of 17 combat veterans compared 12 ses-sions of EMDR to an eyes-fixed procedure inwhich the therapist alternately tapped each legof the subject. The EMDR group reported sig-nificantly greater improvement on measures ofsubjective distress, self-report measures ofPTSD intrusion and avoidance, and psychiatricsymptoms. No effects were observed, however,on other standardized self-report and struc-tured interviews of PTSD. Jensen (1994) com-pared two sessions of EMDR with no treat-ment in 25 combat veterans and found noimprovements in either group on standardizedmeasures of PTSD.

Devilly and Spence (1999) found a statisticaland effect size advantage (Cohen’s d = 0.67) for aCBT intervention including exposure and cogni-tive elements compared to EMDR. These resultswere maintained at 3-month follow-up assess-ment. Taylor and colleagues (2003) also foundthat exposure therapy was statistically superiorto EMDR, which was not significantly differentfrom relaxation therapy in a mixed sample. Theproportion of individuals achieving clinicallysignificant reductions in their PTSD symptoms

(i.e., improvement greater than two standard de-viations on clinician assessment of PTSD) wasgreater in the exposure group.

Several studies have compared EMDR tovarious combinations of CBT. Three of thesestudies report effect size advantages for EMDRin those who complete treatment. Ironson,Freund, Strauss, and Williams (2002) com-pared EMDR to PE in a relatively small mixedsample and found no statistical differences be-tween the two treatments. However, there wasan effect size advantage of 0.65 for EMDRcompared to prolonged exposure in completeranalyses. Lee, Gavriel, Drummond, Richards,and Greenwald (2002) compared EMDR to acombination of SIT and PE in another smallsample and found no statistical differencesbetween the treatments, with exception of astatistical advantage of EMDR in intrusivesymptoms and an overall effect size advantage(d = 0.62) in completer analyses. In a largemixed-trauma sample, Power and colleagues(2002) compared EMDR to a combination ofexposure therapy and cognitive restructuring,and found no significant differences. They re-ported effect size advantages for EMDR in thefrequency and intensity of PTSD symptomsclusters.

Most recently, Rothbaum, Astin, andMarsteller (2005), compared EMDR, PE, and awaiting-list control group in a very well-controlled study of 74 women with rape-related PTSD. They found no differences be-tween the two active conditions on PTSD atposttreatment or 6-month follow-up; bothgroups improved substantially. They onlyfound differences among the subsample withmultiple comorbidities, in which the EMDRsample with multiple comorbidities did notfare as well as the PE sample participants withsuch comorbidity.

While Shapiro has maintained that lateraleye movements are an essential therapeuticcomponent of EMDR, studies that have exam-ined this have had mixed results. Renfrey andSpates (1994) treated a sample of 23 heteroge-neous trauma survivors with standard EMDR,a variant in which lateral eye movements wereengendered via a light-tracking task, or anothervariant in which no lateral eye movementswere induced and subjects were instructed tofix their visual attention. All three groups im-proved significantly on measures of PTSD, de-pression, anxiety, heart rate, and subjectiveunits of distress (SUDs) scores at posttreatment


and at 1- to 3-month follow-up. No differenceswere found among treatments. In contrast, in18 heterogeneous trauma survivors, Wilson,Silver, Covi, and Foster (1996) found signifi-cant improvements in subjects who receivedEMDR, but not in subjects who were in-structed to fix their visual attention, or in sub-jects who alternated tapping their right and leftthumbs to a metronome. Physiological mea-sures and SUDs, but no standardized distressmeasures, were used to measure outcome.Pitman and colleagues (1993), as noted earlier,found mixed results.

Given these results, it is not clear whetherlateral eye movements are an essential compo-nent of EMDR. EMDR forces clients to thinkabout the trauma, to identify the negativecognitions associated with the trauma, and towork toward positive cognitions as they pro-cess the traumatic memory. Without the lateraleye movements, EMDR is quite similar to aform of CPT that facilitates the processing ofthe traumatic memory. Therefore, any efficacydemonstrated by EMDR may be more attribut-able to engagement of the traumatic memoryand the facilitation of cognitive reassessmentthan to eye movements. Nevertheless, EMDRappears to be as effective as other exposure andcombination CBT packages in these relativelysmall studies.

In a comparison of exposure and a combina-tion protocol, Paunovic and Öst (2001) com-pared a package of cognitive-behavioral inter-ventions, including exposure therapy, cognitiverestructuring, and controlled breathing, to ex-posure therapy only in a relatively small sampleof refugees. No statistical differences werefound between the two conditions, but therewas a small effect size advantage (0.13) for thecombination of interventions.

Cloitre and colleagues (2002) compared theSTAIR/MPE protocol for adults who had beensexually abused as children and a waiting-listcontrol group. Compared to the waiting-listcondition, this combination of treatments wasfound to be efficacious, with an effect size of d= 1.3 at posttreatment. The treatment had sig-nificant effects on affect regulation and inter-personal measures, as well as PTSD. Only 23%of the treated group still met criteria for PTSDafter treatment. Cloitre and colleagues alsofound that the participants continued to im-prove from posttreatment to 3-month follow-up and then again at 9-month follow-up. In afurther analysis of the study, Cloitre, Stovall-

McClough, Miranda, and Chemtob (2004) ex-amined the effect of therapeutic alliance andaffect regulation (Phase 1) on the outcome ofPhase 2 exposure treatment. They found thattherapeutic alliance led to greater affect regula-tion in Phase 1, which then resulted in bettertreatment outcome following the modified PE.

Resick and her colleagues (2002) con-ducted a large randomized controlled trialcomparing individually delivered CPT, PE,and minimal attention waiting-list conditionswith survivors of completed rape. Participantsin both CPT and PE conditions showed largedecreases in PTSD symptoms (75% decreaseon average) compared to the waiting-list con-trol condition, which did not change over thesame period. A small effect size difference fa-vored CPT over PE. Eighty percent of partici-pants in both treatment groups remitted fromtheir PTSD diagnosis. CPT was more effec-tive with specific guilt cognitions than PE,although both conditions improved signifi-cantly. In addition to PTSD, there were simi-lar decreases in depression, anger, dissocia-tion, and other indicators of complex PTSD(Resick, Nishith, & Griffin, 2003).

Several studies compared active treatmentsto nonspecific treatments rather than awaiting-list control. These designs control fornonspecific effects of treatment and are not ex-pected to produce the large differences typi-cally found with comparisons to waiting-listcontrol groups. For example, Blanchard andcolleagues (2003) compared CBT to supportivecounseling and a waiting list in a sample ofMVA survivors. They found that CBT (relax-ation, written exposure, in vivo exposure, cog-nitive restructuring, and behavioral activation)was significantly better than supportive coun-seling, which was significantly better thanwaiting list. There was a 0.63 effect size advan-tage for CBT compared to supportive counsel-ing in those who completed the treatment.Neuner, Schauer, Klaschik, Karunakara, andElbert (2004) also used supportive counseling,as well as a psychoeducation intervention forcontrol groups, and compared them to theirnarrative exposure therapy. In this form oftherapy, clients are asked to construct their en-tire autobiography verbally, including all of thetraumatic events they experienced. In a sampleof Sudanese refugees, there were no differencesamong the three conditions at posttreatment.However, narrative exposure therapy was sig-nificantly better than both conditions, which


were equal to one another in effectiveness, at 1-year follow-up.

In one of the largest studies published todate, a multisite trial of group CBT treatmentwas compared to a present-centered problem-solving group treatment (Schnurr et al., 2003).Three hundred sixty male veterans were ran-domly assigned to one of the two conditionsand provided weekly therapy for 7 months.The CBT treatment included education andself-management of symptoms, autobiogra-phies, war-zone scene identification, imaginalexposures, cognitive restructuring, and relapseprevention. Using intention-to-treat (ITT) anal-yses with the entire randomized sample, about38% of each group exhibited clinically signifi-cant improvements after treatment and 44% ofeach group 1 year after the beginning of treat-ment, but there were no differences betweenthe two treatment conditions on any measure.When examining only those clients who re-ceived an “adequate dose” of 24 sessions, therewere some significant differences in PTSDsymptoms between conditions, and it appearedthat the differences were larger in the latter co-horts of the study than in the first cohort, per-haps a practice effect by therapists. In consider-ing the modest findings and lack of differencesbetween the CBT and control condition, the in-vestigators questioned whether the number ofexposures in the group setting was adequate,whether more experienced CBT therapistsmight have made a difference, or whether vet-erans, with their multiple comorbid conditions,might be a more difficult population to treat.

Ehlers and colleagues (2003) conducted arandomized control trial comparing cognitivetherapy, a self-help booklet, and repeated as-sessments with victims of MVA survivors aftera period of self-monitoring. They found that asmall percentage of patients (12%) improvedby self-monitoring alone. The remaining pa-tients with PTSD were randomized into one ofthe three conditions approximately 3 monthsafter the MVA. Although the 64-page bookletincluded cognitive-behavioral principles andeducation about PTSD, it was found to be lesseffective (as were repeated assessments only)compared to cognitive therapy. The cognitivetherapy was highly effective and had no drop-outs. In contrast, the educational self-help ap-proach did not differ from repeated assess-ment, although both groups did improve. Itshould be pointed out, though, that this sampleof patients with PTSD could still have been in

the natural recovery stage, so the importantfinding is the clear superiority of the cognitivetherapy.

Bryant, Moulds, Guthrie, Dang, and Nixon(2003) compared imaginal exposure, a combi-nation of imaginal exposure and cognitive re-structuring, and a supportive counseling ther-apy with 58 civilian survivors of mixedtraumas. The participants received eight 90-minute sessions. In the combined condition,therapists began with the cognitive therapy fo-cusing on catastrophic beliefs about the traumaand skills training to challenge participants’ be-liefs. They spent 25 minutes of each session onthe cognitive component. Bryant and col-leagues found that the combination resulted inlower PTSD scores and better end-state func-tioning than the imaginal exposure conditiononly, which performed better than the support-ive psychotherapy. They concluded that the ad-dition of cognitive therapy enhanced treatmenteffectiveness.

In contrast, a recent study compared PE andPE plus present-focused cognitive restructuring(Foa et al., 2005). This study examined treat-ment in a sample of female sexual or physicalassault survivors treated in either a communityrape crisis center or an academic treatment cen-ter. Foa and her colleagues found no differencesbetween PE and PE plus cognitive restructur-ing, although both showed marked improve-ment compared to the waiting-list control.They also found no differences in the effects oftreatment based on the setting in which thetreatment was provided. In a second study thatexamined a larger range of Axis II disorders,Hembree, Cahill, and Foa (2004) examineddata from the Foa and colleagues (2005) study.A second purpose of the study was to comparethe effectiveness of expert research therapistsand community counselors on these comorbidconditions. They found that 39% of the 75participants met criteria for at least one person-ality disorder, that there were no significant dif-ferences in PTSD outcome at posttreatment,that those without personality disorders tendedto have better end-state functioning, but thatcommunity counselors tended to have betterresults than the research therapists with comor-bid clients. The main results of this study con-trast those of Bryant and colleagues (2003) inwhich the cognitive intervention enhanced theresults of exposure alone. However, it is possi-ble that the cognitive intervention in Bryantand colleagues was more trauma-focused than


the present-oriented cognitive restructuring ofthe Foa and colleagues study, and that the dif-ferent type of cognitive intervention may makea difference.

Chard (2005) expanded CPT to work withthe range of problems observed in adults whowere sexually abused as children (CPT-SA). Inaddition to the core CPT protocol, Chard pro-vided the treatment in a combination of groupand individual therapy in a 17-week protocol.The group supports the cognitive therapy andcompliance with homework, whereas the de-tailed written exposure and cognitive challeng-ing occurred in the individual therapy. Thisprotocol also added modules for examiningpreexisting family “rules” and developmentalabilities, communication skills, and social sup-port. Compared with the waiting-list controlgroup, which did not change, the CPT-SA wasvery successful. Sixty percent of the ITT and93% of the treatment completer samples remit-ted from their PTSD by posttreatment, andtheir treatment gains were maintained throughthe 1-year follow-up.

Monson and colleagues (2006) conducted awaiting-list controlled study of CPT in maleand female veterans with chronic, military-related PTSD. CPT was superior to waiting-listin reducing PTSD and comorbid symptoms;40% of the ITT sample receiving CPT no lon-ger met criteria for a PTSD diagnosis at theend of treatment. Monson and colleagues alsofound that PTSD-related disability status wasnot associated with the outcomes. This trialprovides some of the most encouraging resultsto date in the treatment of veterans withmilitary-related PTSD.

Finally, another multisite study conductedwithin the VA with women veterans comparedPE to a present-centered therapy (PCT) controlcondition (Schnurr et al., 2007). The PCT con-dition was an active treatment that providednonspecific control for attending therapy andhaving a relationship with a therapist but didnot have a specific intervention beyond focus-ing on present life circumstances. Unlike thefirst multisite VA study (Schnurr et al., 2003),which used an untested CBT group treatment,this study implemented treatment individuallywith PE, which had been established to be effi-cacious. Similar to the Monson and colleagues(2006) study, 41% of the female veterans re-mitted from their PTSD diagnosis as a result oftreatment. Schnurr and colleagues (2003) alsofound that neither having a service-connected

disability as a result of PTSD nor being a survi-vor of military sexual trauma specificallyaffected the outcomes of treatment. The PE andPCT conditions differed at posttreatment andat 3-month follow-up but not at 6-monthfollow-up on overall PTSD symptoms. Thislack of statistical difference at 6-month follow-up was not because the PE group worsened;rather, the PCT group continued to improve.These findings speak to the power of the non-specific elements of good psychotherapy. Fu-ture studies will need to focus on parsing outthe active intervention components of the ther-apeutic relationship and other nonspecific fac-tors, such as repeated assessments.

Therapist, Client, and Setting Variables

Gender and Ethnicity

According to the recent National ComorbiditySurvey Replication study (Kessler, Berglund, etal., 2005) lifetime prevalence rates of PTSD arenearly three times higher in women (9.7%)compared with men (3.6%). The recent Tolinand Foa (2006) meta-analysis of sex differencesin risk for potentially traumatic events andPTSD indicated that females were more likelythan males to meet criteria for PTSD but lesslikely to have experienced potentially traumat-ic events. Females were more likely than malesto experience sexual assault and child sexualabuse but less likely to experience accidents,nonsexual assaults, witnessing death and in-jury, disaster or fire, and war-related trauma.Within specific types of traumatic events, fe-males still exhibited greater PTSD, suggestingthat risk of exposure to particular types oftrauma only partially explains the differentialPTSD risk in males and females.

Because sexual assault is predominantly per-petrated by men, and is a highly personal andintimate crime, survivors of sexual assault of-ten distrust men. Consequently, the issue oftherapist gender can be relevant. Frequently,clients prefer or insist on a female therapist.The effectiveness of male therapists has notbeen studied specifically, but they may be quiteeffective, if well trained (Resick, Jordan,Girelli, Hutter, & Marhoeder-Dvorak, 1988).Issues for male therapists, discussed bySilverman (1977) and Koss and Harvey (1991),include the tendency for men to view rape moreas a sexual crime than as a crime of violence(Burt, 1980), therefore focusing too much on


sexual aspects of the experience and its after-math.

In treating survivors of sexual assault, re-gardless of therapist gender, it is essential thatthe therapist be knowledgeable about rape andPTSD. This includes the literature on reactionsto rape, rape myths, and attitudes about rape.Therapists bring their culturally learned per-ceptions with them, as do clients, and these caninterfere with their effectiveness if they adhereto any of the common misperceptions aboutrape (e.g., rape is primarily about sex, mostrapists are strangers, it is not rape unless thewoman actively resists). Sexual assault survi-vors are extremely sensitive to insinuations thatthey might have been to blame, for example,and many drop out of treatment if they sensethat the therapist might harbor victim-blamingattributions.

Similarly, combat veterans are often hesitantto start treatment with novice therapists whomay not know particular war details or rele-vant history. It is important for therapists to ac-knowledge their level of familiarity and workwith clients to understand more fully theirtrauma experiences and the context surround-ing them.

The role of ethnicity in CBT for PTSD hasreceived little attention in outcome research.This limitation, unfortunately, is not restrictedto PTSD treatment research; the minority-focused supplement of the Surgeon General’sreport on mental health (U.S. Department ofHealth and Human Services, 2001) made clearthat there is a paucity of empirical research ontreatment for depression and anxiety in minori-ties. The few studies of prevalence rates amongethnic groups have shown mixed results that,in part, may reflect differential rates of traumaexposure (Breslau, Davis, & Andreski, 1995;Norris, 1992).

Two program evaluation studies comparedAfrican American and European Americanmale veterans with PTSD. Rosenheck,Fontana, and Cottrol (1995) found less im-provement among African American veteranson some measures. However, Rosenheck andFontana (1996) did not support this finding.Only one study to date has examined the effi-cacy of CBT with African American womenwith PTSD. Zoellner, Feeny, Fitzgibbons, andFoa (1999) compared African American andEuropean American women who were survi-vors of either sexual or nonsexual assault.Treatment comprised PE, SIT, or a combination

of the two. There were no ethnic group differ-ences in treatment efficacy. These results wereachieved in spite of an inability to match clientsand therapists on ethnicity. Although these re-sults are encouraging, continued attention toethnic and cultural issues in treatment is impor-tant (see McNair & Neville, 1996).

Vicarious Traumatization

Working with trauma victims can have nega-tive effects on therapists that are similar toproblems experienced by their clients. Thishas been labeled “secondary” or “vicarious”traumatization. McCann and Pearlman (1990)discussed this impact as disruption of the thera-pist’s own cognitive schemas about self and theworld. Hearing clients’ traumatic experiencesmay be shocking and lead to lasting alterationsin assumptions and expectations, which in turnaffect therapists’ feelings, behaviors, and rela-tionships. Working with trauma victims maychallenge therapists’ assumptions about per-sonal invulnerability and safety, as well as be-liefs that the world is a meaningful, orderlyplace filled with trustworthy people. Accordingto McCann and Pearlman’s model, an individ-ual therapist’s reaction depends on the degreeof discrepancy between the survivor’s traumaand the therapist’s cognitive schemas. Forexample, if the therapist’s own complex experi-ences have led to the development of safetyassumptions (schemas) as central to his or herwell-being, working with trauma survivorsmay be distressing due to a heightened sense ofvulnerability. In addition, the therapist’s mem-ory system may be altered to incorporate trau-matic imagery that can become intrusive.

To counteract the effects of vicarious trau-matization, therapists should be prepared torecognize and acknowledge these effects andtake steps to deal with them. McCann andPearlman (1990) recommend the use of one’sprofessional network as a source of supportand to prevent isolation. Talking to other pro-fessionals who work with trauma survivors isespecially useful, because they can help thetherapist to recognize the effects of vicarioustraumatization and to normalize these reac-tions. Other coping strategies suggested byMcCann and Pearlman include balancing one’scaseload with trauma and nontrauma cases,engaging in other professional and personalactivities, recognizing one’s own limitations,working for social change, and focusing on the


positive personal impact of work with traumasurvivors and ways it can enrich one’s life.

“Resistance”

Clients with PTSD can be notoriously difficultdue to their ambivalence about therapy. Theywant help, but they fear confronting theirmemories and have difficulty trusting others,including therapists. They may also havestrong feelings of shame about the traumaticevent that interfere with their willingness todisclose information that they feel may lead torejection by others. Of course, it is importantto remember that avoidance behaviors, includ-ing avoiding thoughts about the trauma, arepart of the criteria for PTSD. Therefore, it is tobe expected that avoidance will also occur inthe context of treatment. No-shows are com-mon, and both subtle and obvious avoidancebehaviors are seen throughout the beginningstages of therapy. If possible, therapists shouldconsider treatment as starting on the telephoneprior to the first session. The no-show rate islikely to decrease if the therapist expresses un-derstanding of the client’s hesitance to come inand encourages attendance. Early in therapy,the therapist should describe avoidance as asymptom of PTSD and an ineffective, thoughunderstandable, means of coping. Labeling thisas “resistance” likely only increases judg-mentalness toward the client and hinders thera-pist effectiveness. This and other challenges inworking with rape trauma survivors are dis-cussed by Koss and Harvey (1991) and Kilpat-rick and Veronen (1983). Shay and Munroe(1999) discuss the challenges of working withcombat veterans with complex PTSD.

Multiply Traumatized Victims

The treatment approaches presented here havebeen shown to produce significant improve-ment in civilian trauma victims within a brieftime. Moreover, in a secondary analysis ofResick and colleagues’ (2002) trial, Resick andcolleagues (2003) showed that patients withchildhood sexual abuse experiences in additionto their adult index trauma had significant im-provements in their PTSD. In recent years,there has been a surge in evidence of treatmentsuccess in individuals with more chronic or re-peated trauma histories, such as domestic vio-lence (Kubany et al., 2004), combat (Monsonet al., 2006), and long-standing childhood sex-

ual abuse (Chard, 2005; Cloitre et al., 2004).This research points to the utility of evidence-based treatment for complex cases and de-bunks the myth that cognitive-behavioral treat-ments only work for “simple” cases of a singletrauma. That said, therapists should be awareof attending to special issues in working withthese cases. For example, some research sug-gests that PTSD in sexual assault trauma survi-vors may play a role in repeat traumatization(e.g., Kilpatrick et al., 1987), although this re-lationship appears to be complex (Wilson,Calhoun, & Bernat, 1999). Survivors of childsexual abuse may present additional challenges(Cloitre, 1998), because their traumatizationmay have interfered with processes of normaldevelopment. Given that the abuse often in-volves a relative or trusted adult, it represents aserious betrayal by someone on whom the childdepended for basic safety and protection. Theymay need more help with skills development aswell, especially interpersonal and emotion reg-ulation skills. In some cases, sexual dysfunc-tions must be addressed. This can be added toan individual treatment program or the clientmay be referred to a sex therapy specialist, butonly after the treatment for other trauma-related problems is complete. Veterans maypresent with issues specific to committing actsof violence or killing others, distrust of the gov-ernment and authority figures, ethnic/racialstereotyping, and protracted grief reactions(Monson, Price, & Ranslow, 2005).

Group Treatment

The decision whether to use a group or an indi-vidual format for treatment is usually made onthe basis of clinical judgment and practicality.There is scant research comparing the two pro-cedures. Most interventions that have beenused with trauma victims are adaptable for usein either format. Recent research has shownthat standardized group treatments may besuccessful for combat veterans and child sexualabuse trauma survivors (Chard, 2005;Creamer, Morris, Biddle, & Elliott, 1999).

Group treatment has several advantages thatmake it popular among both trauma survivorsand professionals. Koss and Harvey (1991) dis-cussed a number of these. Group treatment re-duces the sense of isolation felt by most survi-vors, who withdraw from interactions andbelieve that others cannot understand theirfeelings. It provides social support that is un-


ambiguous and nonblaming. It helps to vali-date and to normalize feelings and reactions tothe trauma. Group treatment confirms the real-ity of the traumatic experience and allows shar-ing of coping strategies. It counteracts self-blame and promotes self-esteem. Because it ismore egalitarian than individual therapy, grouptreatment can promote reempowerment anddecrease dependency. It provides a safe envi-ronment for developing attachment and inti-macy with others and an opportunity for shar-ing grief and loss. Finally, group treatment canhelp trauma survivors assign meaning to theevent, promoting cognitive processing.

Group approaches have drawbacks as well,and care should be taken to screen clients to as-sess their readiness for joining a group.McCann and Pearlman (1990) suggest that cli-ents with severe PTSD should be in individualtherapy simultaneously with group treatment,because groups may elicit strong affect andmemories that can overwhelm an unpreparedclient. For similar reasons, Resick andMarkaway (1991) warn against having groupmembers share their rape experiences duringthe first few sessions. Although important forrecovery, the sharing of “war stories” shouldbe done later in the group process or in individ-ual sessions, to avoid frightening other groupmembers or sensitizing them to other vulnera-ble situations. Poor candidates for group treat-ment, as suggested by Koss and Harvey (1991)and McCann and Pearlman (1990), are suicidalclients, those with heavy substance abuse prob-lems, self-mutilating or substance-abusing cli-ents with a borderline personality disorderdiagnosis, clients with very unstable, disorga-nized lives, and clients who have never beforespoken about the trauma or whose memory ofit is incomplete.

CASE STUDY

“Tom” is a 23-year-old single, white male whopresented for treatment approximately 1 yearafter a traumatic event that occurred during hismilitary service in Iraq. Tom received CPTwhile on active duty in the Army.

Background

Tom was born the third of four children to hisparents. He described his father as an alcoholicwho was frequently absent from the home due

to work travel prior to his parents’ divorce.Tom indicated that his father was always emo-tionally distant from the family, and especiallyafter the divorce. Tom had close relationshipswith his mother and siblings. He denied havingany significant mental health or physical healthproblems in his childhood. However, he de-scribed two significant traumatic events in hisadolescence. Specifically, he described witness-ing his best friend commit suicide by gunshotto the head. Tom indicated that this event se-verely affected him, as well as his entire Mid-western community. He went on to report thathe still felt responsible for not preventing hisfriend’s suicide. The second traumatic eventwas the death of Tom’s brother in an automo-bile accident when Tom was 17 years old. Tomdid not receive any mental health treatmentduring his childhood or after these traumaticevents, though he indicated that he began usingalcohol and illicit substances after these trau-matic events in his youth. He admitted to usingcannabis nearly daily during high school, aswell as daily use of alcohol, drinking as muchas a 24-pack of beer per day until he passedout. Tom reported that he decreased his alcoholconsumption and ceased using cannabis afterhis enlistment.

Tom served in the Infantry. He went to BasicTraining and then attended an advanced train-ing school prior to being deployed directly toIraq. While in Iraq, Tom witnessed and experi-enced a number of traumatic incidents. Hespoke about fellow soldiers who were killedand injured in service, as well as convoys thathe witnessed being hit by improvised explosivedevices (IEDs). However, the traumatic eventthat he identified as most distressing andanxiety-provoking was shooting a pregnantwoman and child.

Tom described this event as follows: Therehad been suicide bombers who detonated sev-eral bombs in the area where Tom served, and acontrol point had been set up to contain thearea. During the last few days of his deploy-ment, Tom was on patrol at this control point.It was dark outside. A car began approachingthe checkpoint, and officers on the ground sig-naled for the car to stop. The car did not stopin spite of these warnings. It continued to ap-proach the control point, entering the areawhere the next level of Infantrymen wereguarding the entrance. Per protocol, Tom fireda warning shot into the air to stop the ap-proaching car, but the car continued toward


the control point. About 25 yards from thecontrol point gate, Tom and at least one othersoldier fired upon the car several times.

After a brief period of disorientation, a cry-ing man with clothes soaked with bloodemerged from the car with his hands in the air.The man quickly fell to his knees, with hishands and head resting on the road. Tom couldhear the man sobbing. According to Tom, thesobs were guttural and full of despair. Tomlooked over to find in the pedestrian seat adead woman who was apparently pregnant. Asmall child in the backseat was also dead. Tomnever confirmed this, but he and his fellow sol-diers believed that the man crying on the roadwas the husband of the woman and the fatherof the child and fetus.

Tom was immediately upset by the event,and a Combat Stress Control unit in the fieldeventually had him air-lifted from the field be-cause of his increasing reexperiencing andhypervigilance symptoms that were beginningto have a paranoid quality. Tom was brought toa major Army hospital and received individualCPT within this setting.

Tom was administered the CAPS at pretreat-ment; his score was in the severe range, and hemet diagnostic criteria for PTSD. He also com-pleted the Beck Depression Inventory (BDI)and the State–Trait Anxiety Inventory (STAI).His depression and anxiety symptoms at pre-treatment were in the severe range. Tom wasprovided feedback about his assessment resultsin a session focused on an overview of his psy-chological assessment results and on obtaininghis informed consent for a course of CPT. Afterproviding feedback about his assessment, thetherapist gave Tom an overview of CPT, withan emphasis on its trauma-focused nature, ex-pectation of homework compliance, and theclient’s active role in getting well. Tom signed a“CPT Treatment Contract” detailing this infor-mation, and was provided a copy of the con-tract for his records. The CPT protocol beganin the next session.

Session 1

Tom arrived 15 minutes prior to his first sched-uled appointment of CPT. He sat down in thechair that the therapist gestured that he sit in,but he was immediately restless and reposi-tioned frequently. Tom quickly asked to moveto a different chair in the room, so that his backwas not facing the exterior door and his gaze

could monitor both the door and the window.He asked the therapist how long his sessionwould take and whether he would have to “feelanything.” The therapist responded that thissession would last 50–60 minutes, and that,compared to other future sessions, she wouldbe doing most of the talking. She added that, asdiscussed during the treatment contracting ses-sion, the focus would be on Tom’s feelings inreaction to the traumatic event, but that thecurrent session would focus less on this. Thetherapist also explained that she would havethe treatment manual in her lap, and would re-fer to it throughout to make sure that she deliv-ered the psychotherapy as it was prescribed.The therapist also encouraged Tom to ask anyquestions he might have as the session un-folded.

The therapist explained that at the beginningof each session, they would develop an agendafor the session. The purposes of the first ther-apy session were to (1) describe the symptomsof PTSD; (2) give Tom a framework for under-standing why these symptoms had not remit-ted; (3) present an overview of treatment tohelp Tom understand why homework comple-tion and therapy attendance were important, toelicit cooperation, and to explain the progres-sive nature of the therapy; (4) build rapport be-tween the client and therapist; and (5) give theclient an opportunity to talk briefly aboutTom’s worst traumatic event or other issues.

The therapist then proceeded to give didacticinformation about the symptoms of PTSD. Sheasked Tom to provide examples of the variousclusters of PTSD symptoms that he was experi-encing, emphasizing how reexperiencing symp-toms are related to hyperarousal symptoms,and how hyperarousal symptoms elicit a desireto avoid or become numb. The paradoxical ef-fect of avoidance in maintaining, or even in-creasing, PTSD symptoms was also discussed.Tom indicated that this was the first time some-one had specifically explained the symptoms ofPTSD to him, and put them “in motion” by de-scribing how they interact with one another.

The therapist transitioned to describe trau-ma aftereffects within an information process-ing framework. She described how traumasmay be schema-discrepant events; traumaticevents often do not fit with prior beliefs aboutoneself, others, or the world. To incorporatethis event into one’s memory, the person mayalter one’s perception of the event (assimilatethe event into an existing belief system). Exam-


ples of assimilation include looking back on theevent and believing that some other course ofaction should have been taken (“undoing” theevent) or blaming oneself because it occurred.The therapist went on to explain that Tomcould have also attempted to radically changehis prior belief system to overaccommodate forthe event. “Overaccommodation” was de-scribed as changing beliefs too much as a resultof the traumatic event (e.g., “I can’t trust my-self about anything”). She explained that sev-eral areas of beliefs are often affected by trau-ma, including safety, trust, power/control,esteem, and intimacy. The therapist describedhow people have beliefs in these areas regard-ing themselves and others, and that beliefs inboth areas could be affected. The therapist alsopointed out that if Tom had negative beliefsprior to the traumatic event relative to any ofthese topics, the event could serve to strengthenthese preexisting negative beliefs.

At this point, Tom described his childhoodand adolescent experiences, and how they hadcontributed to his premilitary trauma beliefs.The therapist noted that Tom tended to blamehimself and to internalize the bad things thathad happened in his family, and the suicide ofhis friend. She also noted his comment, “I won-der if my father drank to cope with me and mysiblings.” In Tom’s case, it seemed likely thatthe traumatic experience served to confirm fur-ther his preexisting beliefs that he had causedor contributed to bad things happening aroundand to him.

Tom then spent some time describing howdrastically things had changed after his militarytraumas. Prior to his military experiences and,specifically, the shooting of the woman andchild, Tom described himself as “proud of be-ing a soldier” and for “pulling his life to-gether.” He indicated that the military struc-ture had been very good for him in developingself-discipline and improving his self-esteem.He indicated that he felt good about “the mis-sion to end terrorism” and was proud to servehis country. He felt camaraderie with his fellowsoldiers and considered a career in the military.He denied any authority problems and in factbelieved that his commanding officers had beenrole models of the type of leader he wished tobe. Prior to his deployment to Iraq, Tom metand married his wife, and they appeared tohave a stable, intimate relationship. After hisreturn from Iraq, Tom indicated that he did nottrust anyone, especially anyone associated with

the U.S. Federal government. Tom expressedhis disillusionment with the war effort and dis-trust of the individuals who commanded hisunit. He also articulated distrust of himself: “Ialways make bad decisions when the chips aredown.” He stated that he felt completely un-safe in his environment. At times in his immedi-ate postdeployment, Tom had believed sniperson the base grounds had placed him in theircrosshairs to kill him. He indicated that heminimally tolerated being close to his wife, in-cluding sexual contact between the two ofthem.

The therapist introduced the notion of“stuck points,” or ways of making sense of thetrauma or of thinking about himself, others,and the world, getting in the way of Tom’s re-covery from the traumatic events. The therapistnoted that a large number of individuals are ex-posed to trauma. In fact, military personnel areamong the most trauma-exposed individuals.However, most people recover from their trau-ma exposure. Thus, a primary goal of the ther-apy was to figure out what had prevented Tomfrom recovering (i.e., how his thinking had gothim “stuck,” leading to the maintenance of hisPTSD symptoms).

The therapist then asked Tom to provide a 5-minute account of his worst traumatic event.Tom immediately responded, “There were somany bad things over there. How could I pickone?” The therapist asked, “Which of thoseevents do you have the most thoughts or im-ages about? Which of those events do you dis-like thinking about the most?” The therapistindicated that Tom did not need to provide afine-grained description of the event, but rathera brief overview of what happened. Tom pro-vided a quick account of the shooting of thewoman and child. The therapist praised Tomfor sharing the account with her and askedabout his feelings as a result of sharing the in-formation. Tom said that he felt anxious andwanted the session to be over. The therapistused this as an opportunity to describe the dif-ferences between “natural” and “manufac-tured” emotions.

The therapist first described “natural” emo-tions as those feelings that occur in response toevents that are normal or that occur naturally.For example, if we perceive that someone haswronged us, it is natural to feel anger. If we en-counter a threatening situation, it is natural tofeel fear. Natural emotions have a self-limitedand diminishing course. If we allow ourselves


to feel these natural emotions, they will natu-rally dissipate. The therapist used the analogyof the energy contained in a bottle of carbon-ated soda to illustrate this concept. If the top ofthe bottle is removed, the pressure initiallycomes out with some force, but that forcesubsides and eventually has no energy forth-coming. On the other hand, there are “manu-factured” emotions, or emotions that a personhas a role in making. Our thoughts contributeto the nature and course of these emotions. Themore that we “fuel” these emotions with ourself-statements, the more we can increase the“pressure” of these emotions. For example, if aperson tells himself over and over that he is astupid person and reminds himself of more andmore situations in which he perceived that hemade mistakes, then he is more likely to havemore and more anger toward himself. Thetherapist reiterated that the goals of the ther-apy were (1) to allow Tom to feel the naturalemotions that he has “stuffed,” which keephim from recovering from his trauma; and (2)to figure out how Tom was manufacturingemotions that were not helpful to him.

The therapist summarized for Tom the threemajor goals of the therapy: (1) to rememberand to accept what happened to him by notavoiding those memories and associated emo-tions; (2) to allow himself to feel his naturalemotions and let them run their course, so thememory could be put away without suchstrong feelings still attached, and (3) to balancebeliefs that had been disrupted or reinforced sothat Tom did not manufacture unhelpful emo-tions.

The therapist made a strong pitch for the im-portance of homework compliance before as-signing Tom the first practice assignment. Thetherapist told Tom that there appeared to be nobetter predictor of response to the treatmentthan how much effort a patient puts into it. Shepointed out that of the 168 hours in a week,Tom would be spending 1–2 hours of that weekin psychotherapy sessions (Note. We havefound it helpful to do twice-weekly sessions, atleast in the initial portion of the therapy, to fa-cilitate rapport building, to overcome avoid-ance, and to capitalize on early gains in thetherapy.) If Tom only spent the time during psy-chotherapy sessions focused on these issues, hewould be spending less than 1% of his week fo-cused on his recovery. To get better, the thera-pist used daily worksheets or other writing as-signments to promote needed skills into Tom’s

daily life and decrease his avoidance. The ther-apist also pointed out that at the beginning ofeach session they would review the homeworkTom has completed. The therapist asked Tom ifthis made sense, and he responded, “Sure. Itmakes sense that you get out of it what you putinto it.”

Tom’s first assignment was to write an Im-pact Statement about the meaning of the eventto determine how he had made sense of thetraumatic event, and to help him begin to de-termine what assimilation, accommodation,and overaccommodation had occurred sincethe event. Stuck points that get in the way of re-covery are identified with this first assignment.Tom was instructed to start writing the assign-ment later that day to address directly anyavoidance about completing the assignment.He was specifically reminded that this was nota trauma account (that would come later) andthat this assignment was specifically designedto get at the meaning of the event in his life,and how it had impacted his belief systems.

The specific assignment was as follows:

Please write at least one page on what itmeans to you that you that this traumatic ex-perience happened. Please consider the ef-fects that the event has had on your beliefsabout yourself, your beliefs about others,and your beliefs about the world. Also con-sider the following topics while writing youranswer: safety, trust, power/competence, es-teem, and intimacy. Bring this with you tothe next session.

Session 2

The purposes of the second session are (1) todiscuss the meaning of the event and (2) to helpTom begin to recognize thoughts, label emo-tions, and see the connection between what hesays to himself and how he feels. Tom arrivedwith obvious anger and appeared defensivethroughout most of the session. He stated thathe had been feeling quite angry all week, andthat he was “disgusted” with society and par-ticularly politicians, who were “all self-interested or pandering to those with money.”He expressed a great deal of anger over the re-ports of alleged torture at Abu Ghraib prison,which was a major news item during his ther-apy. The therapist was interested in the think-ing behind Tom’s anger about the events at AbuGhraib. However, she first reviewed Tom’s


practice assignment, writing the first ImpactStatement, to reinforce the completion ofhomework and to maintain the session struc-ture she had outlined in the first session.

The therapist asked Tom to read his ImpactStatement aloud. Clients in individual CPT arealways asked to read their homework assign-ments aloud. Should the therapist read them,the client could dissociate or otherwise avoidhis or her own reactions to their material. Tomhad written:

The reason that this traumatic event hap-pened is because I was friggin’ stupid andmade a bad decision. I killed an innocentfamily, without thinking. I murdered a man’swife and child. I can’t believe that I did it. Itook that man’s wife and child, and oh, yeah,his unborn child, too. I feel like I don’t de-serve to live, let alone have a wife and childon the way. Why should I be happy whenthat man was riddled with despair, and thatinnocent woman, child, and unborn childdied? Now, I feel like I’m totally unsafe. Idon’t feel safe even here on the hospitalgrounds, let alone in the city or back homewith my family. I feel like someone is watch-ing me and is going to snipe at me and myfamily because the terrorists had informa-tion about the situation and passed it on. Ialso don’t feel that people are safe aroundme. I might go off and hurt someone, andGod forbid it be my own family. With mywife pregnant, I am really concerned that Imight hurt her. I don’t trust anyone aroundme, and especially the government. I don’teven trust the military treating me. I alsodon’t trust myself. If I made a bad decisionat that time, who is to say that I won’t makea bad decision again? About power and con-trol, I feel completely out of control of my-self, and like the military and my command-ing officer have complete control over me.My self-esteem is in the toilet. Why wouldn’tit be given the crappy things that I havedone? I don’t think there are many positivethings that I’ve done with my life, and whenthe chips are down, I always fail and let oth-ers down. I’m not sure what other-esteem is,but I do like my wife. In fact, I don’t thinkshe deserves to have to deal with me, and Ithink they would be better without mearound. I don’t want to be close to my wife,or anyone else for that matter. It makes mewant to crawl out of my skin when my wife

touches me. I feel like I’ll never get over this.It wasn’t supposed to be like this.

The therapist asked Tom what it was like towrite and then read the Impact Statementaloud. Tom responded that it had been very dif-ficult, and that he had avoided the assignmentuntil the evening before his psychotherapy ses-sion. The therapist immediately reinforcedTom for his hard work in completing the as-signment. She also used the opportunity togently address the role of avoidance in main-taining PTSD symptoms. She asked specific So-cratic questions aimed at elucidating the dis-tress associated with anticipatory anxiety, andwondered aloud with Tom about what it wouldhave been like to have completed the home-work earlier in the week. She also asked So-cratic questions aimed at highlighting the factthat Tom felt better, not worse, after complet-ing the assignment.

Tom’s first Impact Statement and the infor-mation he shared in the first session made evi-dent the stuck points that would have to bechallenged. In CPT, areas of assimilation areprioritized as the first targets of treatment. As-similation is targeted first, because changes inthe interpretation of the event itself are inte-grally related to the other, more generalized be-liefs involved in overaccommodation. In Tom’scase, he was assimilating the event by blaminghimself. He used the term “murderer” to de-scribe his role in the event, disregarding impor-tant contextual factors that surrounded theevent. These beliefs would be the first priorityfor challenging. Tom’s overaccommodation isevident in his general distrust of society and au-thority figures, and in his belief that he madebad decisions in difficult situations. His over-accommodation is also evident in his sense ofthreat in his environment (e.g., snipers), diffi-culty being emotionally and physically intimatewith his wife, and low esteem for others andhimself.

The therapist returned to Tom’s anger aboutAbu Ghraib to get a better sense of possiblestuck points, and also to experiment withTom’s level of cognitive rigidity or openness tocognitive restructuring. The following ex-change ensued between Tom and the therapist:

THERAPIST: Earlier you mentioned that youwere feeling angry about the reports fromAbu Ghraib. Can you tell me what makesyou angry?


TOM: I can’t believe that they would do that tothose prisoners.

THERAPIST: What specifically upsets you aboutAbu Ghraib?

TOM: Haven’t you heard the reports? I can’t be-lieve that they would humiliate and hurtthem like that. Once again, the U.S. mili-tary’s use of force is unacceptable.

THERAPIST: Do you think your use of force as amember of the U.S. military was unaccept-able?

TOM: Yes. I murdered innocent civilians. I amno different than those military people atAbu Ghraib. In fact, I’m worse, because Imurdered them.

THERAPIST: “Murder.” That’s a strong word.

TOM: Yeah?

THERAPIST: From what you’ve told me, it seemslike you killed some people who may or maynot have been “innocent.” Your shooting oc-curred in a very specific place and time, andunder certain circumstances.

TOM: Yes, they died at my hands.

THERAPIST: Yes, they died, and it seems, at leastin part, because of your shooting. Does thatmake you a murderer?

TOM: Innocent people died and I pulled thetrigger. I murdered them. That’s worse thanwhat happened at Abu Ghraib.

THERAPIST: (Quietly) Really, you think it isworse?

TOM: Yes. In one case, people died, and in an-other they didn’t. Both are bad, and bothwere caused by soldiers, but I killed peopleand they didn’t.

THERAPIST: The outcomes are different. I’m cu-rious whether how it happened matters.

TOM: Huh?

THERAPIST: Does it matter what the soldiers’ in-tentions were in those situations, never mindthe outcome?

TOM: No. The bottom line is killing versus nokilling.

THERAPIST: (Realizing that there was minimalflexibility at this point) I agree that there isno changing the fact that the woman andchild died, and that your shooting had some-thing to do with that. However, I think wemight disagree on the use of the term “mur-

der.” It is clear that their deaths have been avery difficult thing for you to accept, andthat you are trying to make sense of that.The sense that you appear to have made oftheir deaths is that you are a “murderer.” Ithink this is a good example of one of thosestuck points that has prevented you from re-covering from this traumatic event. We’lldefinitely be spending more time together onunderstanding your role in their deaths. I’mnot sure “murder” is the right word to de-scribe what happened.

In addition to testing Tom’s cognitive flexi-bility, the therapist also wanted to plant theseeds of a different interpretation of the event.She was careful not to push too far and re-treated when it was clear that Tom was notamenable to an alternative interpretation. Hewas already defensive and somewhat angry,and she did not want exacerbate his defensive-ness or possibly contribute to dropout from thetherapy.

From there, the therapist described how im-portant it was to be able to label emotions andto begin to identify what Tom was saying tohimself. The therapist and Tom discussed howdifferent interpretations of events can lead tovery different emotional reactions. They gener-ated several examples of how changes inthoughts result in different feelings. The thera-pist also reminded Tom that some interpreta-tions and reactions follow naturally from situa-tions and do not need to be altered. Forexample, Tom indicated that he was saddenedby the death of the family; the therapist did notchallenge that statement. She encouraged Tomto feel his sadness and to let it run its course.He recognized that he had lost something, andit was perfectly natural to feel sad as a result.At this point Tom responded, “I don’t like tofeel sad. In fact, I don’t like to feel at all. I’mafraid I’ll go crazy.” The therapist gently chal-lenged this belief. “Have you ever allowedyourself to feel sad?” Tom responded that heworked very hard to avoid any and all feelings.The therapist encouraged Tom. “Well, giventhat you don’t have much experience with feel-ing your feelings, we don’t know that you’regoing to go crazy if you feel your feelings,right?” She also asked him whether he had no-ticed anyone in his life who had felt sad, andhad not gone crazy. He laughed. The therapistadded, “Not feeling your feelings hasn’t been


working for you so far. This is your opportu-nity to experiment with feeling these very natu-ral feelings about the traumatic event to seewhether it can help you recover now from whathas happened.”

Tom was given a number of A-B-C Sheets ashomework to begin to identify what he wastelling himself and his resulting emotions. Inthe first column under A, “Something hap-pens,” Tom was instructed to write down anevent. Under the middle column, B, “I tell my-self something,” he was asked to record histhoughts about the event. Under column C, “Ifeel and/or do something,” Tom was asked towrite down his behavioral and emotional re-sponses to the event. The therapist pointed outthat if Tom says something to himself a lot, itbecomes automatic. After a while, he does notneed to think the thought consciously, he cango straight to the feeling. It is important to stopand recognize automatic thoughts to decidewhether they either make sense or should bechallenged and changed.

Session 3

Tom handed the therapist his homework assoon as he arrived. The therapist went over theindividual A-B-C Sheets that Tom had com-pleted and emphasized that he had done a goodjob in identifying his feelings and recognizinghis thoughts. Some of the homework is shownin Figure 2.1.

The purpose of reviewing this homework atthis point in the therapy is to identify thoughtsand feelings, not to heavily challenge the con-tent of those thoughts. The therapist did a mi-nor correction of Tom’s identification of thethought “I feel like I’m a bad person” (boldedin Figure 2.1) as a feeling. She commented thatfeelings are almost always one word and whatyou feel in your “gut,” and that adding thestem “I feel . . . ” does not necessarily make it afeeling. The therapist noticed the pattern ofthoughts that Tom tended to record (i.e., inter-nalizing and self-blaming), as well as the char-acteristic emotions he reported.

The therapist noted the themes of assimilationthat again emerged (i.e., self-blame) and chose tofocus on mildly challenging these relatedthoughts. She specifically chose to focus onTom’s thoughts and feelings related to his wife’spregnancy, which ultimately seemed to be re-lated tohis assimilationof the traumatic event.

THERAPIST: You don’t think you deserve tohave a family? Why is that?

TOM: Why should I get to have a family when Itook someone else’s away?

THERAPIST: OK, so it sounds like this relates tothe first thought that you wrote down on theA-B-C Sheet about being a murderer. Whenyou say to yourself, “I took someone else’sfamily away,” how do you feel?

TOM: I feel bad.

THERAPIST: Let’s see if we can be a bit more pre-cise. What brand of bad do you feel? Re-member how we talked about the primarycolors of emotion? Which of those mightyou feel?

TOM: I feel so angry at myself for doing what Idid.

THERAPIST: OK. Let’s write that down—angerat self. So, I’m curious, Tom, do the otherpeople that you’ve told about this situation,or who were there at the time, think whatyou did was wrong?

TOM: No, but they weren’t the ones who did it,and they don’t care about the Iraqi peoplelike I do.

THERAPIST: Hmm . . . that makes me thinkabout something, Tom. I’m curious, in thecombat zone in which you were involved inIraq, how easy was it to determine who youwere fighting?

TOM: Not always particularly easy. There werelots of insurgents who looked like everydaypeople.

THERAPIST: Like civilians? Innocent civilians?(pause)

TOM: I see where you are going. I feel like it isstill wrong because they died.

THERAPIST: I believe you when you say that itfeels that way. However, feeling a certainway doesn’t necessarily mean that it is basedon the facts or the truth. We’re going towork together on seeing whether that feelingof guilt or wrongdoing makes sense when welook at the situation very carefully in ourwork together.

Because the goal is for Tom to challenge anddismantle his own beliefs, the therapist probedand planted seeds for alternative interpreta-tions of the traumatic event but did not pursue


94

AC

TIV

AT

ING

EV

EN

TB

EL

IEF

CO

NS

EQ

UE

NC

E

AB

C

“Som

ethi

ngha

ppen

s”“I

tell

mys

elf

som

ethi

ng”

“Ife

elso

met

hing

”

Iki

lled

anin

noce

ntfa

mily

.“I

ama

mur

dere

r.”

Ife

ellik

eI’m

ab

adp

erso

n.

Avo

idta

lkin

gab

out

it.

My

wife

ispr

egna

nt.

“Ido

n’t

dese

rve

toha

vea

fam

ily.”

Gui

lty

Abu

Ghr

aib

“The

gove

rnm

ent

suck

s.”

Ang

ry

Goi

ngto

ther

apy

“I’m

wea

k.I

shou

ldn’

tha

veP

TS

D.

PT

SD

ison

lyfo

rth

ew

eak.

”A

ngry

Are

my

thou

ghts

inB

real

istic

?

Yes.

Wha

tca

nyo

ute

llyo

urse

lfon

such

occa

sion

sin

the

futu

re?

?

FIG

UR

E2.

1.A

-B-C

Shee

t.

the matter too far. Although Tom did movesome from his extreme stance within the ses-sion, the therapist was not expecting any dra-matic changes. She focused mostly on beingsupportive, building rapport, and helping Tomget the connections among thoughts, feelings,and behaviors.

The therapist praised Tom for his ability torecognize and label thoughts and feelings, andsaid that she wanted Tom to attend to bothduring the next assignment, which was writingabout the traumatic event. Tom was asked towrite as homework a detailed account of theevent, and to include as many sensory details aspossible. He was asked to include his thoughtsand feelings during the event. He was in-structed to start as soon as possible on the as-signment, preferably that day, and to pick atime and place where he would have privacyand could allow himself to experience his natu-ral emotions. Wherever he had to stop writinghis account of the event, he was asked to drawa line. (The place where the client stops is oftenthe location of a stuck point in the event, wherethe client gave up fighting, where somethingparticularly heinous occurred, etc.) Tom wasalso instructed to read the account to himselfevery day until his next session. The therapistpredicted that Tom would want to avoid writ-ing the account, and procrastinate until as lateas possible. She asked him why it would be im-portant for him to do the assignment, and do itas soon as possible. This was a technique to de-termine how much Tom was able to recountthe rationale for the therapy, and to strengthenhis resolve to overcome avoidance. Tom re-sponded that he needed to stop avoiding, or hewould remain scared of his memory. The thera-pist added that the assignment was to help Tomget his full memory back, to feel his emotionsabout it, and for the therapist and client to be-gin to look for stuck points. She also reassuredTom that although doing so could be difficultfor a relatively brief period of time, it wouldnot continue to be so intense, and he wouldsoon be over the hardest part of the therapy.

Session 4

During the settling-in portion of the session,Tom indicated that he had written the accountof the event the evening before, although hehad thought about and dreaded it every dayprior to that. He admitted that he had beenavoidant due to his anxiety. The therapist

asked Tom to read his account aloud to her. Be-fore starting, Tom asked why it was importantto read it in the session. The therapist remindedTom of what they had talked about the previ-ous session, and added that the act of readingaloud would help him to access the wholememory and his feelings about it. Tom readwhat he wrote quickly, like a police report, andwithout much feeling:

There were several of us who were assignedto guard a checkpoint south of Baghdad. Wewere there because insurgents were begin-ning to take over the particular area, and wewere there to contain the area. I was placedon top of the checkpoint, about 10 feet in theair. It was dusk. It had been a fairly routineday, with people coming through the check-point like they were going through a tollbooth. Off in the distance I noticed a small,dark car that was going faster than mostcars. I could tell it was going faster, becausethere was more sand smoke kicking up be-hind it. Men out in front of the checkpointwere motioning for the car to slow down,but it didn’t seem to be slowing down. Some-one shot into the air to warn them, but theykept on coming. I could see two heads in thecar coming toward us. We had been told toshoot at any vehicle that came within 25yards of the gate to protect those around thegate, and the area beyond the gate. The carkept coming. I shot a bunch of rounds at thecar.

At least one other person shot, too. Therewas so much chaos after that. I rememberfeeling my gun in my hand as I stood there.After a few moments, I also remember mylegs carrying me down to the car. I don’t re-ally remember how I got there, but I did.Several men had surrounded the car, and aman got out of it. The man was crying. No,sobbing. He was speaking fast while hecried. He turned toward the car, resisting themen who attempted to remove him from thescene. I turned to see what the man waslooking at and saw them for the first time. Isaw the woman first.

There was blood everywhere, and her facehad been shot. Then I saw the little girl in thebackseat slumped over, holding a doll. Therewas blood all over her too. I saw the gun-shots through the car. I looked back at the


woman, but avoided looking at her face. Isaw a bump under her dress. She was preg-nant.

I don’t remember much else after that. Iknow I went back to camp and basically fellapart. They took me off duty for a couple ofdays, but eventually they sent me back homebecause I was such a mess.

After reading the account, Tom quicklyplaced it in his binder of materials and closedthe binder as if to indicate that he was ready tomove onto something else. The therapist askedTom what he was feeling, and he indicated thathe was feeling “nothing” The therapist fol-lowed up, saying, “Nothing at all?” Tom reluc-tantly admitted that he was feeling anxious.The therapist then asked him to read the ac-count again, but this time to slow down hisreading rate, and allow himself to experiencethe emotions he had felt at the time of theevent.

After reading the account for the secondtime, the therapist sought to flush out details ofthe event that Tom had “glossed” over and tofocus on what appeared to be the most difficultaspects of the situation.

THERAPIST: What part of what you just read tome is the most difficult?

TOM: It is all difficult. The whole thing is horri-ble.

THERAPIST: What is the worst of it, though?

TOM: I guess the worst of it is seeing that smallgirl in the backseat of the car.

THERAPIST: What did she look like when yousaw her?

(Tom describes his memory of the girl when hearrived at the car.)

THERAPIST: Therapist: What are you feelingright now?

TOM: I feel sick to my stomach. I feel like I didat the time—that I want to throw up. I amalso disgusted and sad. I killed an innocentchild. There are so many things I couldhave done differently not to have taken herlife.

(The therapist is aware of the assimilation pro-cess in Tom’s use of hindsight bias. She storesthat information away for future reference, be-cause she wants to make sure that Tom is feel-

ing strongly as many of his natural emotions aspossible about the traumatic event.)

THERAPIST: Continue to feel those feelings.Don’t run away from them. Anything elsethat you’re feeling?

TOM: I feel mad at myself and guilty.

THERAPIST: Were you feeling mad at yourselfand guilty at the time?

TOM: No. I was horrified.

THERAPIST: OK, let’s stay with that feeling.

TOM: (Pauses.) I don’t want to feel this any-more.

THERAPIST: I know you don’t want to feel thisanymore. You’re doing a great job of notavoiding your feelings here. In order to notfeel like this for a long time, you need to feelthese absolutely natural feelings. Let themrun their course. They’ll decrease if you staywith them.

After a period in which Tom experienced hisfeelings related to the situation and allowedthem to dissipate, a discussion ensued regard-ing how hurtful it was to Tom to hear otherpeople’s reaction to the war. He expressed spe-cific frustration with the presidential adminis-tration and its policy on the war. The therapistgently redirected Tom’s more philosophical dis-cussion of international policy to the effects ofthe trauma on him. He then told a story of howhe had shared his traumatic experience with ahigh school friend. Tom felt that this personhad a negative reaction to him as a result ofsharing the story. Tom felt judged and unsup-ported by this friend. Since this experience withhis friend, Tom had refrained from telling oth-ers about his combat experience. Using So-cratic questioning, the therapist asked Tom ifthere might be any reason, outside of him, thatsomeone might have a negative reaction tohearing about the shooting. Through this ex-change, Tom was able to recognize that whenothers hear about traumatic events, they alsoare trying to make sense of these experiences inlight of their existing belief systems. In otherwords, others around him might fall prey to the“just world” belief that bad things only happento bad people. They also might not take intoaccount the entire context in which Tom shotthe passengers in the car. This recognition re-sulted in Tom feeling less angry at his friend forthis perceived judgment. He was also some-what willing to admit that his interpretation of


his friend’s reaction might have been skewed byhis own judgment of himself. In fact, laterin the therapy, when Tom was able to ask hisfriend directly about his reaction, the friendindicated that it had been hard for him tohear, but that he had not been judging Tomat all. In actuality, he was thinking about theterrible predicament Tom had endured at thetime.

The therapist asked Tom what stuck pointshe had identified in writing and reading his ac-count. The following dialogue then occurred:

TOM: I’m not sure what the stuck points are,but from what you’ve been asking me, Iguess you question whether or not I mur-dered this family.

THERAPIST: That’s true. I think it is worthwhilefor us to discuss the differences betweenblame and responsibility. Let’s start with re-sponsibility. From your account, it soundslike you were responsible for shooting thefamily. It sounds like other people may havebeen responsible, too, given that you werenot the only person who shot at them.

(The therapist stores this fact in her mind tochallenge Tom later about the appropriatenessof his actions. This also provides a good oppor-tunity to reinforce Tom for performing well ina stressful situation.)

The bottom line is that responsibility isabout your behavior causing a certain out-come. Blame has to do with your intention-ality. It has to do with your motivations atthe time. In this case, did you go into the sit-uation with the motivation and intention tokill a family?

TOM: No, but the outcome was that they weremurdered.

THERAPIST: Some died. From what you’veshared, if we put ourselves back into the situ-ation at the time, it was not at all your inten-tion for them to die. They were barrelingdown the road, not responding to the veryclear efforts to warn them to stop. Your ownand others’ intentions were to get them tostop at the checkpoint. Your intention at thetime did not seem to be to kill them. In fact,wasn’t your intention quite the opposite?

TOM: Yes (begins to cry).

THERAPIST: (Pauses until Tom’s crying subsidessomewhat.) Your intention was not to killthem at all. Thus, the word “blame” is not

appropriate. Murder or considering yourselfa murderer is not appropriate in this situa-tion. The reason I’ve questioned the term“murder” or “murderer” all along was be-cause it doesn’t seem like your intention atall was to have to shoot at them.

TOM: But why do I feel like I am to blame?

THERAPIST: That’s a good question. What’syour best guess about why that is?

TOM: (Still crying) If someone dies, someoneshould take responsibility.

THERAPIST: Do you think it is possible to takeresponsibility without being to blame? Whatwould be a better word for a situation that isyour responsibility, but that you didn’t in-tend to happen? If a person shot someonebut didn’t intend to do that, what would wecall that?

TOM: An accident, I guess.

THERAPIST: That’s right. In fact, what wouldyou call shooting a person when you are try-ing to protect something or someone?

TOM: Self-defense.

THERAPIST: Yes, very good. Weren’t you re-sponsible for guarding the checkpoint?

TOM: Yeah.

THERAPIST: So, if you were responsible forguarding that checkpoint, and they contin-ued through, wouldn’t that have put the areaat risk?

TOM: Yes, but it was a family—not insurgents.

THERAPIST: How do you know that?

TOM: There was woman and child in the car.

THERAPIST: Did you know that at the time?

TOM: No.

THERAPIST: So only in hindsight do you knowthat it was a family that might have had nobad intention. We actually don’t know thefamily’s intention, do we? They didn’t heedthe several warnings, right?

TOM: Yes. (pauses) I hadn’t thought that theywould be looking to do something bad witha woman and child in the car.

THERAPIST: We don’t know, and won’t know,bottom line. However, what we do know iswhat you knew at the time. What you knewat the time is that they did not heed thewarnings, that you were responsible for se-curing the checkpoint, and that you took ac-tion when you needed to take action to pro-


tect the post. Thinking about those facts ofwhat happened and what you knew at thetime, how do you feel?

TOM: Hmm . . . I guess I’d feel less guilty.

THERAPIST: Therapist: You’d feel less guilty, oryou feel less guilty?

TOM: When I think through it, I do feel lessguilty.

THERAPIST: There may be points when youstart feeling guiltier again. It will be impor-tant for you to hold onto the facts of whathappened versus going to your automatic in-terpretation that you’ve had for awhile now.Is there any part of it that makes you proud?

TOM: Proud?

THERAPIST: Yes. It seems like you did exactlywhat you were supposed to do in a stressfulsituation. Didn’t you show courage underfire?

TOM: It’s hard for me to consider my killingthem as courageous.

THERAPIST: Sure. You haven’t been thinkingabout it in this way before, but it is some-thing to consider.

The therapist’s Socratic questioning was de-signed to help Tom consider the entire contextin which he was operating. She also began toplant seeds that Tom not only did nothingwrong but he also did what he was supposed todo to protect the checkpoint. Whenever possi-ble, pointing out acts of heroism or couragecan be powerful interventions with trauma sur-vivors.

Prior to ending the session, the therapistchecked Tom’s emotional state to make sure hewas calmer than he had been during the ses-sion. She also inquired about his reaction to thetherapy session. He commented that it hadbeen very difficult, but that he felt better thanhe expected in going into the “nitty-gritty” ofwhat happened. He also noted that there werethings that he had not considered about theevent that were “food for thought.” The thera-pist praised Tom for doing a great job on thewriting assignment and reinforced the impor-tance of not quitting now. She commented thathe had completed one of the hardest steps ofthe therapy, which would help him recover.

The therapist took the first account of thetrauma and gave Tom his next homework as-signment: to write the entire account again.

The therapist asked Tom to add any details hemight have left out of the first account and toprovide even more sensory details. She alsoasked him to record any thoughts and feelingsthat he was having in the here and now in pa-rentheses, along with his thoughts and feelingsat the time of the event.

Session 5

Tom arrived at Session 5 looking brighter andmaking more eye contact with the therapist. Heindicated that he had written the accountagain, right after the previous session. He com-mented that the writing was hard, but not ashard as the first time. The therapist used this asan opportunity to reinforce how natural emo-tions resolve naturally as they are allowed ex-pression. Tom noted that he had talked with hiswife more this week, avoiding her less. Theirincreased communication allowed Tom’s wifeto express her concerns about Tom’s well-being. She shared that he seemed disinterestedin her and in their unborn child. Tom had pre-viously told his wife about the incident, but hehad not shared the specific detail that thewoman in the vehicle appeared to be pregnant.Tom perceived his wife as having a very goodreaction to his disclosure about the pregnantwoman. He noted that she asked him ques-tions, and that her comments indicated that shedid not blame him for his actions. For example,she asked, “How could you have known at thetime that it was a family?” She also reportedlysaid, “It’s hard to know with terrorism if theywere actually just a family traveling.” Tomlaughed when he reported that their conversa-tion sounded like his last psychotherapy ses-sion.

The therapist asked Tom to read his secondaccount out loud, with as many emotions aspossible. Tom had written more about theevent, and the therapist noted that he had in-cluded more information about what he andthe other guards had done to warn the passen-gers in the car to slow down for the checkpoint.Tom read the second account more slowly andwas not as tense as he had been the first time heread aloud. Tom’s second essay included muchmore detail and focused more on the vehicleand its occupants after he had fired upon them.

THERAPIST: I notice that you wrote more aboutthe car and the family. What are you feelingabout that right now?


TOM: I feel sad.

THERAPIST: Do you feel as sad as you felt thefirst time you wrote about it?

TOM: I think I may feel sadder about it now.

THERAPIST: Hmm . . . Why do you think thatis?

TOM: I think it’s like what I wrote in the paren-thesis about what I’m thinking now. Now,instead of feeling so much guilt that I shotthem, I think it’s sad that they didn’t heed thewarnings.

THERAPIST: You mentioned that you’re feelingless guilt now. Why is that?

TOM: I’m beginning to realize that I was notthe only one there that was trying to stopthem. Several of us were trying to get themto stop. There is still some guilt that I wasthe one who shot them.

THERAPIST: If one of the other guards had shotthem, would you blame him or her for theshooting? Would you expect him or her tofeel guilty for their behavior?

TOM: (Laughs.) I started thinking about thatthis week. It made me wonder if it was reallyme who shot them. As I was writing andthinking about it more, I realized that thereis a possibility that another of the guardsmay have been shooting at the same time.

THERAPIST: What would it mean if he or shewas shooting at the same time?

TOM: If he was shooting at the same time, itmeans that he thought that shooting at themmight be the right thing to do in that situa-tion.

THERAPIST: Might have been the right thing todo?

TOM: (Smiling) Yeah, I still have questions thatwe might have been able to do somethingelse.

THERAPIST: It seems like you’re still trying to“undo” what happened. I’m curious, whatelse could you have done?

TOM: Not have shot at them.

THERAPIST: Then what would have happened?

TOM: They might have stopped. (pauses) Or Iguess they could have gone through thecheckpoint and hurt other people past thecheckpoint. I guess they could have also beenequipped with a car bomb that could havehurt many other people. That seems hard to

believe, though, because of the woman andchild in the car.

THERAPIST: It is impossible for us to know theirintentions, as we discussed before. The bot-tom line is that you’ve tended to assume thatdoing something different, or doing nothing,would have led to a better outcome.

TOM: That is true. I still feel sad.

THERAPIST: Sure you do—that’s natural. I takeit as a good sign that you feel sad. Sadnessseems like a very natural and appropriate re-action to what happened—much more con-sistent with what happened than the guiltand self-blame that you’ve been manufactur-ing.

Tom and the therapist discussed how thegoal of the therapy was not to forget what hadhappened, but to have the memory without allof the anxiety, guilt, and other negative emo-tions attached to it. Tom indicated that he wasbecoming less afraid and more able to toleratehis feelings, even when they were intense. Tomacknowledged that reading his account, talkingabout his trauma, and coming to psychothera-py sessions were becoming easier and that hisnegative feelings were beginning to diminish.

After discussing Tom’s reactions to his mem-ories, with a focus on how he had attempted toassimilate the memory into his existing beliefs,the therapist began to discuss areas of over-accommodation. One area of overaccommo-dation was Tom’s beliefs about the U.S. mili-tary. He had entered the service with a verypositive view of the military. Tom had a familyhistory of military service and believed in ser-vice to the country and the “rightfulness” ofthe military. Subsequent to his traumatic eventand military service in Iraq, he developed anegative view of the military that had extendedto the Federal government in general. The ther-apist used this content to introduce the first se-ries of tools to help challenge Tom’s stuckpoints. She also emphasized how he wouldgradually be taking over as his own therapist,capable of challenging his own dysfunctionalpatterns of thinking that kept him “stuck.”

THERAPIST: It seems that you have some verystrong beliefs about the military and the U.S.government since your service. I’d like to usethose beliefs to introduce some new materialthat will be helpful to you in starting to chal-lenge stuck points on your own. You’ve done


an outstanding job of considering the waythat you think and feel about things. You’vebeen very open to considering alternative in-terpretations of things. Starting in this ses-sion, I’m going to help you to become yourown therapist and to attack your own stuckpoints directly.

TOM: OK.

THERAPIST: Today we will cover the first set ofskills. We’re going to be building your skillsover the next few sessions. The first tool isa sheet called the Challenging QuestionsSheet. Our first step is to identify a single be-lief that you have that may be a stuck point.As I mentioned before, I’d like us to use yourbeliefs about the Federal government now.So, if you were to boil down what you be-lieve about the Federal government or themilitary, what is it?

TOM: I don’t know. I’m not sure. I guess I’d saythat the U.S. military is extremely corrupt.

THERAPIST: Good. That is very clear and to thepoint. So let’s go over these questions andanswer them as they relate to this belief. Thefirst question you ask yourself is, “What’sthe evidence for and against this idea?”

TOM: The evidence for this is Abu Ghraib. Canyou believe that they would do that? I wouldhave also put my own shooting under the“for” list, but I’m beginning to questionthat.

THERAPIST: What other evidence is there of cor-ruption?

TOM: Oh, and these defense contractors . . .what a scam! That leads me to the currentadministration and its vested interests in go-ing to war to make money on defense con-tracting. And, oh, of course, to make moneyon the oil coming out of these countries!

THERAPIST: OK. Sounds like you have some“for” evidence. What about the “against”evidence?

TOM: Well, some of my fellow soldiers werevery good. They were very committed intheir service and to the mission. I also hadmostly good leaders, although some of themwere real pigs. Some were really power-hungry a- -holes, frankly.

THERAPIST: So, it sounds like you have somepros and cons that support your belief thatthe U.S. military is completely corrupt. In theprocess of changing, it is not uncommon to

have thoughts on both sides. That is greatnews! It means that you are considering dif-ferent alternatives, and are not “stuck” onone way of seeing things. Let’s take the nextone. . . .

The therapist spent the balance of the sessiongoing over the list of questions to make surethat Tom understood them. Although most ofthe questions focused on the issue of corrup-tion in the military, other issues were alsobrought in to illustrate the meaning of thequestions. For example, the therapist intro-duced the probability questions with the exam-ple from Tom’s life, in which he was shot by asniper. These questions are best illustrated withregard to issues of safety. The therapist pointedout that perhaps not all of the questions ap-plied to the belief on which Tom was working.The question “Are you thinking in all-or-noneterms?” seemed to resonate with Tom the most,because it applied to his belief about the mili-tary. He commented that he was applying a fewexamples of what seemed to be corruption tothe entire military. Tom also indicated that hisdescription of the military as “extremely” cor-rupt was consistent with the question “Are youusing words or phrases that are extreme or ex-aggerated?” Indicative of his grasp of theworksheet, Tom also noticed that the question“Are you taking selected examples out of con-text?” applied to his prior view of his behavioras a murder in the traumatic event.

For homework prior to Session 6, Tomagreed to complete one Challenging QuestionsSheet each day. He and the therapist brain-stormed about potential stuck points prior tothe end of the session to facilitate homeworkcompliance. These stuck points included “Idon’t deserve to have a family,” “I murdered aninnocent family,” and “I am weak because Ihave PTSD.”

Session 6

Tom completed Challenging Questions Sheetsabout all of the stuck points he and the therapisthad discussed. The therapist reviewed theseworksheets to determine whether Tom had usedthe questions as designed. She asked Tom whichof the worksheets he had found least helpful. Heresponded that he had had the most difficultycompleting the sheet about deserving to have afamily. The therapist then reviewed this sheet indetail with Tom (see Figure 2.2).


THERAPIST: So, I notice that in your answerabout the evidence for and against this ideaabout deserving a family, you included as ev-idence that you took some other man’s fam-ily. I’m glad to see that you didn’t include theword “murder”—that’s progress. But, howis that evidence for you not deserving a fam-ily?

TOM: It is evidence, because I feel like I tooksomeone else’s; therefore, I don’t deserve onefor myself. It seems fair.

THERAPIST: Remind me to make sure and look

what you put for item 9 about confusingfeelings and facts. For now, though, help meunderstand the math of why you don’t de-serve your family, and your happiness aboutyour family, because of what happened?

TOM: I don’t know—it just seems fair.

THERAPIST: Fair? That implies that you didsomething bad that requires you to be pun-ished.

TOM: As I’ve been thinking about it more, Idon’t think I did something wrong when I re-ally look at it, but it still feels like I did some-


Challenging Questions Sheet

Below is a list of questions to be used in helping you challenge your maladaptive or problematic beliefs.Not all questions will be appropriate for the belief you choose to challenge. Answer as many questionsas you can for the belief you have chosen to challenge below.

Belief: I don’t deserve to have a family.1. What is the evidence for and against this idea?

FOR: I took some other man’s family.

AGAINST: I didn’t want to have to shoot anyone. An “eye for an eye” does not apply here.

2. Is your belief a habit or based on facts?It is a habit for me to think this way. The facts are that I didn’t do something wrong to deserve to bepunished in this way.

3. Are your interpretations of the situation too far removed from reality to be accurate?My interpretation of the original situation has been fairly unrealistic, which is where I get this belief.

4. Are you thinking in all-or-none terms?N/A

5. Are you using words or phrases that are extreme or exaggerated? (i.e., always, forever, never, need,should, must, can’t, and every time)I guess maybe “deserve” could be an extreme word.

6. Are you taking the situation out of context and only focusing on one aspect of the event?Yes, like #3, I tend to forget what all was going on at the time of my shooting.

7. Is the source of information reliable?No, I’m not very reliable these days.

8. Are you confusing a low probability with a high probability?N/A

9. Are your judgments based on feelings rather than facts?I’m feeling guilty like I did something wrong when the truth is that I did what I was supposed to do.

10. Are you focused on irrelevant factors?Maybe my deserving a family has nothing to do with someone else losing theirs?

FIGURE 2.2. Challenging Questions Sheet.

thing wrong and that I shouldn’t have some-thing good like a wife and child in my life.

THERAPIST: Maybe we should look at your re-sponse to item 9 now. What did you put inresponse to the question “Are your judg-ments based on feelings rather than facts?”

TOM: I said that they were. I wrote, “I’m feel-ing guilty like I did something wrong whenthe truth is that I did what I was supposed todo.” I try to remember what we talkedabout, and what my wife also has said to meabout them not responding to the warningsand my shooting them, which may have pre-vented something else that was bad. I stillfeel bad—not as bad as I did—but I still feellike I did something wrong.

(The therapist uses this as an opportunity totalk about the need for practicing new alter-native thoughts in order to elicit emotionalchange.)

THERAPIST: You are well on your way, Tom, togetting unstuck and recovering. Your head isstarting to get it, and your feelings need tocatch up. You’ve been thinking about whathappened and what you did in a certain wayfor awhile now. You blamed yourself overand over and over again, telling yourself thatyou did something wrong. You gave yourselfa steady diet of that type of thinking, whichresulted in you feeling guilty about whathappened. It is like a well-worn rut of think-ing in your brain that automatically leadsyou down the path of feeling guilty. Whatyou need to do now is start a new road ofmore realistic and truthful thinking aboutthe situation that will eventually be a well-worn path. What is the more realistic view ofyour role in this event?

TOM: (Tearfully) I had to shoot at the car, andpeople died.

THERAPIST: That’s right. And, let’s pretend thatyou really do believe that thought. If so,what would you feel?

TOM: I’d feel so much lighter. I wouldn’t feelguilty. I’d continue to feel sad about this hor-rible situation, but I wouldn’t blame myself.

THERAPIST: Let’s take it the next step. If youdidn’t blame yourself and feel guilty, thenwould you believe that you deserve to behappy with your wife and the baby that willsoon be here?

TOM: Sure.

THERAPIST: So, Tom, your work is to practice,practice, practice this new and more accu-rate way of looking at what happened andyour role in it. With practice, your feelingswill start matching the truth about whathappened and the fact that you are not toblame.

TOM: It is kind of like training to use a weapon.They made us do certain things with ourguns over and over and over again, until itwas automatic. It was very automatic afterawhile.

THERAPIST: That’s right. There are other ques-tions on this sheet that might be helpful inconvincing you of the truth about this inyour practice. What did you put for thequestion “Is your belief a habit or based on afact?”

This dialogue illustrates a common occur-rence at this stage in the therapy. Tom wasstarting to experience cognitive change but hisemotional change was lagging. The therapistreinforced the need to practice the new ways ofthinking to feel different. It is also important tohighlight clients’ gains in changing their think-ing, even if their feelings have not changed orare ambivalent. A change in thinking is framedas more than halfway to a change in feeling. Ineffect, changed thinking involves competingthoughts or learning, and with more repetitionsof the new thought, the associated feeling fol-low and eventually win out.

In the latter portion of this session the thera-pist introduced the Patterns of ProblematicThinking Sheet and provided an explanation ofhow this list was different from the Challeng-ing Questions Sheet (see Figure 2.3). More spe-cifically, she indicated that the Patterns ofProblematic Thinking Sheet pertains to moregeneral patterns of thinking versus challengingindividual thoughts that Tom might have. ThePatterns of Problematic Thinking Sheet listsseven types of faulty thinking patterns (e.g.,oversimplifying, overgeneralizing, and emo-tional reasoning). Tom and therapist wentthrough the list and generated examples foreach of the patterns. For example, for “Disre-garding important aspects of a situation,” thetherapist pointed out something that Tom hadbrought up several times during therapy. Ini-tially Tom had not included the important in-formation that he and the other guards had at-tempted to stop the car before shooting at it.


She also pointed out that emotional reasoningwas similar to confusing a feeling with a fact,which had been a primary focus of the session.

When they got to the item “Overgeneralizingfrom a single incident,” Tom said that hehad noticed he was beginning to change histhoughts about the government and its leaders.He commented that it had been very powerfulfor him to consider that in a number of in-stances his fellow soldiers had operated withintegrity and were committed to the mission,and to the safety and protection of others. Tomsaid spontaneously, “I guess that is also kind oflike drawing conclusions when evidence islacking or even contradictory.” He said that hehad started stereotyping after the traumaticevent—applying negative attributes and opin-ions to everyone in the military and the govern-ment too broadly. Tom and the therapist dis-cussed how the goal of the therapy was to have

a balanced and realistic view of things versusan overly ideal version he had pretrauma or theoverly pessimistic version he had posttrauma.In other words, the goal was to find shades ofgray and balance in his thinking about the gov-ernment, the military, and their leadership.Tom added an example of this thinking: “Thereare at least some people in government whowant to do good for others.”

Tom was given the homework assignment toread over the list in the Patterns of ProblematicThinking Sheet and to note examples of timeshe used each of the problematic thinking pat-terns.

Session 7

Tom began the session stating that he was feel-ing better, and that his wife had also noted adifference in him and was feeling less con-


Patterns of Problematic Thinking

Listed below are several types of patterns of problematic thinking that people use in different lifesituations. These patterns often become automatic, habitual thoughts that cause us to engage in self-defeating behavior. Considering your own stuck points, find examples for each of these patterns. Write inthe stuck point under the appropriate pattern and describe how it fits that pattern. Think about how thatpattern affects you.

1. Jumping to conclusions when the evidence is lacking or even contradictory.I tend to jump to the conclusion that I have done something wrong when bad things happen. Iassume things are my fault.

2. Exaggerating or minimizing a situation (blowing things way out of proportion or shrinking theirimportance inappropriately).I minimize the things that I have done well in the military.

3. Disregarding important aspects of a situation.In the past I have tended to neglect the important aspect that several of us tried to stop the car fromgoing through the checkpoint.

4. Oversimplifying things as good–bad or right–wrong.I can sometimes think of all Iraqis as all bad.

5. Overgeneralizing from a single incident (a negative event is seen as a never-ending pattern).I have assumed that because of my traumatic event, I could not be safe with my baby to be born.

6. Mind reading (you assume people are thinking negatively of you when there is no definite evidencefor this).I assume that everyone thinks I am a terrible person, a murderer, because of what I did.

7. Emotional reasoning (you have a feeling and assume there must be a reason).This one is easy—I feel guilty, and therefore I must be.

FIGURE 2.3. Patterns of Problematic Thinking Sheet.

cerned about the therapy making him worserather than better. The therapist had given Toma PCL and BDI to complete while he was wait-ing for his appointment. She quickly scoredthese assessment measures and gave Tom feed-back about his scores at the beginning of thissession. His PCL score had decreased from 68to 39, which was a clear and clinically mean-ingful change in his PTSD symptomatol-ogy. She noticed that his avoidance andreexperiencing symptoms had decreased themost; his hyperarousal symptoms had also de-creased, but less so. His score on the BDI haddecreased from 28 to 14, clearly indicating areduction in his depressive symptoms.

The therapist asked whether Tom had com-pleted his practice assignment, the Patterns ofProblematic Thinking Sheet. He indicated thathe had not, but that he had thought about itover the week. He also laughed, and said thathe had noticed the thinking patterns in his wifeand others. The therapist asked Tom to com-plete some of the sheet in session. At this pointin therapy, the therapist was sitting back moreas Tom took on the role of challenging his owncognitions. The therapist provided both mini-mal clarification and also additional examplesthat she had noticed in working with Tom.

In this session, the therapist introduced theChallenging Beliefs Worksheet. She was carefulto point out that the worksheet integrated all ofthe previous work that Tom had done andadded a few new elements. The following dia-logue illustrates the introduction of this sheet(see Figure 2.4).

THERAPIST: I want to show you the finalworksheet that we’re going to be using.

TOM: OK. Wow—that looks complicated!

THERAPIST: Actually, you’ve done pretty mucheverything on this worksheet already. Thisworksheet brings together into one place ev-erything that we’ve been working on.

TOM: I’ll take your word for it, Doc.

THERAPIST: Remember the A-B-C Sheets fromway back when?

TOM: Yes.

THERAPIST: (pointing to first three columns onthe Challenging Beliefs Worksheet) This is A,B, and C. You have in column A the situa-tion, or “Activating Event” that you had onthe A-B-C Sheet. In column B you have “Au-tomatic Thoughts,” which is the “Belief”

portion of the A-B-C Sheet. Last, column C,“Emotions,” the “Consequence” portion ofA-B-C Sheet.

TOM: OK. So far, so good.

THERAPIST: Column D is where you identify the“Challenging Questions” from that sheetthat apply to the thought or stuck point thatyou’re working on. In column E, you iden-tify the type of “Patterns of ProblematicThinking” that apply to the thought or stuckpoint that you’re working on. Make sense?

TOM: Yes.

THERAPIST: So, only column F, “AlternativeThought,” is new. Here you identify alterna-tive thoughts that you could have about thesituation. In other words, we’re looking foralternative statements that you tell yourselfor different interpretations of the event. Incolumns G and H, you get to see how yourbelief in your original thoughts may changeand how the new thoughts affect your feel-ings.

TOM: OK.

THERAPIST: So, let’s pick a stuck point and startusing this Challenging Beliefs Worksheet.We’re going to be talking about safety as oneof the first topics of the next few sessions.Can you think of a stuck point that relates toyour ability to keep yourself safe or to howsafe others are around you?

TOM: Well, I still wonder if there are people outin the world who want to hurt me, even if Inow realize that no sniper is going to takeme out.

THERAPIST: So, let’s pick a specific event—themore specific, the better.

TOM: I was in the grocery store, and I had myuniform on. There was this guy who seemedto have a chip on his shoulder about it—likehe hated me or something.

THERAPIST: So, write down the event in Col-umn A. (pauses) What was your thought?You’ve already mentioned one of them.

TOM: This guy has a chip on his shoulder aboutme, because I’m in the military.

THERAPIST: Good. How strongly do you believethat thought?

TOM: 100%

THERAPIST: OK, let’s write that next to thethought. We are now rating how much youbelieve in your thoughts, because you’re go-


105

A.S

ituat

ion

B.T

houg

hts

D.C

halle

ngin

gT

houg

hts

E.P

robl

emat

icP

atte

rns

F.A

ltern

ativ

eT

houg

ht

Des

crib

eth

eev

ent,

thou

ght,

orbe

lief

lead

ing

toth

eun

plea

sant

emot

ion(

s).

Writ

eth

ough

t(s)

rela

ted

toC

olum

nA

.

Rat

ebe

lief

inea

chth

ough

tbe

low

from

0–10

0%.

(How

muc

hdo

you

belie

veth

isth

ough

t?)

Use

Ch

alle

ng

ing

Qu

esti

on

sto

exam

ine

your

auto

mat

icth

ough

tsfr

omC

olum

nB

.

Isth

eth

ough

tba

lanc

edan

dfa

ctua

lor

extr

eme?

Use

the

Pro

ble

mat

icT

hin

kin

gP

atte

rns

shee

tto

deci

deif

this

ison

eof

your

prob

lem

atic

patte

rns

ofth

inki

ng.

Wha

tel

seca

nI

say

inst

ead

ofC

olum

nB

?

How

else

can

Iin

terp

ret

the

even

tin

stea

dof

Col

umn

B?

Rat

ebe

lief

inal

tern

ativ

eth

ough

t(s)

from

0–10

0%.

At

sto

rein

un

iform

“Th

isg

uy

has

ach

ipo

nh

issh

ou

lder

bec

ause

Iam

inth

em

ilita

ry.”

(100

%)

Evi

denc

e?

Hab

ito

rFa

ct?

Hab

itto

thin

kev

eryo

ne

dis

likes

me

bec

ause

Iw

asin

Iraq

.

Inte

rpre

tatio

nsno

tac

cura

te?

All

orno

ne?

Ext

rem

eor

exag

gera

ted?

Out

ofco

ntex

t?

So

urc

eu

nre

liab

le?

Me

Low

vers

ushi

ghpr

obab

ility

?

Bas

edon

feel

ings

orfa

cts?

Irre

leva

ntfa

ctor

s?

Jum

ping

toco

nclu

sion

s

Exa

gger

atin

gor

min

imiz

ing

Dis

rega

rdin

gim

port

ant

aspe

cts

Ove

rsim

plify

ing

Ove

rgen

eral

izin

g

Min

dre

adin

gI

amas

sum

ing

that

he

isth

inki

ng

the

wo

rst

of

me .

Em

otio

nal

reas

onin

g

“Id

on

’tkn

owif

he

has

chip

on

his

sho

uld

er.”

(60%

)

“If

he

do

esh

ave

ach

ipo

nh

issh

ou

lder

,Id

on

’tkn

oww

hat

itis

abo

ut

–m

ayb

eit

isn

’tev

enab

ou

tm

e,le

tal

on

eh

avin

gse

rved

inIr

aq.”

(80%

)

G.R

erat

eO

ldT

houg

hts

C.E

mot

ion(

s)R

erat

eho

wm

uch

you

now

belie

veth

eth

ough

tin

Col

umn

Bfr

om0–

100%

.

35%

Spe

cify

sad,

angr

y,et

c.,

and

rate

how

stro

ngly

you

feel

each

emot

ion

from

0–10

0%.

An

ger

(80%

)F

ear

(30%

)

H.E

mot

ion(

s)

Now

wha

tdo

you

feel

?0–

100%

An

ger

(20%

)F

ear

(15%

)

FIG

UR

E2.

4.C

halle

ngin

gB

elie

fsW

orks

heet

com

plet

edin

sess

ion.

ing to see at the end how much your thoughthas changed. What feeling or feelings are as-sociated with that thought?

TOM: Definitely anger.

THERAPIST: Makes sense given your thought.How much anger from 0 to 100%, with100% being as much anger as you couldpossibly imagine having?

TOM: Hmm . . . I’d say 80%.

THERAPIST: Any other feelings? You can havemore than one.

TOM: I guess when I stop and think about it,there is some fear there, too.

THERAPIST: That makes sense, too. How muchfear from 0 to 100%?

TOM: Oh, maybe 30%. It’s not the strongestfeeling, but it’s there, because I’m wonderingif he is going to say something or do some-thing.

THERAPIST: Nice job. Let’s move onto the nextcolumn that relates to the Challenging Ques-tions Sheet you’ve already done. Take a lookat this list. What questions might apply here?

TOM: I guess I might be confusing a habit witha fact. It seems like it is a habit for me to as-sume that everyone dislikes me because I wasin Iraq. I really don’t know if that is why heseemed to have a chip on his shoulder. Iguess I also don’t know for sure if he had achip on his shoulder. He didn’t say anythingto me. (pauses) I guess that is also an exam-ple of the source of information being unreli-able and that source is me! (Laughs.)

THERAPIST: While you were talking, I wasthinking that the same things applied. Soyou’d write those in this column. You canalso pick out other challenging questionsthat might apply, but usually two or threewill do the trick. In the next column, we’regoing to refer to the Patterns of ProblematicThinking Sheet. What might fit here?

TOM: I guess one jumps out—mind reading.

THERAPIST: How so?

TOM: I’m assuming that he is thinking theworst about me and about my having servedmy country in this war. I’m good at that.

THERAPIST: Write that down. You can add oth-ers later if something seems to apply. Thenext column is very important. This is whereyou start coaching yourself to come up withalternative thoughts or perceptions about

the situation. Based on having asked yourselfthese questions and noticing the problematicthinking patterns, what other ways mightyou think about this situation?

TOM: I guess one thing I could say to myself is,“I don’t know if he has chip on his shoul-der.” I could also say, “If he does have a chipon his shoulder, I don’t know what it isabout—maybe it isn’t even about me, letalone having served in Iraq.”

THERAPIST: Wow! You’re doing great at this.Let’s get those written down. Let’s also addhow much you believe those two newthoughts. Below those alternative thoughts isthe column that asks you to reconsider howmuch you believe your original thoughtsover here in column B. How much do youbelieve them after walking through this pro-cess? Before you said 100%.

TOM: Oh, I’d say now it is only about 35%.

THERAPIST: That is a big change. You wentfrom 100% certainty to 35% certainty thathe had a chip on his shoulder, because youfought in the war.

TOM: I’m a little surprised by that myself.

THERAPIST: Let’s take it the final step. Howabout your feelings now? Let’s rerate thosehere.

TOM: My anger is way down—I’d say onlyabout 20%. The anxiety is still there, be-cause I really wouldn’t want to have to pro-tect myself, and he might have had a chip onhis shoulder at me. It is down a little,though, because I realize I’m not 100% cer-tain he was out to get me. I’d say maybe15% on fear.

THERAPIST: Do you have questions about whatwe just did here?

TOM: Not at the moment. I’ll get back to you.

THERAPIST: I’m going to ask that you do one ofthese sheets about stuck points per day, untilI see you again. I’m also going to give yousome example sheets other patients havedone that might be helpful to you.

TOM: OK. Should be interesting. . . .

The therapist reminded Tom that he mightfind that he is not using problematic thinking,and in that case, no change in feelings would beexpected. She also cautioned Tom that heshould not expect his beliefs and feelings tochange completely in the process of doing the


sheet. The old thought would need to be com-pletely dismantled and the new thought wouldneed to become more habitual for him to see amore permanent change. The therapist sug-gested that Tom read the sheets he completedover to himself a number of times to facilitatethe process.

The Safety module was then introduced.Safety is the first of five modules (two- to three-page handouts) that also include trust, power,esteem, and intimacy. The therapist orientedTom to the format of the module, which in-cluded discussion about how beliefs about theself and others in this area can be “shattered”or seemingly confirmed after a traumatic eventdepending on one’s history prior to the trau-matic event. The modules describe how theseproblematic beliefs are manifested emotionallyand behaviorally (e.g., not leaving one’s homebecause of the belief that the world is unsafe).It also provides alternative self-statements thatare more balanced and realistic in each area.

Tom had felt safe with others before thetraumatic event occurred, and this sense ofsafety about others had been disrupted, as evi-denced by his sense that others around himwere out to get him. Pretrauma, Tom had alsofelt as though he was not a danger to others.Posttrauma, he believed that he could not besafe with others, which specifically manifestedin his concerns about being around his preg-nant wife. The therapist suggested that Tomcomplete at least one worksheet on his stuckpoints about others being safe, as well as hisbeing a possible danger to others. The therapistalso reminded Tom that he needed to finish thePatterns of Problematic Thinking Sheet home-work from last session.

Session 8

Tom arrived at the session having completedthe Patterns of Problematic Thinking Sheet, aswell as two Challenging Beliefs Worksheets.The therapist spent a little time looking at hisanswers to the Patterns of ProblematicThinking Sheet, because she did not want tosend the message inadvertently that completinghomework was unimportant. She asked Tom toread the patterns that he had completed athome, as opposed to those in their previous ses-sion.

Tom completed two Challenging BeliefsWorksheets related to the topic of safety, as thetherapist had instructed. He did one each on

self and other safety beliefs. He did not seem tounderstand that he could use the ChallengingBeliefs Worksheets on everyday events thatwere distressing or even positive for him. Thus,the therapist emphasized how Tom might usethis process more generally in his day-to-daylife, and highlighted how more practice wouldlead to more results. She noted that using theprocess on less emotionally distressing topicscould actually be very helpful in getting theprocess down. It is always easier to learn some-thing when one is not dealing with the mostchallenging circumstances. She used a militaryanalogy with Tom, about learning to load andshoot a gun—best learned in a nonconflict situ-ation, so that it is a more rote behavior whenunder fire.

The therapist skimmed the two sheets Tomhad completed and noticed that he had strug-gled most coming up with alternative state-ments related to his own sense of dangerous-ness related to his wife’s impending delivery oftheir child. The following dialogue ensued (seeFigure 2.5):

THERAPIST: I notice that you might have hadthe most trouble coming up with alternativethoughts about how safe you can be withyour wife and your child that is about to beborn.

TOM: Yeah, I don’t really like to talk about it. Itfreaks my wife out. I’m uncomfortable beingaround my wife, which makes her feel bad,but I’m just afraid I’m going to hurt her orthe child.

THERAPIST: Let’s take your first thought, be-cause it is kind of general. How is it that youthink you’re going to hurt them? Are wetalking physically or mentally?

TOM: Oh, physically is what I mean. I don’tknow how exactly, but somehow, someway,I guess.

THERAPIST: That makes it a bit more concrete.How do you physically think you’re going tohurt them? Do you think you’ll shoot them,given your trauma history?

TOM: No. Absolutely not. There are no fire-arms in my house, and I don’t go hunt-ing or have friends or family that hunt—nothing that would make guns a part of ourlife.

THERAPIST: So, what have you considered inyour mind?


108

A.S

ituat

ion

B.T

houg

hts

D.C

halle

ngin

gT

houg

hts

E.P

robl

emat

icP

atte

rns

F.A

ltern

ativ

eT

houg

ht

Des

crib

eth

eev

ent,

thou

ght,

orbe

lief

lead

ing

toth

eun

plea

sant

emot

ion(

s).

Writ

eth

ough

t(s)

rela

ted

toC

olum

nA

.

Rat

ebe

lief

inea

chth

ough

tbe

low

from

0–10

0%.

(How

muc

hdo

you

belie

veth

isth

ough

t?)

Use

Ch

alle

ng

ing

Qu

esti

on

sto

exam

ine

your

auto

mat

icth

ough

tsfr

omC

olum

nB

.

Isth

eth

ough

tba

lanc

edan

dfa

ctua

lor

extr

eme?

Use

the

Pro

ble

mat

icT

hin

kin

gP

atte

rns

shee

tto

deci

deif

this

ison

eof

your

prob

lem

atic

patte

rns

ofth

inki

ng.

Wha

tel

seca

nI

say

inst

ead

ofC

olum

nB

?

How

else

can

Iin

terp

ret

the

even

tin

stea

dof

Col

umn

B?

Rat

ebe

lief

inal

tern

ativ

eth

ough

t(s)

from

0–10

0%.

Bei

ng

aro

un

dm

yw

ifean

dch

ild

“Ou

to

fn

owh

ere,

I’ll

get

phy

sica

llyvi

ole

nt.”

(80%

)

Evi

denc

e?

Hab

itor

fact

?

Inte

rpre

tatio

nsno

tac

cura

te?

All

orno

ne?

Ext

rem

eor

exag

gera

ted?

Out

ofco

ntex

t?

Sou

rce

unre

liabl

e?

Lo

wve

rsu

sh

igh

pro

bab

ility

?G

iven

my

his

tory

,it

isac

tual

lya

low

pro

bab

ility

no

th

igh

.

Bas

edon

feel

ings

orfa

cts?

Irre

leva

ntfa

ctor

s?

Jum

ping

toco

nclu

sion

s

Exa

gg

erat

ing

or

min

imiz

ing

I’mex

agg

erat

ing

the

likel

iho

od

that

I’db

evi

ole

nt.

Dis

rega

rdin

gim

port

ant

aspe

cts

Ove

rsim

plify

ing

Ove

rgen

eral

izin

gI’m

assu

min

gb

ecau

seI

sho

to

nce

ina

cert

ain

situ

atio

n,I

’llb

evi

ole

nt

ing

ener

al.

Min

dre

adin

g

Em

otio

nal

reas

onin

g

“It

isu

nlik

ely

that

I’ll

hu

rtm

yfa

mily

,an

dev

enm

ore

un

likel

yth

atit

will

be

sudd

enan

du

nex

pec

ted

.”(9

5%)

G.R

erat

eO

ldT

houg

hts

C.E

mot

ion(

s)R

erat

eho

wm

uch

you

now

belie

veth

eth

ough

tin

Col

umn

Bfr

om0–

100%

.

10%

Spe

cify

sad,

angr

y,et

c.,

and

rate

how

stro

ngly

you

feel

each

emot

ion

from

0–10

0%.

Fea

r(8

5%)

H.E

mot

ion(

s)

Now

wha

tdo

you

feel

?0–

100%

Fea

r(<

10%

)

FIG

UR

E2.

5.C

halle

ngin

gB

elie

fsW

orks

heet

rega

rdin

gsa

fety

.

TOM: I guess I’m worried that, out of nowhere,I’ll get physically violent.

THERAPIST: OK, now we’re cooking. Let’s writethat down. “Out of nowhere I’ll get violent.”I noticed that in column C you didn’t men-tion anything about probabilities. Safety is-sues are almost always about gauging proba-bilities. The world is not a completely safeplace, and every day we all make calculatedrisks about our safety based on the probabil-ity of bad things happening to us or to some-one else. How do you think the probabilityquestions might apply?

TOM: Are you getting at the idea that I’m con-fusing a low probability with a high proba-bility?

THERAPIST: Precisely. How do you think thatapplies here?

TOM: I’m convinced that “somehow, some-way” I’m going to hurt my family, so I be-lieve that it is a high probability that it willhappen and not a low probability. I thinkyou think that the probability I will do thatis low. But, I’m still concerned about it.

THERAPIST: Let’s talk about the actual probabil-ity. How often have you hurt your familyphysically?

TOM: Never. Are you kidding?

THERAPIST: I thought as much, but you made itsound like it was very likely to happen. Iguess that’s part of the problem, right?

TOM: You’re right.

THERAPIST: How often have you been physi-cally violent against anyone?

TOM: I haven’t. And it surely hasn’t been unex-pected. Now that we’re talking through it, itfeels a little silly.

THERAPIST: So, it sounds like figuring out theactual probability of this is right where weneeded to go. Given what we’ve talkedabout, what is an alternative statement thatyou can tell yourself and how much do youbelieve it?

TOM: It is unlikely that I’ll hurt my family, andeven more unlikely that it will be sudden andunexpected given that it has never happened.

THERAPIST: Let’s keep going to see how thatmight change how you feel. You wrote thatyou had 85% fear. What is that rating now?

TOM: Less than 10%. There is some fear nowthat I know that I am capable of hurting a

family, but like we’ve talked about before—and what I have to remember—is that it oc-curred in a certain situation and not in myeveryday life now as a civilian in my family.

This exchange between Tom and the thera-pist illustrates the hallmark role of probabilityin assessments and beliefs about safety. It is im-portant to realize that there are some objec-tively unsafe situations or behaviors, and thoseshould not be minimized or restructured. Ifthere are unreasonable safety precautions orbeliefs, the actual probability of harm shouldbe carefully evaluated, keeping in mind that100% safety is rarely, if ever, guaranteed.

The therapist transitioned the session to in-troduce the Trust module. Tom noted that hehad pretty good trust of himself and othersprior to his best friend committing suicidewhen they were in high school. Tom said thatafter that experience, he sometimes did nottrust his judgments about other people and thathe felt responsible for not anticipating hisfriend’s suicide. The military traumatic eventserved to confirm his belief that he could nottrust his judgments about others’ intentions.Tom’s concerns about his ability to be safe withhis wife and unborn child also dovetailed withthe issue of trust. The therapist and Tom wentover the information in the Trust module hand-out, and Tom seemed to resonate with all of thepotential effects. He reported that he had reallybeen trying to open up with his wife and notavoid her. He noted that they were communi-cating more, and that made both of them morerelaxed and comfortable in the final days of herpregnancy.

The therapist closed the session by assigningdaily Challenging Beliefs Worksheets, askingTom to do at least one on the topic of trust. Shereminded him that, like other areas, the goal isto develop balanced alternative thoughts. Inthe case of trust, she noted that stuck pointsabout trust often revolve around making all-or-none judgments, either trusting or not. Thegoal is to consider trust as multidimensional,with different types of issues resulting in differ-ent levels of trust in different situations.

Session 9

Tom arrived at this session having completeda number of Challenging Beliefs Worksheets.Several of them were about trust, including hislevel of trust of his government and trust of


himself in being a father. He had also used theworksheets on non-trust-related topics relativeto his daily life. He commented that the work-sheets had been helpful in working out histhinking before he behaved impulsively or feltmiserable.

The therapist praised Tom for completingthe worksheets so well, and asked Tomwhether he needed assistance on any of theworksheets he felt he could use some assistanceon. Tom quickly responded that he wanted tofocus on the sheet about fatherhood, becausehe was experiencing so much anxiety about hischild’s impending birth. In turning their atten-tion to this worksheet, the therapist immedi-ately noticed that Tom had probably struggledwith this worksheet, because he had listed somany different types of thoughts that werefueling his anxiety about becoming a father.She used this as an opportunity to fine-tuneTom’s use of the worksheets. The therapist’schoice in thoughts to challenge first also illus-trates the prioritization of treatment targets inthe therapy. She chose to go after the more di-rectly trauma-related thoughts that containedremnants of assimilation. Tom’s thoughtsabout deserving to be happy about starting afamily, given the death of the woman, fetus,and child, suggested that he had not fully ac-cepted the traumatic event and the circum-stances surrounding it. Thus, she addressed thisfirst thought (see Figure 2.6).

THERAPIST: Wow, you’ve got lots of thoughtsgoing on in your head about becoming a fa-ther, don’t you? I’m going to suggest that weuse different worksheets for each of the clus-ters of thoughts you’re having on this topic. Ithink that will make your use of the Chal-lenging Beliefs Worksheet better. It seemsthat some thoughts are directly related toyour traumatic experience, others are specif-ically related to your wife’s labor and deliv-ery, and still others are related more gener-ally to being a parent. Let’s focus on thosethat are directly related to your trauma. Youwrote that one of your feelings was guilt(85%), and I’m assuming that it is related toyour thought that it isn’t right that you’rehappy with a soon-to-be-born baby givenwhat happened.

TOM: That’s right. If I’m really honest, I stillfeel guilty that the Iraqi woman was preg-nant and getting ready to have a child, and

the shooting deprived her of the ability tohave that child and be happy, and I’m gettingready to have that happiness.

THERAPIST: We’ve talked about this before, butwe’ve been more focused on the man in-volved in the situation.

TOM: Yeah, I think the closer my wife gets todelivery, the more I think about the Iraqiwoman. I’ve been imagining that she wasn’tpart of a potential plot for terrorist activityand was more an innocent participant. Then,I go back and forth, thinking that she mighthave actually been involved and didn’t carethat she was pregnant. Or maybe it was justan accident, and they truly didn’t understandthat they needed to stop. Uggghhhh, it is ex-hausting.

THERAPIST: And you’ll never know. If yourfriend were saying all of this to you, whatwould be your response to him?

TOM: I’d be telling him to quit beating himselfup and feeling guilty.

THERAPIST: Easier said than done. Anythingelse? Maybe it would help to look at theChallenging Questions and Patterns of Prob-lematic Thinking Sheets. I’m wondering ifyou are focusing on irrelevant factors—item10 on the Challenging Questions Sheet.

TOM: Hmm . . . what is irrelevant in this case?

THERAPIST: How relevant are her intentions todeserving to be happy yourself about havinga child?

TOM: (Pauses) I’m going to have to think aboutthat for a second.

THERAPIST: Aren’t your intentions in that situa-tion what is relevant? Were your intentionsat that time to deprive her of the right tobear her child and live happily ever after?

TOM: No, not at all.

THERAPIST: So, why the guilt? What did you dowrong that you should be punished about?

TOM: Oh, wow. I hadn’t thought of that. Herintentions are irrelevant. It only makes mecrazy to try to get in her head. I guess thatwould be mind reading, now wouldn’t it?

THERAPIST: Very good—a different spin onmind reading. So what is the alternative,more balanced and realistic thought?

TOM: My intentions are what matter. I didn’tintend for her to lose her own or her baby’slife.


111

A.S

ituat

ion

B.T

houg

hts

D.C

halle

ngin

gT

houg

hts

E.P

robl

emat

icP

atte

rns

F.A

ltern

ativ

eT

houg

ht

Des

crib

eth

eev

ent,

thou

ght,

orbe

lief

lead

ing

toth

eun

plea

sant

emot

ion(

s).

Writ

eth

ough

t(s)

rela

ted

toC

olum

nA

.

Rat

ebe

lief

inea

chth

ough

tbe

low

from

0–10

0%.

(How

muc

hdo

you

belie

veth

isth

ough

t?)

Use

Ch

alle

ng

ing

Qu

esti

on

sto

exam

ine

your

auto

mat

icth

ough

tsfr

omC

olum

nB

.

Isth

eth

ough

tba

lanc

edan

dfa

ctua

lor

extr

eme?

Use

the

Pro

ble

mat

icT

hin

kin

gP

atte

rns

shee

tto

deci

deif

this

ison

eof

your

prob

lem

atic

patte

rns

ofth

inki

ng.

Wha

tel

seca

nI

say

inst

ead

ofC

olum

nB

?

How

else

can

Iin

terp

ret

the

even

tin

stea

dof

Col

umn

B?

Rat

ebe

lief

inal

tern

ativ

eth

ough

t(s)

from

0–10

0%.

Kill

ing

ap

reg

nan

tIr

aqi

wo

man

and

her

son

.

“It

isn

’tri

gh

tth

atI’m

hap

pyw

ith

ab

aby

on

the

way

,giv

enw

hat

hap

pen

ed.”

(80%

)

“Sh

em

igh

tn

ot

hav

eb

een

par

to

fa

terr

ori

stp

lot,

but

just

ap

asse

ng

er.”

(50%

)

Evi

denc

e?

Hab

itor

fact

?

Inte

rpre

tatio

nsno

tac

cura

te?

All

orno

ne?

Ext

rem

eor

exag

gera

ted?

Out

ofco

ntex

t?

Sou

rce

unre

liabl

e?

Low

vers

ushi

ghpr

obab

ility

?

Bas

edon

feel

ings

orfa

cts?

Irre

leva

nt

fact

ors

?H

erin

ten

tio

ns

are

no

tre

leva

nt.

Min

ear

e.

Jum

ping

toco

nclu

sion

s

Exa

gger

atin

gor

min

imiz

ing

Dis

rega

rdin

gim

port

ant

aspe

cts

Ove

rsim

plify

ing

Ove

rgen

eral

izin

g

Min

dre

adin

gI’m

tryi

ng

tofi

gu

reo

ut

wh

atw

asin

her

hea

d.

Em

otio

nal

reas

onin

g

“My

inte

nti

on

sar

ew

hat

mat

ter.

Id

idn

’tin

ten

dto

do

anyt

hin

gto

dep

rive

som

eon

eel

seo

ffa

mily

hap

pin

ess.

”(8

5%)

G.R

erat

eO

ldT

houg

hts

C.E

mot

ion(

s)R

erat

eho

wm

uch

you

now

belie

veth

eth

ough

tin

Col

umn

Bfr

om0–

100%

.

15%

(2n

dd

oes

n’t

mat

ter)

Spe

cify

sad,

angr

y,et

c.,

and

rate

how

stro

ngly

you

feel

each

emot

ion

from

0–10

0%.

Gu

ilt(8

5%)

H.E

mot

ion(

s)

Now

wha

tdo

you

feel

?0–

100%

Gu

ilt(5

%)

Hap

py(1

0%)

FIG

UR

E2.

6.C

halle

ngin

gB

elie

fsW

orks

heet

rega

rdin

gtr

aum

a.

THERAPIST: Go on . . . don’t you have everyright in the world to experience happiness?

TOM: I guess so. It just feels weird.

THERAPIST: Sure it does. It is different thanwhat you’ve been thinking about it forawhile. I’m curious—what would you feel ifyou said to yourself, “I did not intentionallydo anything to deprive someone else of fam-ily happiness. I deserve to be happy in be-coming a father.”

TOM: I’d feel less guilty for sure, and evenhappy.

THERAPIST: Let’s get this all written down. Nowyou have the job of holding on to these newinsights and practicing them. Read over thisworksheet every day until you see me again.I’d also like you to take these other thoughtson your original Challenging Beliefs Work-sheet about this topic and put them on sepa-rate worksheets and work through them.Can you commit to doing that?

TOM: Yes, I already feel lighter.

THERAPIST: This is an exciting time—you’vegot to continue to work on this, so that youcan have the enjoyment you deserve!

At this point, the therapist introduced thePower/Control module. Tom admitted thatprior to the traumatic event, he was someonewho liked to be in control. He did not like un-predictability, and he noticed that this tendencyhad gotten especially bad after his friend’s sui-cide. The military lifestyle seemed to be con-gruent with this tendency. Tom indicated thathe had not had authority issues prior to thetraumatic event, but he had noticed himselfquestioning authority much more since his mil-itary trauma. As with previous sessions, Tomwas given as homework to complete Challeng-ing Beliefs Worksheets every day prior to thenext session, and at least one was assigned onpower/control.

Session 10

Tom began the session saying that his wife hadgone to her obstetrician on the previous day,and that her labor would be induced in 1 weekif she did not go into natural labor before then.Tom indicated that the last session had beenvery good in helping him to become happierabout his child’s impending birth, and that hehad read the Challenging Questions Worksheet

about deserving to be happy several times sincethe last session, and he believed it more andmore. He stated that he was still having someanxiety about becoming a father, and about ev-erything going OK with his wife’s labor and de-livery. The therapist normalized some of Tom’sanxiety, stressing how it was very natural for afirst-time father, and Tom was able to recognizethe typicality of this anxiety in others he hadwitnessed becoming parents.

Tom stated that since reading the Power/Control module after the last session, he hadstarted to realize that not everyone in authorityover him had wielded his or her authority ma-levolently. This was very important in light ofTom’s preexisting history of desiring to exertcontrol; he had directly confronted his illusionof control. The therapist and Tom went overthis worksheet.

Tom went on to describe how his belief thathe could and should have control over every-thing had resulted in low self-esteem. In gen-eral, when things did not go as he desired, Tomfelt as though he was a failure for not control-ling the outcome. This belief structure led himto think that he should have been able to con-trol his friend and stop him from committingsuicide. It also led him to believe that he shouldhave been able to create a positive outcome inthe military traumatic event. This discussionserved as a natural segue to the next topic—esteem. Tom admitted that he had becomesomeone who thrived too much on accomplish-ment. This had affected his self-esteem and wasespecially relevant to his belief that he had notaccomplished his goal in the military, becausehe had to be airlifted out of country after thetraumatic event at the checkpoint.

After reviewing the Esteem module, the ther-apist asked Tom to complete Challenging Be-liefs Worksheets on his remaining stuck points,as well as any stuck points relating to esteem.He was also given two other assignments: topractice giving and receiving compliments ev-ery day, and to do one nice thing for himself ev-ery day that was not contingent on “achieving”something. These assignments were to help himwith his self- and other-esteem.

Session 11

Tom completed a worksheet on his self-esteem,related to his belief that he had not achieved hisgoal within the military. The therapist and Tomwent over this worksheet, and both noted that


he had made significant progress by using theworksheet to change the way that he thoughtand felt about himself. He asserted that he wasbeginning to see that people are much morethan their professional accomplishments. Theyalso have other activities and relationships withtheir families, friends, and themselves.

The therapist inquired about the homeworkof giving and receiving compliments. Tom re-plied that it had gone well, even though it felt abit awkward and forced. He was even able tonotice that when he gave compliments and wasmore positive toward other people, he seemedto get more positive responses back from them.The therapist noticed that several of the com-pliments were to his wife, and she pointed outthat Tom seemed more connected to his wife.He said that he was actually beginning to feelglimmers of excitement about the birth of theirchild. He reported that he was still feeling someanxiety about becoming a father, and abouthow the labor and delivery would go, but thatthe anxiety was less and more manageable.When the therapist asked about Tom receivingcompliments, he reported more difficulties. Sheasked what Tom typically did when he receivedcompliments, and it became clear that he oftendeflected or minimized them. Correspondingly,Tom also said that he had only done one nicething for himself since the last session, and thatit had felt uncomfortable. This pattern seemedto fit with Tom’s overall schema of being un-worthy and undeserving. The following dia-logue related to this:

THERAPIST: It seems like you have a hard timeletting someone be nice to you and beingnice to yourself.

TOM: Yes.

THERAPIST: Why do you think that is?

TOM: I don’t know. (pauses) I don’t like it. Itfeels like they shouldn’t be nice to me, and Ishouldn’t be nice to me.

THERAPIST: Hmm . . . I wonder if there is any-thing “off” about that thinking? What doyou think?

TOM: As I hear myself say it, it sounds a littleweird. It sounds like I don’t deserve to havenice things for me. Kind of like not deservingto have a family . . .

THERAPIST: This seems like a larger tendency inyour life—one of those problematic thinkingpatterns. What pattern do you hear in your

thinking? Look at the worksheet if you wantto.

TOM: Maybe emotional reasoning. I feel like Idon’t deserve it; therefore, I must not deserveit. That seems like the best one. Maybe I’malso drawing a conclusion when the evidenceis lacking.

THERAPIST: I agree. Given how much you seemto follow this pattern of thinking, I’m bettingit has been around for awhile—maybe evenbefore the shooting occurred in Iraq.

TOM: It has. I think it had to do with my dad,his alcoholism, and not being close to me. Asa kid, I always thought I had done somethingwrong, or that I was so bad that he didn’twant to be around me.

THERAPIST: Now, with adult eyes, what do youthink about your dad not being close to you?

TOM: I figure that he drank for a reason, andthat it might have been me and my otherbrothers and sisters.

THERAPIST: Why do you assume that he drankbecause of you kids?

TOM: I don’t know. I figure it was stressful hav-ing four kids.

THERAPIST: It probably was at times, but as youhear yourself talk about this, what is amissin how you’ve made sense of his drinkingand being close to you?

TOM: I know other people who have had fourkids and didn’t have drinking problems.There were a lot of big families where I grewup. Plus, I know that he and my mom hadmoney problems when we were young, andthat they fought a lot.

THERAPIST: So, again, why do you assume itwas you then that caused his drinking andalienation?

TOM: When we talk about it, I guess I see thatit might not have been me alone.

THERAPIST: Or not even you at all. Everybodyhas a choice about how they handle theirstress, and it seems that he is distant from ev-eryone, not just you.

TOM: True. It still feels that way.

THERAPIST: There seems to be a well-worn pathin your brain to assume that when some-thing goes wrong, you blame yourself. Thenext step is that you deserve to be punished,or at least you don’t deserve anything good. Idon’t think this tendency is going to change


overnight. You’re going to need to workhard at talking to yourself more rationally tochange how you feel. For that new path toget worn, you’re going to have to walk downit a number of times. Pretty soon, the pathwill be more worn and automatic. It willtake some effort, but you can change theway you automatically feel. I’d like you todo a Challenging Beliefs Worksheet aboutwhat we’ve just talked about. Once we get agood one about it, you can read and refer toit as part of forging that new path. Can youdo that?

TOM: Yes. I think it would be good.

This exchange regarding Tom’s dad dove-tailed nicely with the final module, Intimacy.The therapist noted that people tend to think ofintimacy as it relates to romantic relationships,and especially in terms of sexual intimacy. Shestressed that there are all kinds of intimacy withdifferent people. In essence, intimacy relates tohow close and open we feel with other people.She went on to discuss the notion of self-intimacy, or how well we take care of, support,and soothe ourselves. In other words, it reflectshow good a relationship we have with our-selves. Tom admitted that he struggled with be-ing close to other people, which had most obvi-ously manifested in the work he had done abouthis wife and child, who was to be born at anymoment. As noted earlier, Tom also struggledwith doing nice things and taking good care ofhimself. Both of these areas seemed to be af-fected by Tom’s underlying schema that he wasundeserving and unworthy. The therapist as-signed daily Challenging Beliefs Worksheets andrequested that he do worksheets on being nice tohimself and being close to his wife. In addition,she asked Tom to write a final Impact State-ment, specifically about his understanding ofthe trauma now, after all the work he had done.The therapist asked him to write about his cur-rent thoughts/beliefs in the areas of safety, trust,power/control, esteem, and intimacy.

Session 12

The day after Session 11, Tom left a messageindicating that his wife had delivered a healthybaby girl. He indicated in his voice mail mes-sage that he felt happy and relieved. He wenton about how beautiful the baby was, how well

his wife had done in labor and delivery, andhow he had enjoyed holding his daughter in hisarms the first time. The 12th session was de-layed an extra week because of the baby’s ar-rival.

Tom’s wife and new daughter accompaniedhim to the final session. The therapist spentsome time admiring Tom’s new baby and con-gratulating his wife before the final sessionstarted. Tom seemed genuinely proud andhappy about his daughter, and noted that be-coming a father had been more natural than hehad anticipated. He commented that he hadbeen worried that he would not want to holdthe infant for fear of hurting her or because hewould do something wrong. Instead, he foundit almost “instinctual” to hold her, and thatsoothing her had come more naturally than heexpected. Tom seemed surprised about hownatural his role as a father had become.

The therapist inquired about how thehomework had gone. Tom said that he hadnot done as much as he had hoped given thebaby’s arrival, but that he had doneworksheets about his father and about beingclose to his wife. The therapist looked overthese worksheets, which Tom had done verywell. She asked Tom about how helpful theyhad been, and he reported that they had beenvery helpful. He added that he was still strug-gling about his father, but that he was begin-ning to think that it was not all about him,which had made him feel better about himselfand less guilty in general. He mentioned thathe was considering writing a letter to his fa-ther about his daughter’s arrival, and that hewas thinking about asking his father aboutwhy he drank and distanced himself from hisfamily. The therapist reinforced Tom for con-sidering this and for not blindly making as-sumptions about his role in his father’s drink-ing. However, she also attempted to inoculateTom to the possibility that his father couldblame him or his siblings for his alcoholism(given that she did not know his father or hishistory), and that this did not necessarilymean that it was true. She reminded him thathe needed to consider the source of informa-tion, and that any good detective would getmultiple reporters. Tom seemed to like theidea of getting more information from others,mentioning that he and his siblings had neverreally talked about his belief that they wereto blame for their father’s alcoholism.


Tom also shared that he better understoodthe idea of having intimacy, without sex, in hisrelationship with his wife. He said that sincethe birth of their child, he felt closer to his wifeand had generally been more open and presentto her. The therapist asked him about doingnice things for himself, and Tom laughed andsaid that he was more open to that but wasfinding less time to do it with a new baby.

The therapist then asked Tom to read what hehad written about the meaning of the event forhimafter thework thathehaddone.Hewrote:

There is no doubt that this traumatic eventhas deeply impacted me. My thoughts aboutmyself, others, and the world were changed.When I started therapy, I believed that I wasa murderer. I blamed myself completely.Now, I believe that I shot a family, but I didnot murder them. I realize that I and othersaround me had to do what we did at thetime, and that we chose to shoot because wehad to. I will never know what that man, ormaybe even the family, was trying to do bygoing through that checkpoint, but I knownow that I had no choice but to shoot to stopthem. Regarding safety, I used to think thatthere were people out to get me, but now Irealize that the probability of that is slim. Istill feel a little anxious about myself, mywife, and now my daughter, getting hurt, butnot by a sniper. That seems unlikely. Now Iworry about the stuff that everyone worriesabout—like crazy drivers, illness, or someaccident. About safety, I used to worry that Iwas going to go “off” and hurt my family. Idon’t believe that I will do that, because I’venever done that before and basically thistrauma messed with my head about howlikely I would be to hurt someone unless Ihad to. I’m trusting myself more in terms ofthe decisions I make, and I have some morefaith and trust in my government now that Irealize I really needed to shoot in that situa-tion. I think I may always struggle withwanting to have power and control overthings, but I’m working on not having con-trol over everything. The fact is, I don’t havecontrol, even though I like to think that I do.My self-esteem is improving. I have to re-member that not every bad thing that hap-pens is my fault, and that I deserve to behappy even if I don’t fully believe it yet. Oneof the biggest things that seems to be chang-

ing is that I’m enjoying being close to mywife and my new daughter. I used to avoidmy wife, because I thought I didn’t deserveto be happy and that I might hurt her. SlowlyI’m realizing that it is not very likely that I’llhurt my wife or my new daughter, or at leasthurt them intentionally. My wife seems muchhappier now. I want to hold on to this timein my life and provide a good life for mydaughter and wife. I’m happy to know thatmy daughter is not going to know someonewho thought that snipers were out to gethim, and who was anxious, avoiding every-thing and everyone. It sounds silly, but I’mkind of glad that I went through this, be-cause I think I’m going to be better becauseof it.

Tom was a bit teary as he finished reading.The therapist asked Tom whether he remem-bered what he wrote the first time. Tom saidno, so the therapist read to him his first ImpactStatement. She pointed out that Tom had comea long way, and he agreed. The therapist andTom reviewed the whole therapy process, whatthey had covered, and the “stuck points” thatTom had challenged. Tom said that he was go-ing to continue using the worksheets, becausethey had been so helpful in making him slowdown to think about things instead of just re-acting. They did some lapse planning, and thetherapist asked Tom what he could do if hesensed that he was struggling with PTSD or de-pressive symptoms, or second-guessing his newways of thinking. He mentioned that he wasgoing to share the materials with his wife, be-cause she was very good at helping him to “gethis head on straight.” He also included on hislist a review of the materials he had completedduring the course of therapy. The therapy ses-sion ended with a discussion of Tom’s goal towrite his father a letter and to increase his con-tact with his siblings. He was planning to usethese contacts to discover more about the rea-sons his father was alcoholic and had seemedto abandon the family. Tom also shared hisgoals about the type of father and husband hehoped to be, and what his professional futureheld as he left the military. The therapist con-gratulated Tom on his willingness to do thehard work to recover from what happened tohim and wished him the best with his familyand future. Tom expressed his appreciation forthe therapy.


REFERENCES

Amdur, R. L., & Liberzon, I. (2001). The structure ofposttraumatic stress disorder symptoms in combatveterans: A confirmatory factor analysis of the Im-pact of Event Scale. Journal of Anxiety Disorders,15, 345–357.


Andrews, B., Brewin, C. R., Rose, S., & Kirk, M.(2000). Predicting PTSD symptoms in victims of vio-lent crime: The role of shame, anger, and childhoodabuse. Journal of Abnormal Psychology, 109, 69–73.

Asmundson, G. J. G., Frombach, I., McQuaid, J.,Pedrelli, P., Lenox, R., & Stein, M. B. (2000).Dimensionality of posttraumatic stress symptoms: Aconfirmatory factor analysis of DSM-IV symptomclusters and other symptom models. Behaviour Re-search and Therapy, 38, 203–214.

Astin, M. C., Layne, C. M., Camilleri, A. J., & Foy, D.W. (1994). Posttraumatic stress disorder invictimization-related traumata. In I. J. Briere (Ed.),Assessing and treating victims of violence: New di-rections for mental health services (pp. 39–51). SanFrancisco: Jossey-Bass.

Becker, J. V., Skinner, L. J., Abel, G. G., Axelrod, R., &Cichon, J. (1984). Sexual problems of sexual assaultsurvivors. Women Health, 9, 5–20.

Beckham, J. C., Feldman, M. E., & Kirby, A. C. (1998).Atrocities exposure in Vietnam combat veterans withchronic posttraumatic stress disorder: Relationshipto combat exposure, symptom severity, guilt, and in-terpersonal violence. Journal of Traumatic Stress, 11,777–785.

Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek,D. G., Gusman, F. D., Charney, D. S., et al. (1995).The development of a Clinician-Administered PTSDScale. Journal of Traumatic Stress, 8, 75–90.

Blanchard, E. B., Hickling, E. J., Buckley, T. C., & Tay-lor, A. E. (1996). Psychophysiology of posttraumaticstress disorder related to motor vehicle accidents:Replication and extension. Journal of Consulting andClinical Psychology, 64, 742–751.

Blanchard, E. B., Hickling, E. J., Devinei, T., Veazey, C.H., Galovski, T. E., & Mundy, E. (2003). A con-trolled evaluation of cognitive behavioral therapy forposttraumatic stress in motor vehicle accident survi-vors. Behaviour Research and Therapy, 41, 79–96.

Boudewyns, P. A., Hyer, L. A., Peralme, L., Touze, J., &Kiel, A. (1995, August). Eye movement desensitiza-tion and reprocessing (EMDR) and exposure therapyin the treatment of combat-related PTSD: An earlylook. Paper presented at the annual meeting of theAmerican Psychological Association, New York.

Bowen, G. R., & Lambert, J. A. (1986). Systematic de-sensitization therapy with post-traumatic stress dis-order cases. In C. R Figley (Ed.), Trauma and itswake: Vol. II. Traumatic stress theory, research, and

intervention (pp. 280–291). New York: Brunner/Ma-zel.

Breslau, N., Davis, G. C., & Andreski, P. (1995). Riskfactors for PTSD-related traumatic events: A pro-spective analysis. American Journal of Psychiatry,152, 529–535.

Brewin, C. R., Dalgleish, T., & Joseph, S. (1996). A dualrepresentation theory of posttraumatic stress disor-der. Psychological Review, 103, 670–686.

Briere, J. (1995). The Trauma Symptom Inventory(TSI): Professional manual. Odessa, FL: Psychologi-cal Assessment Resources.

Brom, D., Kleber, R. J., & Defares, P. B. (1989). Briefpsychotherapy for PTSD. Journal of Consulting andClinical Psychology, 57, 607–612.

Bryant, R. A., Harvey, A. G., Dang, S. T., Sackville, T.,& Basten, C. (1998). Treatment of acute stress disor-der: A comparison of cognitive-behavioral therapyand supportive counseling. Journal of Consultingand Clinical Psychology, 66, 862–866.

Bryant, R. A., Moulds, M. L., Guthrie, R. M., Dang, S.T., & Nixon, R. D. V. (2003). Imaginal exposurealone and imaginal exposure with cognitive restruc-turing in treatment of posttraumatic stress disorder.Journal of Consulting and Clinical Psychology, 71,706–712.

Buckley, T. C., Blanchard, E. B., & Hickling, E. J.(1998). A confirmatory factor analysis of posttrau-matic stress symptoms. Behaviour Research andTherapy, 36, 1091–1099.

Burt, M. R. (1980). Cultural myths and supports forrape. Journal of Personality and Social Psychology,38, 217–230.

Chard, K. M. (2005). An evaluation of cognitive pro-cessing therapy for the treatment of posttraumaticstress disorder related to childhood sexual abuse.Journal of Consulting and Clinical Psychology, 73,965–971.

Chemtob, C., Roitblat, H. L., Hamada, R. S., Carlson,J. G., & Twentyman, C. T. (1988). A cognitive actiontheory of post-traumatic stress disorder. Journal ofAnxiety Disorders, 2, 253–275.

Cloitre, M. (1998). Sexual revictimization: Risk factorsand prevention. In V. M. Follette, J. I. Ruzek, & F. R.Abueg (Eds.), Cognitive-behavioral therapies fortrauma (pp. 278–304). New York: Guilford Press.

Cloitre, M., Koenen, K. C., Cohen, L. R., & Han, H.(2002). Skills training in affective and interpersonalregulation followed by exposure: A phase-basedtreatment for PTSD related to childhood abuse. Jour-nal of Consulting and Clinical Psychology, 70, 1067–1074.

Cloitre, M., Stovall-McClough, K. C., Miranda, R., &Chemtob, C. M. (2004). Therapeutic alliance, neg-ative mood regulation, and treatment outcome inchild abuse-related posttraumatic stress disorder.Journal of Consulting and Clinical Psychology, 72,411–416.

Creamer, M., Burgess, P., & Pattison, P. (1992). Reac-


tions to trauma: A cognitive processing model. Jour-nal of Abnormal Psychology, 101, 452–459.

Creamer, M., Morris, P., Biddle, D., & Elliott, P. (1999).Treatment outcome in Australian veterans withcombat-related posttraumatic stress disorder: Acause for cautious optimism? Journal of TraumaticStress, 12, 545–558.

Dalgleish, T. (2004). Cognitive approaches to posttrau-matic stress disorder: The evolution ofmultirepresentational theorizing. Psychological Bul-letin, 130, 228–260.

Denny, N., Robinowitz, R., & Penk, W. (1987). Con-ducting applied research on Vietnam combat-relatedpost-traumatic stress disorder. Journal of ClinicalPsychology, 43, 56–66.

Derogatis, L. R. (1983). SCL-90-R: Administration,scoring and procedures manual-II. Towson, MD:Clinical Psychometric Research.

Devilly, G. J., & Spence, S. H. (1999). The relative effi-cacy and treatment distress of EMDR and acognitive-behavior trauma treatment protocol in theamelioration of posttraumatic stress disorder. Jour-nal of Anxiety Disorders, 13, 131–157.

Dohrenwend, B. P., Turner, J. B., Turse, N. A., Adams,B. G., Koenen, K. C., & Marshall, R. (2006). Thepsychological risks of Vietnam for U.S. veterans: Arevisit with new data and methods. Science, 313,979–982.

Ehlers, A., & Clark, D. M. (2000). A cognitive model ofposttraumatic stress disorder. Behaviour Researchand Therapy, 38, 319–345.

Ehlers, A., Clark, D. M., Hackmann, A., McManus, F.,Fennell, M., Herbert, C., et al. (2003). A randomizedcontrolled trial of cognitive therapy, a self-help book-let, and repeated assessments as early interventionsfor posttraumatic stress disorder. Archives of GeneralPsychiatry, 60, 1024–1032.

Ehlers, A., Hackmann, A., Steil, R., Clohessy, S.,Wenninger, K., & Winter, H. (2002). The nature ofintrusive memories after trauma: The warning signalhypothesis. Behaviour Research and Therapy, 40,995–1002.

Falsetti, S. A., Resnick, H. S., Resick, P. A., & Kilpat-rick, D. G. (1993). The Modified PTSD SymptomScale: A brief self-report measure of posttraumaticstress disorder. Behavior Therapist, 16, 161–162.

First, M. B., Spitzer, R. L., Williams, J. B. W., & Gibbon,M. (1995). Structured Clinical Interview for DSM-IV—Patient Edition (SCID-P). Washington, DC:American Psychiatric Press.

Foa, E. B. (1995). Posttraumatic Stress Diagnostic Scale(manual). Minneapolis, MN: National ComputerSystems.

Foa, E. B. (1996, November). Conceptualization ofpost-trauma psychopathology as failure in emotionalprocessing. Information Processing Theories: Varia-tions on a Theme symposium conducted at the meet-ing of the International Society for Traumatic StressStudies, San Francisco.

Foa, E. B., Dancu, C. V., Hembree, E. A., Jaycox, L. H.,Meadows, E. A., & Street, G. P. (1999). A compari-son of exposure therapy, stress inoculation training,and their combination for reducing posttraumaticstress disorder in female assault victims. Journal ofConsulting and Clinical Psychology, 67, 194–200.

Foa, E. B., Hembree, E. A., Cahill, S. E., Rauch, S. A.M., Riggs, D. S., Feeny, N. C., et al. (2005). Random-ized trial of prolonged exposure for posttraumaticstress disorder with and without cognitive restructur-ing: Outcome at academic and community clinics.Journal of Consulting and Clinical Psychology, 73,953–964.

Foa, E. B., Riggs, D. S., Dancu, C. V., & Rothbaum, B.O. (1993). Reliability and validity of a brief instru-ment for assessing post-traumatic stress disorder.Journal of Traumatic Stress, 6, 459–473.

Foa, E. B., Rothbaum, B., Riggs, D., & Murdock, T.(1991). Treatment of posttraumatic stress disorder inrape victims: A comparison between cognitive-behavioral procedures and counseling. Journal ofConsulting and Clinical Psychology, 59, 715–723.

Foa, E. B., & Rothbaum, B. O. (1998). Treating thetrauma of rape: Cognitive-behavioral therapy forPTSD. New York: Guilford Press.

Foa, E. B., Steketee, G., & Rothbaum, B. O. (1989).Behavioral/cognitive conceptualizations of post-traumatic stress disorder. Behavior Therapy, 20,155–176.

Frank, E., Anderson, B., Stewart, B. D., Dancu, C.,Hughes, C., & West, D. (1988). Efficacy of cognitivebehavior therapy and systematic desensitization inthe treatment of rape trauma. Behavior Therapy, 19,403–420.

Frank, E., & Stewart, B. D. (1983). Treating depressionin victims of rape. Clinical Psychologist, 36, 95–98.

Frank, E., & Stewart, B. D. (1984). Depressive symp-toms in rape victims: A revisit. Journal of AffectiveDisorders, 1, 269–277.

Glover, H., Pelesky, C. A., Bruno, R., & Sette, R.(1990). Post-traumatic stress disorder conflicts inVietnam combat veterans: A confirmatory factor an-alytic study. Journal of Traumatic Stress, 3, 573–591.

Green, B. L. (1996). Trauma History Questionnaire. InB. H. Stamm (Ed.), Measurement of stress, trauma,and adaptation (pp. 366–369). Lutherville, MD:Sidran Press.

Griffin, M. G., Resick, P. A., & Mechanic, M. B. (1997).Objective assessment of peritraumatic dissociation:Psychophysiological indicators. American Journal ofPsychiatry, 154, 1081–1088.

Griffin, M. G., Uhlmansiek, M. H., Resick, P. A., & Me-chanic, M. B. (2004). Comparison of the Posttrau-matic Stress Disorder Scale versus the Clinician-Administered Posttraumatic Stress Disorder Scale indomestic violence survivors. Journal of TraumaticStress, 17, 497–504.

Gutner, C., Rizvi, S. L., Monson, C. M., & Resick, P. A.(2006). Changes in coping strategies, relationship to


the perpetrator, and posttraumatic stress disorder infemale crime victims. Journal of Traumatic Stress,19, 813–823.

Hembree, E. A., Cahill, S. E., & Foa, E. B. (2004). Im-pact of personality disorders on treatment outcomefor female assault survivors with chronic posttrau-matic stress disorder. Journal of Personality Disor-ders, 18, 117–127.

Henning, K. R., & Frueh, B. C. (1997). Combat guiltand its relationship to PTSD symptoms. Journal ofClinical Psychology, 53, 801–808.

Hoge, C. W., Auchterlonie, J. L., & Milliken, C. S.(2006). Mental health problems, use of mental healthservices, and attrition from military service after re-turning from deployment to Iraq or Afghanistan.Journal of the American Medical Association, 295,1023–1032.

Hoge, C. W., Castro, C. A., Messer, S. C., McGurk, D.,Cotting, D. I., & Koffman, R. L. (2004). Combatduty in Iraq and Afghanistan, mental health prob-lems, and barriers to care. New England Journal ofMedicine, 351, 13–22.

Holmes, M. R., & St. Lawrence, J. S. (1983). Treatmentof rape-induced trauma: Proposed behavioral con-ceptualization and review of the literature. ClinicalPsychology Review, 3, 417–433.

Horowitz, M. J. (1986). Stress response syndromes (2nded.). New York: Aronson.

Horowitz, M. J., Wilner, N., & Alvarez, W. (1979). Im-pact of Event Scale: A measure of subjective stress.Psychosomatic Medicine, 41, 209–218.

Ironson, G., Freund, B., Strauss, J. L., & Williams, J.(2002). Comparison of two treatments for traumaticstress: A community-based study of EMDR and Pro-longed Exposure. Journal of Clinical Psychology, 58,113–128.

Janoff-Bulman, R. (1985). The aftermath of victimiza-tion: Rebuilding shattered assumptions. In C. R.Figley (Ed.), Trauma and its wake: Vol. I. The studyand treatment of posttraumatic stress disorder (pp.15–35). New York: Brunner/Mazel.

Janoff-Bulman, R. (1992). Shattered assumptions: To-wards a new psychology of trauma. New York: FreePress.

Jensen, J. A. (1994). An investigation of eye movementdesensitization and reprocessing (EMD/R) as a treat-ment for posttraumatic stress disorder (PTSD) symp-toms of Vietnam combat veterans. Behavior Therapy,25, 311–325.

Jordan, B. K., Marmar, C. R., Fairbank, J. A., Schlenger,W. E., Kulka, R. A., Hough, R. L., et al. (1992).Problems in families of male Vietnam veterans withposttraumatic stress disorder. Journal of Consultingand Clinical Psychology, 60, 916–926.

Keane, T. M., Brief, D. J., Pratt, E. M., & Miller, M. W.(2007). Assessment and its comorbidities in adults. InM. J. Friedman, T. M. Keane, & P. A. Resick (Eds.),Handbook of PTSD: Science and practice (pp. 279–305). New York: Guilford Press.

Keane, T. M., Caddell, J. M., & Taylor, K. L. (1988).

Mississippi Scale for Combat-Related PosttraumaticStress Disorder: Three studies in reliability and valid-ity. Journal of Consulting and Clinical Psychology,56, 85–90.

Keane, T. M., Fairbank, J. A., Caddell, J. M., &Zimering, R. T. (1989). Implosive (flooding) therapyreduces symptoms of PTSD in Vietnam combat veter-ans. Behavior Therapy, 20, 245–260.

Keane, T. M., Kaloupek, D. G., & Kolb, L. C. (1998).VA Cooperative Study #334: I. Summary of findingson the psychological assessment of PTSD. PTSD Re-search Quarterly, 9, 1–4.

Keane, T. M., Kolb, L. C., Kaloupek, D. G., Orr, S. P.,Blanchard, E. B., Thomas, R. G., et al. (1998). Utilityof psychophysiology measurement in the diagnosis ofposttraumatic stress disorder: Results from a Depart-ment of Veteran’s Affairs cooperative study. Journalof Consulting and Clinical Psychology, 66, 914–923.

Keane, T. M., Malloy, P. F., & Fairbank, J. A. (1984).Empirical development of an MMPI subscale for theassessment of combat-related posttraumatic stressdisorder. Journal of Consulting and Clinical Psychol-ogy, 52, 888–891.

Keane, T. M., Zimering, R. T., & Caddell, J. M. (1985).A behavioral formulation of posttraumatic stress dis-order. Behavior Therapist, 8, 9–12.

Kessler, R. C., Berglund, P., Demler, O., Jin, R.,Merikangas, K. R., & Walters, E. E. (2005). Lifetimeprevalence and age-of-onset distributions of DSM-IVdisorders in the National Comorbidity Survey Repli-cation. Archives of General Psychiatry, 62, 593–602.

Kessler, R. C., Chiu, W. T., Demler, O., Merikangas, K.R., & Walters, E. E. (2005). Prevalence, severity, andcomorbidity of 12-month DSM-IV disorders in theNational Comorbidity Survey Replication. Archivesof General Psychiatry, 62, 617–627.

Kessler, R. C., Sonnega, A., Bromet, E., Hughes, M., &Nelson, C. B. (1995). Posttraumatic stress disorder inthe National Comorbidity Survey. Archives of Gen-eral Psychiatry, 52, 1048–1060.

Kilpatrick, D. G. (1983). Rape victims: Detection, as-sessment and treatment. Clinical Psychologist, 36,92–95.

Kilpatrick, D. G., & Amick, A. E. (1985). Rape trauma.In M. Hersen & C. Last (Eds.), Behavior therapycasebook (pp. 86–103). New York: Springer.

Kilpatrick, D. G., Edmunds, C. N., & Seymour, A. K.(1992). Rape in America: A report to the nation.Arlington, VA: National Victim Center.

Kilpatrick, D. G., Resnick, H. S., & Freedy, J. R. (1991).The Potential Stressful Events Interview. Unpub-lished instrument, National Crime Victims Researchand Treatment Center, Medical University of SouthCarolina, Charleston.

Kilpatrick, D. G., Saunders, B. E., Veronen, L. J., Best,C. L., & Von, J. M. (1987). Criminal victimization:Lifetime prevalence, reporting to police, and psycho-logical impact. Crime and Delinquency, 33, 479–489.

Kilpatrick, D. G., & Veronen, L. J. (1983). Treatment


for rape-related problems: Crisis intervention is notenough. In L. H. Cohen, W. L. Claiborn, & G. A.Specter (Eds.), Crisis intervention (pp. 165–185).New York: Human Sciences Press.

Kilpatrick, D. G., Veronen, L. J., & Best, C. L. (1985).Factors predicting psychological distress among rapevictims. In C. R. Figley (Ed.), Trauma and its wake:Vol. I. The study and treatment of posttraumaticstress disorder (pp. 114–141). New York: Brunner/Mazel.

Kilpatrick, D. G., Veronen, L. J., & Resick, P. A. (1982).Psychological sequelae to rape: Assessment and treat-ment strategies. In D. M. Doleys, R. L. Meredith, &A. R. Ciminero (Eds.), Behavioral medicine: Assess-ment and treatment strategies (pp. 473–497). NewYork: Plenum Press.

King, D. W., Leskin, G. A., King, L. A., & Weathers, F.W. (1998). Confirmatory factor analysis of theClinician-Administered PTSD Scale: Evidence for thedimensionality of posttraumatic stress disorder. Psy-chological Assessment, 10, 90–96.

Koss, M. P., & Harvey, M. R. (1991). The rape victim:Clinical and community interventions (2nd ed.).Thousand Oaks, CA: Sage.

Kubany, E. S., Abueg, F. R., Owens, J. A., Brennan, J.M., Kaplan, A. S., & Watson, S. B. (1995). Initial ex-amination of a multidimensional model of trauma-related guilt: Applications to combat veterans andbattered women. Journal of Psychopathology andBehavioral Assessment, 17, 353–376.

Kubany, E. S., Haynes, S. N., Abueg, F. R., & Manke, F.P. (1996). Development and validation of theTrauma-Related Guilt Inventory (TRGI). Psychologi-cal Assessment, 8, 428–444.

Kubany, E. S., Haynes, S. N., Leisen, M. B., Owens, J.A., Kaplan, A. S., Watson, S. B., et al. (2000). Devel-opment and preliminary validation of a brief broad-spectrum measure of trauma exposure: The Traumat-ic Life Events Questionnaire. Psychological Assess-ment, 12, 210–224.

Kubany, E. S., Hill, E. E., Owens, J. A., Iannce-Spencer,C., McCaig, M. A., & Tremayne, K. J. (2004). Cog-nitive trauma therapy for battered women withPTSD (CTT-BW). Journal of Consulting and ClinicalPsychology, 72, 3–18.

Kubany, E. S., Leisen, M. B., Kaplan, A. S., & Kelly, M.P. (2000). Validation of a brief measure of posttrau-matic stress disorder: The Distressing Event Ques-tionnaire (DEQ). Psychological Assessment, 12,197–209.

Kulka, R. A., Schlenger, W. E., Fairbank, J. A., Hough,R. L., Jordan, B. K., Marmar, C. R., et al. (1990).Trauma and the Vietnam war generation: Report offindings from the National Vietnam Veterans Read-justment Study. New York: Brunner/Mazel.

Lang, P. J. (1968). Fear reduction and fear behavior:Problems in treating a construct. In J. M. Schlien(Ed.), Research in psychotherapy (pp. 90–102).Washington, DC: American Psychological Associa-tion.

Lang, P. J. (1977). Imagery in therapy: An informationprocessing analysis of fear. Behavior Therapy, 8,862–886.

Lauterbach, D., & Vrana, S. R. (1996). Three studies onthe reliability and validity of a self-report measure ofposttraumatic stress disorder. Assessment, 3, 17–25.

Lee, C., Gavriel, H., Drummond, P., Richards, J., &Greenwald, R. (2002). Treatment of PTSD: Stress in-oculation training with prolonged exposure com-pared to EMDR. Journal of Clinical Psychology, 58,1071–1089.

Leskela, J., Dieperink, M., & Thuras, P. (2002). Shameand posttraumatic stress disorder. Journal of Trau-matic Stress, 15, 223–226.

Mahoney, M. J., & Lyddon, W. J. (1988). Recent devel-opments in cognitive approaches to counseling andpsychotherapy. Counseling Psychologist, 16, 190–234.

Marks, I., Lovell, K., Noshirvani, H., Livanou, M., &Thrasher, S. (1998). Treatment of posttraumaticstress disorder by exposure and/or cognitive restruc-turing: A controlled study. Archives of General Psy-chiatry, 55, 317–325.

McCann, I. L., & Pearlman, L. A. (1990). Vicarioustraumatization: A framework for understanding thepsychological effects of working with victims. Jour-nal of Traumatic Stress, 3, 131–149.

McFall, M., Fontana, A., Raskind, M., & Rosenheck,R. (1999). Analysis of violent behavior in Vietnamcombat veteran psychiatric inpatients with posttrau-matic stress disorder. Journal of Traumatic Stress, 12,501–517.

McNair, L. D., & Neville, H. A. (1996). African Ameri-can women survivors of sexual assault: The intersec-tion of race and class. In M. Hill & E. D. Rothblum(Eds.), Classism and feminist therapy: Counting costs(pp. 107–118). New York: Haworth Press.

Meichenbaum, D. H. (1985). Stress inoculation train-ing. Elmsford, NY: Pergamon Press.

Miller, M. W., Kaloupek, D. G., Dillon, A. L., & Keane,T. M. (2004). Externalizing and internalizing sub-types of combat-related PTSD: A replication and ex-tension using the PSY-5 scales. Journal of AbnormalPsychology, 113, 636–645.

Miller, M. W., Kaloupek, D. G., & Keane, T. M. (1999,October). Antisociality and physiological hypo-responsivity during exposure to trauma-related stim-uli in patients with PTSD. Poster presented at the39th Annual Meeting of the Society for Psycho-physiological Research, Granada, Spain.

Monson, C. M., Price, J. L., & Ranslow, E. (2005, Oc-tober). Treating combat PTSD through cognitive pro-cessing therapy. Federal Practitioner, 75–83.

Monson, C. M., Schnurr, P. P., Resick, P. A., Friedman,M. J., Young-Xu, Y., & Stevens, S. P. (2006). Cogni-tive processing therapy for veterans with military-related posttraumatic stress disorder. Journal of Con-sulting and Clinical Psychology, 74, 898–907.

Mowrer, O. H. (1947). On the dual nature of learning—a re-interpretation of “conditioning” and “problem-


solving.” Harvard Educational Review, 14, 102–148.

Neuner, F., Schauer, M., Klaschik, C., Karunakara, U.,& Elbert, T. (2004). A comparison of narrative expo-sure therapy, supportive counseling, and psychoedu-cation for treating posttraumatic stress disorder in anAfrican refugee settlement. Journal of Consultingand Clinical Psychology, 72, 579–587.

Norris, F. (1990). Screening for traumatic stress: A scalefor use in the general population. Journal of AppliedSocial Psychology, 20, 1704–1718.

Norris, F. H. (1992). Epidemiology of trauma: Fre-quency and impact of different potentially traumaticevents on different demographic groups. Journal ofConsulting and Clinical Psychology, 60, 409–418.

Oquendo, M. A., Friend, J. M., Halberstam, B.,Brodsky, B. S., Burke, A. K., Grunebaum, M. F., et al.(2003). Association of comorbid posttraumatic stressdisorder and major depression with greater risk forsuicidal behavior. American Journal of Psychiatry,160, 580–582.

Orr, S. P., Lasko, N. B., Metzger, L. J., Berry, N. J.,Ahern, C. E., & Pitman, R. K. (1998).Psychophysiologic assessment of women with post-traumatic stress disorder resulting from childhoodsexual abuse. Journal of Consulting and Clinical Psy-chology, 66, 906–913.

Orr, S. P., Metzger, L. J., Miller, M. W., & Kaloupek, D.G. (2004). Psychophysiological assessment of PTSD.In J. P. Wilson & T. M. Keane (Eds.), Assessing psy-chological trauma and PTSD (2nd ed., pp. 289–343).New York: Guilford Press.

Paunovic, N., & Öst, L. G. (2001). Cognitive-behaviortherapy versus exposure therapy in the treatment ofPTSD in refugees. Behaviour Research and Therapy,39, 1183–1197.

Pitman, R. K., Orr, S. P., Altman, B., Longpre, R. E.,Poire, R. E., & Lasko, N. B. (1993, October). A con-trolled study of eye movement desensitization/repro-cessing (EMDR) treatment of post-traumatic stressdisorder. Presentation at International Society forTraumatic Stress Studies, San Antonio, TX.

Pitman, R. K., Orr, S. P., Forgue, D. F., & Altman, B.(1990). Psychophysiologic responses to combat im-agery of Vietnam veterans with posttraumatic stressdisorder versus other anxiety disorders. Journal ofAbnormal Psychology, 99, 49–54.

Pitman, R. K., Orr, S. P., Forgue, D. F., de Jong, J., &Claiborn, J. M. (1987). Psychophysiologic assess-ment of posttraumatic stress disorder imagery inVietnam combat veterans. Archives of General Psy-chiatry, 44, 970–975.

Power, K., McGoldrick, T., Brown, K., Buchanan, R.,Sharp, D., & Swanson, V. (2002). A controlled com-parison of eye movement desensitization and repro-cessing versus exposure plus cognitive restructuringversus wait list in the treatment of posttraumaticstress disorder. Clinical Psychology and Psychothera-py, 9, 299–318.

Prins, A., Ouimette, P., Kimerling, R., Camerond, R. P.,

Hugelshofer, D. S., Shaw-Hegwer, J., et al. (2004).The Primary Care PTSD Screen (PC-PTSD): Develop-ment and operating characteristics. Primary CarePsychiatry, 9, 9–14.

Renfrey, G., & Spates, C. R. (1994). Eye movement de-sensitization: A partial dismantling study. Journal ofBehavior Therapy and Experimental Psychiatry, 25,231–239.

Resick, P. A., Jordan, C. G., Girelli, S. A., Hutter, C. K.,& Marhoeder-Dvorak, S. (1988). A comparative out-come study of behavioral group therapy for sexualassault victims. Behavior Therapy, 19, 385–401.

Resick, P. A., & Markaway, B. E. (1991). Clinical treat-ment of adult female victims of sexual assault. In C.R. Hollin & K. Howells (Eds.), Clinical approachesto sex offenders and their victims (pp. 261–284).London: Wiley.

Resick, P. A., Nishith, P., & Griffin, M. G. (2003). Howwell does cognitive-behavioral therapy treat symp-toms of complex PTSD?: An examination of childsexual abuse survivors within a clinical trial. CNSSpectrums, 8, 340–355.

Resick, P. A., Nishith, P., Weaver, T. L., Astin, M. C., &Feuer, C. A. (2002). A comparison of cognitive pro-cessing therapy, prolonged exposure and a waitingcondition for the treatment of posttraumatic stressdisorder in female rape victims. Journal of Con-sulting and Clinical Psychology, 70, 867–879.

Resick, P. A., & Schnicke, M. K. (1992). Cognitive pro-cessing therapy for sexual assault victims. Journal ofConsulting and Clinical Psychology, 60, 748–756.

Resick, P. A., & Schnicke, M. K. (1993). Cognitive pro-cessing therapy for rape victims: A treatment manual.Newbury Park, CA: Sage.

Resnick, H. S., Kilpatrick, D. G., Dansky, B. S.,Saunders, B. E., & Best, C. L. (1993). Prevalence ofcivilian trauma and posttraumatic stress disorder in arepresentative national sample of women. Journal ofConsulting and Clinical Psychology, 61, 984–991.

Resnick, H. S., Kilpatrick, D. G., & Lipovsky, J. A.(1991). Assessment of rape-related posttraumaticstress disorder: Stressor and symptom dimensions[Special section: Issues and methods in assessment ofposttraumatic stress disorder]. Psychological Assess-ment, 3, 561–572.

Robins, L. N., Helzer, J. E., Croughan, J., & Ratcliff, K.S. (1981). National Institute of Mental Health Diag-nostic Interview Schedule: Its history, characteristics,and validity. Archives of General Psychiatry, 38,381–389.

Rosenheck, R., & Fontana, A. (1996). PTSD andcommunity-based treatment: A commentary on“PTSD diagnosis and treatment for mental health cli-nicians.” Community Mental Health Journal, 32,191–193.

Rosenheck, R., Fontana, A., & Cottrol, C. (1995). Ef-fect of clinician–veteran racial pairing in the treat-ment of posttraumatic stress disorder. AmericanJournal of Psychiatry, 152, 555–563.

Rothbaum, B. O. (1998, March). Virtual reality expo-


sure therapy for Vietnam veterans with PTSD: Pre-liminary results. Paper presented at the Lake GeorgeConference on Posttraumatic Stress Disorder, BoltonLanding, NY.

Rothbaum, B. O., Astin, M. C., & Marsteller, F. (2005).Prolonged exposure versus eye movement desensiti-zation and reprocessing (EMDR) for PTSD rape vic-tims. Journal of Traumatic Stress, 18, 607–616.

Rothbaum, B. O., & Foa, E. B. (1992). Exposure ther-apy for rape victims with post-traumatic stress disor-der. Behavior Therapist, 15, 219–222.

Rothbaum, B. O., & Foa, E. B. (1993). Subtypes ofposttraumatic stress disorder and duration of symp-toms. In J. R. T. Davidson & E. B. Foa (Eds.), Post-traumatic stress disorder: DSM-IV and beyond (pp.23–35). Washington, DC: American PsychiatricPress.

Saunders, B. E., Arata, C. M., & Kilpatrick, D. G.(1990). Development of a crime-related post-traumatic stress disorder scale for women within theSymptom Checklist–90—Revised. Journal of Trau-matic Stress, 3, 439–448.

Schindler, F. E. (1980). Treatment by systematic desensi-tization of a recurring nightmare of a real life trauma.Journal of Behavior Therapy and Experimental Psy-chiatry, 11, 53–54.

Schnurr, P. P., Friedman, M. J., Engel, C. C., Foa, E. B.,Shea, T., Chow, B. K., et al. (2007). Cognitive-behavioral therapy for posttraumatic stress disorderin women: A randomized controlled trial. Journal ofthe American Medical Association, 297(8), 820–830.

Schnurr, P. P., Friedman, M. J., Foy, D. W., Shea, M. T.,Hsieh, F. Y., Lavori, P. W., et al. (2003). Randomizedtrial of trauma-focused group therapy for posttrau-matic stress disorder: Results from a Department ofVeterans Affairs cooperative study. Archives of Gen-eral Psychiatry, 60, 481–489.

Shalev, A. Y., Orr, S. P., & Pitman, R. K. (1992).Psychophysiologic response during script-driven im-agery as an outcome measure in posttraumatic stressdisorder. Journal of Clinical Psychiatry, 53, 324–326.

Shapiro, F. (1989). Eye movement desensitization: Anew treatment for post-traumatic stress disorder.Journal of Behavior Therapy and Experimental Psy-chiatry, 20, 211–217.

Shapiro, F. (1995). Eye movement desensitization andreprocessing: Basic principles, protocols, and proce-dures. New York: Guilford Press.

Shay, J., & Munroe, J. (1999). Group and milieu ther-apy for veterans with complex posttraumatic stressdisorder. In P. A. Saigh & J. D. Bremner (Eds.), Post-traumatic stress disorder: A comprehensive text(pp. 391–413). Needham Heights, MA: Allyn & Ba-con.

Shin, L. M., Rauch, S. L., & Pitman, R. K. (2005).Structural and functional anatomy of PTSD: Findingsfrom neuroimaging research. In J. J. Vasterling & C.R. Brewin (Eds.), Neuropsychology of PTSD: Biolog-ical, cognitive, and clinical perspectives (pp. 59–82).New York: Guilford Press.

Silverman, D. (1977). First do no more harm: Femalerape victims and the male counselor. American Jour-nal of Orthopsychiatry, 47(1), 91–96.

Simms, L. J., Watson, D., & Doebbeling, B. N. (2002).Confirmatory factor analyses of posttraumatic stresssymptoms in deployed and nondeployed veterans ofthe Gulf War. Journal of Abnormal Psychology, 111,637–647.

Smith, M. Y., Redd, W., DuHamel, K., Vicksberg, S. J.,& Ricketts, P. (1999). Validation of the PTSDChecklist—Civilian version in survivors of bone mar-row transplantation. Journal of Traumatic Stress, 12,485–499.

Sutker, P. B., Uddo-Crane, M., & Allain, A. N. (1991).Clinical and research assessment of posttraumaticstress disorder: A conceptual overview [Special sec-tion: Issues and methods in assessment of posttrau-matic stress disorder]. Psychological Assessment, 3,520–530.

Tarrier, N., Pilgrim, H., Sommerfield, C., Faragher, B.,Reynolds, M., Graham, E., et al. (1999). A random-ized trial of cognitive therapy and imaginal exposurein the treatment of chronic posttraumatic stress dis-order. Journal of Consulting and Clinical Psychology,67, 13–18.

Tarrier, N., & Sommerfield, C. (2004). Treatment ofchronic PTSD by cognitive therapy and exposure: 5year follow-up. Behavior Therapy, 35, 231–246.

Taylor, S., Kuch, K., Koch, W. J., Crockett, D. J., &Passey, G. (1998). The structure of posttraumaticstress symptoms. Journal of Abnormal Psychology,107, 154–160.

Taylor, S., Thordarson, D. S., Maxfield, L., Fedoroff, I.C., Lovell, K., & Orgodniczuk, J. (2003). Compara-tive efficacy, speed, and adverse effects of three PTSDtreatments: Exposure therapy, EMDR, and relax-ation training. Journal of Consulting and ClinicalPsychology, 71, 330–338.

Tolin, D. F., & Foa, E. B. (2006). Sex differences in trau-ma and posttraumatic stress disorder: A quantitativereview of 25 years of research. Psychological Bulle-tin, 132, 959–992.

Turner, S. M. (1979, December). Systematic desensitiza-tion of fears and anxiety in rape victims. Paper pre-sented at the Association of Advancement of Behav-ior Therapy, San Francisco.

U.S. Department of Health and Human Services.(2001). Mental health: A report of the Surgeon Gen-eral. Washington, DC: Author.

Vrana, S., & Lauterbach, D. (1994). Prevalence of trau-matic events and post-traumatic psychological symp-toms in a nonclinical sample of college students.Journal of Traumatic Stress, 7, 289–302.

Weathers, F., Litz, B. T., Herman, D. S., Huska, J. A., &Keane, T. M. (1993, October). The PTSD Checklist(PCL): Reliability, validity, and diagnostic utility.Presentation at the International Society for Trau-matic Stress Studies, San Antonio, TX.

Weathers, F. W., & Keane, T. M. (1999). Psychologicalassessment of traumatized adults. In P. A. Saigh & J.


D. Bremner (Eds.), Posttraumatic stress disorder: Acomprehensive text (pp. 219–247). Boston: Allyn &Bacon.

Weathers, F. W., Keane, T. M., & Davidson, J. R.(2001). Clinician-Administered PTSD Scale: A re-view of the first ten years of research. Depression andAnxiety, 13, 132–156.

Weathers, F. W., Ruscio, A. M., & Keane, T. M.(1999). Psychometric properties of nine scoringrules for the Clinician-Administered PosttraumaticStress Disorder Scale. Psychological Assessment,11, 124–133.

Weiss, D. S., & Marmar, C. R. (1997). The Impact ofEvent Scale—Revised. In J. P. Wilson & T. M. Keane(Eds.), Assessing psychological trauma and PTSD(pp. 399–411). New York: Guilford Press.

Wilson, A. E., Calhoun, K. S., & Bernat, J. A. (1999).Risk recognition and trauma-related symptomsamong sexually revictimized women. Journal of Con-sulting and Clinical Psychology, 67, 705–710.

Wilson, D., Silver, S. M., Covi, W., & Foster, S. (1996).Eye movement desensitization and reprocessing: Ef-fectiveness and automatic correlates. Journal ofBehavior Therapy and Experimental Psychiatry, 27,219–229.

Wong, M. R., & Cook, D. (1992). Shame and its contri-bution to PTSD. Journal of Traumatic Stress, 5, 557–562.

Zoellner, L. A., Feeny, N. C., Fitzgibbons, L. A., & Foa,E. B. (1999). Response of African American andCaucasian women to cognitive behavioral therapyfor PTSD. Behavior Therapy, 30, 581–595.


C H A P T E R 3

Social Anxiety Disorder

CYNTHIA L. TURKRICHARD G. HEIMBERG

LEANNE MAGEE

Many people are very shy and somewhat inhibited. For this reason, the suffering associ-ated with social anxiety disorder is often minimized as a common trait in the populationthat does not require a heavy artillery of formalized treatment interventions (either drugsor psychological treatments). Nothing could be further from the truth. For members of avery large segment of the population with debilitating social anxiety (over 12% and in-creasing), at some point in their lives the seemingly simple process of interacting with peo-ple or forming relationships provokes overwhelming terror and is often avoided. The ef-fects on career and quality of life can be devastating. This chapter examines the latestiteration of an established psychological treatment for social anxiety disorder. Asis increasingly true of our new generation of psychological interventions, cognitive-behavioral group therapy has proven to be significantly better than equally credible butless focused psychological interventions, and its effect is increasingly powerful over time.As such, this treatment is among the best of the new generation of psychological treat-ments characterized by power and specificity. The case of Josie, new to this edition, alsoillustrates the maturity and clinical sophistication of this remarkable approach to socialanxiety.—D. H. B.

The National Comorbidity Survey ReplicationStudy (NCS-R), which assessed over 9,000noninstitutionalized individuals throughoutthe United States, found that 12.1% of peoplehave social anxiety disorder at some point dur-ing their lives (Kessler et al., 2005). In this sur-vey, social anxiety disorder (also known associal phobia; Liebowitz, Heimberg, Fresco,Travers, & Stein, 2000) was the fourth mostcommon psychiatric disorder, with only majordepressive disorder, alcohol abuse, and specificphobia being more prevalent. More conserva-tive lifetime prevalence estimates suggest thatclinically significant social anxiety affects a

compelling but more modest 4% of the popula-tion (Narrow, Rae, Robins, & Regier, 2002).

When social anxiety disorder was first in-cluded as a diagnostic category in the third edi-tion of the Diagnostic and Statistical Manual ofMental Disorders (DSM-III; American Psychi-atric Association, 1980), it was thought to re-sult in only minimal disruption in role func-tioning. Research has since revealed that socialanxiety disorder can be quite incapacitating.The vast majority of individuals with socialanxiety disorder report that their career, aca-demic, and general social functioning havebeen seriously impaired by their fears

123

(Katzelnick et al., 2001; Schneier et al., 1994;Turner, Beidel, Dancu, & Keys, 1986). In onestudy, despite a mean age in the early 30s, 50%of the individuals with social anxiety disorderhad never married, compared to 36% of indi-viduals with panic disorder and agoraphobia,and 18% of those with generalized anxiety dis-order (Sanderson, Di Nardo, Rapee, & Barlow,1990). In another study, individuals with socialanxiety disorder were more likely than theirnonanxious counterparts to work at a job be-low their level of educational attainment and tobelieve that their supervisors did not think thatthey fit into the work environment (Bruch,Fallon, & Heimberg, 2003). Symptoms of so-cial anxiety are also associated with low lifesatisfaction, even after taking into account thelevel of disability engendered by these symp-toms (Hambrick, Turk, Heimberg, Schneier, &Liebowitz, 2003).

Social anxiety disorder most commonly be-gins during early childhood or adolescence(Schneier, Johnson, Hornig, Liebowitz, &Weissman, 1992) and typically follows an un-remitting course (Chartier, Hazen, & Stein,1998; Reich, Goldenberg, Vasile, Goisman, &Keller, 1994). Nevertheless, most individualswith social anxiety disorder do not seek treat-ment unless they develop an additional disor-der (Schneier et al., 1992). Unfortunately, rec-ognition of social anxiety disorder appears tobe poor among health care professionals(Bisserbe, Weiller, Boyer, Lepine, & Lecrubier,1996; Katzelnick et al., 2001; Weiller, Bisserbe,Boyer, Lepine, & Lecrubier, 1996).

Approximately 70–80% of individuals withsocial anxiety disorder meet criteria for addi-tional diagnoses, and, in most cases, social anx-iety disorder predates the onset of the comor-bid condition (W. Magee, Eaton, Wittchen,McGonagle, & Kessler, 1996; Schneier et al.,1992). In community samples, the most com-mon additional diagnoses include specific pho-bia, agoraphobia, major depression, and alco-hol abuse and dependence (W. Magee et al.,1996; Schneier et al., 1992). Compared to indi-viduals with uncomplicated social anxiety dis-order, persons with social anxiety disorder andcomorbid disorders have higher rates of suicideattempts, are more likely to report significantrole impairment, and more often use medica-tion to control their symptoms (W. Magee etal., 1996; Schneier et al., 1992). Comorbidity isalso associated with more severe impairmentbefore and after cognitive-behavioral therapy

(CBT), although individuals with and withoutcomorbid conditions make similar gains(Erwin, Heimberg, Juster, & Mindlin, 2002).Comorbid mood disorders appear to be morestrongly associated with impairment than docomorbid anxiety disorders (Erwin et al.,2002).

SUBTYPES OF SOCIAL ANXIETYDISORDER AND AVOIDANTPERSONALITY DISORDER

Socially anxious individuals are a heteroge-neous group in terms of the pervasiveness andseverity of their fears. In the current diagnosticsystem, the generalized subtype is specified forindividuals who fear most social situations.These individuals often simultaneously have avariety of social interaction fears (e.g., dating,joining an ongoing conversation, being asser-tive), performance fears (e.g., public speaking,playing a musical instrument in front of oth-ers), and observation fears (e.g., working infront of others, walking down the street). Thenongeneralized subtype is “a heterogeneousgroup that includes persons who fear a singleperformance situation as well as those who fearseveral, but not most, social situations” (Amer-ican Psychiatric Association, 1994, p. 413). Forexample, individuals who fear public speakingor eating in front of others but otherwise feelcomfortable interacting with and being ob-served by others would be assigned to thenongeneralized subtype.

Relative to the nongeneralized subtype, thegeneralized subtype has been associated withearlier age of onset, decreased educationalattainment, higher rates of unemployment,and a greater likelihood of being unmarried(Heimberg, Hope, Dodge, & Becker, 1990b;Mannuzza et al., 1995). This group also expe-riences more depression, social anxiety, avoid-ance, fear of negative evaluation, and func-tional impairment (e.g., Brown, Heimberg, &Juster, 1995; Herbert, Hope, & Bellack, 1992;Turner, Beidel, & Townsley, 1992). Neverthe-less, clients with generalized social anxiety dis-order improve as much as clients with thenongeneralized subtype after CBT (Brown etal., 1995; Hope, Herbert, & White, 1995;Turner, Beidel, Wolff, Spaulding, & Jacob,1996). However, because clients with general-ized social anxiety disorder begin treatmentwith greater impairment, they remain more im-


paired after receiving the same number of treat-ment sessions. Therefore, clients with general-ized social anxiety disorder may require alonger course of treatment to achieve outcomessimilar to those of clients with nongeneralizedsocial anxiety.

The generalized subtype of social anxietydisorder has many features in common withavoidant personality disorder (APD). In thecurrent diagnostic system, APD is character-ized by a long-standing pattern “of social inhi-bition, feelings of inadequacy, and hypersensi-tivity to negative evaluation” (AmericanPsychiatric Association, 1994, p. 664). Giventhe similarity between the descriptions of thetwo disorders, it is not surprising that many in-dividuals who meet criteria for generalized so-cial anxiety disorder also meet criteria forAPD. However, there is little scientific evidenceto suggest that individuals meet criteria forAPD without also meeting criteria for socialanxiety disorder (Widiger, 1992). The mostparsimonious description of the relationshipbetween social anxiety disorder and APD isthat they are not different disorders and thatindividuals meeting criteria for both disordersare simply the most severely impaired personswith social anxiety disorder (Heimberg, Holt,Schneier, Spitzer, & Liebowitz, 1993). With re-gard to treatment outcome, some studies havefound that clients with and without comorbidAPD make similar gains (Brown et al., 1995;Hofmann, Newman, Becker, Taylor, & Roth,1995; Hope, Herbert, & White, 1995), al-though others have found that comorbid APDis associated with a poorer treatment response(Chambless, Tran, & Glass, 1997; Feske, Perry,Chambless, Renneberg, & Goldstein, 1996).As with individuals meeting criteria for gener-alized social anxiety disorder, clients with APDmay require a longer course of treatment toachieve an optimal outcome.

OVERVIEW OF THE TREATMENTOUTCOME LITERATURE

Researchers have investigated the efficacy of abroad range of treatments for social anxietydisorder, including social skills training, cogni-tive therapy, relaxation training, exposure, in-terpersonal psychotherapy, dynamically ori-ented supportive psychotherapy, and variouspharmacotherapies. This review examines onlystudies that have tested the efficacy of com-

bined exposure and cognitive treatment. Thecombination of exposure and cognitive restruc-turing has been the most frequently studiedform of psychosocial intervention for socialanxiety disorder and represents the focus of theintervention described in this chapter (abroader review of the treatment outcome liter-ature is provided by Rodebaugh, Holaway, &Heimberg, 2004).

Researchers have often sought to demon-strate that the efficacy of exposure is improvedby the addition of cognitive restructuring. Thedata from such efforts are mixed. For example,in an early study, Butler, Cullington, Munby,Amies, and Gelder (1984) compared in vivo ex-posure to in vivo exposure plus an anxietymanagement program that comprised distrac-tion, relaxation, and rational self-talk, and awaiting-list control. At posttest, both exposuretreatments were more effective than thewaiting-list control across a variety of mea-sures, with few differences between the activetreatments. However, at follow-up, partici-pants who received exposure plus anxiety man-agement training (which included a cognitivecomponent) fared better than participants whoreceived exposure alone. Similarly, Mattick andcolleagues (Mattick & Peters, 1988; Mattick,Peters, & Clarke, 1989), in two studies exam-ining the efficacy of a combined package of ex-posure and cognitive restructuring, providedevidence for a somewhat greater advantage forcombined exposure and cognitive restructuringrelative to exposure alone. However, Taylor etal. (1997a) found no differences between par-ticipants randomized initially to receive cogni-tive restructuring or a control therapy prior toboth groups receiving exposure treatment; thatis, exposure with and without prior cognitiverestructuring produced similar improvement inoutcomes for the socially anxious participantsin this study.

Heimberg, Dodge, and colleagues (1990)conducted the first controlled trial of his grouptreatment for social anxiety disorder, whichcomprised integrated exposure, cognitive re-structuring, and homework assignments (i.e.,cognitive-behavioral group therapy [CBGT]).This treatment was compared to an attentioncontrol treatment that comprised educationabout social anxiety disorder and nondirectivesupportive group therapy. CBGT participantsreported less anxiety during an individualizedbehavioral test and were more likely than con-trol participants to be rated as improved by a

Social Anxiety Disorder 125

clinical assessor. At 6-month follow-up, the in-ternal dialogue of CBGT participants was char-acterized by a “positive dialogue,” a state asso-ciated with good mental health according toSchwartz and Garamoni’s (1989) states ofmind (SOM) model. The internal dialogue ofattention control participants, in contrast, rep-resented a “negative monologue,” the mostpathological state of mind described in theSOM model (Bruch, Heimberg, & Hope,1991). A 5-year follow-up of a subset of partic-ipants from the original sample indicated thatindividuals who had received CBGT were morelikely than comparable attention control par-ticipants to maintain their gains (Heimberg,Salzman, Holt, & Blendell, 1993).

In a component analysis of CBGT, Hope,Heimberg, and Bruch (1995) reported thatCBGT and exposure alone were both more ef-fective than a waiting-list control. At posttest,there was evidence that the exposure-alonecondition was more effective than CBGT, butthese differences disappeared at 6-monthfollow-up.

In a two-site study, 133 clients were ran-domly assigned to receive CBGT, the mono-amine oxidase inhibitor phenelzine, pill pla-cebo, or the attention control psychotherapydeveloped by Heimberg, Dodge, and colleagues(1990, 1998). At posttest, independent asses-sors classified 21 of 28 CBGT completers(75%) and 20 of 26 phenelzine completers(77%) as treatment responders. The responserates for phenelzine and CBGT clients were sig-nificantly better than those for clients receivingpill placebo or the attention control psycho-therapy. In the second phase of this study, cli-ents who responded to CBGT or phenelzinewere continued through 6 additional months ofmaintenance treatment and a 6-month follow-up period (Liebowitz et al., 1999). At study’send, 50% of previously responding clients whoreceived phenelzine relapsed, compared to only17% of clients who received CBGT.

A waiting-list controlled study of an individ-ually administered version of this treatmentwas recently completed (Zaider, Heimberg,Roth, Hope, & Turk, 2003). The individualcognitive-behavioral treatment consistentlyoutperformed the waiting list on both self-report and clinician-administered measures ofsocial anxiety. Effect sizes were large, and littledropout occurred.

Controlled trials of Clark’s (1997) version ofcognitive-behavioral treatment for social anxi-

ety disorder have also yielded large effect sizesand appear promising (Clark et al., 2003,2006; Stangier, Heidenreich, Peitz, Lauterbach,& Clark, 2003). Clark’s protocol includes bothcognitive restructuring and exposure elements.However, his treatment also emphasizes proce-dures derived from his theoretical work andbasic research, such as identification of safetybehaviors, shifting focus of attention awayfrom the self (to the social situation), behavior-al experiments in which safety behaviors aredropped, and video feedback on social perfor-mance.

In summary, it appears that a combinedpackage of exposure and cognitive restructur-ing is an effective intervention for social anxi-ety disorder. Whether it is the most effectivetreatment, or more effective than exposurealone, is more difficult to determine given thelimited number of studies and mixed results.However, more important to practitioners isthe mounting evidence that three of four clientswith social anxiety disorder are likely to realizeclinically significant change after a reasonablyintensive trial of combined exposure and cogni-tive restructuring.

AN INTEGRATEDCOGNITIVE-BEHAVIORAL MODELOF SOCIAL ANXIETY DISORDER

Rapee and Heimberg (1997; see also Heimberg& Becker, 2002; Roth & Heimberg, 2001;Turk, Lerner, Heimberg, & Rapee, 2001) de-veloped an integrated cognitive-behavioralmodel that describes how individuals with so-cial anxiety disorder process information whenconfronted with a situation that holds the po-tential for negative evaluation (see Figure 3.1).The process begins when the socially anxiousindividual is in the presence of an audience.The term “audience” captures the sense thatsocially anxious individuals have of being “onstage” in the presence of others, whether thesituation is a speech, social interaction, or an-other situation in which the person may be ob-served by others. Socially anxious persons per-ceive this audience as inherently critical (e.g.,Leary, Kowalski, & Campbell, 1988) and ashaving standards that they are unlikely to meet(e.g., Wallace & Alden, 1991). Trower andGilbert (1989; Gilbert, 2001) also suggest thatsocially anxious individuals view their audi-ence as competitors. According to their


psychobiological/ethological theory, two sys-tems have evolved that allow social groups todevelop and maintain order. The defense sys-tem involves the establishment of dominantand subordinate roles within social groups. In-dividuals are ranked within hierarchies, andthe ultimate goal is to have dominant status,because higher levels in the hierarchy are asso-ciated with greater access to resources (e.g.,mates, food). In contrast, in the safety system, amore recent and highly evolved system of socialinteraction, group members offer signals of re-

assurance and positive reinforcement instead ofthreats to maintain structure. Inevitably, thiscreates less social anxiety between members,decreases the need to maintain a constant stateof social comparison, and encourages a lessthreatening and more cooperative environ-ment. Trower and Gilbert suggest that every-one has access to both of these systems, but so-cially anxious individuals are more prone toutilize the defense system, even when the socialenvironment is nonthreatening. Thus, sociallyanxious individuals view others as threatening


FIGURE 3.1. A model of the generation and maintenance of anxiety in social-evaluative situations. FromRapee and Heimberg (1997). Copyright 1997 by Elsevier Science Ltd. Reprinted by permission.

competitors who have the goal of dominatingthem. In fact, socially anxious individuals havebeen shown to perceive social interactions asmore competitive than do their nonanxiouscounterparts and to view themselves as comingacross to others as submissive (Hope, Sigler,Penn, & Meier, 1998). To illustrate, one clientwho avoided dancing in public made progresson this issue when he reconceptualized dancingas an opportunity for people to have fun to-gether rather than a competition in which ev-eryone judges and criticizes the skills of every-one else.

In the presence of an audience, socially anx-ious individuals construct a mental representa-tion of how they appear to these other people.This mental representation of the self com-prises negative and distorted images in whichthe self is viewed from an observer perspective(Hackmann, Surawy, & Clark, 1998). Theseimages have been found to be related to memo-ries of early adverse social events such as beingbullied (Hackmann, Clark, & McManus,2000). For example, one client described beingmercilessly teased by her peers as a young ado-lescent for being tall and skinny. Although atthe time of treatment she was an attractiveadult, she still described her appearance as“gangly,” “awkward,” and “ugly.” Researchhas also shown that socially anxious individu-als rate their own social behavior more harshlythan do objective observers (Rapee & Lim,1992; Stopa & Clark, 1993) and overestimatethe visibility of their anxiety relative to ratingsby objective observers (Bruch, Gorsky, Collins,& Berger, 1989; Norton & Hope, 2001). Un-doubtedly, these beliefs negatively impact theirimage of how they are coming across to others.The image of the self is further influenced byinternal cues (e.g., somatic sensations of perspi-ration may produce images of sweat drippingdown one’s face) and external cues such as oth-ers’ reactions (e.g., someone glancing at one’sclothing and frowning might lead to images ofappearing disheveled).

Beliefs that they are unacceptable to otherpeople, that other people are inherently critical,and that the evaluation of others is extremelyimportant motivate individuals with socialanxiety disorder to be hypervigilant for earlyindications of disapproval from others (e.g.,frowns, yawns) and aspects of their ownbehavior or appearance that might elicit nega-tive evaluation from others (e.g., making an in-

ane comment, not being dressed appropriately,visibly shaking). The division of attentional re-sources among external social threats, the (dis-torted) mental representation of the self as seenby others, and the demands of the current so-cial task may result in actual performance defi-cits (which may then elicit actual negativesocial feedback). In effect, socially anxiouspersons operate within the equivalent of amultiple-task paradigm, which increases theprobability of disrupted social performance(MacLeod & Mathews, 1991). Therefore,complex social tasks are more likely than lesscomplex tasks to result in poorer performance,due to limited processing resources.

Persons with social anxiety disorder also at-tempt to predict the standards that the audienceholds for them in the situation. Characteristicsof the audience (e.g., importance, attractiveness)and features of the situation (e.g., whether it isformal or informal) influence the projected stan-dards of the audience. Individuals with socialanxiety disorder then attempt to judge the extentto which their current mental representation oftheir appearance and behavior matches the pre-dicted standards of the audience. Of course,given the negative bias present in their mentalself-representation, they are likely to concludethat they are falling short of the audience’s ex-pectations, and painful outcomes such as loss ofsocial status and rejection are likely follow. Neg-ative predictions result in cognitive, behavioral,and physiological symptoms of anxiety thateventually feed back into the negatively biasedmental representation of the self as seen by oth-ers and perpetuate the cycle of anxiety.

Clark and Wells (1995) emphasize that whenin social situations, socially anxious individualsmay engage in safety behaviors. Safety behav-iors are essentially behaviors that the personperforms in an attempt to avoid the negativeoutcomes that he or she fears. For example, aman who is afraid of his hand shaking whiledrinking coffee may grip the cup tightly, hopingto prevent any observable trembling. Suchcompensatory acts are viewed as helpful by theindividual with social anxiety disorder, butthese behaviors often make the fearedoutcome—in this case, shaking—more likely.Furthermore, they increase the likelihood thatindividuals with social anxiety disorder willconclude that they cannot make it through thesocial situation successfully in the absence ofsafety behaviors (Clark & Wells, 1995).


TREATMENT RATIONALEFROM A COGNITIVE-BEHAVIORALPERSPECTIVE

The heart of our approach to treatment in-volves the integration of exposure and cogni-tive restructuring (Heimberg & Becker, 2002;Hope, Heimberg, Juster, & Turk, 2000). Expo-sure is beneficial for several reasons that arecommunicated to the client. From the perspec-tive of the model, one of the most important as-pects of exposure is the opportunity to test dys-functional beliefs (e.g., “I will get too anxiousto finish the speech”) and to generate more re-alistic ways of understanding the self and oth-ers (e.g., “I can continue the speech even if I amanxious”). Exposure also allows clients to ex-perience the natural reduction in anxiety thatcomes with staying in a feared situation for aprolonged period of time on several occasions(i.e., habituation). Last, exposure allows clientsto practice perhaps long-avoided behavioralskills (e.g., asking someone for a date, being as-sertive) and to engage in behaviors consistentwith their long-term, valued goals (e.g., engag-ing in small talk with new people today maylead to friendships in the future).

From the perspective of the model, cognitiverestructuring is also important for several rea-sons communicated to the client. Clients learnto treat their anxiety-provoking thoughts andbeliefs as hypotheses and to explore whetherthere are more helpful or realistic ways of view-ing the situation, the self, and others. As clientscome to view social situations as less threaten-ing, they are often more willing to confrontthose situations in exposures. Furthermore, ad-dressing dysfunctional cognitions often freesup attentional resources that can be used to in-crease focus on the social task at hand and po-tentially improve performance. Additionally,cognitive restructuring may help clients to takecredit for successes and cope with disappoint-ments after exposures. The use of cognitiveskills to diffuse postevent rumination may beespecially important given that more rumina-tion following one anxiety-provoking event hasbeen shown to be associated with greater initialanxiety prior to confronting another anxi-ety provoking event (Kocovski, Rector, &Denisoff, 2004). Last, as clients’ cognitive as-sessments of the danger inherent in social situa-tions become more realistic, their physiologicalsymptoms of anxiety often diminish as well.

In-session exposures, with cognitive restruc-turing occurring before, during, and after eachexposure, are viewed as particularly crucial.With the gentle encouragement and emotionalsupport of the therapist, clients are often will-ing to engage in behaviors in session that theyhave avoided outside of session for many years.Successful in-session exposures can provide cli-ents with the confidence and motivation to tryout these behaviors in the real world. In-sessionexposures may also be more easily graduatedthan in vivo exposures. For example, a clientwho is terrified of formal public speakingmight do an initial in-session exposure thatcomprises reading an article aloud from a mag-azine while sitting down. The next exposuremight involve standing up while reading the ar-ticle aloud, then taking questions from a smallaudience. In-session exposures also provide thetherapist with the opportunity to teach clientsabout the principles of effective exposure, suchas staying in a feared social situation for a pro-longed period of time without avoiding, even ifanxiety escalates. In-session exposures also al-low the therapist to see firsthand any subtleavoidance behaviors (i.e., safety behaviors)that the client may be employing to manageanxiety. For example, a client who fears thatother people might reject him if they got toknow him personally may assail a conversationpartner with questions and not reveal any per-sonal information. A client who is afraid of thereaction of others to her speech might neverlook up from her notes, so that she never has tosee any bored or critical faces (and, of course,also never has a chance to see any interested orapproving faces). As noted by Clark and Wells(1995), such safety behaviors may have the un-intended effect of making individuals appearsocially awkward and increase the likelihoodof negative reactions from others. Further-more, clients do not learn that they might havemanaged the social situation as well, if notbetter, without engaging in the safety behav-iors. In-session exposures provide the therapistwith the opportunity to point out these behav-iors and to set goals for the next exposure thatinvolve fully confronting the feared situation(e.g., looking at the audience during the speech,revealing personal information during a con-versation). In-session exposures also providethe opportunity for clients to receive directfeedback on their social performance in a waythat is often not possible in real life. For exam-


ple, a client who believes that she trembles somuch during public speaking that others pityher can ask her exposure audience how notice-able the trembling was and what they thoughtof her (rarely will an audience reply that ananxious person was pitied). Last, in-session ex-posures provide the opportunity for the ther-apist to assist clients in applying cognitiverestructuring skills to an actual anxiety-provoking situation. Once in-session exposuresbegin, homework involves confronting similarsituations outside of session. As with in-sessionexposures, clients are asked to engage in cogni-tive restructuring activities before, during, andafter each in vivo exposure.

TREATMENT-RELATED VARIABLES

Group versus Individual Treatment

Treatment of social anxiety disorder has mostcommonly been conducted in a group format(e.g., Heimberg, Dodge, et al., 1990; Heimberget al., 1998), which has been conceptualized ashaving a number of advantages over the indi-vidual format. It provides opportunities forvicarious learning, support from others withsimilar problems, the availability of multiplerole-play partners, informal exposure stem-ming from group participation, and a range ofpeople to provide evidence to counter distortedthinking (Heimberg & Becker, 2002; Sank &Shaffer, 1984). Nevertheless, the group formathas a number of potential disadvantages aswell. Treatment cannot begin until a group ofapproximately six clients has been assembled,which means that the first few individuals join-ing the group may have to wait for an extendedperiod before starting treatment. Some of theseindividuals lose motivation and drop out be-fore the group begins. Group treatment alsoprovides less flexibility in making up missedsessions. Although group treatment providesopportunities for vicarious learning, this bene-fit may be offset by the reduced time availablefor the idiosyncratic needs of each client.Clinically, some clients appear to get discour-aged and lose motivation when they see othergroup members progressing at a more rapidpace than themselves. Additionally, even whenefforts are made to screen potential groupmembers, a group may include a client who isdomineering, aggressive, or otherwise difficultto manage, and may have a negative impact on

the group. Last, some individuals never fullyhabituate to the group setting, and their anxi-ety may interfere with their ability to processinformation during sessions.

Little available data speak to the issue ofwhether individual or group therapy providesbetter outcomes. In their meta-analysis, Gould,Buckminster, Pollack, Otto, and Yap (1997) re-ported that group and individual cognitivebehavioral interventions for social anxiety dis-order produced similar effect sizes. Group andindividual formats have rarely been comparedwithin the same study. Lucas and Telch (1993)found that group and individual versions ofHeimberg’s treatment resulted in similar gains,both superior to an educational-supportivecontrol group. Scholing and Emmelkamp(1993) examined various combinations of ex-posure and cognitive restructuring delivered ingroup or individual formats and found no clearadvantage for either treatment modality. Simi-larly, Wlazlo, Schroeder-Hartwig, Hand, Kai-ser, and Münchau (1990) found no differencesin outcome between group and individual ex-posure treatment. One study did find that anindividual version of CBT for social anxietydisorder was superior to the group version, andthat both formats were superior to a waitinglist (Stangier et al., 2003). However, the datado not clearly suggest an advantage for eithermodality in the treatment for social anxiety dis-order.

Heimberg and Becker (2002) provide a de-scription of CBGT for social anxiety disorder.This chapter provides a description of the indi-vidual version of this treatment, which hasbeen the focus of our research in recent years(see also Hope et al., 2000; Hope, Heimberg,& Turk, 2006).

Appropriate Clients

In anxiety studies to date, social anxiety disor-der has been the primary diagnosis of the par-ticipants. As previously discussed, social anxi-ety disorder is often comorbid with otherdisorders (W. Magee et al., 1996; Schneier etal., 1992). In such cases, the clinician and clientneed to reach a joint decision that social anxi-ety will be the focus of treatment for a specifiedperiod of time (e.g., 16 weeks). Clinically, it isimportant to clarify this issue prior to begin-ning treatment, because some clients may ex-press a desire to switch the focus of treatment


to another problem area when it is time to be-gin exposures. In general, our policy is to re-mind the client of the original treatment con-tract and goals, and to encourage him or her to“avoid avoidance.” In some cases, the best clin-ical decision is to change the focus of treatmentdue to changes in the client’s life circumstancesor symptoms. Nevertheless, we are especiallywary of altering the treatment plan when theclient’s desire to change the focus of treatmentcoincides with the initiation of exposures.Therapists without a strong background inCBT or exposure therapy may be at risk forchanging the focus of treatment too readily,partially due to their own discomfort with con-ducting in-session exposures. Sometimes acompromise can be reached in which exposuresremain in the treatment plan but some sessiontime is set aside to deal with other issues.

Appropriate Therapists

The ideal therapist has a strong background inthe theoretical underpinnings of CBT for anxi-ety disorders, experience conducting expo-sures, good basic therapy skills, and experiencewith social anxiety in particular. Little isknown about the effectiveness of manualizedtreatments when implemented by clinicianswithout extensive training in the manual’stheoretical underpinnings and procedures(Chambless & Ollendick, 2001). Clearly, moreresearch is needed on this issue. That said, ourclinical experience suggests that with good su-pervision, this treatment can be successfullyimplemented by novice therapists seeing theirvery first clients (in fact, the case study pre-sented later was treated by L. Magee, who atthe time was a novice therapist) and experi-enced therapists with a theoretical orientationother than a cognitive-behavioral one. Ex-perienced clinicians with a good cognitive-behavioral background are likely to do quitewell working independently with thismanualized treatment. Moreover, our clientworkbook (Hope et al., 2000) and therapistguide (Hope et al., 2006) were designed withthe intention of being sufficiently detailed to al-low therapists with a wide variety of back-grounds to provide this treatment, althoughsupervision and/or consultation with an experi-enced cognitive-behavioral therapist on thefirst few cases and the more complex caseswould be beneficial.

Medication and CBTfor Social Anxiety Disorder

Many clients are already taking psychotropicmedications when they present for therapy.Some take medication to control their socialanxiety, whereas others take medications forcomorbid conditions. We do not require clientsto discontinue medications prior to startingCBT (unless dictated by a specific research pro-tocol). However, we do ask clients to stabilizetheir dosage before starting treatment and torefrain from changing their dosage or tryingany new medications during treatment.Clinically, we want clients to attribute positivechanges in their symptoms to the work they aredoing in therapy rather than to changes in theirmedication regimen. Individuals who takemedication on an as-needed basis are asked torefrain from doing so before sessions or expo-sure homework assignments. The same requestis made of individuals who use nonprescriptiondrugs or alcohol to control their anxiety.

PRETREATMENT ASSESSMENTAND PREPARATION FOR TREATMENT

Assessment should play an integral part indiagnosis, case conceptualization, treatmentplanning, and termination decisions. Althoughadministering all of the measures presentedhere is not practical in most clinical settings, wehighly recommend administering a subset ofthe self-report and clinician-rated instrumentsdescribed below, and a behavioral assessment,if at all possible (for a more thorough reviewof assessment of social anxiety disorder, seeHeimberg & Turk, 2002).

Clinical Interview

The Anxiety Disorders Interview Schedule forDSM-IV (ADIS-IV; Brown, Di Nardo, &Barlow, 1994) and the ADIS-IV Lifetime Ver-sion (ADIS-IV-L; Di Nardo, Brown, & Barlow,1994) provide a thorough diagnostic assess-ment of the anxiety disorders and include mod-ules for mood disorders, substance use disor-ders, and disorders that overlap with anxietydisorders in terms of presenting symptoms(e.g., hypochondriasis, somatization disorder).There are screening questions for other majordisorders (e.g., psychosis). A 0–8 clinician’s se-


verity rating (CSR) designates the degree of dis-tress and interference associated with each di-agnosis. A CSR of 4 or higher suggests that theclient meets criteria for the specific DSM-IV di-agnosis. When multiple disorders receive aCSR of 4 or higher, the disorder receiving thehighest CSR is designated as the principal diag-nosis and the other disorders are designated asadditional. In a major reliability study of theADIS-IV-L, 362 individuals seeking treatmentat an anxiety specialty clinic received two in-dependent administrations of the ADIS-IV(Brown, Di Nardo, Lehman, & Campbell,2001). Social anxiety disorder evidenced goodreliability when examined as a principal diag-nosis (kappa = .77), as a clinical (principal oradditional) diagnosis (kappa = .77), and as apast (lifetime) diagnosis (kappa = .72). TheCSR for social anxiety disorder (r = .80) alsoevidenced good interrater reliability.

Our clinics routinely employ the ADIS-IVduring the first client visit. Semistructured in-terviews such as the ADIS-IV assist with differ-ential diagnosis and in providing systematicassessment for a broad range of comorbid con-ditions that may affect the course of treatment.Nevertheless, diagnostic interviews may becostly and time-consuming. Regardless ofwhether a semistructured or unstructured clini-cal interview is conducted, devoting time to athorough and accurate diagnostic evaluation iscrucial to appropriate treatment planning. Cli-ents whose symptoms are better accounted forby another diagnosis may experience, amongother things, little improvement with treatmenttargeting social anxiety, unnecessarily pro-longed suffering as they go without the appro-priate treatment, and increased hopelessnessand financial hardship.

Self-Report Instruments

At the Adult Anxiety Clinic of Temple Univer-sity and the Anxiety Clinic at Washburn Uni-versity, clients complete a packet of ques-tionnaires prior to treatment. Self-reportinstruments are an important complement tothe diagnostic interview. In an interview, at atime when they may experience feelings ofshame and embarrassment, clients may give de-scriptions of their problems that do not fullyreflect the severity or pervasiveness of their so-cial fear and avoidance. Furthermore, norma-tive data on questionnaires allow the clinicianto evaluate the severity of a client’s symptoms

relative to meaningful reference points. Impor-tantly, pretreatment scores provide a baselineagainst which treatment progress can be as-sessed.

Measures of Social Anxiety

A number of questionnaires are available to as-sess social anxiety and avoidance. The SocialInteraction Anxiety Scale (SIAS) and SocialPhobia Scale (SPS) are commonly used com-panion measures designed specifically for theassessment of social anxiety disorder (Mattick& Clarke, 1998). The SIAS assesses fear of in-teracting in dyads and groups, and the SPS as-sesses fear of being observed by others (e.g.,eating in front of others). In most publishedstudies, the SIAS and SPS contain 20 itemseach. For each instrument, respondents ratehow characteristic each statement is of them ona 5-point Likert-type scale (0, Not at all; 4, Ex-tremely). Multiple studies suggest that the SIASand SPS are reliable and valid measures, andsensitive to the effects of CBT (e.g., Brown etal., 1997; Cox, Ross, Swinson, & Direnfeld,1998; Heimberg, Mueller, Holt, Hope, &Liebowitz, 1992; Ries et al., 1998).

Another commonly used instrument devel-oped to assess social anxiety disorder, theSocial Phobia and Anxiety Inventory (SPAI;Turner, Beidel, Dancu, & Stanley, 1989), com-prises a Social Phobia subscale, an Agorapho-bia subscale, and a derived Difference (or To-tal) score (i.e., Social Phobia–Agoraphobiasubscales). The SPAI contains 45 items, 21 ofwhich require multiple responses. For example,for the item that begins “I attempt to avoid so-cial situations where there are . . . ,” the clientseparately rates how frequently he or sheavoids situations involving strangers, authorityfigures, the opposite sex, and people in general.The client makes a total of 109 responses, mak-ing the administration and scoring of the SPAIrelatively time-consuming. Despite the disad-vantages imposed by the length of the scale, thequantity and specificity of information that itelicits can be quite helpful in case formulationand treatment planning. The Social Phobiasubscale assesses somatic, cognitive, andbehavioral responses to a variety of interaction,performance, and observation situations. TheAgoraphobia subscale assesses anxiety in situa-tions commonly feared by individuals withpanic disorder with agoraphobia (e.g., waitingin line). Respondents rate how frequently they


feel anxious in each situation using a 7-pointLikert-type scale (1, Never; 7, Always). TheDifference score is intended to provide an in-dex of social anxiety and avoidance distinctfrom the sometimes similar concerns of clientswith agoraphobia. However, the very high cor-relation between the Difference score and thescore of the Social Phobia subscale (r = .91)(Ries et al., 1998) suggests that there may belittle benefit to this strategy. Multiple studiessuggest that the SPAI is a reliable and valid in-strument (e.g., Beidel, Turner, Stanley, &Dancu, 1989; Herbert, Bellack, & Hope, 1991;Turner et al., 1989), and sensitive to treatment-related change (e.g., Cox et al., 1998; Ries etal., 1998; Taylor, Woody, McLean, & Koch,1997b).

Another scale that may prove to be quite use-ful is the 17-item Social Phobia Inventory, alsoknown as the SPIN (Antony, Coons, McCabe,Ashbaugh, & Swinson, 2006; Connor et al.,2000), which has demonstrated good reliability,significant correlations with related measures,and the ability to discriminate between clientswith social anxiety disorder and those with otheranxiety disorders. A 3-item version of the SPIN,the Mini-SPIN, has also shown substantial util-ity as a screening device for social anxiety disor-der in both general health care (Connor, Kobak,Churchill, Katzelnick, & Davidson, 2001) andanxiety specialty clinic (Weeks, Spokas, &Heimberg, in press) settings.

Although developed and validated with col-lege students prior to the inclusion of socialanxiety disorder in the DSM, the original Fearof Negative Evaluation scale (FNE; Watson &Friend, 1969) and the brief version of the scale(BFNE; Leary, 1983) continue to be widelyused, because they target the core construct ofthe disorder. The FNE comprises 30 items andemploys a true–false format. In treatment stud-ies, changes on the FNE have been found topredict end-state functioning (Mattick & Pe-ters, 1988; Mattick et al., 1989). The FNE ap-pears to be sensitive to the effects of treatment,although changes are typically small in magni-tude (Heimberg, 1994). The BFNE contains 12items, uses a 5-point Likert-type format (1, Notat all characteristic of me; 5, Extremely charac-teristic of me), and correlates highly with theoriginal scale (r = .96) (Leary, 1983). A recentstudy by Rodebaugh, Woods, and colleagues(2004) that utilized a large sample of under-graduates found that the BFNE was more sen-sitive than the FNE to differing degrees of fear

of negative evaluation. A study by Weeks andcolleagues (2005) that utilized a large sampleof clients with social anxiety disorder foundthat the BFNE had high internal consistency,correlated with measures of social anxiety,showed good discriminant validity, and wassensitive to the effects of CBT. Both of thesestudies suggest quite compellingly that thereverse-scored items of the BFNE may detractfrom its validity, and we now score only thestraightforwardly worded items. Others havetaken a similar approach by rewording thereverse-scored items to be consistent with thestraightforwardly worded items (e.g., Collins,Westra, Dozois, & Stewart, 2005).

In a recently developed measure for the as-sessment of session-by-session change in thetreatment of social anxiety disorder, the SocialAnxiety Session Change Index (SASCI; Hayes,Miller, Hope, Heimberg, & Juster, in press), cli-ents use a 7-point Likert-type scale to indicatehow much they feel that they have changedsince the beginning of therapy on four dimen-sions: anxiety, avoidance, concern about hu-miliation and embarrassment, and interference.More specifically, the SASCI measures howanxious the respondent becomes in anticipa-tion of or when he or she is in social or perfor-mance situations; how much the respondentavoids social or performance situations; howconcerned the respondent is about embarrass-ing or humiliating him- or herself in front ofothers; and how much the respondent’s anxietyinterferes with work or social activities. A totalscore of 16 indicates no change since the begin-ning of treatment. Scores of 4–15 indicate im-provement, whereas scores of 17–28 indicatedeterioration. The internal consistency of theSASCI across sessions ranged from .84 to .94across sessions (M = .89) in the study by Hayes,Miller, and colleagues (in press). Change on theSASCI was related to change in fear of negativeevaluation and to clinician-rated improvement,but not to ratings of anxiety sensitivity or de-pression. The use of the SASCI and the BFNEas session-by-session measures is demonstratedin the case study reported later in this chapter.

Other Self-Report Measures

In addition to measures targeting social anxi-ety, we administer questionnaires assessingother constructs relevant to case conceptualiza-tion and treatment outcome. We routinely ad-minister the Beck Depression Inventory–II


(BDI-II; Beck, Steer, & Brown, 1996), whichassesses symptoms of depression, including theaffective, cognitive, behavioral, somatic, andmotivational components, as well as suicidalwishes. We also use the Liebowitz Self-RatedDisability Scale (Schneier et al., 1994), whichassesses impairment across 11 domains (e.g.,school, work, alcohol abuse), and the Qualityof Life Inventory (Frisch, 1994), which assessesthe client’s overall sense of well-being and satis-faction with life.

Feedback and Treatment Contract Interview

At the session following the intake session, theinterviewer reviews the assessment data withthe client, explains diagnoses, answers ques-tions, and offers treatment recommendations.For clients with comorbid diagnoses, the bene-fits of initially making social anxiety the pri-mary focus of treatment for 12–16 weekly ses-sions is discussed, as noted earlier. Last, clientsare assured that the severity of their social anx-iety symptoms and any comorbid conditionswill be reassessed following this initial courseof treatment and, if necessary, additional treat-ment recommendations will be made at thattime. Additional assessments may be con-ducted at this visit, or clients may be scheduledfor one additional session to conduct furtherassessments before treatment formally begins.The additional assessments may includeclinician-administered measures of social anxi-ety and behavior tests.

Clinician-Administered Measuresof Social Anxiety

The two most commonly used clinician-administered measures of social anxiety are theLiebowitz Social Anxiety Scale (LSAS;Liebowitz, 1987) and the Brief Social PhobiaScale (BSPS; Davidson et al., 1991, 1997). Bothmeasures have been shown to have good psy-chometric properties (Heimberg & Turk,2002). Our clinics routinely utilize the LSAS,so we describe it here. This measure is com-pleted at the second clinic visit largely to avoidmaking the initial intake session undulylengthy. The LSAS separately evaluates Fearand Avoidance of 11 interaction (e.g., talkingto people in authority) and 13 performance(e.g., working while being observed) situationsusing 4-point Likert-type scales. Summing theFear and Avoidance ratings for all 24 items

yields an index of overall severity. The LSAShas been shown to have good reliability andconvergent/discriminant validity (e.g., Cox etal., 1998; Heimberg et al., 1999). It discrimi-nates well between clients with social anxietydisorder and generalized anxiety disorder(Heimberg & Holaway, in press). It hasalso demonstrated sensitivity to cognitive-behavioral and pharmacological treatment ofsocial anxiety disorder (e.g., Heimberg et al.,1998). When the LSAS is accompanied by de-tailed instructions, it appears that it also hasgood reliability and validity as a self-report in-strument (Fresco et al., 2001).

Behavioral Assessment

Behavior tests, or role-played re-creations ofsocial situations relevant to the client, can be avery useful addition to the self-report andclinician-administered measures describedhere. Although these tests may be quite anxietyprovoking for the client, the information ob-tained during a behavior test is unique and im-portant in several ways. Individuals with socialanxiety disorder are likely to describe their so-cial behavior as inadequate (e.g., Rapee & Lim,1992; Stopa & Clark, 1993) and their anxietyas highly visible to others (e.g., Norton &Hope, 2001). However, behavior tests oftendemonstrate that these reports are largely inac-curate and are essentially examples of distortedbeliefs. Furthermore, level of anxiety and qual-ity of performance exhibited by a client duringbehavior tests may be used to calibrate the dif-ficulty of exposures and increase the likelihoodthat the therapist creates a first exposure thatprovides both a challenge and a successful ex-perience for the client.

During behavior tests, the clinician may askfor anxiety ratings before, at several pointsduring, and after the role play. Other assess-ments, such as asking clients to rate the qualityof their performance or list the thoughts theyhad during the role play, are easily incorpo-rated. Affordable ambulatory physiologicalequipment, particularly pulse monitors, is in-creasingly available and adds a useful dimen-sion. The clinician can choose to test the limitsof a client’s ability to perform when anxious byasking him or her to remain in the situation fora certain period of time (typically 4–5 minutes).On the other hand, the clinician may give theclient explicit permission to stop (e.g., Ries etal., 1998) when the anxiety is excessive, and la-


tency to escape the situation becomes a mea-sure of avoidance.

Behavior tests may either be standardized orindividualized. Standardized role plays allowfor the observation of differences across pa-tients for a particular task. Commonly usedtasks include a conversation with a same-sexstranger, a conversation with an opposite-sexstranger, a conversation with two or more peo-ple, and a speech to a small audience. In con-trast, the advantage of individualized behaviortests is that they may be designed more pre-cisely to target idiosyncratic fears. For exam-ple, a client with a fear of displaying a handtremor when eating in front of others may ex-perience little anxiety during a one-on-one con-versation. However, if that client is asked to en-gage in a one-on-one conversation while eatinga bowl of soup, his or her anxiety is more likelyto be activated.

INDIVIDUAL CBT FOR SOCIALANXIETY DISORDER

Principles and Logistics

For our studies on individual treatment ofsocial anxiety disorder, we have establishedguidelines to which our therapists must ad-here (Hope et al., 2006). One issue facingmanualized treatments is that it is unclear towhat extent one may take liberties with howthe treatment is carried out and still expect out-comes similar to those obtained during efficacystudies. Therefore, although we recognize thatthis treatment may be applied more flexibly inclinical settings, we present these guidelines tohelp therapists understand the treatment as ithas been most thoroughly studied.

The treatment comprises 16 weekly 1-hoursessions within a period of 16–20 weeks. Al-though completing 16 sessions within 16 weeksmight be ideal, up to 20 weeks are allowed totake into consideration sessions missed due toillness, vacations, holidays, and so forth.Taking longer than 20 weeks to complete 16sessions may compromise the momentum oftherapy. The two to three pretreatment sessionsdescribed earlier, as well as posttreatmentassessment and feedback sessions, are notcounted as part of the 16-session treatmentphase.

Our individual treatment requires that theclient use a workbook (Hope et al., 2000) as heor she works with the therapist during each

session and for homework assignments. Thisworkbook comprises 14 chapters, but thetreatment program does not follow a one-chapter-per-session format. Rather, treatmentis divided into five segments, and the therapisthas flexibility in terms of the pace and, to someextent, the content of these segments. The de-scription of the treatment that follows is orga-nized around these segments. We ask clients toread assigned chapters before coming to thesession.

Clients are asked to bring their workbook toevery session. Therapists typically bring theirown copy of the workbook and/or a copy ofthe therapist guide to session as well. The ther-apist also makes sure that there is something towrite on that the therapist and client can lookat together during the session. In our clinics, anewsprint easel or a dry-erase board mountedto the wall is typically used (for ease, we referto the use of an easel hereafter). However, attimes, we have also just pulled two chairs sideby side, so that both the therapist and clientcan write on and view the same piece of paperon a clipboard. Importantly, every session in-troduces key concepts that are written down(e.g., a therapist may draw a continuum ofanxiety in Session 1; client automatic thoughtsand rational responses are always recorded).Writing things down during session helps cli-ents better to track and process the informationbeing covered and is a required component ofthe treatment.

The therapeutic relationship is actively fos-tered throughout treatment. As Kendall, Chu,Gifford, Hayes, and Nauta (1998) have pointedout, some critics assume that the therapeutic re-lationship is not considered relevant when amanual is utilized. In fact, in our research stud-ies, our therapist adherence manual (Hope, VanDyke, Heimberg, Turk, & Fresco, 2002) in-cludes explicit ratings of the therapeutic rela-tionship within each segment of treatment.Therapists are rated on how well they engage inactive listening, respond to verbal and nonverbalcues, facilitate the client’s investment in treat-ment, and communicate support of and invest-ment in the client. Additionally, contrary to thecriticism that CBTs encourage blanket suppres-sion of affect, therapists are expected to help cli-ents to experience and deepen affect as appropri-ate to the situation at hand. For example, atherapist would be expected to help a client to beopen to the experience of anxiety during expo-sures and to help the client process and grieve


losses that are a consequence of many years ofsocial avoidance. In other situations, the thera-pist would be expected to help the client use un-comfortable feelings as cues for additional cog-nitive coping practice. For example, thetherapist would help a client who experiences adeep sense of shame or guilt after making a smallsocial mistake (e.g., spilling a drink, callingsomeone by the wrong name) to engage in cogni-tive restructuring to change the meaning of theevent eliciting such emotions. Therapists are ac-tively supervised on these aspects of treatmentand not attending to the therapeutic relationshipor appropriately responding to what is going onin session is considered a serious failure to followprotocol.

Recent data suggest that the therapeutic rela-tionship is indeed related to successful out-comes in our protocol (Hayes, Hope, VanDyke, & Heimberg, 2007). Clients’ ratings ofthe therapeutic alliance were positively relatedto their perceptions of session helpfulness. In-terestingly, however, the relationship betweenthe therapeutic alliance and anxiety reductionduring in-session exposures was curvilinear;that is, when ratings of the alliance were higheror lower, clients showed less anxiety reductionthan when the ratings of the alliance were mod-erate. The effect for low ratings of the alliancemay not require explanation, but we have con-jectured that if the client feels too comfortablewith the therapist, then this may possibly inter-fere with anxiety evocation during exposuresand thereby interfere with emotional process-ing. Maintaining the “right” therapeutic alli-ance is important in the CBT for social anxietydisorder (Hayes et al., 2007).

Therapists are also expected to engage in ef-ficient time management. Our therapist adher-ence manual rates therapists’ ability to coverrequired topics flexibly, without rushing orboring the client by belaboring various points(Hope et al., 2002). Two types of problemstend to crop up with regard to time manage-ment (and may negatively impact the therapeu-tic relationship as well). Some therapists fallinto the trap of turning each session into a psy-choeducational lecture, and this tendency maybe aggravated if the therapist is working with aclient who is very quiet in the early sessions.The primary recommendation in this situationis to ask open-ended questions that prompt theclient to relate his or her personal experiencesto the material. The client’s verbalizations in re-sponse to these questions are then reinforced

through active listening skills, which elicitmore client verbalizations. The other trap intowhich some therapists fall is allowing the clientto dominate the session with long, detailed sto-ries about experiences with social anxiety. Insome cases, the therapist may have inadver-tently taught the client to lead the session inthis way, by indiscriminately using active listen-ing skills and nonverbal cues, even when theclient is discussing tangential or repetitive ma-terial. In other cases, clients may be accus-tomed to dominating sessions through priorexperiences with traditional talk therapy, orthey may even intentionally attempt to domi-nate the session to avoid threatening material,particularly in-session exposures. In this situa-tion, gentle verbal redirection is required. Insome cases, a straightforward discussion oftime management issues and negotiation of ses-sion time may be beneficial.

Last, we want to emphasize our belief thatmanualized treatments such as this one do notrepresent an inflexible set of procedures ap-plied in an indiscriminant fashion to all clients.Rather, we expect therapists to develop an indi-vidualized conceptualization of each client’sparticular presenting problem. This conceptu-alization is continually refined over the courseof treatment and incorporates the unique char-acteristics of the client (e.g., cultural group,sex, educational background). Treatment pro-cedures are then tailored to address the uniqueaspects of each case.

Segment 1: Psychoeducation

Segment 1 covers the material in Chapters 1–4of the client workbook (Hope et al., 2000). Es-tablishing a good working relationship is a crit-ical goal for this segment of therapy. Therefore,if the therapist was not also the initial inter-viewer, three to four sessions are typicallyneeded for the psychoeducation segment toallow adequate attention to building rapport. Ifa good working alliance has been establishedduring the course of the initial assessment andif the client is especially bright and motivated,then it may be possible to cover two chaptersper session, finishing this portion of treatmentin as little as 2 weeks. Therapists should not as-sign more than two chapters per week at anypoint in therapy, because clients are unlikely toprocess the material fully and make the most ofthe associated exercises and self-monitoringhomework.


With regard to homework during this seg-ment, one or two chapters are assigned at theend of each session. Each chapter includes a va-riety of forms that the client attempts to com-plete as well. At the beginning of the next ses-sion, the therapist asks the client for reactionsto and questions about the readings, and thera-pist and client review corresponding formscompleted by the client. If the client has notmade an attempt to complete a form, it is com-pleted at the beginning of the session duringhomework review.

Therapists are required to review certaintopics from each chapter in the client work-book in session; other topics are optional de-pending on the needs of the individual client.For Chapter 1, the therapist must review twotopics: (1) normal versus problematic socialanxiety, and (2) the investment that treatmentwill require. The discussion of normal andproblematic social anxiety conveys to the clientthat social anxiety exists along a continuum,with some people rarely experiencing any so-cial anxiety, other people frequently experienc-ing intense social anxiety, and most people fall-ing somewhere between these extremes. Peoplewho fall in the middle of the continuum experi-ence social anxiety in some situations, such as afirst date or a job interview, but the anxiety istypically manageable and decreases as the per-son remains in the situation. Factors such as in-tensity and duration of anxiety before and dur-ing social situations, the number of situationseliciting anxiety, and the degree of life impair-ment differentiate between normal social anxi-ety and problematic social anxiety for which adiagnosis may be assigned. Therapists often en-gage clients in this material by asking whattypes of situations elicit social anxiety for manypeople and what purpose social anxiety mightserve—to illustrate the notion that social anxi-ety has an adaptive value (e.g., caring aboutwhat other people think might cause anxiety,but it also helps people to treat each other morekindly). A therapist typically uses this opportu-nity to launch a discussion about what the cli-ent should expect in terms of treatment out-come. First, the therapist points out that socialanxiety is part of the human condition; thus, itis impossible to remove it completely, even ifthis were a desirable goal. Therefore, total free-dom from social anxiety is not a reasonable ex-pectation for treatment. However, it is reason-able to expect to move down the continuumfrom problematic levels of social anxiety to a

level of social anxiety that is more typical ofmost people. Not all clients are willing to ac-cept that social anxiety exists on a continuum,or that it can have an adaptive value this earlyin treatment. This first session is merely an in-troduction to these concepts, and arguing witha client about these issues is likely to be unpro-ductive. Rather, the therapist empathizes withthe client’s current worldview (e.g., “You feelyou are either calm or extremely anxious, andthat no middle ground is possible for you”) butdoes not agree with it.

The discussion of the required investment intreatment involves instilling hope in the clientthat he or she will be able to make significantchanges and conveying the time, effort, andemotional energy that treatment activities re-quire. Research has shown that positive expec-tancies are associated with more favorable out-comes for CBT for social anxiety (Chambless etal., 1997; Safren, Heimberg, & Juster, 1997b).The therapist shares with the client that instudies of individuals who completed CBT, ap-proximately 75–80% were rated by indepen-dent clinical interviewers as having experiencedmeaningful reductions in their social anxiety(e.g., Heimberg et al., 1998). The therapist alsoemphasizes that attending sessions regularly,doing homework, being willing to experienceanxiety during exposures, and being open tonew ways of looking at the world, other peo-ple, and oneself are factors that substantiallyinfluence whether someone becomes a treat-ment responder, and that these factors arelargely under the client’s control. During thistime, clients are encouraged to speak abouttheir fears regarding treatment and their goalsfor a better life after treatment.

Next, the therapist reviews with the clientthe three components of anxiety, which are pre-sented in Chapter 2 of the client workbook(Hope et al., 2000). Each of the three compo-nents of anxiety is described: (1) the physiolog-ical component comprises bodily reactions,such as a pounding heart; (2) the cognitivecomponent comprises what a person says tohim- or herself, such as “I look foolish”; and(3) the behavioral component comprises whata person does when he or she is anxious, suchas avoiding eye contact. When presenting thebehavioral component, the therapist empha-sizes how escape and avoidance behaviors mayreduce anxiety in the short term but lead to aless satisfying life and to other negative feel-ings, such as guilt and shame, in the long term.


The therapist then helps the client to practiceidentifying each of these components with ahypothetical example. Over the years, we havefound that using a hypothetical example ini-tially is much more conducive to learning thanstarting with an example from the client’s ownexperience. Issues such as fear of negative eval-uation, defensiveness, anxiety, and lack of ob-jectivity can result in a client having difficultyidentifying the three components from a per-sonal experience this early in treatment (e.g., “Iwasn’t thinking anything”). One option is forthe therapist simply to review the hypotheticalexamples from the workbook that illustrate thethree components, but we generally recom-mend using a different hypothetical example insession to give the client more practice. Oftenwe use the example from the group treatmentversion (Heimberg & Becker, 2002), in whichthe client is asked to imagine a person waitingin the lobby for a job interview that is ex-tremely attractive in terms of responsibilities,pay, and hours. The client is told that the per-son is feeling anxious about the interview, thenis asked what that person might be experienc-ing as he or she anxiously waits to be called into the office. Using the easel, the therapistwrites the client’s responses under the appro-priate heading (e.g., “feeling nauseous” is writ-ten under “Physiological Component”). Afterclassifying the client’s spontaneous responses,the therapist then asks questions to elicit sev-eral examples of each of the three components(e.g., “Might the person have any thoughts re-lating to her sweaty palms? What might she sayto herself about that?”). The therapist makesan effort to draw out how each of the threecomponents interacts with the others, resultingin an escalating spiral of anxiety (e.g., sweatypalms might lead to the thought “The inter-viewer will think I am anxious,” which mightlead to a behavior of wiping sweaty hands onone’s suit). The therapist concludes that treat-ment involves learning ways to disrupt this spi-ral of anxiety. Assuming that the client doesreasonably well with this exercise, the therapistthen helps the client to repeat the exercise witha recent, personal experience of social anxiety.At this point in therapy, the therapist makes noeffort to help the client to challenge anxiety-provoking cognitions, only to identify the threecomponents of anxiety and to understand howthey interact with each other. In addition toreadings from the workbook, the client is giventhe homework assignment of identifying all

three components of anxiety experienced dur-ing a social situation during the week.

Next, the therapist describes for the clientfactors that are thought to be related to the de-velopment of problematic social anxiety, whichare also presented in Chapter 3 of the work-book. The therapist tells the client that researchsuggests that genetics play a role in the etiologyof social anxiety disorder. The therapist em-phasizes that what is inherited is probably atendency to be a sensitive and emotionally re-active individual rather than a gene that inevi-tably results in social anxiety (Barlow, 2002).This inherited sensitivity might not be at allproblematic, and it may even be a good thing.For example, the person might be especiallyempathic toward others and have a greater ca-pacity to experience joy, as well as anxiety.However, if a biologically sensitive individual isexposed to important social experiences duringdevelopment and learns that others are threat-ening and hurtful, then the biological vulnera-bility may contribute to the development ofproblematic social anxiety. The therapist mightinquire whether other family members seem tohave problems with social anxiety or othertypes of anxiety, suggesting a genetic predispo-sition. The therapist then presents experienceswithin the family as another potential factorcontributing to social anxiety. For example, asocially anxious mother might teach a childthat the opinions of others are extremely im-portant, and that negative evaluation is to beavoided at all cost. Alternatively, a verballyabusive father might provide experiences thatteach the child that other people are dangerous,and that avoidance is a good strategy for self-protection. Again, the therapist asks the clientwhether this information fits with his or herpersonal experiences. Last, the therapist de-scribes how experiences outside of the familycan similarly teach a person that others are asource of threat. For example, teasing by peersseems to be related to later social anxiety(Roth, Coles, & Heimberg, 2002). Moreover,throughout the discussion, the therapist em-phasizes the idea that because social anxiety islargely learned through experience, it can bechanged through experience.

This discussion of the etiology of social anxi-ety leads into a discussion of dysfunctionalthoughts; that is, therapists might ask clientswhat lessons they learned about themselvesand other people through a negative early-lifesocial experience, such as being teased (e.g., “I


learned that I don’t fit in”). In this way, thera-pists educate clients about the origins of manyof their automatic thoughts, which is impor-tant given that many clients view their ten-dency to have negative cognitions as a sign ofinherent personal flaws rather than a logicalconsequence of their experiences. They thenask clients to think about how these dysfunc-tional ways of thinking color the kinds ofthoughts and reactions they now have whenconfronting social situations (e.g., thoughts ofnot fitting in at parties, expecting to be madefun of by coworkers). At this point in therapy,clients are asked simply to consider the possi-bility that changing some of these automaticthinking biases might change the experience ofanxiety. Therapists then set forth the rationalefor primary interventions of cognitive restruc-turing, exposures, and homework.

Referring to the material covered in Chapter4 of the client workbook, the therapist teachesthe client how to use the Subjective Units ofDiscomfort Scale (SUDS; Wolpe & Lazarus,1966) and to construct a fear and avoidance hi-erarchy. Self-monitoring of daily overall anxi-ety, overall depression, and social anxiety isalso introduced and continued throughouttreatment.

Segment 2:Training in Cognitive Restructuring

Training in cognitive restructuring typically re-quires two to three sessions. The material ispresented in Chapters 5 and 6 of the clientworkbook (Hope et al., 2000). After three ses-sions, a small minority of clients may still bestruggling to master the cognitive concepts dueto factors such as low educational attainmentand/or poor abstract thinking skills. Our expe-rience suggests that rather than forcing the is-sue with such clients, it is better simply to moveon to the next segment of treatment after threesessions. In this circumstance, Heimberg andBecker (2002) recommend that, for the remain-der of therapy, the cognitive portion of treat-ment be deemphasized in favor of relying moreheavily on repeated exposure. Additionally,such clients may be able to generate realisticself-statements (e.g., “I have done this before”)or self-instructions (e.g., “Start by introducingmyself”) informally, with help from the thera-pist, for use in anxiety-provoking situationsrather than relying on more formal cognitiverestructuring.

As in the previous segment, each session be-gins with a homework review and ends withhomework assignment. Homework assign-ments include of readings from the client work-book, associated forms, and self-monitoring.Some clients begin to attempt in vivo exposuresspontaneously at this point in treatment, with-out being specifically assigned to do so. If theseexposures go well, then they may be beneficialto the client. However, therapists should notactively encourage exposures as homework un-til completion of training in cognitive restruc-turing and one in-session exposure. At thispoint in treatment, clients still have few adap-tive strategies for managing anxiety or copingwith social situations that have disappointingoutcomes. They are also likely to escape a situ-ation at the first sign of trouble. Therefore, thepossibility of a less than positive exposureexperience at this phase of treatment couldlead to resistance to attempting future expo-sures.

In this segment of treatment, clients are ex-plicitly introduced to the basic tenet of thecognitive-behavioral model, which proposesthat emotional reactions occur as a result of in-terpretations of situations, not situations them-selves. “Automatic thoughts” are definedas negative, distorted, or irrational thoughtsabout oneself, the world, or the future that leadto or increase the experience of problematicanxiety. These concepts are first demonstratedwith a hypothetical example rather than anemotionally-charged example from the client’sown life. The client workbook describes thevery different thoughts of two young men aftermeeting a woman. Negative thoughts lead oneman to give up any attempts to get to know thewoman better and to experience several moreunpleasant thoughts and emotions. The moreneutral and realistic thoughts of the other manfacilitate his continued efforts at making con-versation. Additional examples are providedfor the therapist and client to examine if the cli-ent appears to need further instruction on theconcept of automatic thoughts. The client thencompletes an exercise, listing any automaticthoughts he or she may have about startingCBT for social anxiety. For homework, the cli-ent monitors his or her automatic thoughts andthe emotions they cause in one or two naturallyoccurring situations during the week.

At the next session, “thinking errors” are de-fined as common categories into which auto-matic thoughts may fall. In other words, people


often find that they are particularly prone toengage in certain types of distorted thinking(e.g., mind reading, fortune-telling, disqualify-ing the positive), and identifying these habitualthinking errors can assist in challenging auto-matic thoughts (e.g., “There I go, mind readingagain. I have no reason to assume that they arethinking the worst”). Clients are introduced toan adaptation of the list of thinking errors out-lined by Judith Beck (1995) in her book Cogni-tive Therapy: Basics and Beyond. Ideally, theclient has reviewed this list in the course of do-ing the readings for homework, and the thera-pist and client can briefly discuss which think-ing errors the client personally found mostrelevant. If the client has not done the home-work, each thinking error is briefly reviewed bythe therapist and client in session. The therapistand client can practice identifying both thethinking errors in workbook examples and theclient’s automatic thoughts recorded as part ofself-monitoring homework, and further exam-ine automatic thoughts the client might haveabout upcoming situations.

The next step in cognitive restructuring is tochallenge the automatic thoughts and developrational responses to them. The process ofchallenging automatic thoughts is begun withthe assistance of disputing questions, originallyadapted from the work of Sank and Shaffer(1984). These generic questions can be broughtto bear on automatic thoughts (e.g., “Do Iknow for certain that ?”; “Whatevidence do I have that ?; Whatevidence do I have that the opposite is true?”;“What is the worst that could happen? Howbad is that? How can I cope with that?”;“What does mean? Doesreally mean that I am a(n) ?”).Before having the client attempt to apply thesequestions to his or her own thoughts, however,the therapist and client review two hypotheti-cal examples from the client workbook. There-after, the client’s automatic thoughts examinedin the earlier discussion of thinking errors areaddressed. Together the client and therapistquestion the automatic thoughts using the dis-puting questions and answer the disputingquestions to arrive at alternative, more realisticways of viewing the situation. The answers tothe disputing questions are then summarizedinto one or two statements that serve as ratio-nal responses. Alternatively, the most compel-ling answer to one of the disputing questionsmay be adopted as a rational response. The

therapist explains that a rational response pro-vides a realistic view of the situation that ismore positive than that provided by the auto-matic thoughts. Rational responses are also rel-atively short and concise. For the next home-work assignment, the client is now ready torecord his or her automatic thoughts from ananxiety-provoking situation during the week,to identify the thinking errors, to dispute theautomatic thoughts, and to develop a rationalresponse.

Toward the end of the final session in thissegment, the therapist informs the client thatexposures will begin the following week. To theextent that beginning exposures causes the cli-ent to feel anxious, the client is encouraged to“avoid avoidance” by coming to the next ses-sion and to identify automatic thoughts relatedto the exposure, challenge them, and arrive at arational response. The therapist reassures theclient that a challenging, but not overwhelm-ing, situation will be used for the first expo-sure. More experienced therapists often discussa few possibilities for the first exposure duringthis session. Therapists new to this treatmentapproach often discuss details of the exposurewith the client the following week, after thetherapist has had supervision.

Segment 3: Exposures

Exposures should begin no later than Session 8and may begin as early as Session 5, dependingon the rate at which the earlier material wascovered. Chapter 7 from the client workbookreviews the rationale for exposures and intro-duces the concept of setting achievable behav-ioral goals. The content of Chapter 7 is onlybriefly touched upon; many of these conceptshave been covered earlier (e.g., why exposuresare important), and the new material is demon-strated during the exposure itself. Similarly,when Chapter 8 (Ongoing Exposures) and oneor more of the specific topic chapters (Chapter9, Observational Fears; Chapter 10, Interac-tion Fears; Chapter 11, Public Speaking Fears)are assigned during this segment, therapistsrarely devote an entire session to discussingeach chapter. Rather, the content is only brieflyaddressed during the homework review and anin-session exposure is conducted; that is, dur-ing this phase of treatment, therapists strive forthe goal of completing an exposure almost ev-ery session. To follow protocol in our researchstudies, a minimum of 4 in-session exposures


must be completed. More commonly, at leastsix in-session exposures are completed, and themore the better. Therefore, although the proto-col does allow the therapist the freedom to ause a couple of sessions during this segment todo cognitive restructuring alone or to review atopic chapter in detail, our belief is that thetime is best devoted to repeated in-session ex-posures integrating cognitive restructuring.

All exposure sessions follow the same basicformat (see Table 3.1). The session begins witha homework review and ends with a home-work assignment. The homework assignmenttypically comprises self-monitoring and in vivoexposure tasks, and associated cognitive re-structuring that logically build upon the in-session exposure. The first in-session exposuretypically includes a role play with the therapistof a situation that the client assigned a SUDS

rating of at least 50. The rule of thumb is thatclients should perceive exposures as challeng-ing but not overwhelming or beyond their abil-ity to use cognitive coping skills. Therapistsmake exposures as realistic as possible by rear-ranging furniture, using props, and instructingrole-play partners (i.e., therapist assistantsbrought into session to assist with exposures)to behave in particular ways. A bit of effort inmaking the situation more realistic may meanthe difference between an exposure that elicitssignificant anxiety and one that is too artificialto be relevant. Commonly used props includefood or drink for individuals with fears of eat-ing, drinking, or serving food in front of others,and client-prepared notes for presentations.Careful attention should be paid to aspects thatmake the situation more or less anxiety-provoking. For example, a client who fears eat-ing in front of others may be more anxiouswhen eating something that is easy to spill (e.g.,soup) than when eating finger foods.

The essence of the treatment is the coordina-tion between the cognitive restructuring workand the exposure (see Table 3.1). The therapistbegins the exposure preprocessing by brieflydescribing an exposure situation. The clientmay suggest modifications or alternatives.However, the therapist needs to balance ap-proaching the exposure task collaborativelywith being highly vigilant for overly detaileddiscussions that largely serve to avoid gettingstarted with the exposure. The therapist thenelicits the client’s automatic thoughts regardingthe chosen situation and records the thoughtson the easel. The therapist picks one or two au-tomatic thoughts to dispute rather than at-tempting to help the client to dispute all auto-matic thoughts. The client takes the lead inidentifying thinking errors, challenging thethoughts with disputing questions, and devel-oping a rational response. The therapist onlyassists as needed. The rational response is writ-ten on a fresh sheet of paper, so that the clientmay refer to it during the exposure, without be-ing distracted by automatic thoughts listed onthe same page.

Next, behavioral goals for the exposure areset. Without guidance, clients with socialanxiety disorder tend to set unrealistic,perfectionistic goals (e.g., “I won’t get anx-ious” or “I will never stumble over my words”)or goals that are based on the reactions of otherpeople and not under their control (e.g., tomake a good impression). The therapist helps


TABLE 3.1. Outline of Exposure Sessions

1. Review homework from previous week.

2. Complete in-session exposure.

a. Exposure preprocessing• Briefly negotiate details of exposure• Elicit automatic thoughts• Client labels thinking errors in automatic

thoughts• Client disputes one or two automatic

thoughts• Client develops a rational response• Client sets nonperfectionistic, behavioral

goals

b. Conduct role play for approximately 10minutes• Request SUDS and the rational response

every minute• Track client’s progress on behavioral goals

c. Exposure postprocessing• Review whether goals were attained• Discuss how well the rational response

worked• Discuss any unexpected or new automatic

thoughts• Client, therapist, and role players react to

client’s performance• Graph and interpret SUDS ratings pattern• Client states what was learned from the

exposure

3. Assignment of homework

a. Related in vivo exposures with associatedcognitive restructuring

b. Continued self-monitoring of depression,general anxiety, and social anxiety

c. Other assignments as appropriate

the client to develop two or three goals relatingto what he or she wants to accomplish in thesituation (e.g., asking at least three questions toget to know the person better, stating an opin-ion, talking about three different points duringa speech). These goals are recorded below therational response, so that the client can refer tothem during the exposure.

During the exposure, the therapists requestsSUDS ratings at 1-minute intervals and when-ever anxiety appears to increase or decrease. Ateach SUDS prompt, the client also reads his orher rational response aloud. Clients quickly ad-just to this disruption, particularly if the roleplayer(s) help to reorient them with a verbalcue (e.g., “You were talking about . . . ”). Theexposure should continue until anxiety has be-gun to decrease or plateau and the goals havebeen met, typically at around 10 minutes. Thetherapist should be the one to end the exposureto help the client overcome any tendencies toescape the situation when it becomes difficult.

The postprocessing phase of the exposure in-cludes a review of goal attainment, a discussionof the usefulness of the rational response, andidentification of any unexpected automaticthoughts that should be addressed in the fu-ture. Rather than asking the client about howthe role play went from his or her perspective,the therapist asks the whether the client’sagreed-upon goals were accomplished, provid-ing no opening for the client to revert to a dis-cussion about failure to attain other goals, suchas not becoming anxious. The therapist alsoshares his or her opinion on goal attainmentand how well the exposure went, responding toconcerns raised by the client and not allowingthe client to disqualify the positive. In later ex-posures, outside role players may also be askedto share their reactions and answer any ques-tions the client has about how the exposurewent. Feedback from the therapist and any roleplayers should emphasize information thatcounters the client’s negative beliefs. For exam-ple, the client might believe that he blushesbright red, is boring, appears inadequate toothers, and cannot carry on a 10-minute con-versation. Perhaps, objectively, the client didblush at the beginning of the exposure, only re-sponded to questions from the role player, andreported very high SUDS for the whole 10 min-utes (90 or higher). However, the client met hisgoals of sharing something about himself andstaying in the exposure until the therapist saidto stop. In this situation, the therapist would

point out how the client met his goals and howmany of the client’s feared consequences (e.g.,being too anxious to carry on a conversation)did not occur. Additionally, only in rare casesdo clients look as anxious as they think theydo, so it is often quite appropriate to acknowl-edge that the client showed some symptoms ofanxiety, but that the anxiety was not as notice-able as the client’s SUDS ratings might suggest,the anxiety did not detract from enjoyment ofthe conversation, and so forth. Therapists whodeny seeing any anxiety in a client who was ob-viously anxious risk damaging their credibility.Especially early in treatment, therapists shouldthink seriously before providing a laundry listof client behaviors that detracted from thequality of the conversation (e.g., the client onlyresponded to questions and did not ask any, theclient could have engaged in more eye contact).Many clients discount the positive aspects ofthe exposure and focus only on the negativefeedback. Therefore, it is better to set this in-formation aside temporarily and to make surethat these behaviors are included as goals forthe next role play—if the client does not spon-taneously suggest one of these problematicbehaviors as a goal. For clients who are self-critical during the postprocessing for an objec-tively poor behavior such as never asking anyquestions, the most common strategy is toshare with the client the conceptualization ofthe problematic behavior as an avoidance re-sponse to significant anxiety (rather than as ev-idence of poor social skills or some other inade-quacy within the client). Working from thisconceptualization, the therapist may then in-quire about automatic thoughts leading to theproblematic behavior. The client’s responsesare often quite illuminating (e.g., “I might in-advertently ask a question about a sensitivesubject that will offend the other person”). Thetherapist would then plan with the client towork on these automatic thoughts and ap-proach the feared social behavior by includingit as a goal in a future exposure.

Next, the pattern of SUDS ratings is quicklygraphed for the client. Different SUDS patternsmay be used to make different points. The mostcommon pattern is for the initial SUDS ratingsto be rather high, then to decline over time. Inthese cases, the therapist can point out that itseems like getting started is the hardest part,and that if the client hangs in there, things seemto get easier. If this pattern holds across futureexposures, it can lead to a new rational re-


sponse (e.g., “It gets easier”; “The hardest partis getting started”). For clients with high SUDSratings throughout the entire exposure whostill meet their goals, the therapist can pointout that a person may simultaneously be quiteanxious and still do what he or she needs to doin a given social situation (e.g., “I can be anx-ious and still share my opinion” might be a fu-ture rational response). The therapist also in-quires about any automatic thoughts that mayhave interfered with habituation, so that thesemight be addressed prior to the next exposure.Another common pattern is decreasing orsteady SUDS ratings that spike to a high levelwhen the client perceives a difficulty arisingduring the exposure. Sometimes the therapist isable to observe what led to the spike in anxiety(e.g., a pause in the conversation). At othertimes, it may not be as obvious, and the clientneeds to explain what happened (e.g., therewas no pause, but the client had an automaticthought that the discussion of the current topichad reached its limits, and feared that he or shewould be unable to continue the conversation).Such discussions may lead to new goals (e.g.,“Allow yourself to pause twice during the con-versation, so that you can learn that you can re-cover from pauses and not to fear them”) andtargets for cognitive restructuring (i.e., the be-lief that pauses reflect social incompetencemight be challenged and lead to a rational re-sponse, such as “Pauses are a normal part ofconversations” or “I am only 50% responsiblefor ending a pause”). Finally, the client is askedwhat he or she has learned from the exposurethat can be applied to life outside of the clinic.

Homework assignments are a very impor-tant part of treatment in this segment, becausethe client is asked to enter situations similar tothose that have been targeted during in-sessionexposures. To assist them in engaging in cogni-tive restructuring prior to the actual conduct ofthe homework exposure, clients are providedcopies of the Be Your Own Cognitive Coachworksheet (BYOCC), which is also discussed insession and exemplified in the client work-book. This form (or a slightly modified versionof it from Heimberg & Becker, 2002) leads cli-ents through each of the steps of cognitive re-structuring described earlier and is often usedfor in-session exposures, as well as those as-signed for homework. After the homework ex-posure is completed, the second portion of theBYOCC assists clients in completing a cogni-tive debriefing and consolidating the learning

that has occurred during and after the home-work experience.

Segment 4:Advanced Cognitive Restructuring

After three or four exposure sessions, the thera-pist and client should begin to notice commonthemes in the client’s automatic thoughts. Atthis point, it is time to move on to advancedcognitive restructuring (Chapter 12 of the cli-ent workbook). This segment assists the clientin moving beyond situation-specific automaticthoughts by applying the downward arrowtechnique (see J. S. Beck, 1995) to thoughtsthat have frequently recurred over the course oftreatment. To accomplish this task efficiently,the client and therapist review all previouswritten homework assignments that involvedthe recording of automatic thoughts, as well asthe thoughts recorded in advance of in-sessionexposures and work through the Peeling YourOnion worksheet from the client workbook.This worksheet provides a systematic approachto questioning automatic thoughts and furtherprobing the responses to these questions, untilthe client’s core belief(s) have been identified.

The beginning of advanced cognitive restruc-turing does not signal the end to exposures.Rather, the therapist and client create in-session and in vivo exposures to challenge corebeliefs. For instance, if a client has a core beliefalong the lines of “I need to be perfect to be ac-cepted by others,” the client might be engagedin exposures that involve making mistakes orsocial missteps (e.g., spilling a drink, turning ina report with a typo) and noticing whether oth-ers continue to be accepting. Core beliefs maybe challenged in other ways as well. For in-stance, in addition to exposures, the client maybe asked to notice instances in which others’mistakes actually make them easy to identifywith or more endearing. One session is typi-cally devoted completely to helping the clientutilize the downward arrow technique andchallenge core beliefs. Additional sessions notonly continue to explore core beliefs but alsoinclude an in-session exposure.

Segment 5: Termination

As treatment approaches Session 16, the issueof termination and/or reevaluation of the treat-ment contract will have been raised informallyby the therapist several times. After Session 15,


Chapter 13 in the client workbook is assigned.Session 16 formally focuses on assessment ofprogress, relapse prevention, and the issue oftermination. The therapist and client go over aworksheet in which the client reports thingsthat he or she has learned during treatment(e.g., how to identify and challenge automaticthoughts, the importance of avoiding avoid-ance). The therapist asks the client to providecurrent ratings for all items from the fear andavoidance hierarchy developed earlier in treat-ment; a discussion follows regarding progressthat has been made and where further work isneeded. When the therapist and client mutuallyagree that termination is in order, risk factorsfor relapse are discussed (e.g., social pressurefrom others, new situations that arise as im-provement continues). Plans to deal with signsof relapse are also made (e.g., the client can re-view chapters in the client workbook or call thetherapist). Mixed emotions (e.g., pride, sadnessover the termination of the therapeutic rela-tionship) associated with termination are alsoprocessed.

In our clinics, the same self-report question-naires, clinician-administered measures, andbehavior tests that were administered at pre-treatment are repeated after Session 16. Theclient is given feedback on these assessments,with data presented in the context of the initialpretreatment assessment results. At the end of16 sessions, many clients still experience prob-lematic social anxiety in a few domains. As westated in the first session, elimination of socialanxiety is not the criterion for termination. In-stead, if the client has stopped avoiding key so-cial situations, has experienced a meaningfulreduction of anxiety in a few areas, and be-lieves that he or she can use the skills gained intherapy to continue to work independently,then the client is ready to stop treatment. Mostindividuals who respond to treatment in thisway are likely to continue to make progress af-ter termination. Follow-up appointments arerecommended to monitor the client’s clinicalstatus (e.g., 1- and 6-months posttreatment).

For some individuals, treatment gains will beevident, but the anxiety and avoidance con-tinue to be too severe and pervasive for the cli-ent to continue on alone. This phenomenon ismost common among clients with quite severesocial anxiety at treatment onset, typically, cli-ents with generalized social anxiety and/orAPD. Clients with significant comorbidities

may also take longer to make sufficient treat-ment gains. In such cases, continued treatmentis recommended. The new treatment contract,as with the original one, should be for 16 ses-sions or less, at which point another assessmentshould occur. The treatment plan typicallyconsists of more cognitive restructuring andexposures, although new domains are oftenintroduced (e.g., moving from working onfriendships to dating relationships).

CASE STUDY

To better illustrate how individual CBT for so-cial anxiety disorder is implemented, a case ex-ample is presented. First, background and pre-treatment assessment data are given, followedby a description of the client’s progress through16 weeks of treatment and her status 1-yearposttreatment.

Josie, a 22-year-old woman, presented withsignificant anxiety concerning both socialinteraction and performance situations. Shewas studying music at a local college and livingwith a roommate. At the time she entered treat-ment, Josie was a full-time student. She was notemployed despite significant financial need.Josie described herself as shy and having trou-ble connecting with others. Josie reported thatshe socialized with her roommate and her boy-friend, but she often turned down opportuni-ties to socialize with their friends because ofanxiety. Josie reported that she had no closefriends of her own and that her anxiety kepther from forming close friendships, especiallywith women her age. Josie’s social anxiety hadalso resulted in occupational and academic im-pairment. For example, although Josie hadgone on numerous job interviews and had evenbeen offered positions, she was currently un-employed because she had difficulty acceptingjob offers because of fears that she would getfired. She reported great anxiety in her classesand did not participate unless directly asked aquestion, even if participation was a significantpart of her grade. She often declined to enrollin classes that interested her because of partici-pation requirements. In addition, Josie was re-quired to do recitals and in-class critiques ofher music compositions several times a semes-ter. Although she never avoided any of these re-citals, she worried about them for weeks in ad-vance and suffered through them with great


anxiety. Josie feared that her social anxietywould cause her even greater difficulty aftergraduation, when she would need to go on jobinterviews and auditions.

Pretreatment Assessment

Josie came to the clinic seeking help for socialanxiety, but she also had difficulties with gen-eral worry and tension, depression, and panicattacks. She was administered the ADIS-IV-Land a series of self-report measures. Based onthis information, Josie was assigned a diagno-sis of generalized social anxiety disorder, withan ADIS-IV-L clinician severity rating of 5,which indicates moderate to severe symp-toms. She received additional diagnoses ofgeneralized anxiety disorder, recurrent majordepressive disorder of moderate severity, post-traumatic stress disorder related to a recentautomobile accident, and panic disorder withagoraphobia.

With regard to measures of social anxiety,Josie’s scores indicated significant fear in bothsocial interaction and performance/observationsituations (see Table 3.2). Heimberg and col-leagues (1992) suggested cutoff scores of 34 forthe SIAS (interaction fears) and 24 for the SPS(performance/observation fears) to differenti-ate between individuals with and without so-cial anxiety disorder. Josie’s scores on bothmeasures exceeded these cutoffs. Similarly, onthe LSAS, Josie reported fear and avoidance ofboth social interaction and performance situa-tions, and her total score was well above theempirically derived cutoff score for the pres-

ence of social anxiety disorder (Mennin et al.,2002).

Ideally, treatment should result in symptomreduction and improved functionality, as wellas promote the client’s overall sense of well-being and life satisfaction. Safren, Heimberg,Brown, and Holle (1997) reported a meanscore of 0.8 on the Quality of Life Inventoryfor clients with social anxiety disorder, which issignificantly lower than the mean score of thenonclinical adult sample reported by Frisch(1994; M = 2.6, SD = 1.3). Josie’s score indi-cated that her degree of life satisfaction wasvery low, but similar to that of others with so-cial anxiety disorder.

The disruption in functioning caused by so-cial anxiety and the associated poor quality oflife frequently results in dysphoria. Josie’s scoreon the BDI-II indicated a level of depressionmore severe than is typically seen among indi-viduals in our treatment program (Elting,Hope, & Heimberg, 1997). Josie did not en-dorse having any thoughts of suicide. Based onclinical interview, we determined that Josie’scurrent low mood was related to recurrent ma-jor depression. This most recent depressive epi-sode had begun approximately 6 months priorto her assessment and was associated with neg-ative social experiences while studying abroad.Josie reported that her depressed mood wasslightly less distressing and impairing than hersocial anxiety. Thus, the depression was judgedto be secondary to her social anxiety and un-likely to have an adverse impact on her treat-ment. Similarly, we determined that her otherdiagnoses, although clinically significant, did


TABLE 3.2. Self-Report and Clinician-Administered Assessments at Pretreatmentand 1-Year Follow-Up

Measure Pretreatment1-year

follow-up

Self-reportBrief Fear of Negative Evaluation Scale 49 23Social Interaction Anxiety Scale 48 22Social Phobia Scale 38 12Beck Depression Inventory–II 25 8Quality of Life Inventory –0.9 1.1

Clinician-ratedADIS-IV-L Clinician Severity Rating 5 3Liebowitz Social Anxiety Scale—Total 48 37

Note. ADIS-IV-L, Anxiety Disorders Interview Schedule for DSM-IV—Lifetime Version.

not rise to a level of concern that might inter-fere with or derail the treatment of her socialanxiety disorder.

Treatment

Psychoeducational Segment

SESSIONS 1–2

The first four sessions were devoted to layingthe groundwork for treatment, establishingrapport, educating Josie about the cognitive-behavioral model of social anxiety disorder,and outlining her goals for treatment.

In the first two meetings, Josie was very soft-spoken and made little eye contact, and attimes expressed feelings of hopelessness abouther ability to get better. She asked many ques-tions about how other clients had fared withthe treatment program, and the therapiststressed that outcome is often dependent on theclient’s ability to put in consistent effortthroughout treatment. The therapist used someof Josie’s own experiences to parallel the chal-lenges and rewards she might face in treatment.The therapist asked Josie how long she hadbeen playing piano and whether she had no-ticed any changes over that time. Josie reportedthat when she started, the process was veryeffortful, challenging, and even disheartening,but that with ongoing practice and education,she had reached a point where it was enjoyableand easy for her. With this, Josie appeared to bevery encouraged about her prognosis, report-ing that she was eager to begin treatment andmotivated to work very hard. The majority ofthe first session was spent establishing rapportand introducing psychoeducational materialabout social anxiety and the components oftreatment. To encourage Josie to speak abouther own experiences with social anxiety, thetherapist employed open-ended questions,asked if the examples sounded familiar to Josie,and requested that she describe some situationsin which her anxiety was better or worse thanothers. For example, although Josie felt she hada significant amount of anxiety when perform-ing in front of an audience, she felt that thisanxiety was typical for most people in that situ-ation. However, her anxiety while interactingwith others at receptions before and after per-formances was severe, and she often avoidedthese situations, arriving at recitals the lastminute and leaving as soon as she finished. By

participating in this way, Josie demonstrated anunderstanding of the important concept thatsocial anxiety exists on a continuum.

SESSION 3

In the third session, the development ofJosie’s social anxiety and how it was main-tained over the years was explored. After ask-ing her mother, Josie learned that she wasvery shy even as a young child, and that hermother often had to push her to participatein social activities with other children. Josiealso reported that her mother always had atendency to keep to herself and minimize in-teractions with others, which Josie may havemodeled from a young age. In addition, Josiereported struggling with feelings of fallingshort from a young age. For example, she re-ported that she never qualified for elite youthsoccer teams, although she was a skilled ath-lete and never received more than third placeor honorable mention in recitals and competi-tions. The therapist discussed with Josie howsuch experiences may have shaped her beliefsthat she is not good enough, and also high-lighted how Josie’s tendency to view the nega-tive aspects of situations (e.g., not gettingfirst place) often led her to disqualify othersuccesses (e.g., being among the top five per-formers in a large competition).

SESSIONS 4–5

In the early part of the fourth session, the ther-apist and Josie completed a discussion fromSession 3 on the role that perfectionistic stan-dards and low self-efficacy played in Josie’s ex-perience of social anxiety. The remainder ofthis session and the next were spent developingand refining Josie’s fear and avoidance hierar-chy (see Table 3.3). Josie reported that interact-ing with people in person was easier than talk-ing over the phone, because face-to-facecontact allowed her to interpret body languageand facial expressions to determine how theinteraction was going for the other person. Inaddition, Josie felt that it was easier to interactwith males than with females, who she believedwere more critical. Josie was also more anxiousaround individuals she respected and admiredthan she was with strangers. She respondedwell to the rationale regarding the importanceof self-monitoring her symptoms of anxiety, de-


pression, and social anxiety in two situations.Josie decided to monitor her peak anxiety dur-ing class critiques of her music compositionand casual conversations with friends.

Cognitive Restructuring Training Segment

SESSION 6

In this session, Josie was introduced to cogni-tive restructuring. Though she demonstratedan understanding of the importance ofthoughts in producing feelings of anxiety, Josiereported having continued difficulty identify-ing her own automatic thoughts. Specifically,she tended to identify questions that passedthrough her mind in anticipation of or during

anxiety-provoking situations. For example,when recalling a recent anxiety-provoking situ-ation in which she had to present her work to aclass, Josie identified the thought, “What willother people think of me?” The therapisthelped Josie to restate this question as a state-ment of what she feared, asking Josie to let her“anxious self” answer the question. She wasable to come up with automatic thoughts, suchas “Other people will think I’m a moron” and“Other people will think I am unprepared forthis presentation.” She identified otherthoughts, such as “I am wasting their time,”“They will get bored with me,” and “My pro-fessor will be disappointed that I didn’t do abetter job.” Many of the automatic thoughtsappeared to be related to a general sense of let-


TABLE 3.3. Fear and Avoidance Hierarchy with Pretreatment, Posttreatment, and 1-YearFollow-Up Ratings

Situation Pretreatment Posttreatment 1-year follow-up

1 Riding public transportation aloneFear 100 20 9Avoidance 100 20 6

2 Staying after recitals to socializeFear 95 7 3Avoidance 95 8 0

3 Going to recitalsFear 90 7 3Avoidance 70 8 0

4 Making phone calls to people I used to be close withFear 75 5 3Avoidance 60 5 1

5 Class critiques and talking about my compositions in classFear 65 6 1Avoidance 50 1 0

6 Talking with professors and authority figures I respectFear 65 4 5Avoidance 50 2 5

7 Making/maintaining conversations with friends and people I respectFear 60 3 4Avoidance 60 1 2

8 Making the first phone callFear 55 2 4Avoidance 5 0 1

9 Hanging out with boyfriend’s friends’ girlfriendsFear 50 6 3Avoidance 85 2 0

10 Talking with old friends I’ve lost touch withFear 40 7 4Avoidance 70 4 2

ting others down. Regarding self-monitoring ofanxiety, depression, and peak anxiety in twosituations, Josie had only practiced self-monitoring for 1 day over the past week, re-porting that she was overwhelmed with theamount of homework and struggling to “cor-rectly” identify her automatic thoughts. Timewas spent discussing the importance of regu-larly monitoring her anxiety and depression,along with her automatic thoughts, so that thetherapist would be able to assess her anxietyand progress throughout treatment more accu-rately. The therapist encouraged Josie that,with continued practice, she would get better atrecognizing automatic thoughts.

SESSION 7

In this session, Josie again struggled to identifyautomatic thoughts, but with prompting, wasincreasingly able to do so. She also had somedifficulty identifying the thinking errors con-tained in her automatic thoughts, often insist-ing that all the thoughts qualified as maladap-tive thoughts—that is, true but unhelpful. Thetherapist asked Josie some disputing questionsto illustrate that the thoughts were not true ortotally accurate, and that other thinking errorscould apply. By the end of session, Josie had abetter handle on thinking errors and was ableto start using some disputing questions success-fully. Josie reported that the cognitive restruc-turing work was much more challenging thanshe expected, but that she was eager to put thework into action during exposures.

SESSION 8

The therapist decided to take an additional ses-sion to continue work on cognitive restructur-ing before moving onto exposures, so that Josiewas maximally prepared. Cognitive restructur-ing practice was focused on a recent situationin which Josie turned down tickets, offered bya professor at school, to a show by her favoriteartist. Josie was much better able to identifyher automatic thoughts related to this situa-tion, which included the following:

1. “Other students will be mad at me if I takethe tickets.”

2. “I will be too scared to go and would wastethe tickets.”

3. “I will not be able to take public transporta-tion to get there.”

4. “If I do take public transportation, I will getlost and miss the show.”

5. “Other students want the tickets more thanI do.”

Josie was able to identify thinking errorssuch as fortune-telling, mind reading, andcatastrophizing, among others. However, therewas evidence that Josie was critical of herselfand her occasional difficulty grasping the con-cepts of cognitive restructuring. At first, Josiestruggled with labeling her automatic thoughtsas wrong, and trying to search for the right ra-tional response. The therapist encouraged Josiethat there was no right or wrong way to chal-lenge automatic thoughts or to develop a ratio-nal response, but Josie continued to evaluateher suggestions as not good enough. Further-more, Josie grew extremely frustrated and pes-simistic when she used disputing questions tochallenge her automatic thoughts. Specifically,she had trouble challenging the thought thatgetting lost on the subway would have to leadto missing the show entirely. She reported thatshe doubted her own ability to read the subwaymap or ask for help, and predicted that shewould get so lost that she would either turnaround and go home or arrive at the show afterit had already started. Josie’s anxiety about en-tering the performance late would then, shepredicted, lead her to go home anyway. Josiereported that she would feel disappointed, em-barrassed, and ashamed if she lost her way onpublic transportation, because it would meanshe had failed to attend the performance and indoing so would anger her classmates, who shepredicted would feel as though she wasted theopportunity. Josie grew increasingly frustratedwith the strength of her anxious thoughts andreported that she felt “stupid” for not beingable to challenge them successfully. This high-lighted some patterns noted throughout treat-ment, including Josie’s tendency to rely on per-fectionist standards and all-or-nothing thinking(e.g., evaluating her thoughts and behavioralresponses as all right or all wrong), and impa-tience with her own progress in treatment. De-spite this, she reported being willing to keeppracticing and move on to in-session expo-sures. The therapist worked with Josie for theremainder of session to challenge her auto-matic thoughts. It was particularly helpful forJosie to identify that she had no evidence thatshe would get lost and to recognize that even ifshe did get lost, it would be important enough


for her to attend the performance that shewould ask for additional help or directions.

Exposure Segment

SESSION 9

During homework review, Josie reported thatshe was able to use her cognitive restructuringskills when she ran into an old roommate at acoffee shop. Josie reported that the exchangehad gone much more smoothly than she antici-pated and had encouraged her to feel moreconfident in her understanding of and abilitiesto use cognitive restructuring skills actively insocial situations. With regard to the in-sessionexposure, the therapist wanted to choose a sit-uation that would seem relevant to her treat-ment goals, elicit a moderate amount of anxi-ety, and be one in which Josie was likely toperform reasonably well. The in-session expo-sure was a follow-up to Josie’s recent interac-tion with her classmate Anne, who mentionedto Josie that some people from class would begetting together one weekend to hang out.However, Anne did not call Josie to make spe-cific plans; neither did Josie follow throughwith making plans to attend the event, and sheended up missing it. Josie thought that Annedid not call because she might have interpretedJosie’s anxiety symptoms (e.g., shyness, quiet-ness, lack of eye contact) as disinterest in so-cializing. Josie’s exposure involved her talkingwith Anne, whose role was played by the thera-pist, about this misunderstanding. The follow-ing excerpts are from the cognitive restructur-ing prior to the exposure.

THERAPIST: First, let’s go to the BYOCCworksheet and work through the situationbefore the exposure happens. When youthink about this situation happening now,what kind of automatic thoughts come upfor you?

JOSIE: I’m not going to talk to her, because Iwon’t know what to say. Also, I shouldn’tsay anything about her not calling me, be-cause it would make everything awkward. Ishould be really friendly and smile the wholetime.

THERAPIST: Are you making any predictionsabout how this interaction will go?

JOSIE: Oh, yeah. I’m going to choke up, and sheis going to judge me. I think that’s it.

THERAPIST: OK, good job identifying those au-tomatic thoughts. Now what kinds of emo-tions come up when you think thosethoughts?

JOSIE: I feel sad. The anxiousness almost goesaway, and I just don’t feel like talking to her.I also feel a little bit of anger and frustrationat myself, because I can’t confront her aboutthis and I don’t want to make her feel un-comfortable.

THERAPIST: So it sounds like that kind of think-ing might lead to you not saying anything atall then, avoiding the interaction?

JOSIE: Yeah, I’d definitely avoid it.

THERAPIST: Well this exposure will allow us topractice confronting those automaticthoughts, but first we need to do some cogni-tive restructuring practice. Why don’t westart by identifying some thinking errors?

JOSIE: Right away I see should statements andfortune-telling. And to think that she’s judg-ing me, that’s mind reading.

THERAPIST: Good job. Now let’s try to disputesome of these thoughts. Let’s start with theautomatic thought: “I’m not going to talk toher, because I won’t have anything to say.”Do you know this for certain?

JOSIE: (Smiling) No, actually I do want to talkwith her, and I could always ask her littlethings. I have no evidence that I’ll have noth-ing to say.

THERAPIST: So you could come up with some-thing to say?

JOSIE: Yeah, even if it’s just small talk, I couldcome up with something. I do want to askher about what happened, but I don’t wantto come across as too aggressive or con-frontational. I feel bad when I confront peo-ple.

THERAPIST: So let’s figure out some ways you’dfeel comfortable asking her to hang out,ways that allow you to get your messageacross without feeling aggressive or blunt.This can help us start to identify somebehavioral goals for the exposure. Let’s takea look at what it is you want to ask her ortell her.

JOSIE: Well, I want to ask her if we can hangout some other time, since she didn’t call melast time. But I have the automatic thoughtthat I shouldn’t say anything, because it willmake things awkward. I guess if I challenge


that, I don’t really know that I will makethings awkward.

THERAPIST: What might be another outcome ifyou say something?

JOSIE: She might apologize and ask me if I wantto come, and I can tell her that I’d love tocome the next time she invites me. She didinvite me in the first place. She just nevercalled, because she thought I didn’t want togo.

THERAPIST: Good. So we identified some auto-matic thoughts, and you said they made youfeel sad and frustrated. Then we challengedthem and came up with some copingthoughts. How do those thoughts make youfeel?

JOSIE: That’s better. I feel a lot better.

THERAPIST: How could we sum these up into arational response? Does anything stand outto you?

JOSIE: “Saying something might have a positiveoutcome” really stands out to me. I’d feelbetter and things might turn out better if I re-mind myself of that. It will help me to chal-lenge the thought that I’ll choke up or runout of things to say.

The therapist selected that automaticthought for cognitive restructuring, because ithad been a recurrent one for Josie in home-work and seemed relevant to the upcomingexposure. Also, the therapist made an effortto take Josie through the cognitive restructur-ing relatively quickly, because the longer theexposure was delayed, the more anxious shewould get and the more difficult it would befor her to focus on the cognitive restructuringbecause of her anxiety. The next step was toset goals for the exposure. Josie decided thather achievable behavioral goals for this expo-sure would include telling Anne that she wasinterested in the last social event, letting Anneknow that she would be interested in hangingout some time soon, and giving Anne herphone number.

The exposure began with both Josie and thetherapist standing up to simulate more closelythe interaction that might occur as they passedeach other in the hallway at school. Josie stoodwhere she would be able to see her rational re-sponse and behavioral goals written on the ea-sel.

THERAPIST: Josie, what is your initial SUDS rat-ing?

JOSIE: About a 45.

THERAPIST: OK, and what is your rational re-sponse?

JOSIE: Saying something might have a positiveoutcome.

Josie and the therapist began to chat aboutclass, and Josie soon brought up the recentlypast social event. The therapist let the exposurego on for approximately 5 minutes. Upon be-ing prompted by the therapist, Josie gave SUDSratings at 1-minute intervals throughout theinteraction and read her rational responsealoud. After the exposure had ended, postpro-cessing began.

THERAPIST: Did you reach your behavioralgoals?

JOSIE: I told her I wanted to hang out and askedher to call me again.

THERAPIST: So you met your goals. Great job.Did any new automatic thoughts come up?

JOSIE: I noticed that I fidgeted a lot. I had thethought “This is scary,” but also that sheseemed pretty cool about the interaction. Ithought I really shouldn’t be nervous.

THERAPIST: Did the thoughts we identified ear-lier come up?

JOSIE: Yeah. I thought I wouldn’t know what tosay, and there was a pause in the conversa-tion but it was fine when I said something. Iworried that I’d make this awkward for her,but I think it was more awkward for me, be-cause I was so nervous.

THERAPIST: How did your rational responsework?

JOSIE: It worked pretty well; it got easier to re-member and I believed it was true when Isaid it. It really seemed to help me thinkmore about rational thoughts instead of anx-ious ones.

THERAPIST: Let’s take a look at your SUDS. Youstarted at a 45 and went up to 55 when con-versation started. When you said somethingto her about wishing she’d call you, SUDSwent up to a 60. By the time she told youwhat happened, the SUDS dropped to 40.After some small talk, it went down to a 20.That’s a pretty significant drop.


JOSIE: It was a lot better after I got over thatbump, when I told her I wanted to hang out.

THERAPIST: What thoughts do you rememberhaving then?

JOSIE: I shouldn’t say anything. I should walkaway. I have to say something.

THERAPIST: So what happened then?

JOSIE: I said my rational response when youasked, right after I asked her about whathappened, and I felt a lot better. It felt likethings were going to be OK. I feel reallygood about how it went.

After processing the exposure, the therapisthelped Josie to plan an in vivo exposure to doon her own that week. She planned to meet herold roommate for coffee, using the BYOCCforms to guide her cognitive restructuring, theexposure, and her own postprocessing of theexperience.

SESSION 10

In the 10th session, Josie reported that she hada very challenging but rewarding week. She re-called an in-class critique of her work, and,though her anxiety was intense, she was able tomanage it and remain in class by remindingherself of her rational response: “I can learnsomething from their feedback.” Josie alsonoted that she had completed two in vivo expo-sures, including having coffee with her oldroommate and a conversation with her class-mate Anne, just as she practiced in last week’ssession. Josie seemed very encouraged by theoutcomes of these situations and her ability touse cognitive restructuring techniques beforeand during anxiety-provoking events.

For exposure this session, Josie planned tointeract with two women playing the roles ofher boyfriend’s friends’ fiancées, to whom Josiefelt she could not relate. The therapist antici-pated that this exposure would be more diffi-cult for Josie, because it involved a fairly un-structured interaction and likely a conversationabout topics with which Josie was relativelyunfamiliar, such as wedding planning. Josie re-ported the following automatic thoughts in an-ticipation of the exposure:

1. “I’m not married yet, so what do I knowabout relationships and weddings?”

2. “I will offend them with my views.”

3. “They are going to think I am weird.”4. “They are going to think I act too young.”5. “I’m not going to give the right answers to

their questions.”

As with automatic thoughts in her first expo-sure, Josie was able to identify thinking errors,including fortune-telling, mind reading, label-ing, and all-or-nothing thinking. By using thedisputing questions, Josie was able to challengethe automatic thoughts she identified. In re-sponse to the first thought, Josie determinedthat although she was not married, she hadbeen in a relationship for a number of years, soshe would be able to relate to the other womenon that level. In addition, she had helped hersister plan a wedding, so she was familiar withsome aspects of wedding planning. Regardingher fear of being offensive, Josie noted that shewas not sure she would offend them with herviews, and that her different perspective mightlead to an interesting conversation. The auto-matic thought, that the other women wouldthink she was weird, was more difficult forJosie to challenge, but she was satisfied whenshe determined that she would be able to livewith herself if they thought she was a little un-usual, and that it was unlikely to affect her re-lationships with them. In reference to her fearsthat she would act too young, Josie used thechallenge that she was only 2 years youngerthan these women, but she had chosen a differ-ent path for herself that involved continued ed-ucation and saving marriage until she felt moresettled in life, a decision she felt was very ma-ture. In response to the automatic thought thatshe would not give the right answer to theirquestions, Josie argued that perhaps there wasno right or wrong answer, and that even if shegave an answer that was inconsistent with theother women’s views, she could recover fromthose mistakes, and differences of opinioncould lead to interesting conversation.

Josie decided that an appropriate rational re-sponse would be one that helped her feel thatshe could relate to these two women despitetheir different circumstances. Her rational re-sponse for this exposure was “I’m not so differ-ent.” Josie’s goals were to talk about her owninterests (e.g., music), ask a question, offer anopinion, and stay in the conversation until theexposure was over. Josie’s SUDS ratings startedout at 45, decreased to about 20 by the fourthminute of the exposure, and went as low as 15


by the final minute of the 8-minute exposure.Josie met all of her behavioral goals, andthough she was quiet compared to the roleplayers early in the conversation, she appearedto grow more at ease as time passed. Josie re-ported that the exposure went fairly well andwas pleased that her sharing a story during theconversation led to a change of subject and ad-ditional conversation among the three women.Although Josie did not share their experiencesand opinions, she was able to find a way to re-late to them and to participate in the conversa-tion. She reported having thoughts of leaving,but she convinced herself to stay by remindingherself that she would benefit from staying inthe situation. Josie’s homework was to call anold music composition classmate, have a con-versation over the phone with him, and askhim to listen to her practice a piece she wasworking on for an upcoming show.

SESSION 11

Prior to this session, Josie had gotten very busywith class work and preparation for an upcom-ing recital of her compositions as part of a finalexam. Consequently, she had missed severalsessions. Josie reported that over the course ofthe past month, she had interviewed for a jobat a local bookstore, was offered the position,and accepted it. She said she experienced a sig-nificant amount of anxiety on the interviewand in the first few days of work, but that shewas feeling much more confident about thingsnow. Although she had made dramatic im-provements in her self-reported anxiety andcontinued to improve during the 1-month hia-tus from regular treatment, the therapist dis-cussed the importance of completing the re-maining 5–6 weeks of the treatment program.Josie committed to following through withtreatment and doing her homework more con-sistently. Though Josie had not formally beendoing her homework and an in-session expo-sure was not conducted, the therapist reviewedher progress so far in relation to her fear andavoidance hierarchy. Socializing with a groupof friends and riding public transportationalone were identified as areas for continuedwork. For homework, Josie was assigned thetask of handing out fliers to friends and familyfor the upcoming recital, which was to involverepeated social interactions and a discussion ofher work. She was also given the assignment ofreading the chapter in the client workbook re-

lated to social interactions and making smalltalk.

SESSION 12

In the Session 12 homework review, Josie saidshe interacted with many people on campusand around town, and that the conversationshad gone much more smoothly than antici-pated. She said it was helpful to remind herselfthat she was an expert on her own composi-tions, so she could talk at least some about theperformance she was advertising. The third in-session exposure, completed in this session,was designed to help Josie work on some of herautomatic thoughts related to her upcoming re-cital that weekend. The therapist had plannedan exposure during which Josie was to interactwith audience members at a reception follow-ing her recital, and staff members from theclinic volunteered to play these roles. Josieagreed that this exposure was very relevant toher current concerns and would make her veryanxious. Josie reported the following auto-matic thoughts in anticipation of the exposure:

1. “I’m going to sound pretentious if I talkconfidently about my performance.”

2. “They are only talking to me because theyfeel bad for me.”

3. “I have to impress them.”4. “They will think this is boring.”5. “I won’t know what to say.”

Josie was able to identify thinking errors,including all-or-nothing thinking, labeling,fortune-telling, and mind reading. During thecognitive restructuring, Josie was encouragedto look at her successful experiences in the twoprevious exposures as evidence that she hadthings to say that were of interest to others.With the help of the therapist, Josie arrived atthis rational response: “I can talk about myperformance, because I created it.”

Josie’s goals were to talk slowly, to provideinformation about herself, to answer questionsposed to her, and to stay in a given interactionuntil the other person ended it. Josie’s initialSUDS rating started out at 80, but it had de-creased to 30 within 3 minutes. At the end ofthe 10-minute exposure, Josie’s SUDS ratingwas below 20. Once again, Josie met herbehavioral goals, and her performance was ob-jectively skilled. Josie said that in addition toher rational response, it was helpful to tell her-


self that she did not have to take everythingothers said to heart. Josie thought the exposurewent very well and even reported having funinteracting with the audience members and be-ing the center of attention. She felt that thispractice was going to be very helpful when sheencountered the real situation at that week-end’s recital. Josie’s homework was to use theBYOCC forms before and after her actual re-cital to prepare herself to cope with anxiousthoughts that might arise before, during, andafter the performance.

SESSION 13

Josie came to the session, reporting that heranxiety at the reception following last week’srecital was much more manageable than it hadbeen in the past, and that she was actually ableto enjoy conversations she had with family,music instructors, and fellow students. She re-ported feeling confident about her perfor-mance, which helped her to feel confident andcomfortable while interacting with others atthe reception. Though Josie did report experi-encing significant anxiety as she began to so-cialize with family and friends after the recital,she said that her anxiety dropped substantiallyas she continued to interact with others abouther performance, and at no point did she havethoughts of leaving the reception. Josie felt thatthe recital and reception were a success, andthat her anxiety did not prevent her from en-joying these situations as it had in the past.

For an in-session exposure, Josie said itwould be helpful to practice receiving feed-back, both positive and negative, about hermusical and academic performances. This situ-ation was relevant because, as part of a finalexam in Josie’s music composition course, shewas to perform an original composition infront of students and her instructor, then re-ceive their feedback. Though her ability to dothis had improved on a smaller scale during hercoursework, this particular class critique elic-ited a considerable amount of anxiety for Josie,and she wanted some direct help with the cog-nitive restructuring and behavioral practice.The therapist decided to conduct two smallerexposures—one in which Josie received neutralor positive feedback from class members andher instructor, and another in which she re-ceived negative feedback. Each exposure wasdesigned to go on for approximately 5–10 min-utes to allow Josie to discuss her composition

and performance adequately and to receive andrespond to feedback. A number of clinic staffmembers volunteered to assist with this expo-sure.

Prior to the exposure, Josie identified the fol-lowing automatic thoughts for both exposures:

1. “I am going to mess this up.”2. “I must appear professional.”3. “I have to perform perfectly to impress

them.”4. “I won’t know what to say.”5. “I won’t understand their questions.”

Josie was able to identify thinking errorssuch as fortune-telling, all-or-nothing thinking,catastrophizing, and “should” statements. Shealso noted how her perfectionistic standardsand feelings of low self-efficacy were quicklyelicited by this situation involving overt evalua-tion of her performance and interaction skills.After challenging her automatic thoughts, Josiedecided that a helpful rational response in eachof the exposures would be “I know my work,because I did it,” which she felt would encour-age her to continue talking confidently abouther own composition and performance, regard-less of the feedback she received. Josie’s behav-ioral goals were to answer questions posed toher and to ask follow-up questions, if sheneeded clarification.

The exposures were set up to take place im-mediately after a performance and began withJosie saying, “Thank you, I’d like to hear yourthoughts about the piece.” In the first expo-sure, she received a mix of neutral and positivefeedback about her composition, her perfor-mance, and her ability to conduct herself dur-ing the critique. Josie’s SUDS score started at45 and dropped down to 15 within 3 minutes.She met all of her behavioral goals. The secondexposure began the same way, but this time shereceived a mix of neutral and negative feed-back; listeners pointed out mistakes she madein her performance, told her that she lookednervous, and indicated that they did not likethe music she wrote. Josie’s SUDS score againstarted at a 45 and went up as high as 60 whenshe received her first negative comment. By thefifth minute of the exposure, her SUDS scorewas 30, and by the end of the exposure, 20.Josie again met her behavioral goals and re-ported that she handled both situations verywell. Josie noted that although her anxietypeaked when she received negative feedback,


she felt she responded to the comments politelyand was even able to learn from them. She feltthe exposure was very helpful in preparing forher upcoming feedback sessions in class. Forhomework, Josie was asked to read the clientworkbook chapter on core beliefs and to do anin vivo exposure that would involve her takingpublic transportation to and from school witha friend. She was also asked to complete aBYOCC form before and after her in-class per-formance and evaluation.

Advanced Cognitive Restructuring Segment

SESSION 14

Josie completed her homework exposure ofriding the bus to and from school with a friendand reported that she was not very anxious.She said that next time she would take the buson her own and work toward going to placesfarther away and less familiar. Josie also re-ported that things went well with her in-classevaluation, that her performance of the musicshe composed went well, but that she thoughtshe messed up once or twice. However, shenoted that no students gave her feedback indi-cating that they noticed the mistake. Josie re-ported feeling incredibly anxious after she fin-ished playing, and that right before thefeedback session she even had thoughts of ex-cusing herself for a few minutes. Remindingherself of her rational response, “I can learnsomething from their feedback,” encouragedJosie to stay in class and complete the evalua-tion. Though she received some challengingquestions about how she composed the musicand who had influenced her, Josie thought shehad done a decent job, although she had stum-bled over her words a few times. She noted thata few students commented about not particu-larly liking the style of music she chose, butJosie said she was able not to take such com-ments personally. Overall Josie felt that she hadlearned a lot from the students’ and instructor’sfeedback and that, although anxious, felt it wasan acceptable level of anxiety given the perfor-mance and evaluation aspects of the situation.

Josie did not read the chapter on core beliefs,so the therapist spent time in session introduc-ing this concept. Josie worked hard in this ses-sion to identify her core beliefs, using thePeeling Your Onion worksheet. She realizedthat her most common automatic thoughts

across situations were related to labeling her-self, feeling as though she would not live up toothers’ expectations, and fears that otherswould judge her negatively because of her so-cial or performance skills. Josie felt that thetheme of “not measuring up” was familiar andrelated potentially to her beliefs about nevermeasuring up to her parents’ standards and notdoing anything right, which led to more gen-eralized perfectionism. Josie identified thethought, “I’m not measuring up,” as related toa fear that she would not be successful in life,and that to be successful, she felt she must beperfect. Identifying this chain of thinking lefther feeling sad, frustrated, and disappointed.From here, Josie worked through the work-sheet and identified the core belief, “If I’m notperfect, I’m worthless.” Time was spent brain-storming exposures Josie could do on her ownto test and challenge this core belief. For home-work, the therapist asked Josie to mess up in-tentionally while playing the piano for somefamily members, so as to challenge her core be-lief that anything short of perfectionism repre-sents a failure. Josie also identified another sit-uation in which she could intentionally messup, while using the intercom system at her job.The therapist also challenged Josie to identifyways that her imperfections might actuallycontribute to her value as a person. Josie washesitant to do this but agreed when she re-minded herself of the importance of facingfeared situations.

SESSION 15

Josie arrived at the 15th session and reportedthat she had been unable to make a mistake in-tentionally in her musical performance in frontof family members, but she was able to make amistake while using the intercom system atwork. The therapist continued to use cognitiverestructuring to explore Josie’s core belief, “IfI’m not perfect, I’m worthless.”

THERAPIST: So, last week we started to talkabout some core issues, and I felt like youleft here feeling a little overwhelmed. Howare you feeling about things now?

JOSIE: It was a difficult session, because I wasfeeling really good about overcoming hardstuff in the beginning, like learning how topick out automatic thoughts, see how they


were wrong, then finding a really good, chal-lenging response to them and actually believ-ing it. I felt good because I can do that now,almost automatically, or at least without toomuch effort. I felt like I was done, and thenwe ended up having to face some really scarystuff again. I feel like I took a giant step up,to a harder level of work.

THERAPIST: The things we talked about lastweek were harder. That’s a good sign thatwe’ve really worked our way to your corebeliefs, because it feels so different. But re-member all the work and practice that youput into identifying and challenging everyother automatic thought or situation thatwe’ve worked on. It didn’t all happen rightaway. In terms of core beliefs, we’re going todo the same thing. All we’ve done is identifyit. Now we will spend time challenging itand testing out the thought, “If I’m not per-fect, I’m worthless.” We’ll start here in ses-sion today, but this is work you’ll continueon your own. What might be some thinkingerrors? What’s limiting in thinking that any-thing less than 100% is a failure?

JOSIE: That’s catastrophizing. Probably also a“should” statement, because I feel like Imust be perfect. And also maybe fortune-telling, because I feel like I’m assumingthings will go wrong eventually. That’s it.

THERAPIST: What about all-or-nothing think-ing?

JOSIE: Oh yeah, definitely. In my mind, it’s likethere’s 0% success or 100% success.

THERAPIST: So what are some ways we canchallenge this? It won’t be easy, but let’s try.

JOSIE: Umm, I guess simple challenges, like“How do I know I’m going to be worthless ifI’m not perfect?” I don’t know. I guess notbeing perfect doesn’t have to equal beingworthless.

THERAPIST: Tell me more about that.

JOSIE: Some things are maybe valuable evenwhen they’re not perfect. Like people. Orart.

THERAPIST: So sometimes people’s flaws, orquirks, make them more interesting?

JOSIE: Yeah, that’s what I like best about myboyfriend. Those things other people arebothered by I find endearing. I like thingsthat are imperfect, that have their own per-

sonality to them that’s not quite right. I don’treally have a problem when the imperfectionis with someone else. But with me, it’s differ-ent.

THERAPIST: Are you a harsher critic of yourselfthan others?

JOSIE: Definitely. Too bad I can’t see myself theway I see others.

THERAPIST: Maybe you can, maybe we can trythat. It’s hard, because you’ve been lookingat yourself critically for a long time. It’s go-ing to be a change, and it will take testingand practice. But what might be the rewardfor trying something besides being a harsherjudge of yourself than others?

JOSIE: I guess the work will pay off, like it didwith everything else. With practice, I evenhad fun at my show. I didn’t believe that Iwas going to be a failure. I believed I coulddo it. It was a huge success.

THERAPIST: Were there any minor flaws withthe evening? Things that went wrong, butthat you coped with?

JOSIE: Oh, yeah, tons. Like I forgot my friend’sname when I was introducing her to my par-ents. But it didn’t ruin the evening. Overallthings were good, and I felt good about it. Icould skip over those flaws and just enjoy it.

THERAPIST: So when thinking about that situa-tion, how true does it feel that if you are notperfect, you are worthless?

JOSIE: When I came in, I felt overwhelmed byeverything; now that we can see all the workI’ve done, I feel better. I think I can challengethis. I don’t think it feels totally true any-more.

THERAPIST: OK, so what’s some evidence thatthis core belief may not be true?

JOSIE: Well, it’s OK for everyone else to not beperfect. Maybe a little less OK for me, but Iguess it’s still somewhat OK. And some peo-ple are more valuable to me because of theirimperfections. I guess being perfect is boring.Being perfect is impossible.

THERAPIST: So that seems like a high standardto hold yourself to.

JOSIE: (Laughs.) Yeah. I don’t know anyonewho’s perfect. Everyone makes mistakes.Even me. Especially me. Everyone in the his-tory of the world has been imperfect.


THERAPIST: Right. Now let’s get back to you,not just others. Do we have evidence thatyou may be valuable even if you’re imper-fect? Are there upsides of imperfection?

JOSIE: I’ll be more unique, more of an individ-ual. It makes for interesting challenges; itcould make for a really interesting life. Icould learn a lot about different ways to dothings, try out all kinds of new things. Makelots of mistakes, and just learn to do thingsbetter the next time. Not perfect, but justbetter. And make room for failure.

THERAPIST: So how much do you believe that ifyou’re not perfect, you’re worthless?

JOSIE: Maybe like 20%. Right now, at least,only 20%.

THERAPIST: So how can we sum up some ofthese challenges that have gotten you to be-lieving this only about 20% ? What might bea rational response to this core belief, “If I’mnot perfect, I’m worthless.”

JOSIE: I don’t know; this is hard. If I’m not per-fect, I’m worthy? I don’t believe it, but itmakes me laugh. I guess maybe it’s OK if I’mnot perfect.

THERAPIST: Does that feel convincing?

JOSIE: Not really. I’m drawing a blank withcoming up with the right rational response.

THERAPIST: What if we recognized that this ra-tional response won’t be perfect? Maybe weneed to find one that’s good enough for now.So let’s keep at this.

JOSIE: Oh I see, there’s the perfectionistic think-ing again. I guess really, it is not humanlypossible to be perfect. And I don’t think Iwould even want to be that person who fi-nally was perfect.

THERAPIST: So it seems like a lot of work to bechasing perfection when it kind of doesn’texist. It sounds like no matter how hard youtry, there would always be the attempt to dobetter than perfect.

JOSIE: Yeah, like no matter what, I’d never beable to be perfect. It’s not humanly possible.I guess if I had to make this a rational re-sponse, I’d tell myself something like “Thereis no such thing as perfect.” I like that. Ithink that will work for now. I’m going tothink about this more. This is a challenge,but I want to spend time on this.

THERAPIST: Ok, so maybe we can try out this

rational response this week by having you dosome exposures that will allow you to chal-lenge the core belief, “If I’m not perfect, I’mworthless.”

JOSIE: Yeah. I need to do that every day, chal-lenge this, become more accepting of my im-perfections. It might be fun to mess up onpurpose. But a lot of the time I mess up justnaturally, but I guess even then I’m notworthless. Mistakes are going to happen, be-cause there is no such thing as perfect.

For homework this session, Josie was as-signed the task of intentionally making mis-takes in conversation (e.g., forgetting people’snames, asking them to repeat something theyhave said), at work (e.g., pressing the wrongbutton when using the intercom, asking forhelp with a task), and in performances (e.g.,playing the wrong notes while playing the pi-ano for her family and friends).

Termination Segment

SESSION 16

In the final session, Josie and the therapistdiscussed her progress and the challenges thatlay ahead. Josie rerated her fear and avoid-ance hierarchy, and the decrease in her rat-ings was considered significant and meaning-ful (see Table 3.3). She reported that, over thecourse of treatment, she learned new skillssuch as recognizing and challenging auto-matic thoughts and seeing all aspects ofherself—including imperfection—as accept-able. The most important change she notedwas being able to get and to keep a job at abookstore, where she must interact with cus-tomers and coworkers all day. Josie reportedthat as a result of her treatment, for the firsttime she was able to see herself as equalrather than inferior to others, and she wasoptimistic about her future goals, both per-sonal and professional.

Session-by-Session Assessment

Figure 3.2 displays Josie’s session-by-sessionscores on the SASCI and the BFNES. Recallthat a score of 16 on the SASCI corresponds tono change, with lower scores indicative of in-creasingly greater improvement relative tobaseline levels. Josie started to show modestimprovement after Session 3. Her score spiked


at Session 8, then began a steady decline, reach-ing the minimum score of 4 by Session 12 andremaining there until the end of treatment. Forthe BFNE, Weeks and colleagues. (2005) re-ported a mean score of 46.91 (SD = 9.27) in alarge sample with social anxiety disorder and amean of 26.81 (SD = 4.78) among normal con-trols. Early in treatment, Josie’s session scoresexceeded the mean of the clinical sample by asmuch as 10 points. However, her scores de-clined consistently thereafter, and by the end oftreatment were just below the mean of the con-trol sample.

One-Year Follow-Up Assessment

After 1 year with no treatment, Josie showedconsiderable improvement on all self-reportand clinician-administered measures comparedto her pretreatment assessment (see Table 3.2).In fact, she no longer met diagnostic criteria forsocial anxiety disorder. She was considering ap-plications to graduate programs to continueher study of music and had plans to auditionfor the local orchestra. Josie reported that shehad been making friends at her job and felt thatshe now had a core of social supports on whomshe could rely.


FIGURE 3.2. Weekly scores on the (a) Social Anxiety Session Change Index (SASCI) and (b) Brief Fear ofNegative Evaluation Scale (BFNE).

CLINICAL PREDICTORSOF SUCCESS AND FAILURE

Josie’s treatment turned out well, as it does formany clients. However, treatment response is avariable phenomenon, and researchers have in-vestigated a number of predictors of CBT out-come (see a more thorough review of this topicby L. Magee, Erwin, & Heimberg, in press).

Expectancy for improvement, subtype of so-cial anxiety disorder, and comorbidity withother anxiety or mood disorders, as well asAPD, have been discussed in other sections ofthis chapter. Here we briefly review three othervariables—adherence to assigned CBT home-work, anger, and cognitive change.

Adherence to prescribed homework assign-ments has been associated with positive treat-ment outcome, and evidence suggests thatadherence to particular components of CBThomework assignments may be differentiallypredictive of outcome. For instance, adherenceto between-session cognitive restructuring andexposure assignments predicts posttreatmentoutcome better than adherence to assignmentsrelated more to psychoeducation early in treat-ment (Leung & Heimberg, 1996). Other stud-ies have not replicated these effects immedi-ately after treatment (Edelmann & Chambless,1995; Woody & Adessky, 2002). However, 6-month follow-up assessments revealed thathomework-compliant individuals reportedfewer avoidant behaviors, less fear of negativeevaluation, and less anxiety when givinga speech compared to less compliant clients(Edelman & Chambless, 1995).

Anger is a significant predictor of CBT out-come as well. In a study by Erwin, Heimberg,Schneier, and Liebowitz (2003), individualswith high levels of trait anger were more likelyto terminate treatment prematurely. Also, lev-els of state and trait anger and anger suppres-sion before treatment were significantly cor-related with posttreatment severity of socialanxiety.

Finally, in a study comparing group CBT, ex-posure group therapy, and waiting-list controlconditions, changes in estimated social cost,or negative cognitive appraisal, mediated pre-treatment to posttreatment changes in both ac-tive treatment groups (Hofmann, 2004). Fur-thermore, only the group receiving cognitivetechniques, in addition to exposure, continued

to show improvement from posttreatment tothe 6-month follow-up assessment. Continuedbenefit was associated with an overall reduc-tion in estimated social cost from pre- toposttreatment assessments, suggesting that thecognitive-behavioral intervention is associatedwith greater treatment gains that are mediatedthrough changes in estimated social cost(Hofmann, 2004).

SUMMARY AND CONCLUSIONS

Our primary purpose in this chapter was toprovide a step-by-step analysis of the conductof individual CBT for social anxiety disorderfrom initial assessment to long-term follow-up.Discussion of procedural difficulties that mayarise in the conduct of CBT for social anxietydisorder were not be discussed in detail herebecause of space limitations but are discussedthoroughly by Heimberg and Becker (2002).Individuals like Josie, who make dramatic im-provements in their lives, provide the impetusfor continued research on the treatment of so-cial anxiety disorder—a potentially debilitat-ing disorder with an increasingly encouragingprognosis.

REFERENCES

American Psychiatric Association. (1980). Diagnosticand statistical manual of mental disorders (3rd ed.).Washington, DC: Author.


Antony, M. M., Coons, M. J., McCabe, R. E.,Ashbaugh, A., & Swinson, R. P. (2006). Psychomet-ric properties of the Social Phobia Inventory: Furtherevaluation. Behaviour Research and Therapy, 44,1177–1185.

Barlow, D. H. (2002). Anxiety and its disorders: The na-ture and treatment of anxiety and panic (2nd ed.).New York: Guilford Press.

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). BeckDepression Inventory manual (2nd ed.). San Anto-nio, TX: Psychological Corporation.

Beck, J. S. (1995). Cognitive therapy: Basics and be-yond. New York: Guilford Press.

Beidel, D. C., Turner, S. M., Stanley, M. A., & Dancu, C.V. (1989). The Social Phobia and Anxiety Inventory:Concurrent and external validity. Behavior Therapy,20, 417–427.

Bisserbe, J. C., Weiller, E., Boyer, P., Lepine, J.-P., &


Lecrubier, Y. (1996): Social phobia in primary care:Level of recognition and drug use. International Clin-ical Psychopharmacology, 11, 25–28.

Brown, E. J., Heimberg, R. G., & Juster, H. R. (1995).Social phobia subtype and avoidant personality dis-order: Effect on severity of social phobia, impair-ment, and outcome of cognitive-behavioral treat-ment. Behavior Therapy, 26, 467–486.

Brown, E. J., Turovsky, J., Heimberg, R. G., Juster, H.R., Brown, T. A., & Barlow, D. H. (1997). Validationof the Social Interaction Anxiety Scale and the SocialPhobia Scale across the anxiety disorders. Psycholog-ical Assessment, 9, 21–27.

Brown, T. A., Di Nardo, P. A., & Barlow, D. H. (1994).Anxiety Disorders Interview Schedule for DSM-IV.Albany, NY: Graywind.

Brown, T. A., Di Nardo, P. A., Lehman, S. L., &Cambell, L. A. (2001). Reliability of DSM-IV anxietyand mood disorders: Implications for the classifica-tion of emotional disorders. Journal of AbnormalPsychology, 110, 49–58.

Bruch, M. A., Fallon, M., & Heimberg, R. G. (2003).Social phobia and difficulties in occupational adjust-ment. Journal of Counseling Psychology, 50, 109–117.

Bruch, M. A., Gorsky, J. M., Collins, T. M., & Berger, P.(1989). Shyness and sociability reexamined: A multi-component analysis. Journal of Personality and So-cial Psychology, 57, 904–915.

Bruch, M. A, Heimberg, R. G., & Hope, D. A. (1991).States of mind model and cognitive change in treatedsocial phobics. Cognitive Therapy and Research, 15,429–441.

Butler, G., Cullington, A., Munby, M., Amies, P., &Gelder, M. (1984). Exposure and anxiety manage-ment in the treatment of social phobia. Journal ofConsulting and Clinical Psychology, 52, 642–650.

Chambless, D. L., & Ollendick, T. H. (2001). Em-pirically supported psychological interventions: Con-troversies and evidence. Annual Review of Psychol-ogy, 52, 685–716.

Chambless, D. L., Tran, G. Q., & Glass, C. R. (1997).Predictors of response to cognitive-behavioral grouptherapy for social phobia. Journal of Anxiety Disor-ders, 11, 221–240.

Chartier, M. J., Hazen, A. L., & Stein, M. B. (1998).Lifetime patterns of social phobia: A retrospectivestudy of the course of social phobia in a nonclinicalpopulation. Depression and Anxiety, 7, 113–121.

Clark, D. M. (1997). Cognitive therapy for social pho-bia: Some notes for therapists. Unpublished manu-script, University of Oxford, Department of Psychia-try.

Clark, D. M., Ehlers, A., Hackmann, A., McManus, F.,Fennell, M., Grey, N., et al. (2006). Cognitive ther-apy versus exposure and applied relaxation in socialphobia: A randomized controlled trial. Journal ofConsulting and Clinical Psychology, 74, 568–578.

Clark, D. M., Ehlers, A., McManus, F., Hackmann, A.,

Fennell, M., Campbell, H., et al. (2003). Cognitivetherapy versus fluoxetine in generalized social pho-bia: A randomized placebo-controlled trial. Journalof Consulting and Clinical Psychology, 71, 1058–1067.

Clark, D. M., & Wells, A. (1995). The cognitive modelof social phobia. In R. G. Heimberg, M. R.Leibowitz, D. A. Hope, & F. R. Schneier (Eds.), So-cial phobia: Diagnosis, assessment, and treatment(pp. 69–93). New York: Guilford Press.

Collins, K. A., Westra, H. A., Dozois, D. J. A., & Stew-art, S. H. (2005). The validity of the brief version ofthe Fear of Negative Evaluation Scale. Journal ofAnxiety Disorders, 19, 345–359.

Connor, K. M., Davidson, J. R. T., Churchill, E.,Sherwood, A., Foa, E. B., & Weisler, R. H. (2000).Psychometric properties of the Social Phobia Inven-tory (SPIN): A new self-rating scale. British Journalof Psychiatry, 176, 379–386.

Connor, K. M., Kobak, K. A., Churchill, E., Katzelnick,D., & Davidson, J. R. T. (2001). Mini-SPIN: A briefscreening assessment for generalized social anxietydisorder. Depression and Anxiety, 14, 137–140.

Cox, B. J., Ross, L., Swinson, R. P., & Direnfeld, D. M.(1998). A comparison of social phobia outcome mea-sures in cognitive-behavioral group therapy. Behav-ior Modification, 22, 285–297.

Davidson, J. R. T., Miner, C. M., De Veaugh-Geiss, J.,Tupler, L. A., Colket, J. T., & Potts, N. L. (1997). TheBrief Social Phobia Scale: A psychometric evaluation.Psychological Medicine, 27, 161–166.

Davidson, J. R. T., Potts, N. L., Richichi, E. A.,Krishnan, R. R., Ford, S. M., Smith, R. D., et al.(1991). The Brief Social Phobia Scale. Journal ofClinical Psychiatry, 52, 48–51.

Di Nardo, P. A., Brown, T. A., & Barlow, D. H. (1994).Anxiety Disorders Interview Schedule for DSM-IV:Lifetime version (ADIS-IV-L). Albany, NY: Gray-wind.

Edelmann, R. E., & Chambless, D. L. (1995). Adher-ence during session and homework in cognitive-behavioral group treatment of social phobia. Behav-iour Research and Therapy, 33, 537–577.

Elting, D. T., Hope, D. A., & Heimberg, R. G. (1997).Inter-relationships among measures commonly usedin research on social phobia. Depression and Anxi-ety, 4, 246–248.

Erwin, B. A., Heimberg, R. G., Juster, H. R., & Mindlin,M. (2002). Comorbid anxiety and mood disordersamong persons with social phobia. Behaviour Re-search and Therapy, 40, 19–35.

Erwin, B. A., Heimberg, R. G., Schneier, F. R., &Liebowitz, M. R. (2003). Anger experience and ex-pression in social anxiety disorder: Pretreatment pro-file and predictors of attrition and response tocognitive-behavioral treatment. Behavior Therapy,34, 331–350.

Feske, U., Perry, K. J., Chambless, D. L., Renneberg, B.,& Goldstein, A. J. (1996). Avoidant personality dis-


order as a predictor for severity and treatment out-come among generalized social phobics. Journal ofPersonality Disorders, 10, 174–184.

Fresco, D. M., Coles, M. E., Heimberg, R. G.,Liebowitz, M. R., Hami, S., Stein, M. B., et al.(2001). The Liebowitz Social Anxiety Scale: A com-parison of the psychometric properties of self-reportand clinician-administered formats. PsychologicalMedicine, 31, 1025–1035.

Frisch, M. B. (1994). Manual and treatment guide forthe Quality of Life Inventory. Minneapolis, MN: Na-tional Computer Systems.

Gilbert, P. (2001). Evolution and social anxiety: Therole of attraction, social competition, and social hier-archies. Psychiatric Clinics of North America, 24,723–751.

Gould, R. A., Buckminster, S., Pollack, M. H., Otto, M.W., & Yap, L. (1997). Cognitive-behavioral andpharmacological treatment for social phobia: Ameta-analysis. Clinical Psychology: Science and Prac-tice, 4, 291–306.

Hackmann, A., Clark, D. M., & McManus, F. (2000).Recurrent images and early memories in social pho-bia. Behaviour Research and Therapy, 38, 601–610.

Hackmann, A., Surawy, C., & Clark, D. M. (1998).Seeing yourself through others’ eyes: A study of spon-taneously occurring images in social phobia. Behav-ioural and Cognitive Psychotherapy, 26, 3–12.

Hambrick, J. P., Turk, C. L., Heimberg, R. G., Schneier,F. R., & Liebowitz, M. R. (2003). The experience ofdisability and quality of life in social anxiety disorder.Depression and Anxiety, 18, 46–50.

Hayes, S. A., Hope, D. A., Van Dyke, M., & Heimberg,R. G. (2007). Working alliance for clients with socialanxiety disorder: Relationship with session helpful-ness and within-session habituation. Cognitive Be-haviour Therapy, 36, 34–42.

Hayes, S. A., Miller, N. A., Hope, D. A., Heimberg, R.G., & Juster, H. R. (in press). Assessing client prog-ress session-by-session: The Social Anxiety SessionChange Index. Cognitive and Behavioral Practice.

Heimberg, R. G. (1994). Cognitive assessment strategiesand the measurement of outcome of treatment for so-cial phobia. Behaviour Research and Therapy, 32,269–280.

Heimberg, R. G., & Becker, R. E. (2002). Cognitive-behavioral group therapy for social phobia: Basicmechanisms and clinical strategies. New York:Guilford Press.

Heimberg, R. G., Dodge, C. S., Hope, D. A., Kennedy,C. R., Zollo, L. J. & Becker, R. E. (1990). Cognitive-behavioral group treatment for social phobia: Com-parison with a credible placebo control. CognitiveTherapy and Research, 14, 1–23.

Heimberg, R. G., & Holaway, R. M. (in press). Exami-nation of the known-groups validity of the LiebowitzSocial Anxiety Scale. Depression and Anxiety.

Heimberg, R. G., Holt, C. S., Schneier, F. R., Spitzer, R.L., & Liebowitz, M. R. (1993). The issues of sub-

types in the diagnosis of social phobia. Journal ofAnxiety Disorders, 7, 249–269.

Heimberg, R. G., Hope, D. A., Dodge, C. S., & Becker,R. E. (1990). DSM-III-R subtypes of social phobia:Comparison of generalized social phobics and publicspeaking phobics. Journal of Nervous and MentalDisease, 178, 172–179.

Heimberg, R. G., Horner, K. J., Juster, H. R., Safren, S.A., Brown, E. J., Schneier, F. R., et al. (1999). Psycho-metric properties of the Liebowitz Social AnxietyScale. Psychological Medicine, 29, 199–212.

Heimberg, R. G., Liebowitz, M. R., Hope, D. A.,Schneier, F. R., Holt, C. S., Welkowitz, L. A., et al.(1998). Cognitive behavioral group therapy vs.phenelzine therapy for social phobia: 12-week out-come. Archives of General Psychiatry, 55, 1133–1141.

Heimberg, R. G., Mueller, G. P., Holt, C. S., Hope, D.A., & Liebowitz, M. R. (1992). Assessment of anx-iety in social interaction and being observedby others: The Social Interaction Anxiety Scale andthe Social Phobia Scale. Behavior Therapy, 23, 53–73.

Heimberg, R. G., Salzman, D. G., Holt, C. S., &Blendell, K. A. (1993). Cognitive-behavioral grouptreatment for social phobia: Effectiveness at five-yearfollowup. Cognitive Therapy and Research, 17, 325–339.

Heimberg, R. G., with Turk, C. L. (2002). Assessmentof social phobia. In R. G. Heimberg & R. E. Becker(Eds.), Cognitive-behavioral group therapy for socialphobia: Basic mechanisms and clinical strategies (pp.107–126). New York: Guilford Press.

Herbert, J. D., Bellack, A. S., & Hope, D. A. (1991).Concurrent validity of the Social Phobia and AnxietyInventory. Journal of Psychopathology and Behav-ioral Assessment, 13, 357–368.

Herbert, J. D., Hope, D. A., & Bellack, A. S. (1992). Va-lidity of the distinction between generalized socialphobia and avoidant personality disorder. Journal ofAbnormal Psychology, 101, 332–339.

Hofmann, S. G. (2004). Cognitive mediation of treat-ment change in social phobia. Journal of Consultingand Clinical Psychology, 72, 392–399.

Hofmann, S. G., Newman, M. G., Becker, E., Taylor, C.B., & Roth, W. T. (1995). Social phobia with andwithout avoidant personality disorder: Preliminarybehavior therapy outcome findings. Journal of Anxi-ety Disorders, 9, 427–438.

Hope, D. A., Heimberg, R. G., & Bruch, M. A. (1995).Dismantling cognitive-behavioral group therapy forsocial phobia. Behaviour Research and Therapy, 33,637–650.

Hope, D. A., Heimberg, R. G., Juster, H., & Turk, C. L.(2000). Managing social anxiety: A cognitive-behavioral therapy approach (client workbook).New York: Oxford University Press.

Hope, D. A., Heimberg, R. G., & Turk, C. L. (2006).Managing social anxiety: A cognitive-behavioral


therapy approach (therapist guide). New York: Ox-ford University Press.

Hope, D. A., Herbert, J. D., & White, C. (1995). Diag-nostic subtype, avoidant personality disorder, and ef-ficacy of cognitive behavioral group therapy for so-cial phobia. Cognitive Therapy and Research, 19,399–417.

Hope, D. A., Sigler, K. D., Penn, D. L., & Meier, V.(1998). Social anxiety, recall of interpersonal infor-mation, and social impact on others. Journal of Cog-nitive Psychotherapy: An International Quarterly,12, 303–322.

Hope, D. A., Van Dyke, M., Heimberg, R. G., Turk, C.L., & Fresco, D. M. (2002). Cognitive-behavioraltherapy for social anxiety disorder: Therapist adher-ence scale. Unpublished manuscript, available fromRichard G. Heimberg, Adult Anxiety Clinic, Depart-ment of Psychology, Temple University, Philadelphia,PA 19122-6085.

Katzelnick, D. J., Kobak, K. A., DeLeire, T., Henk, H.J., Greist, J. H., Davidson, J. R. T., et al. (2001). Im-pact of generalized social anxiety disorder in man-aged care. American Journal of Psychiatry, 158,1999–2007.

Kendall, P. C., Chu, B., Gifford, A., Hayes, C., &Nauta, M. (1998). Breathing life into a manual: Flex-ibility and creativity with manual-based treatments.Cognitive and Behavioral Practice, 5, 177–198.

Kessler, R. C., Berglund, P. D., Demler, O., Olga, J. R.,Merikangas, K. R., & Walters, E. E. (2005). Lifetimeprevalence and age-of-onset distributions of DSM-IVdisorders in the National Comorbidity Survey Repli-cation. Archives of General Psychiatry, 62, 593–602.

Kocvoski, N. L., Rector, N. A., & Denisoff, E. (2004,November). Post-event processing during CBGT forsocial anxiety disorder. Poster presented at the meet-ing of the Association for the Advancement of Behav-ioral Therapy, New Orleans, LA.

Leary, M. R. (1983). A brief version of the Fear of Nega-tive Evaluation Scale. Personality and Social Psychol-ogy Bulletin, 9, 371–375.

Leary, M. R., Kowalski, R. M., & Campbell, C. D.(1988). Self-presentational concerns and social anx-iety: The role of generalized impression expectan-cies. Journal of Research in Personality, 22, 308–321.

Leung, A. W., & Heimberg, R. G. (1996). Homeworkcompliance, perceptions of control, and outcome ofcognitive-behavioral treatment for social phobia. Be-haviour Research and Therapy, 34, 423–432.

Liebowitz, M. R. (1987). Social phobia. Modern Prob-lems in Pharmacopsychiatry, 22, 141–173.

Liebowitz, M. R., Heimberg, R. G., Fresco, D. M.,Travers, J., & Stein, M. B. (2000). Social phobia orsocial anxiety disorder: What’s in a name? Archivesof General Psychiatry, 57, 191–192.

Liebowitz, M. R., Heimberg, R. G., Schneier, F. R.,Hope, D. A., Davies, S., Holt, C. S., et al. (1999).Cognitive-behavioral group therapy versus

phenelzine in social phobia: Long-term outcome. De-pression and Anxiety, 10, 89–98.

Lucas, R. A., & Telch, M. J. (1993, November).Group versus individual treatment of social phobia.Paper presented at the annual meeting of the Asso-ciation for Advancement of Behavior Therapy, At-lanta, GA.

MacLeod, C., & Mathews, A. (1991). Biased cognitiveoperations in anxiety: Accessibility of information orassignment of processing priorities? Behaviour Re-search and Therapy, 29, 599–610.

Magee, L., Erwin, B. A., & Heimberg, R. G. (in press).Psychological treatment of social anxiety disorderand specific phobia. In M. M. Antony & M. B. Stein(Eds.), Handbook of anxiety and the anxiety disor-ders. New York: Oxford University Press.

Magee, W. J., Eaton, W. W., Wittchen, H.-U.,McGonagle, K. A., & Kessler, R. C. (1996). Agora-phobia, simple phobia, and social phobia in the Na-tional Comorbidity Survey. Archives of General Psy-chiatry, 53, 159–168.

Mannuzza, S., Schneier, F. R., Chapman, T. F.,Liebowitz, M. R., Klein, D. F., & Fyer, A. J. (1995).Generalized social phobia: Reliability and validity.Archives of General Psychiatry, 52, 230–237.

Mattick, R. P., & Clarke, J. C. (1998). Development andvalidation of measures of social phobia scrutiny fearand social interaction anxiety. Behaviour Researchand Therapy, 36, 455–470.

Mattick, R. P., & Peters, L. (1988). Treatment of severesocial phobia: Effects of guided exposure with andwithout cognitive restructuring. Journal of Con-sulting and Clinical Psychology, 56, 251–260.

Mattick, R. P., Peters, L., & Clarke, J. C. (1989). Expo-sure and cognitive restructuring for social phobia: Acontrolled study. Behavior Therapy, 20, 3–23.

Mennin, D. S., Fresco, D. M., Heimberg, R. G.,Schneier, F. R., Davies, S. O., & Liebowitz, M. R.(2002). Screening for social anxiety disorder in theclinical setting: Using the Liebowitz Social AnxietyScale. Journal of Anxiety Disorders, 16, 661–673.

Narrow, W. E., Rae, D. S., Robins, L. N., & Regier, D.A. (2002). Revised prevalence estimates of mentaldisorders in the United States: Using a clinical signifi-cance criterion to reconcile two surveys’ estimates.Archives of General Psychiatry, 59, 115–123.

Norton, P. J., & Hope, D. A. (2001). Kernels of truth ordistorted perceptions: Self and observer ratings of so-cial anxiety and performance. Behavior Therapy, 32,765–786.

Rapee, R. M., & Heimberg, R. G. (1997). A cognitive-behavioral model of anxiety in social phobia. Behav-iour Research and Therapy, 35, 741–756.

Rapee, R. M., & Lim, L. (1992). Discrepancy betweenself- and observer ratings of performance in socialphobics. Journal of Abnormal Psychology, 101, 728–731.

Reich, J., Goldenberg, I., Vasile, R., Goisman, R., &Keller, M. (1994). A prospective follow-along study


of the course of social phobia. Psychiatry Research,54, 249–258.

Ries, B. J., McNeil, D. W., Boone, M. L., Turk, C. L.,Carter, L. E., & Heimberg, R. G. (1998). Assessmentof contemporary social phobia verbal report instru-ments. Behaviour Research and Therapy, 36, 983–994.

Rodebaugh, T. L., Holaway, R. M., & Heimberg, R. G.(2004). The treatment of social anxiety disorder.Clinical Psychology Review, 24, 883–908.

Rodebaugh, T. L., Woods, C. M., Thissen, D. M.,Heimberg, R. G., Chambless, D. L., & Rapee, R. M.(2004). More information from fewer questions: Thefactor structure and item properties of the originaland Brief Fear of Negative Evaluation Scale. Psycho-logical Assessment, 16, 169–181.

Roth, D. A., Coles, M. E., & Heimberg, R. G. (2002).The relationship between memories for childhoodteasing and anxiety and depression in adulthood.Journal of Anxiety Disorders, 16, 149–164.

Roth, D. A., & Heimberg, R. G. (2001). Cognitive-behavioral models of social anxiety disorder. Psychi-atric Clinics of North America, 24, 753–771.

Safren, S. A., Heimberg, R. G., Brown, E. J., & Holle,C. (1997). Quality of life in social phobia. Depres-sion and Anxiety, 4, 126–133.

Safren, S. A., Heimberg, R. G., & Juster, H. R. (1997).Client expectancies and their relationship to pretreat-ment symptomatology and outcome of cognitive-behavioral group treatment for social phobia. Jour-nal of Consulting and Clinical Psychology, 65, 694–698.

Sanderson, W. C., Di Nardo, P. A., Rapee, R. M., &Barlow, D. H. (1990). Syndrome comorbidity in pa-tients diagnosed with a DSM-III-R anxiety disorder.Journal of Abnormal Psychology, 99, 308–312.

Sank, L. I., & Shaffer, C. S. (1984). A therapist’s manualfor cognitive behavior therapy in groups. New York:Plenum Press.

Schneier, F. R., Heckelman, L. R., Garfinkel, R.,Campeas, R., Fallon, B. A., Gitow, A., et al. (1994).Functional impairment in social phobia. Journal ofClinical Psychiatry, 55, 322–331.

Schneier, F. R., Johnson, J., Hornig, C. D., Liebowitz,M. R., & Weissman, M. M. (1992). Social phobia:Comorbidity and morbidity in an epidemiologicsample. Archives of General Psychiatry, 49, 282–288.

Scholing, A., & Emmelkamp, P. M. G. (1993). Exposurewith and without cognitive therapy for generalizedsocial phobia: Effects of individual and group treat-ment. Behaviour Research and Therapy, 31, 667–681.

Schwartz, R. M., & Garamoni, G. L. (1989). Cognitivebalance and psychopathology: Evaluation of an in-formation processing model of positive and negativestates of mind. Clinical Psychology Review, 9, 271–294.

Stangier, U., Heidenreich, T., Peitz, M., Lauterbach, W.,

& Clark, D. M. (2003). Cognitive therapy for socialphobia: Individual versus group treatment. Behav-iour Research and Therapy, 41, 991–1007.

Stopa, L., & Clark, D. M. (1993). Cognitive processesin social phobia. Behaviour Research and Therapy,31, 255–267.

Taylor, S., Woody, S., Koch, W., McLean, P., Paterson,R., & Anderson, K. W. (1997a). Cognitive restruc-turing in the treatment of social phobia: Efficacy andmode of action. Behavior Modification, 21, 487–511.

Taylor, S., Woody, S., McLean, P. D., & Koch, W. J.(1997b). Sensitivity of outcome measures for treat-ments of generalized social phobia. Assessment, 4,181–191.

Trower, P., & Gilbert, P. (1989). New theoretical con-ceptions of social anxiety and social phobia. ClinicalPsychology Review, 9, 19–35.

Turk, C. L., Lerner, J., Heimberg, R. G., & Rapee, R. M.(2001). An integrated cognitive-behavioral model ofsocial anxiety. In S. G. Hofmann & P. M. DiBartolo(Eds.), From social anxiety to social phobia: Multipleperspectives (pp. 281–303). Needham Heights, MA:Allyn & Bacon.

Turner, S. M., Beidel, D. C., Dancu, C. V., & Keys, D. J.(1986). Psychopathology of social phobia and com-parison to avoidant personality disorder. Journal ofAbnormal Psychology, 95, 389–394.

Turner, S. M., Beidel, D. C., Dancu, C. V., & Stanley, M.A. (1989). An empirically derived inventory to mea-sure social fears and anxiety: The Social Phobia andAnxiety Inventory. Psychological Assessment, 1, 35–40.

Turner, S. M., Beidel, D. C., & Townsley, R. M. (1992).Social phobia: A comparison of specific and general-ized subtype and avoidant personality disorder. Jour-nal of Abnormal Psychology, 101, 326–331.

Turner, S. M., Beidel, D. C., Wolff, P. L., Spaulding, S.,& Jacob, R. G. (1996). Clinical features affectingtreatment outcome in social phobia. Behaviour Re-search and Therapy, 34, 795–804.

Wallace, S. T., & Alden, L. E. (1991). A comparison ofsocial standards and perceived ability in anxious andnonanxious men. Cognitive Therapy and Research,15, 237–254.

Watson, D., & Friend, R. (1969). Measurement ofsocial-evaluative anxiety. Journal of Consulting andClinical Psychology, 33, 448–457.

Weeks, J. W., Heimberg, R. G., Fresco, D. M., Hart, T.A., Turk, C. L., Schneier, F. R., et al. (2005). Empiri-cal validation and psychometric evaluation of theBrief Fear of Negative Evaluation Scale in patientswith social anxiety disorder. Psychological Assess-ment, 17, 179–190.

Weeks, J. W., Spokas, M. E., & Heimberg, R. G. (inpress). Psychometric evaluation of the Mini–SocialPhobia Inventory (Mini-SPIN) in a treatment-seekingsample. Depression and Anxiety.

Weiller, E., Bisserbe, J. C., Boyer, P., Lepine, J.-P., &


Lecrubier, Y. (1996). Social phobia in general healthcare: An unrecognised undertreated disabling disor-der. British Journal of Psychiatry, 168, 169–174.

Widiger, T. A. (1992). Generalized social phobia versusavoidant personality disorder: A commentary onthree studies. Journal of Abnormal Psychology, 101,340–343.

Wlazlo, Z., Schroeder-Hartwig, K., Hand, I., Kaiser, G.,& Münchau, N. (1990). Exposure in vivo vs. socialskills training for social phobia: Long-term outcomeand differential effects. Behaviour Research andTherapy, 28, 181–193.

Wolpe, J., & Lazarus, A. A. (1966). Behavior therapytechniques. New York: Pergamon.

Woody, S. R., & Adessky, R. S. (2002). Therapeutic alli-ance, group cohesion, and homework complianceduring cognitive-behavioral group treatment of so-cial phobia. Behavior Therapy, 35, 5–27.

Zaider, T. I., Heimberg, R. G., Roth, D. A., Hope, D., &Turk, C. L. (2003, November). Individual cognitive-behavioral therapy for social anxiety disorder: Pre-liminary findings. Paper presented at the AnnualMeeting of the Association for Advancement ofBehavior Therapy, Boston.


C H A P T E R 4

Obsessive–Compulsive Disorder

MARTIN E. FRANKLINEDNA B. FOA

It will not take the reader long to see that successful therapy for obsessive–compulsivedisorder (OCD) is markedly different in both structure and content from the usual thera-peutic approaches. For this reason, regrettably, few therapists feel self-efficacious enoughto undertake this therapy, yet this approach is clearly the treatment of choice for the mostbeneficial short- and long-term effects in OCD according to recent clinical trials. The infor-mation provided in this detailed chapter should be sufficient for any reasonably well-trained mental health professional to undertake this treatment, particularly if few otheroptions are available. The suffering involved with OCD can be extraordinary, and even im-perfect attempts at therapy can relieve much of this suffering. This chapter describes thedetailed conduct of intensive daily sessions involving both imaginal and direct in vivo prac-tice. Also noticeable is the ingenuity required of therapists (e.g., Where do you find deadanimals?). The importance of involving significant others continues a theme first describedby Craske and Barlow in Chapter 1 of this volume, where spouses/partners or other peo-ple close to the individual with the problem become an important and integral part of treat-ment. Finally, this chapter contains an up-to-date review of the current status of psycho-logical and pharmacological approaches to OCD.—D. H. B.

Advances in cognitive-behavioral and phar-macological treatments in the last three de-cades have greatly improved the prognosis forpatients with obsessive–compulsive disorder(OCD). In this chapter we first discuss diagnos-tic and theoretical issues of OCD and reviewthe available treatments, then describe assess-ment procedures and illustrate in detail how toimplement intensive cognitive-behavioral treat-ment (CBT) involving exposure and ritual pre-vention (EX/RP) for OCD. Throughout thechapter, we use case material to illustrate inter-actions that occur between therapist and pa-tient to demonstrate the process that occursduring treatment.

DEFINITION

According to the text revision of the fourth edi-tion of the Diagnostic and Statistical Manual ofMental Disorders (DSM-IV-TR; American Psy-chiatric Association, 2000), OCD is character-ized by recurrent obsessions and/or compul-sions that interfere substantially with dailyfunctioning. Obsessions are “persistent ideas,thoughts, impulses, or images that are experi-enced as intrusive and inappropriate and causemarked anxiety or distress” (p. 457). Commonobsessions are repeated thoughts about causingharm to others, contamination, and doubtingwhether one locked the front door. Compul-

164

sions are “repetitive behaviors or mental actsof which the goal is to prevent or reduce anxi-ety or distress” (p. 457). Common compulsionsinclude handwashing, checking, and counting.

In DSM-IV-TR, the functional link betweenobsessions and compulsions is emphasized:“Obsessions” are defined as thoughts, images,or impulses that cause marked anxiety or dis-tress, and “compulsions” are defined as overt(behavioral) or covert (mental) actions that areperformed in an attempt to reduce the distressbrought on by obsessions or according to rigidrules. This modification is supported by find-ings from the DSM-IV field trial on OCD, inwhich over 90% of participants reported thatthe aim of their compulsions was either to pre-vent harm associated with their obsessionsor to reduce obsessional distress (Foa et al.,1995).

Data from the DSM-IV field study also in-dicated that the vast majority (over 90%) ofindividuals with OCD manifest both obses-sions and behavioral rituals. When mental rit-uals are included, only 2% of the sample re-ported “pure” obsessions (Foa et al., 1995).Behavioral rituals (e.g., handwashing) areequivalent to mental rituals (e.g., silently re-peating special prayers) in their functional re-lationship to obsessions: Both serve to reduceobsessional distress, to prevent feared harm,or to restore safety. Thus, whereas all obses-sions are indeed mental events, compulsionscan be either mental or behavioral. Identifica-tion of mental rituals is an especially impor-tant aspect of treatment planning, becauseobsessions and compulsions are addressed viadifferent techniques. For example, we oncetreated a patient who described himself as a“pure obsessional,” who would experienceintrusive and unwanted images of harm com-ing to his girlfriend by an animal attack. Thepatient would quickly and intentionally inserthis own image into the scene to become thevictim of the animal mauling, thereby reduc-ing his distress and, in his estimation, reduc-ing the likelihood that some future harmwould come to his girlfriend. The substitutionof his own image into the scene constituted amental ritual, and the success of imaginal ex-posure exercises required that the patient re-frain from this form of compulsion.

A growing consensus about a continuum ofinsight in individuals with OCD (e.g., Foa etal., 1995; Insel & Akiskal, 1986) led to the in-clusion in DSM-IV (American Psychiatric As-

sociation, 1994) of a subtype of OCD “withpoor insight” to include individuals who in-deed have obsessions and compulsions but failto recognize their senselessness; this changewas retained in DSM-IV-TR. Clinically it is im-portant to evaluate the degree of insight priorto initiating CBT, because fixed belief about theconsequences of refraining from compulsionsand avoidance behaviors has been found to beassociated with attentuated treatment outcome(e.g., Foa, Abramowitz, Franklin, & Kozak,1999; Neziroglu, Stevens, Yaryura-Tobias, &McKay, 2000).

To be diagnosed with OCD, obsessions and/or compulsions must be found to be of suffi-cient severity to cause marked distress, be time-consuming, and interfere with daily function-ing. If another Axis I disorder is present, theobsessions and compulsions cannot be re-stricted to the content of that disorder (e.g.,preoccupation with food in the presence of eat-ing disorders). There continues to be muchdiscussion about the presence of an obsessive–compulsive spectrum that includes a wide vari-ety of compulsive and impulsive disorders (e.g.,pathological gambling, trichotillomania; seeHollander & Stein, 2006), but current DSM-IV-TR nosology still specifies that such disor-ders should not be diagnosed as OCD.

PREVALENCE AND COURSE OF OCD

Once thought to be an extremely rare disorder,the 12-month prevalence of OCD was esti-mated at 1.0% in the recent National Comor-bidity Survey Replication involving over 9,000adult participants in the United States (Kessleret al., 2005). Epidemiological studies with chil-dren and adolescents suggest similar lifetimeprevalence rates in these samples (e.g., Flamentet al., 1988; Valleni-Basille et al., 1994).Slightly more than half of adults suffering fromOCD are female (Rasmussen & Tsuang, 1986),whereas a 2:1 male to female ratio has been ob-served in several pediatric clinical samples (e.g.,Hanna, 1995; Swedo, Rapoport, Leonard,Lenane, & Cheslow, 1989). Age of onset of thedisorder typically ranges from early adoles-cence to young adulthood, with earlier onset inmales; modal onset is ages 13–15 in males, andages 20–24 in females (Rasmussen & Eisen,1990). However, cases of OCD have beendocumented in children as young as age 2(Rapoport, Swedo, & Leonard, 1992).

Obsessive–Compulsive Disorder 165

Development of the disorder is usually grad-ual, but acute onset has been reported in somecases. Although chronic waxing and waning ofsymptoms are typical, episodic and deteriorat-ing courses have been observed in about 10%of patients (Rasmussen & Eisen, 1989). Insome cases of pediatric OCD and tic disorders,onset is very sudden and associated with strepinfection; treatment of the infection is associ-ated with substantial reduction of symptoms,but recurrence of infection is again associatedwith symptom exacerbation (Swedo et al.,1998). Presentation of OCD in these cases ismuch more typical in males than in females andhas come to be known as pediatric autoim-mune neuropsychiatric disorders (PANDAS);the prevalence of PANDAS has yet to be deter-mined. OCD is frequently associated with im-pairments in general functioning, such as dis-ruption of gainful employment (Koran, 2000;Leon, Portera, & Weissman, 1995), and inter-personal relationship difficulties (Emmelkamp,de Haan, & Hoogduin, 1990; Riggs, Hiss, &Foa, 1992). Adolescents identified as havingOCD (Flament et al., 1988) reported in a sub-sequent follow-up study that they had with-drawn socially to prevent contamination andto conserve energy for obsessive–compulsivebehaviors (Flament et al., 1990).

Many individuals with OCD suffer for yearsbefore seeking treatment. In one study, individ-uals first presented for psychiatric treatmentover 7 years after the onset of significant symp-toms (Rasmussen & Tsuang, 1986). The disor-der may cause severe impairment in function-ing, resulting in job loss and disruption ofmarital and other interpersonal relationships.Marital distress is reported by approximately50% of married individuals seeking treatmentfor OCD (Emmelkamp et al., 1990; Riggs etal., 1992).

COMORBIDITY

Convergent epidemiological and clinical dataindicate that OCD rarely occurs in isolation:Although the rates of comorbidity differ acrossstudies due to selection of population andmethodology, comorbidity is generally high.For example, Weissman and colleagues (1994)found that 49% of individuals diagnosed withOCD suffered from a comorbid anxiety disor-der and 27% from comorbid major depressivedisorder (MDD). Among studies conducted

specifically within anxiety clinics, there is greatvariability, but comorbid conditions are gener-ally common (for a review, see Ledley, Pai, &Franklin 2007). In the largest of the studiesconducted in the context of an anxiety clinic,Brown, Campbell, Lehman, Grisham, andMancill (2001) found that 57% of 77 adultswith a principal diagnosis of OCD had a cur-rent comorbid Axis I condition; the rate rose to86% for lifetime comorbid Axis I conditions.Notably, when OCD co-occurs with other anx-iety disorders, it is typically the principal diag-nosis (e.g., the diagnosis of greatest severity;see Antony, Downie, & Swinson, 1998). It alsoappears to be the case that MDD onset tends tofollow that of OCD, suggesting that depressionmight be a response to OCD symptoms(Bellodi, Sciuto, Diaferia, Ronchi, & Smeraldi,1992; Diniz et al., 2004).

The data are equivocal with respect to the in-fluence of comorbidity on OCD presentation.In one study, Denys, Tenney, van Megen, deGeus, and Westenberg (2004) found thatcomorbidity did not influence OCD symptomseverity, whereas others (Angst, 1993; Tukel,Polat, Ozdemir, Aksut, & Turksov, 2002)found a relationship between comorbidity andOCD symptom severity. A more consistentfinding is that comorbidity is associated withpoorer quality of life, particularly in the case ofcomorbid depression (Lochner & Stein, 2003;Masellis, Rector, & Richter, 2003).

With respect to the effect of comorbid anxi-ety and depression on treatment outcome, theinfluence of depression has received more em-pirical attention to date. Some studies havefound that higher levels of depression at pre-treatment are related to poorer outcome (e.g.,Keijsers, Hoogduin, & Schaap, 1994; Steketee,Chambless, & Tran, 2001), whereas othershave found little or no effect (Mataix-Cols,Marks, Greist, Kobak, & Baer, 2002;O’Sullivan, Noshirvani, Marks, Monteiro, &Lelliott, 1991; Steketee, Eisen, Dyck, Warshaw,& Rasmussen, 1999). Some have suggestedthat, more specifically, the severity of thecomorbid depression might influence its effectson OCD treatment outcome: Abramowitz,Franklin, Street, Kozak, and Foa (2000) foundthat only severely depressed patients were lesslikely to respond to EX/RP therapy for OCD.Similarly, highly depressed OCD patients seemto be at greater risk for relapse following treat-ment discontinuation (Abramowitz & Foa,2000; Basoglu, Lax, Kasvikis, & Marks,


1998). The influence of comorbid anxiety dis-orders on outcome has received less attentionthus far: One study reported that patients withOCD and comorbid generalized anxiety disor-der (GAD) terminate OCD treatment at higherrates than other patients (Steketee et al., 2001),and another found that the presence of post-traumatic stress disorder (PTSD) in patientswith OCD attenuated response to EX/RP(Gershuny, Baer, Jenike, Minichiello, & Wil-helm, 2002). The mechanisms by which thesecomorbid conditions influence outcome haveyet to be explored.

Tourette syndrome and other tic disordersalso appear to be related to OCD. Estimates ofthe comorbidity of Tourette syndrome andOCD range from 28 to 63% (Comings, 1990;Kurlan et al., 2002; Leckman & Chittenden,1990; Pauls, Towbin, Leckman, Zahner, & Co-hen, 1986). Conversely, up to 17% of patientswith OCD are thought to have Tourette syn-drome (Comings, 1990; Kurlan et al., 2002;Rasmussen & Eisen, 1989). In a recent report,tic comorbidity was associated with poorertreatment outcome in general (Matsunaga etal., 2005), yet in a recent pediatric study wasfound to influence pharmacotherapy treatmentoutcome but not CBT response (March et al.,2007).

DIFFERENTIAL DIAGNOSIS

The high comorbidity of OCD with other dis-orders noted earlier, as well as the similarity be-tween the criteria for OCD and other DSM-IVdisorders, can pose diagnostic quandaries. Be-low we review some of the more common diag-nostic difficulties likely to confront clinicians,and provide recommendations for makingthese difficult diagnostic judgments.

Obsessions versus Depressive Rumination

It is sometimes difficult to differentiate be-tween depressive ruminations and obsessions.The distinction rests primarily on thought con-tent and the patient’s reported resistance tosuch thoughts. Unlike obsessions, ruminationsare typically pessimistic ideas about the self orthe world, and ruminative content frequentlyshifts. Additionally, depressive ruminators tendto not make repeated attempts to suppress theirruminations the way individuals with OCD tryto suppress obsessions. When depression and

OCD co-occur, both phenomena may be pres-ent, but only obsessions should be targetedwith exposure exercises. We have also foundclinically that the generally pessimistic presen-tation of depressed patients can underminehopefulness about improvement during EX/RP; thus, these beliefs may require therapeuticintervention even though they are not obses-sional.

Other Anxiety Disorders

OCD often co-occurs with other anxiety disor-ders, and diagnostic criteria are sometimes sim-ilar among anxiety disorders, but the symp-toms associated with each diagnosis canusually be distinguished. For example, the ex-cessive worries characteristic of GAD may ap-pear similar to those in OCD but, unlike obses-sions, worries are excessive concerns aboutreal-life circumstances and are experienced bythe individual as appropriate (ego-syntonic). Incontrast, obsessive thinking is more likely to beunrealistic or magical, and obsessions are usu-ally experienced by the individual as inappro-priate (ego-dystonic). There are, however, ex-ceptions to this general rule: Individuals witheither GAD or OCD may worry about every-day matters, such as their children getting sick.However, when worried about their childrencatching cold, parents with GAD might focustheir concern on the long-term consequences(e.g., falling behind in school, development of alifelong pattern of debilitation), whereas par-ents with OCD might focus more on the con-tamination aspect of illness (e.g., their child be-ing infested with “cold germs”). The problemof distinguishing between obsessions and wor-ries in a particular patient is most relevantwhen the patient exhibits no compulsions, but,as we mentioned earlier, pure obsessionalscomprise only about 2% of individuals withOCD (Foa et al., 1995).

In the absence of rituals, the avoidance asso-ciated with specific phobias may also appearsimilar to OCD. For example, excessive fear ofgerms and specific phobia both may result inpersistent fear of dogs. However, unlike an in-dividual with OCD, a person with a specificphobia can successfully avoid dogs for the mostpart, or reduce distress quickly by escapingdogs when avoidance is impractical. In con-trast, the individual with OCD who is obsessedwith “dog germs” continues to feel contami-nated even after the dog is gone, and sometimes


knowing that a dog was in the vicinity severalhours earlier can also produce obsessional dis-tress even if there is no possibility that the dogwill return. This distress often prompts subse-quent avoidance behaviors (e.g., taking offclothing that might have been near the contam-inating dog) not typically observed in specificphobias.

Hypochondriasis and BodyDysmorphic Disorder

The health concerns that characterize hypo-chondriasis and the preoccupation with imag-ined physical defects of body dysmorphic dis-order (BDD) are both formally similar to theobsessions of OCD. The best way to differenti-ate between these disorders and OCD is toexamine for content specificity of the fear-provoking thoughts. Most individuals withhypochondriasis or BDD are singly obsessed,whereas most individuals with OCD have mul-tiple obsessions. Moreover, patients with OCDtypically obsess about contracting diseases orbecoming ill in the future, whereas patientswith hypochondriasis focus on physical andpsychological symptoms that lead them to fearthey have already contracted diseases or ill-nesses.

Tourette Syndrome and Tic Disorders

To differentiate the stereotyped motor behav-iors that characterize Tourette syndrome andtic disorders from compulsions, the functionalrelationship between these behaviors and anyobsessive thoughts must be examined. Motortics are generally experienced as involuntaryand are not aimed at neutralizing distressbrought about by obsessions. There is noconventional way to differentiate them from“pure” compulsions, but OCD with “pure”compulsions is extremely rare (Foa et al.,1995). As noted earlier, there appears to be ahigh rate of comorbidity between OCD and ticdisorders (e.g., Pauls et al., 1986); thus, bothdisorders may be present simultaneously in agiven patient. Interestingly, recent research in-dicated that tics were similarly responsive to anexposure plus response prevention protocolwhen compared in a randomized study tohabit-reversal training in which a competingresponse is substituted for the tic; this findingsuggests that the conceptual model underlyingthe treatment of tics might require modifica-

tion (Verdellen, Keijsers, Cath, & Hoogduin,2004).

Delusional Disorder and Schizophrenia

Individuals with OCD may present with obses-sions of delusional intensity (for a review, seeKozak & Foa, 1994). Approximately 5% ofpatients with OCD report complete convictionthat their obsessions and compulsions are real-istic, with an additional 20% reporting strongbut not fixed conviction. Therefore, it is impor-tant to consider the diagnosis of OCD “withpoor insight” even if these beliefs are verystrongly held. The differentiation between de-lusional disorder and OCD can depend on thepresence of compulsions in OCD (Eisen et al.,1998). In OCD, obsessions of delusional inten-sity are usually accompanied by compulsions.

It is also important to recognize that the con-tent of obsessions in OCD may be quite bi-zarre, as in the delusions of schizophrenia, butbizarreness in and of itself does not preclude adiagnosis of OCD. For example, one patientseen at our center was fearful that small bits ofher “essence” would be forever lost if shepassed too close to public trash cans. This pa-tient did not report any other symptoms offormal thought disorder, such as loose associa-tions, hallucinations, flat or grossly inappropri-ate affect, and thought insertion or projection.Following a course of EX/RP that focused onexercises designed to expose the patient to theloss of her “essence” (e.g., driving by the citydump), her OCD symptoms were substantiallyreduced. On occasion patients do meet diag-nostic criteria for both OCD and schizophre-nia, and a dual diagnosis is appropriate underthese circumstances. Importantly, EX/RP withsuch patients should proceed only if the associ-ated treatment exercises do not exacerbate thecomorbid thought disorder symptoms.

COGNITIVE ANDBEHAVIORAL MODELS

Mowrer’s (1939) two-stage theory for the ac-quisition and maintenance of fear and avoid-ance behavior has been commonly adopted toexplain phobias and OCD. As elaborated byMowrer (1960), this theory proposes that inthe first stage, a neutral event becomes associ-ated with fear by being paired with a stimulusthat by its nature provokes discomfort or anxi-


ety. Through conditioning processes, objects,as well as thoughts and images, acquire theability to produce discomfort. In the secondstage of this process, escape or avoidance re-sponses are developed to reduce the anxiety ordiscomfort evoked by the various conditionedstimuli and are maintained by their success indoing so. Dollard and Miller (1950) adoptedMowrer’s two-stage theory to explain thedevelopment of phobias and obsessive–compulsive neurosis. As noted earlier, becauseof the intrusive nature of obsessions, many sit-uations that provoke obsessions cannot readilybe avoided. Passive avoidance behaviors, suchas those utilized by phobics, are also less effec-tive in controlling obsessional distress. Activeavoidance patterns in the form of ritualistic be-haviors are then developed and maintained bytheir success in alleviating this distress.

In light of equivocal empirical support forthe two-stage theory and its limitations in ex-plaining the etiology of obsessions, several cog-nitive explanations have been offered to ac-count for the development and maintenance ofOCD symptoms. Carr (1974) proposed thatindividuals with OCD have unusually highexpectations of negative outcome; theyoverevaluate the negative consequences for avariety of actions. He noted that obsessionalcontent typically includes exaggerations of theconcerns of normal individuals: health, death,welfare of others, sex, religion, and the like.According to this theory, the sources ofobsessive–compulsive concerns are identical tothose found in GAD, agoraphobia, and socialphobia. In this way, Carr’s explanation of OCDis similar to that offered by Beck (1976), whosuggested that the content of obsessions is re-lated to danger in the form of doubt or warn-ing. However, neither account distinguishesbetween threat-related obsessions and threat-related thoughts in phobics.

Salkovskis (1985) offered a more compre-hensive cognitive analysis of OCD. He positedthat intrusive obsessional thoughts are stimulithat may provoke certain types of negative au-tomatic thoughts. Accordingly, an intrusivethought leads to mood disturbances only if ittriggers negative automatic thoughts throughinteraction between the unacceptable intrusionand the individual’s belief system (e.g., onlybad people have sexual thoughts). Accordingto Salkovskis, exaggerated senses of responsi-bility and self-blame are the central themes inthe belief system of a person with OCD.

Neutralization, in the form of behavioral orcognitive compulsions, may be understood asan attempt to reduce this sense of responsibilityand to prevent blame. In addition, frequentlyoccurring thoughts regarding unacceptable ac-tions may be perceived by the individual withOCD as equivalent to the actions themselves,so, for example, even if the person has notsinned, the thought of sinning is as bad as sin-ning itself.

Salkovskis (1985) further proposed that fivedysfunctional assumptions characterize indi-viduals with OCD and differentiate them frompersons without OCD:

(1) Having a thought about an action is like per-forming the action; (2) failing to prevent (or fail-ing to try to prevent) harm to self or others is thesame as having caused the harm in the first place;(3) responsibility is not attenuated by other fac-tors (e.g., low probability of occurrence); (4) notneutralizing when an intrusion has occurred issimilar or equivalent to seeking or wanting theharm involved in that intrusion to actually hap-pen; (5) one should (and can) exercise controlover one’s thoughts. (p. 579)

Thus, while the obsession may be ego-dystonic,the automatic thought it elicits will be ego-syntonic. By extension, this model suggests thattreatment of OCD should largely focus onidentifying the erroneous assumptions andmodifying the automatic thoughts. This theorypaved the way for various elaborations on thecognitive models, experimental studies of themodel, and the development of cognitive thera-pies that derive from the central role of thesekey cognitive factors.

Salkovskis’s (1985) theory sparked examina-tion of the role of responsibility in the psycho-pathology of OCD (Ladoucer et al., 1995;Rachman, Thordarson, Shafran, & Woody,1995; Rhéame, Freeston, Dugas, Letarte, &Ladoucer, 1995). Further attention has beenpaid to what Rachman (1998) referred to asthought–action fusion (TAF), wherein individ-uals believe that simply having an unacceptablethought increases the likelihood of the occur-rence of a feared outcome, and that thoughts ofengaging in repugnant activities are equivalentto actually having done so. Contemporarycognitive theorists would then suggest thatobsessive–compulsive beliefs such as TAF, ex-aggerated responsibility, and intolerance of un-certainty likely result in increased and ulti-mately futile efforts at thought suppression and


other ill-advised mental control strategies,which would then yield increased frequency ofsuch thoughts and associated distress (Purdon& Clark, 2002). Hence a vicious cycle ofavoidance maintains and strengthens the OCD,and the cognitive therapies that derive fromthese contemporary models would directly tar-get these obsessive–compulsive beliefs in an ef-fort to break the cycle.

In an integrated cognitive-behavioral ac-count, Foa and Kozak (1985) conceptualizedanxiety disorders in general as specific impair-ments in emotional memory networks. Follow-ing Lang (1979), they view fear as an informa-tion network existing in memory that includesrepresentation about fear stimuli, fear re-sponses, and their meaning. With regard to thefear content, Foa and Kozak suggested thatfear networks of individuals with anxiety dis-orders are characterized by the presence of er-roneous estimates of threat, unusually highnegative valence for the feared event, and ex-cessive response elements (e.g., physiologicalreactivity), and are resistant to modification.This persistence may reflect failure to accessthe fear network, because of either activeavoidance or because the content of the fearnetwork precludes spontaneous encounterswith situations that evoke anxiety in everydaylife. Additionally, anxiety may persist becauseof some impairment in the mechanism of ex-tinction. Cognitive defenses, excessive arousalwith failure to habituate, faulty premises, anderroneous rules of inference are all impair-ments that would hinder the processing of in-formation necessary for modifying the fearstructure to reduce fear behavior.

Foa and Kozak (1985) suggested severalforms of fear occur in individuals with OCD.The patient who fears contracting venereal dis-ease from public bathrooms and washes to pre-vent such harm has a fear structure that in-cludes excessive associations between thestimuli (e.g., bathroom) and the anxiety/dis-tress responses, as well as mistaken beliefsabout the harm related to the stimulus. Forother individuals with OCD, fear responses areassociated with mistaken meaning rather thanwith a particular stimulus. For example, somepatients who are disturbed by perceived asym-metry, and who reduce their distress by rear-ranging objects, do not fear the objects them-selves, nor do they anticipate disaster from theasymmetry. Rather, they are upset by their view

that certain arrangements of stimuli are “im-proper.”

Like Reed (1985), Foa and Kozak (1985)proposed that in addition to the pathologicalcontent of the obsessions, OCD is distin-guished from other disorders by pathology inthe mechanisms underlying information pro-cessing. Specifically, they suggested that pa-tients with OCD experience impairments inconsidering the rules for making inferencesabout harm, often concluding that a situationis dangerous based on the absence of evidencefor safety, and that they often fail to make in-ductive leaps about safety from informationabout the absence of danger. Consequently, rit-uals performed to reduce the likelihood ofharm can never provide safety and must be re-peated. In an elaboration on emotional pro-cessing theory and the mechanism by which ex-posure works, Foa, Huppert, and Cahill (2006)suggested that in vivo exposure to the fearedstimulus in the absence of the anticipated harmcorrects the exaggerated probability estimates;imaginal exposure not only corrects the exag-gerated cost but also strengthens the discrimi-nation between “thoughts about harm” and“real harm,” thus altering the associations be-tween threat meaning of stimulus and/or re-sponse elements in the fear structure.

In contrast to the more general theories ofOCD described earlier, some theorists haveposed more specific hypotheses to account forthe pathology observed in certain OCD sub-types. For example, clinical observations ledsome investigators to hypothesize that memorydeficits for actions underlie compulsive check-ing (e.g., Sher, Frost, & Otto, 1983). However,the results of experimental investigations ofthis hypothesis are equivocal. Some support foran action–memory deficit was found in non-clinical individuals with checking rituals (e.g.,Rubenstein, Peynircioglu, Chambless, &Pigott, 1993; Sher et al., 1983). In contrast, astudy using a clinical sample found that com-pared to nonpatients, OCD patients withchecking rituals better recalled their fear-relevant actions (e.g., plugging in an iron, un-sheathing a knife), but not fear-irrelevant ac-tions (e.g., putting paper clips in a box;Constans, Foa, Franklin, & Mathews, 1995).From these data, it appears that checking is notmotivated primarily by memory problems;hence, teaching mnemonic strategies to OCDpatients with checking rituals is probably not


the optimal clinical strategy; rather, having pa-tients repeatedly confront the low-risk situa-tions that provoke obsessional distress, whilesimultaneously refraining from checkingbehavior or mental reviewing of actions, is pre-ferred.

TREATMENTS

Exposure and Ritual Prevention

The prognostic picture for OCD has improveddramatically since Victor Meyer (1966) first re-ported on two patients who responded well toa treatment that included prolonged exposureto obsessional cues and strict prevention of rit-uals. This procedure, known at the time as ex-posure and ritual prevention (EX/RP), waslater found to be extremely successful in 10 of15 cases and partly effective in the remainder.Patients treated with this regimen also ap-peared to maintain their treatment gains: At a5-year follow-up only two of these patients hadrelapsed (Meyer & Levy, 1973; Meyer, Levy, &Schnurer, 1974).

As was the case with Meyer’s program, cur-rent EX/RP treatments typically include bothprolonged exposure to obsessional cues andprocedures aimed at blocking rituals. Exposureexercises are often done in real-life settings (invivo), for example, by asking the patient whofears accidentally causing a house fire by leav-ing the stove on, to leave the house withoutchecking the burners. When patients reportspecific feared consequences of refraining fromrituals, these fears may also be addressedvia imaginal exposure. In fact, in vivo andimaginal exposure exercises are designed spe-cifically to prompt obsessional distress. It is be-lieved that repeated, prolonged exposure tofeared thoughts and situations provides infor-mation that disconfirms mistaken associationsand evaluations held by the patients andthereby promotes habituation (Foa & Kozak,1986). Exposure is usually done gradually byconfronting situations that provoke moderatedistress before confronting more upsettingones. Exposure homework is routinely as-signed between sessions, and patients are alsoasked to refrain from rituals.

Since Meyer’s (1966) initial positive reportof the efficacy of EX/RP, many subsequentstudies of EX/RP have indicated that most EX/RP treatment completers make and maintain

clinically significant gains. Randomized con-trolled trials (RCTs) have indicated that EX/RPis superior to a variety of control treatments,including placebo medication (Marks, Stern,Mawson, Cobb, & McDonald, 1980), relax-ation (Fals-Stewart, Marks, & Schafer, 1993),and anxiety management training (Lindsay,Crino, & Andrews, 1997). Foa and Kozak’s(1996) review of 12 outcome studies (N = 330)that reported treatment responder rates indi-cated that 83% of EX/RP treatment completerswere classified as responders at posttreatment.In 16 studies reporting long-term outcome (N =376; mean follow-up interval of 29 months),76% were responders. Moreover, several stud-ies have now indicated that these encouragingfindings for EX/RP are not limited to highly se-lected RCT samples (Franklin, Abramowitz,Kozak, Levitt, & Foa, 2000; Rothbaum &Shahar, 2000; Valderhaug, Larsson, Gotestam,& Piacentini, 2006; Warren & Thomas, 2001).

In general, EX/RP has been found quite ef-fective in ameliorating OCD symptoms and hasproduced great durability of gains followingtreatment discontinuation. In our review of theliterature, it also was apparent that among themany variants of EX/RP treatment, some arerelevant for outcome and others are not. We re-view the literature on the relative efficacy of theingredients that comprise EX/RP to help clini-cians decide which EX/RP components aremost essential.

EX/RP Treatment Variables

EXPOSURE VERSUS RITUAL PREVENTIONVERSUS EX/RP

To separate the effects of exposure and ritualprevention on OCD symptoms, Foa, Steketee,Grayson, Turner, and Latimer (1984) randomlyassigned patients with washing rituals to treat-ment by exposure only (EX), ritual preventiononly (RP), or their combination (EX/RP). Eachtreatment was conducted intensively (15 daily2-hour sessions conducted over 3 weeks) andfollowed by a home visit. Patients in each con-dition were found to be improved at bothposttreatment and follow-up, but EX/RP wassuperior to the single-component treatments onalmost every symptom measure at both assess-ment points. In comparing EX and RP, patientswho received EX reported lower anxiety whenconfronting feared contaminants than patients


who had received RP, whereas the RP group re-ported greater decreases in urge to ritualizethan did the EX patients. Thus, it appears thatEX and RP affected different OCD symptoms.The findings from this study clearly suggestthat EX and RP should be implemented con-currently; treatments that do not include bothcomponents yield inferior outcome. It is impor-tant to convey this information to patients, es-pecially when they are experiencing difficultyeither refraining from rituals or engaging effec-tively in exposure exercises during and betweensessions.

IMPLEMENTATION OF RITUAL PREVENTION

Promoting abstinence from rituals during treat-ment is thought to be essential for successfultreatment outcome, but the preferred methodof RP has changed over the years. In Meyer’s(1966) EX/RP treatment program, hospitalstaff physically prevented patients from per-forming rituals (e.g., turning off the watersupply in a patient’s room). However, physicalintervention by staff or family members to pre-vent patients from ritualizing is no longer typi-cal nor recommended. It is believed that suchprevention techniques are too coercive to be anaccepted practice today. Moreover, physicalprevention by others may actually limitgeneralizability to nontherapy situations inwhich others are not present to intercede. In-stead, instructions and encouragement to re-frain from ritualizing and avoidance are nowrecommended. As noted earlier, although expo-sure in itself can reduce obsessional distress, itis not so effective in reducing compulsions. Tomaximize treatment effects, the patient needsto voluntarily refrain from ritualizing while en-gaging in systematic exposure exercises. Thetherapist should strongly emphasize the impor-tance of refraining from rituals and help the pa-tient with this difficult task by providing sup-port, encouragement, and suggestions aboutalternatives to ritualizing.

USE OF IMAGINAL EXPOSURE

Treatment involving imaginal plus in vivo EX/RP was superior at follow-up to an in vivo EX/RP program that did not include imaginal ex-posure (Foa, Steketee, Turner, & Fischer, 1980;Steketee, Foa, & Grayson, 1982). However, asecond study did not find that the addition of

imaginal exposure enhanced long-term efficacycompared to in vivo exposure only (De Araujo,Ito, Marks, & Deale, 1995). The treatmentprogram in the former study differed from thatof De Araujo and colleagues (1995) on severalparameters (e.g., 90-minute vs. 30-minuteimaginal exposures, respectively); thus, thesource of these studies’ inconsistencies cannotbe identified.

In our clinical work, we have found imaginalexposure to be helpful for patients who reportthat disastrous consequences will result if theyrefrain from rituals. Because many of theseconsequences cannot be readily translated intoin vivo exposure exercises (e.g., burning inhell), imaginal exposure allows the patient anopportunity to confront these feared thoughts.Also, the addition of imagery to in vivo expo-sure may circumvent the cognitive avoidancestrategies used by patients who intentionallytry not to consider the consequences of expo-sure while confronting feared situations invivo. In summary, although imaginal exposuredoes not appear essential for immediate out-come, it may enhance long-term maintenanceand be used as an adjunct to in vivo exercisesfor patients who fear disastrous consequences.For patients who only report extreme distressas a consequence of refraining from rituals andavoidance behaviors, imaginal exposure maynot be needed.

GRADUAL VERSUS ABRUPT EXPOSURES

No differences in OCD symptom reductionwere detected in a study comparing patientswho confronted the most distressing situationsfrom the start of therapy to those who con-fronted less distressing situations first, yet pa-tients preferred the more gradual approach(Hodgson, Rachman, & Marks, 1972). Be-cause patient motivation and agreement withtreatment goals are core elements of successfulEX/RP, situations of moderate difficulty areusually confronted first, followed by several in-termediate steps, before the most distressingexposures are attempted. Thus, we emphasizethat exposure will proceed at a pace that is ac-ceptable to the patient, and that no exposurewill ever be attempted without the patient’s ap-proval. At the same time, it is preferable to con-front the highest item on the treatment hierar-chy relatively early in treatment (e.g., withinthe first week of intensive treatment) to allow


sufficient time to repeat these difficult expo-sures over the later sessions.

DURATION OF EXPOSURE

Duration of exposure was once believed to beimportant for outcome in that prolonged, con-tinuous exposure was found more effectivethan short, interrupted exposure (Rabavilas,Boulougouris, & Perissaki, 1976). Indeed, re-duction in anxiety (habitation) across sessionshas been associated with improvement follow-ing exposure-based treatments for OCD andfor PTSD (e.g., Jaycox, Foa, & Morral, 1998;Kozak, Foa, & Steketee, 1988; van Minnen &Hagenaars, 2002). However, several studieshave not found a strong relationship betweenwithin-session habituation and fear and symp-tom reduction (Jaycox et al., 1998; Kozak etal., 1988; Mathews, Johnston, Shaw, & Gelder,1974; Rowe & Craske, 1998). In an elabora-tion on emotional processing theory, Foa andcolleagues (2006) found that the recent de-emphasis on the relationship between within-session habituation and outcome is not criticalto emotional processing theory, because theproposed mechanism underlying symptom re-duction is the modification of the relevant erro-neous associations through disconfirming in-formation, not through habituation per se. Inpractical terms, this means patients shouldbe instructed that although, optimally, theyshould persist with exposure until the anxietyis substantially reduced, the more importantfactor is repeating the same exposures, to pro-mote reduction of associated anxiety over time.Patients with OCD might be particularly vul-nerable to fears of ending exposures “toosoon,” hence, doing the treatment incorrectly,so this new instruction might help encouragepatients to go about their business withoutritualizing or avoiding, regardless of whetheranxiety still lingers on from an exposure task.

FREQUENCY OF EXPOSURE SESSIONS

Optimal frequency of exposure sessions has yetto be established. Intensive exposure therapyprograms that have achieved excellent results(e.g., Foa, Kozak, Steketee, & McCarthy,1992) typically involve daily sessions over thecourse of approximately 1 month, but quite fa-vorable outcomes have also been achieved withmore widely spaced sessions (e.g., Abram-

owitz, Foa, & Franklin, 2003; De Araujo et al.,1995; Franklin et al., 1998). A recent RCT inpediatric OCD found no difference between in-tensive and weekly treatment (Storch, Geffken,et al., in press). Clinically, we have found thatless frequent sessions may be sufficient forhighly motivated patients with mild to moder-ate OCD symptoms, who readily understandthe importance of daily exposure homework.Patients with very severe symptoms or thosewho for various reasons cannot readily complywith EX/RP tasks between sessions, are typi-cally offered intensive treatment.

THERAPIST-ASSISTED VERSUS SELF-EXPOSURE

Evaluations of the presence of a therapist dur-ing exposure have yielded inconsistent results.In one study, patients with OCD receivingtherapist-assisted exposure were more im-proved immediately posttreatment than thosereceiving clomipramine and self-exposure, butthis difference was not evident at follow-up(Marks et al., 1988). However, these results aredifficult to interpret in light of the study’s com-plex design. A second study using patients withOCD also indicated that therapist-assistedtreatment was not superior to self-exposure atposttreatment or at follow-up (Emmelkamp &van Kraanen, 1977), but the number of pa-tients in each condition was too small to renderthese findings conclusive. In contrast to thenegative findings of Marks and colleagues(1988) and Emmelkamp and van Kraanen(1977), therapist presence yielded superior out-come of a single, 3-hour exposure session com-pared to self-exposure for persons with specificphobia (Ost, 1989). Because specific phobiasare, on the whole, less disabling and easier totreat than OCD, one may surmise that thera-pist presence should also influence treatmentoutcome with OCD. Moreover, using meta-analytic procedures, Abramowitz (1996) foundthat therapist-controlled exposure was associ-ated with greater improvement in OCD andGAD symptoms compared to self-controlledprocedures. A recent study found comparableoutcome for patients receiving EX/RP withtherapist assistance and those who receivedteletherapy (Lovell et al., 2006), which furtherraises the question of whether therapist assis-tance is required for good outcome. In light ofthese inconsistent findings, no clear answer isavailable on the role of therapist assistance


with exposure tasks in OCD treatment. How-ever, we have found clinically that the presenceof a therapist can be useful in helping patientsremain engaged in exposures while anxiety ishigh, to avoid subtle rituals or avoidance be-haviors during exposure (e.g., distraction, men-tal rituals), and to remain sufficiently moti-vated despite distress.

EX/RP versus Other Treatment Approaches

In this section we review the literature on the ef-ficacy of standard individual EX/RP treatmentversus other therapeutic approaches, includinggroup treatment, family-based EX/RP treat-ment, cognitive therapy, and pharmacotherapy.

INDIVIDUAL VERSUS GROUP EX/RP

Intensive individual EX/RP, although effective,can pose practical obstacles such as high costfor treatment, and scheduling problems for pa-tient and therapist alike. Additionally, becauseexperts in EX/RP treatment are few and far be-tween, patients may need to wait for long peri-ods of time or travel substantial distances to betreated. Thus, some researchers have begun toexamine the efficacy of more affordable and ef-ficient treatment modalities. One such alterna-tive is group treatment. Fals-Stewart and col-leagues (1993) conducted a controlled study inwhich patients with OCD were randomly as-signed to individual EX/RP, group EX/RP, ora psychosocial control condition (relaxation).Each of the active treatments was 12 weekslong, with sessions held twice weekly, and in-cluded daily exposure homework. Significantimprovement in OCD symptoms was evidentin both active treatments, with no differencesdetected between individual and group EX/RPimmediately posttreatment or at 6-monthfollow-up. Profile analysis of OCD symptomratings collected throughout treatment did in-dicate a faster reduction in symptoms for pa-tients receiving individual treatment. These re-sults offer evidence for the efficacy of grouptreatment. However, because patients were ex-cluded from this study if they were diagnosedwith any personality disorder or with comor-bid depression, it may be that the sample wassomewhat atypical. In addition, none of theparticipants had received previous OCD treat-ment, which is also unusual for this populationand suggestive of a less symptomatic sample.

Thus, inferences about the broader OCD popu-lation merit caution until these results arereplicated.

More recently, Barrett, Healy-Farrell, andMarch (2004) found that individual and groupCBT were highly and similarly efficacious forchildren and adolescents with OCD relative toa wait-list control; this raises the possibilitythat group interventions might hold particularpromise in the treatment of youth with OCD.Also in youth, Asbahr and colleagues (2005)found group CBT and sertraline comparable atposttreatment but less relapse in the formercondition. More recently, another Australianresearch group found comparable outcomesfor group treatment compared to individualtreatment, both of which were superior to awaiting-list control (Anderson & Rees, 2007);not surprisingly, though, individual treatmentwas associated with more rapid response.

FAMILY INVOLVEMENT VERSUS STANDARDEX/RP TREATMENT

Emmelkamp, de Haan, and Hoogduin (1990)examined whether family involvement in treat-ment would enhance the efficacy of EX/RP forOCD. Patients who were married or living witha romantic partner were randomly assigned toreceive EX/RP either with or without partner in-volvement in treatment. Results indicated thatOCD symptoms were significantly lowered fol-lowing treatment for both groups. No differ-ences between the treatments emerged, and ini-tial marital distress did not predict outcome.However, the reduction in anxiety/distress re-ported for the sample as a whole was modest(33%), which may have resulted from the rela-tively short treatment sessions and absence of invivo exposure exercises in treatment sessions.

Mehta (1990) also examined the effect offamily involvement on EX/RP treatment out-come. To adapt the treatment to serve the largenumbers of young unmarried people seekingOCD treatment and the “joint family system”prevalent in India, Mehta used a family-basedrather than spouse-based treatment approach.Patients who did not respond to previouspharmacotherapy were randomly assigned toreceive treatment by systematic desensitizationand EX/RP, either with or without family assis-tance. Sessions in both conditions were heldtwice per week for 12 weeks; response preven-tion was described as “gradual.” In the family


condition, a designated family member (parent,spouse, or adult child) assisted with homeworkassignments, supervised relaxation therapy,participated in response prevention, and wasinstructed to be supportive. On self-reportedOCD symptoms, a greater improvement wasfound for the family-based intervention atposttreatment and 6-month follow-up. Al-though this study had methodological prob-lems that complicate interpretation of findings(e.g., use of self-report OCD measures only, un-clear description of treatment procedures), itoffers some preliminary evidence that familyinvolvement may be helpful in OCD treatment.Clinically, we routinely enlist the support offamily members in EX/RP, providing psychoed-ucation about the illness and its consequencesduring the early stages of treatment planning,and advice and encouragement in managingthe patient’s request for assurances, avoidantbehaviors, and violation of EX/RP rules be-tween sessions. We also try to reduce familymembers’ criticism of the patient and uncon-structive arguing about OCD and related mat-ters when these issues arise in the therapy.

Published randomized studies of CBT forOCD with youth have each included parents atleast to some extent in treatment (Barrett et al.,2004; de Haan, Hoogduin, Buitelaar, &Keijsers, 1998; Pediatric OCD Treatment StudyTeam, 2004), and a direct comparison of CBT,with and without a family component, using anotherwise identical protocol has yet to be con-ducted in pediatric OCD. Research on whetherfamily involvement enhances individual CBToutcomes in other anxiety disorders has gener-ally yielded mixed findings, however, and alarge, recently completed RCT indicated thatboth forms of treatment are both efficaciousand essentially equivalent to one another(Bogels & Bodden, 2005). Higher family dys-function was associated with poorer long-termoutcome in a recent study (Barrett, Farrell,Dadds, & Boulter, 2005), and at this point itmight be clinically prudent to include a morecomprehensive family component when thefamily is very directly involved in the patient’srituals (e.g., reassurance seeking) or when fam-ily psychopathology threatens generalizabilityof treatment gains to a chaotic home environ-ment. It also may be that greater family in-volvement in treatment is needed when the pa-tient is very young (Freeman et al., 2003,2007).

EX/RP VERSUS COGNITIVE THERAPIES

Increased interest in cognitive therapy (e.g.,Beck, 1976; Ellis, 1962), coupled with dissatis-faction with formulations of treatment medi-ated by processes such as extinction (Stampfl& Levis, 1967) or habituation (Watts, 1973),prompted examination of the efficacy of cogni-tive procedures for anxiety disorders in generaland for OCD in particular. A number of earlystudies found few differences between standardbehavioral treatments and behavioral treat-ments enhanced with various cognitive ap-proaches (e.g., Emmelkamp & Beens, 1991;Emmelkamp, Visser, & Hoekstra, 1988). Re-cent advances in cognitive conceptualizationsof OCD have apparently yielded more effica-cious and durable cognitive treatments.Freeston and colleagues (1997) found acognitive-behavioral intervention efficaciouscompared to a wait-list control group for pa-tients with “pure” obsessions. Several other re-cent studies (Cottraux et al., 2001; McLean etal., 2001; Vogel, Stiles, & Götestam, 2004;Whittal, Thordarson, & McLean, 2005) havesuggested equivalent results for CBT and EX/RP, respectively, although some proceduraloverlap between the two conditions in thesestudies makes their findings difficult to inter-pret. In concert with studies attesting to theutility of cognitively oriented approaches forconditions that are quite similar to OCD, suchas hypochondriasis (Barsky & Ahern, 2004;Warwick, Clark, Cobb, & Salkovskis, 1996), itdoes appear that cognitive therapies holdpromise for the treatment of OCD and mightbe an efficacious potential alternative to EX/RP.

The question of whether cognitive therapyimproves the efficacy of EX/RP is generally dif-ficult to discern, because both exposure ther-apy and cognitive therapy are intended tomodify mistaken cognitions. A randomizedcontrolled study that compared “pure” formsof CT or EX/RP, with or without medication,found similar, yet somewhat attentuated out-comes relative to what might typically be ex-pected from either treatment (van Balkom etal., 1998). Foa and Kozak (1986) argued thatthe disconfirmation of erroneous associationsand beliefs is a crucial mechanism underlyingthe efficacy of exposure treatments, hence dis-puting discussions that mistaken cognitionsfrom EX/RP might be expected to hamper out-


come. For example, a patient and therapist sit-ting on the bathroom floor in a public restroomconducting an exposure to contaminated sur-faces routinely discuss risk assessment, proba-bility overestimation, and so forth, as the ther-apist helps the patient achieve the cognitivemodification necessary for improvement. Thepractical issue of interest is how to maximizeefficacy: Is informal discussion of cognitive dis-tortions during the exposure exercises suffi-cient, or should the therapist engage in formalSocratic questioning of hypothesized distor-tions, such as inflated responsibility? Notably,in a meta-analytic review, cognitive therapiesfor OCD that included some form of exposureto feared stimuli were superior to those that didnot, suggesting that exposure may be necessaryto maximize outcomes (Abramowitz, Franklin,& Foa, 2002).

To further expand upon this point, Hiss,Foa, and Kozak (1994) investigated whetherformal relapse prevention techniques followingintensive EX/RP enhanced maintenance ofgains. Notably, all discussions about cognitivefactors typically included during the core treat-ment (e.g., discussion of lapse vs. relapse,posttreatment exposure instructions, themes ofguilt and personal responsibility, and fearedconsequences) were removed. Patients receivedthis modified EX/RP, followed by either a re-lapse prevention treatment or a psychosocialcontrol treatment (associative therapy). All pa-tients in both conditions were classified as re-sponders at posttreatment (defined as 50% orgreater reduction in OCD symptoms), withtreatment gains better maintained in the re-lapse prevention group than in the associativetherapy condition at 6-month follow-up. Thepercentages of responders at follow-up were75% in the relapse prevention condition and33% in associative therapy. The higher thanusual observed relapse rate in the associativetherapy may have resulted from the removal ofcognitive techniques typically utilized duringthe core treatment, such as discussion of fearedconsequences. These findings, and those dis-cussed earlier, further underscore our beliefthat blended treatment designed to providepatients the opportunity to disconfirm their er-roneous cognitions makes the most sense clini-cally. Accordingly, our approach clearly in-corporates informal cognitive procedures, anddiscussions of the outcome of exposures aregeared toward challenging mistaken beliefs;this is accomplished in the context of a treat-

ment approach that still emphasizes the impor-tance of EX/RP in bringing about suchchanges.

Serotonergic Medications

Effectiveness of Medications

The use of serotonergic medications in thetreatment of OCD has received a great deal ofattention in the past 25 years. Of the tricyclicantidepressants, clomipramine (CMI) has beenstudied most extensively. In controlled trials,CMI has consistently been found to be superiorto placebo (e.g., DeVeaugh-Geiss, Landau, &Katz, 1989). Similar results have been obtainedwith the selective serotonin reuptake inhibitors(SSRIs) fluoxetine, fluvoxamine, and sertraline(see Greist, Jefferson, Kobak, Katzelnick, &Serlin, 1995). Accordingly, each of these medi-cations has been approved by the U.S. Foodand Drug Administration (FDA) as treatmentsfor adult OCD. On the whole, these studiessuggest that up to 60% of patients show someresponse to treatment with SSRIs. However,even the average treatment gain achieved bytreatment responders is moderate at best(Greist, 1990). In addition, amelioration ofobsessive–compulsive symptoms is maintainedonly as long as the drug is continued: For ex-ample, in an early controlled, double-blind dis-continuation study, 90% of patients relapsedwithin a few weeks after being withdrawnfrom CMI (Pato, Zohar-Kadouch, Zohar, &Murphy, 1988). More recent discontinuationstudies with slower taper periods have notyielded such dramatic results, but they never-theless converge to suggest that maintenancetreatment is necessary to sustain achievementsattained with pharmacotherapy alone for OCD(Dougherty, Rauch, & Jenike, 2002).

EX/RP versus Pharmacotherapy

Many controlled studies have indicated that se-rotonergic antidepressants are superior to pla-cebo in ameliorating OCD symptoms (for a re-view, see Greist et al., 1995). However, only afew controlled studies have directly comparedthe relative or combined efficacy of antidepres-sant medications and EX/RP, and several stud-ies that have made such a comparison includedcomplex designs that make it difficult to drawconfident conclusions about relative and com-bined efficacy (e.g., Marks et al., 1980, 1988).


Cottraux and colleagues (1990) comparedfluvoxamine (FLV) with antiexposure instruc-tions, FLV plus weekly EX/RP, and pill placebo(PBO) plus EX/RP, and found FLV + EX/RPand FLV + antiexposure instructions superiorto PBO + EX/RP; there was a trend toward anadvantage for combined treatment, but it failedto reach significance. Hohagen and colleagues(1998) compared EX/RP + FLV to EX/RP +PBO and found that both groups improved sig-nificantly and comparably on compulsions, butthe patients who received EX/RP + FLV weresignificantly better at posttreatment on obses-sions than those who received EX/RP + PBO.Subanalyses indicated that patients with sec-ondary depression also fared better if they werereceiving EX/RP + FLV.

The relative and combined efficacy of CMIand intensive EX/RP was examined in a multi-center, RCT conducted at our center (Penn)and at Columbia University. Findings withboth treatment completer and intention-to-treat (ITT) data indicated at posttreatment thatthe active treatments were superior to placebo,EX/RP was superior to CMI, and the combina-tion of the two treatments was not superior toEX/RP alone (Foa et al., 2005); relapse wasmore evident following treatment discontinua-tion in the CMI group than in either treatmentthat included intensive EX/RP (EX/RP, EX/RP+ CMI; Simpson et al., 2004). However, the de-sign used in the Penn–Columbia study may nothave optimally promoted an additive effect forCMI, because the intensive portion of the EX/RP program was largely completed before pa-tients reached their maximum dose of CMI. Inaddition, combined treatment effects may bemore evident when intensive EX/RP is not used(Foa, Franklin, & Moser, 2002). Notably, anadditive effect for combined treatment wasfound in a recent study in pediatric OCD atPenn, Duke, and Brown (Pediatric OCD Treat-ment Study Team, 2004), although examina-tion of effect sizes by site indicated that theCBT monotherapy effect at Penn was verylarge, and no additive effect for combinedtreatment was found at this site.

In summary, although there is clear evidencethat both pharmaceutical treatment with sero-tonergic medications and EX/RP treatmentsare effective for OCD, information about theirrelative and combined efficacy remains scarce,because most of the studies that examinedthese issues have been methodologically lim-ited. Nevertheless, no study has found clear,

long-term superiority for combined pharmaco-therapy plus EX/RP over EX/RP alone. Theabsence of conclusive findings notwithstand-ing, many experts continue to advocate com-bined procedures as the treatment of choice forOCD (e.g., Greist, 1992). In clinical practice, itis common to see patients in EX/RP treatmentwho are taking SSRIs concurrently. In un-controlled examinations of EX/RP treatmentoutcome for adults (Franklin, Abramowitz,Bux, Zoellner, & Feeny, 2002) and youth(Franklin et al., 1998; Piacentini, Bergman,Jacobs, McCracken, & Kretchman, 2002)treated in OCD outpatient clinics, noposttreatment differences in OCD symptom se-verity were detected between patients who re-ceived EX/RP alone and those who receivedSSRI medication when receiving EX/RP. Fromthese data we can surmise that concomitantpharmacotherapy is not required for every pa-tient to benefit substantially from EX/RP, andthat concomitant pharmacotherapy does notappear to inhibit EX/RP treatment response.More definitive conclusions about the effects ofaugmenting pharmacotherapy with EX/RPawait a more carefully controlled examination,however.

ASSESSMENT

Following a diagnostic interview to ascertainthe presence of OCD, it is advisable to quantifythe severity of the OCD symptoms with oneor more of the instruments described below.Quantification of symptom severity assists thetherapist in evaluating how successful treat-ment was for a given patient. In our clinic, weuse several assessment instruments. As in mostOCD clinical research studies, however, theprimary measure of OCD symptom sever-ity used in our center is the Yale–BrownObsessive–Compulsive Scale (Goodman et al.,1989a, 1989b).

Yale–Brown Obsessive–Compulsive Scale

The Yale–Brown Obsessive–Compulsive Scale(Y-BOCS; Goodman et al., 1989a, 1989b), astandardized, semistructured interview, takesapproximately 30 minutes to complete. The Y-BOCS severity scale includes 10 items (five as-sess obsessions and five compulsions), each ofwhich is rated on a 5-point scale ranging from0 (No symptoms) to 4 (Severe symptoms). As-


sessors rate the time occupied by the obsessionsand compulsions, the degree of interferencewith functioning, the level of distress, attemptsto resist the symptoms, and level of controlover the symptoms. The Y-BOCS has shownadequate interrater agreement, internal consis-tency, and validity (Goodman et al., 1989a,1989b). The Y-BOCS served as the primarymeasure of outcome in most of the publishedOCD pharmacotherapy and CBT treatmentstudies conducted during the 1990s.

Self-Report Measures

Obsessive–Compulsive Inventory—Revised

The Obsessive–Compulsive Inventory—Revised(OCI-R; Foa, Huppert, et al., 2002) is an 18-item, self-report measure that assesses thedistress associated with obsessions and com-pulsions. In addition to the total score, six sep-arate subscale scores are calculated by addingthe three items that comprise each subscale:Washing, Checking, Ordering, Obsessing,Hoarding, and Neutralizing. Foa, Huppert,and colleagues (2002) reported good internalconsistency, test–retest reliability, and discrimi-nant validity in clinical patients with OCD,PTSD, generalized social phobia, and non-anxious controls. The total score ranges from 0to 72, and each subscale ranges from 0 to 12.

Other Self-Report Measures

A few self-report instruments for assessingOCD symptoms, such as the Leyton Obses-sional Inventory (Kazarian, Evans, & LeFave,1977) and the Lynfield Obsessional/Compul-sive Questionnaire (Allen & Tune, 1975), arealso available. These instruments are limitedin that they assess only certain forms ofobsessive–compulsive behavior and/or they in-clude items that are unrelated to OCD symp-toms. More recently, Storch, Murphy, and col-leagues (in press) have developed the Children’sFlorida Obsessive–Compulsive Inventory,which is intended primarily for screening pur-poses.

INITIAL INTERVIEW

After a diagnosis of OCD has been established,and before actually beginning treatment, thetherapist should schedule 4–6 hours of ap-

pointments with the patient. In these sessions,the therapist needs to accomplish three impor-tant tasks. First, the sessions are used to collectthe information necessary to develop a treat-ment plan. Specifically, the therapist must firstidentify specific cues that cause the patient dis-tress (threat cues), avoidance, rituals, andfeared consequences. Second, the therapistshould develop a good rapport with the pa-tient, because the patient will engage in expo-sure exercises designed to elicit anxiety and dis-tress during intensive EX/RP, and the lack of agood relationship between the therapist andthe patient may compromise outcome. Third,the therapist needs to explore the patient’s be-liefs about OCD and the perceived conse-quences of refraining from rituals and avoid-ance, because this information guides theinformal discussions of cognitive processes thattake place throughout EX/RP.

Threat cues may be either (1) tangible ob-jects in the environment or (2) thoughts, im-ages, or impulses that the person experiences(for lack of better terms, we have labeled them“external cues” and “internal cues,” respec-tively). Passive avoidance and ritualistic behav-ior (sometimes called “active avoidance”) bothserve to reduce the distress associated with thethreat cues. Rituals may be further divided intoovert or covert (mental) forms. It is essentialthat patients understand the difference betweenobsessions and mental compulsions, becauseobsessions are treated with systematic expo-sure and mental compulsions, with ritual pre-ventions. During treatment, patients should beinstructed to report any mental compulsions tothe therapist, because performing such compul-sions during exposure exercises attenuates theeffects of these exercises in the same way thatbehavioral compulsions do.

External Fear Cues

Most individuals with OCD experience fear inreaction to specific environmental cues (objects,persons, or situations), but each patient willhave his or her own idiosyncratic threat cues.For example, individuals who fear contamina-tion from toilets may differ as to whether all toi-lets are feared or only those open to the public.One patient may fear only the toilet itself,whereas another patient also fears bathroomfloors, doorknobs, and faucets. Similarly, twoindividuals may experience distress at the pros-pect of a fire burning down their home, but


whereas one experiences the distress only whenshe is the last person to leave the house, theother experiences distress before going to bed atnight when his children are present.

The therapist needs to gather specific infor-mation about cues that elicit the patient’s dis-tress to identify the basic sources of the fear.Identification of the basic source is importantfor planning the treatment program. Con-fronting the source of the fear is essential forsuccessful behavioral treatment of OCD. Of-ten, when such exposure does not take placeduring treatment, relapse will occur. For exam-ple, a patient who feared contamination by herhometown was treated with EX/RP 3,000miles away from the town. Because of the dis-tances involved, direct exposure to the townwas impossible, so treatment comprised expo-sure to objects contaminated directly or indi-rectly by contact with the town. Although thepatient habituated to the objects used in the ex-posure sessions, she continued to fear herhometown. Within 1 year after treatment, shehad developed fears to new objects related toher hometown. Not until she engaged in re-peated exposures to the town itself did she ex-perience lasting improvement.

It is important that the therapist conduct athorough investigation of objects, situations,and places that evoke obsessional distress forthe patient at the time of presentation and atonset. Such information helps to identify thesource of the distress. To facilitate communica-tion with the patient about situations thatevoke distress, a Subjective Units of DiscomfortScale (SUDS) ranging from 0 to 100 is intro-duced. Patients are asked to rate each situationwith respect to the level of distress that they ex-pect to experience upon exposure. The sourceof the distress is expected to be 100. The fol-lowing dialogue between therapist and patientillustrates the process of gathering informationabout distressing situations.

THERAPIST: When do you get the urge to washyour hands?

PATIENT: In a lot of places. There are so manyplaces.

THERAPIST: Are there any places where theurges are particularly strong?

PATIENT: Well when I am sitting in my livingroom, particularly near the fireplace. Also inthe laundry room, which I never go to. Also,when I walk in the park.

THERAPIST: Let’s talk about your living room.How upset are you when you are sitting nextto your fireplace?

PATIENT: That’s bad. I guess about a 90.

THERAPIST: Can you tell me what makes you soupset in your living room?

PATIENT: Well that is a long story . . . and Iknow it doesn’t make sense.

THERAPIST: Go on. It’s important that we un-derstand what makes you uncomfortableand fearful in your living room.

PATIENT: About 2 years ago, I got up in themorning and went into the living room, andI saw a dead squirrel in the fireplace. I guesshe got in through the chimney. So, I figuredthat if the squirrel was dead, he must havebeen sick. I know that a lot of squirrels haverabies, so I thought that if the squirrel diedof rabies, then there are germs all over thechimney.

THERAPIST: Have you tried to have the chimneyand the fireplace cleaned?

PATIENT: Yes, we did have a company come inand clean the whole area, but I’m not surethat they can clean away the germs.

THERAPIST: I understand. How about the laun-dry room, how upsetting is it to be in thelaundry room?

PATIENT: That would be a 100; that’s why Idon’t go in there.

THERAPIST: How did the laundry room becomedangerous?

PATIENT: Oh, that’s another story. Until a yearago, my children used to keep their guineapigs in the laundry room. One day we foundthe female guinea pig dead. So I thought thatit probably died of rabies too.

THERAPIST: Oh, I understand. So you are gener-ally afraid that you will contract rabies ifyou come in contact with things that youthink are contaminated with rabies germs. Isthis true?

PATIENT: Exactly. That’s why I don’t like towalk in the woods or the park. You knowthose places have all kind of animals, andyou can never tell where the germs might be.

It is clear from this conversation that it wasnot living rooms, laundry rooms, or parks perse that the patient feared. Rather, any situationor object that, in her mind, had some prob-


ability of being infested with rabies germsbecame a source of contamination. Somecontamination-fearful patients, however, can-not specify feared consequences of coming intocontact with stimuli they perceive to be con-taminated. For these patients, the primary fearis that they will not be able to tolerate the ex-treme emotional distress generated by beingcontaminated. With such patients, it is also im-portant to probe further to discern whether thepatient has fears about the long-term healthconsequences of experiencing high and unre-mitting anxiety in response to stimuli thatprompt obsessions.

Internal Fear Cues

Anxiety and distress may also be generated byimages, impulses, or abstract thoughts that theindividual finds disturbing, shameful, or dis-gusting. Examples of such cues include im-pulses to stab one’s child, thoughts of one’sspouse injured in an accident, or images of reli-gious figures engaged in sexual activity. Clearly,internal threat cues may be produced by exter-nal situations, such as the sight of a knife trig-gering the impulse to stab one’s child. Somepatients may become distressed when they ex-perience certain bodily sensations, such as mi-nor pains triggering the fear of having cancer.

In many cases, patients may be reluctant toexpress their obsessive thoughts, because theyare either ashamed of them or fear that ex-pressing them will make the consequence morelikely to occur. In these cases, the therapistneeds to encourage the expression of thesethoughts through direct questioning and amatter-of-fact attitude. Sometimes it helps totell the patient that many people with andwithout OCD have unwanted thoughts (asmany as 85% of normal individuals)(Rachman & DeSilva, 1978). It may also behelpful to remind the patient that talking aboutthe obsessions will be a part of therapy; theevaluation session provides an opportunity tobegin this process.

THERAPIST: So tell me, when is it that you feelthe urge to count?

PATIENT: It seems like I’m always countingsomething, but it’s mostly when I thinkabout certain things.

THERAPIST: What kind of things?

PATIENT: I don’t know. Bad things.

THERAPIST: Can you give me some examples ofbad thoughts that will make you want tocount?

PATIENT: (brief silence) I really prefer not totalk about them. It makes things worse.

THERAPIST: You mean it makes the countingworse?

PATIENT: Yes.

THERAPIST: All right, I know now that whenyou think or talk about certain bad things,you have an urge to count, but I still don’tknow what those bad things are. How aboutyou tell me so that I can help you with them?

PATIENT: I’d really rather not. Can’t we talkabout something else?

THERAPIST: It is important that I know whatthe thoughts are to plan your treatment. I’lltry to help you. Do the thoughts involvesomeone being hurt?

PATIENT: Yes.

THERAPIST: Do the thoughts involve only cer-tain people getting hurt or could it be any-one?

PATIENT: Mostly my family.

THERAPIST: OK, what else can you tell meabout the thoughts?

PATIENT: I really don’t want to say any more.

THERAPIST: I know this is scary, but rememberthat facing your fears is what this treatmentis all about.

PATIENT: OK. It’s not always thoughts. Some-times I see pictures in my mind, where mybrother or my mom and dad are killed. I’mafraid when I talk about these thoughts andpictures that they really will die.

THERAPIST: A lot of people have thoughts thatthey don’t like to have. Even people withoutOCD. Just because you have these thoughts,or talk about them, doesn’t mean that badthings will actually happen or that you wantthem to come true.

It is important to reassure the patient thatunpleasant thoughts occur often and to empha-size the distinction between thoughts and real-ity. Many patients with OCD have magicalideas in which the distinction between “think-ing about” and “making things happen” isblurred, a process labeled by Salkovskis (1985)as “thought–action fusion” (TAF). It is impor-tant to point out to the patient that thoughts


are different from actions. Also, many patientsthink that if negative thoughts enter their mind,then it means they wish the bad thing will hap-pen. The therapist should assure the patientthat thinking about bad things does not meanthat one wants them to happen. These sorts ofinformal discussions of mistaken beliefs are anintegral part of correct implementation of EX/RP. Such discussions should accompany thetreatment planning process, and should be reit-erated as needed during exposure exercises. Itis, however, important that such discussions ac-company EX/RP exercises rather than replacethem.

Feared Consequences

Many individuals with OCD are afraid thatsomething terrible will happen if they fail toperform their rituals. Such patients withwashing rituals, for example, typically fearthat they and/or someone else will become illor disabled, or die as a result of being con-taminated. Many patients with checking ritu-als fear that, because of their negligence, cer-tain catastrophes will occur, such as theirhomes burning down or that they might killsomeone while driving. Some patients haveonly a vague notion of what these negativeconsequence might be (e.g., “I don’t knowexactly what will happen, but I feel that if Idon’t count to seven, something bad will hap-pen to my family”). Others do not fear catas-trophes at all, but they cannot tolerate theemotional distress they experience if they donot perform rituals. Some fear that unlessthey ritualize, anxiety will increase continu-ally, until they have a nervous breakdown.Data from the DSM-IV field trial indicatedthat approximately two-thirds of patientswith OCD could clearly identify conse-quences other than emotional distress thatwould follow from refraining from rituals,whereas the remainder could report no suchconsequences (Foa et al., 1995).

It is important to identify the specific detailsof the patient’s feared consequences to plan aneffective exposure program. For example, thecontent of the imaginal exposure of a patientwho checks while driving for fear of having hita pedestrian and being sent to jail differs fromthat of a patient who fears that hitting a pedes-trian will result in punishment by God. Simi-larly, patients who ritualistically place objectsin a specific order may differ with respect to

their feared catastrophes. Some perform the rit-ual to prevent catastrophic consequences (e.g.,death of parents), whereas others do so only toreduce distress elicited by disordered objects.The former would benefit from treatment thatincludes both imaginal and in vivo exposure,whereas the latter is likely to profit from invivo exposure alone.

Strength of Belief

Clinical observations have led to suggestionsthat individuals with OCD who have poor in-sight do not respond well to exposure and re-sponse prevention, although two later studiesfailed to find a linear relationship betweenstrength of belief in feared catastrophes andimprovement following exposure and responseprevention (Foa et al., 1999; Lelliott,Noshirvani, Basoglu, Marks, & Monteiro,1988). Two issues need to be considered inevaluating these collective findings. First, thereliability and validity of the strength of beliefmeasures used in previous studies are un-known. Second, the relationship between over-valued ideation and treatment outcome maynot be linear. Clinical observation suggests thatonly patients who express extreme belief intheir obsessional ideation show poor outcome.Indeed, Foa and colleagues (1999) found thatonly extremely strong belief (fixed belief) wasassociated with attenuated outcome. Such pa-tients may appear delusional when discussingtheir feared catastrophes. We hypothesize thatthe effect of fixed belief on outcome may bemediated by treatment compliance: Patientswho are convinced that feared disasters will en-sue if they engaged in prescribed exercisesprobably will not complete the tasks as as-signed.

When assessing the strength of belief, it isimportant to remember that a patient’s insightinto the senselessness of his or her belief oftenfluctuates. Some patients readily acknowledgethat their obsessional beliefs are irrational, butthey still cause marked distress. A few individu-als firmly believe that their obsessions andcompulsions are rational. In most patients,though, the strength of belief fluctuates acrosssituations, making it difficult to ascertain thedegree to which they believe the obsessions areirrational. The following example is an inquiryinto the strength of a patient’s belief in her ob-sessional fear of contracting acquired immunedeficiency syndrome (AIDS).


THERAPIST: How likely is it that you will con-tract AIDS from using a public restroom?

PATIENT: I’m really terrified that I will get AIDSif I use a bathroom in a restaurant.

THERAPIST: I know that you are afraid of get-ting AIDS, but if you think logically, howlikely do you think you are to get AIDS bysitting on a public toilet?

PATIENT: I think I will get AIDS if I use a publictoilet.

THERAPIST: So do you mean to say that there isa 100% chance of you getting AIDS if you siton a public toilet once?

PATIENT: Well, I don’t know about once, but if Idid it again and again I would.

THERAPIST: What about other people? Will theyget AIDS if they use a public toilet?

PATIENT: I guess so. I’m not sure.

THERAPIST: Since most people use public bath-rooms, almost everyone should have AIDSby now. How do you explain the fact thata relatively small number of people haveAIDS?

PATIENT: Maybe not everybody is as susceptibleto AIDS as I am.

THERAPIST: Do you think that you are moresusceptible than other people?

PATIENT: I don’t know for sure. Maybe the like-lihood of my getting AIDS is only 50%.

Based on the interaction just described, thetherapist concluded that the patient was not an“overvalued ideator”; thus, the prognosis forthis patient is brighter than it would be if shecontinued strongly to hold her original belief.Accordingly, the implementation of EX/RP forthis patient would follow the standard guide-lines.

Avoidance and Rituals

To maximize treatment efficacy, all avoidanceand ritualistic behaviors, even seemingly minorones, should be prevented. Therefore, the ther-apist should gather complete informationabout all passive avoidance and rituals. Whenthe therapist is in doubt as to whether a partic-ular avoidance behavior is related to OCD, heor she might suggest an “experiment” in whichthe patient is exposed to the avoided situation.If the patient experiences anxiety or distress,the avoidance behavior should be prevented as

part of treatment. Similarly, if it is unclearwhether a given action constitutes a ritual, a re-sponse prevention “experiment” may be imple-mented. If refraining from performing the ac-tion evokes distress, the action is identified as aritual and should be addressed in therapy.

Individuals with OCD, like those with spe-cific phobias, often attempt to avoid anxiety-evoking situations. Most passive avoidancestrategies are fairly obvious (e.g., not enteringpublic rest rooms, not preparing meals, and nottaking out the trash). However, the therapistalso needs to be attentive to subtle forms ofavoidance, such as carrying money in one’spockets to avoid opening a wallet, wearingslip-on shoes to avoid touching laces, and usingdrinking straws to avoid contact with a glassor a can. Patients with obsessive–compulsivechecking rituals also engage in subtle avoid-ance behaviors that are important to explore,such as arranging their work schedules to en-sure that they are rarely, if ever, the last personto leave the business, thus ensuring that the re-sponsibility for checking the safe falls on acoworker.

Active rituals, such as passive avoidance,may be explicit (e.g., prolonged washing, re-peated checks of the door, and ordering of ob-jects) and/or subtle (e.g., wiping hands on pantlegs, blinking, and thinking “good” thoughts).It is important that the therapist identify bothexplicit and subtle rituals, so they both may beaddressed during treatment.

Although compulsive rituals are intended toreduce the distress associated with obsessions,patients sometimes report that the performanceof these rituals is aversive in itself. For exam-ple, Ms. S, who was obsessed with the orderli-ness of objects on her shelves, found reorderingthe shelves aversive, because she was unable tofind the “perfect” place for everything. Simi-larly, Mr. J, who felt contaminated by chemi-cals, found the act of decontaminating himselfby repeated handwashing aversive, because hewas unable to decide when his hands were suf-ficiently clean; therefore, he washed until hishands became raw. Rituals may also becomeaversive because of their intrusion into otheraspects of the person’s life. For example, Mr. J,who would take 2-hour-long showers to feeladequately clean, was reprimanded repeatedlyby his supervisor for arriving late to work.

When certain compulsions become aversive,some patients decrease the time they spend per-forming the ritual by increasing avoidance be-


haviors, or by substituting other, less time-consuming rituals. For example, Ms. E, whowas obsessed with fears of contamination byfuneral-related objects (e.g., cemeteries andpeople returning from a funeral), respondedwith hours of showering and handwashing. Sheeventually retreated into her bedroom andavoided all contact with the outside world. Mr.J, described earlier, avoided taking a shower fordays at a time, but between showers he wipedhis hands compulsively and avoided touchinghis wife. In some cases, seemingly “new” ritu-als may develop during the course of treatmentto function in the place of those previouslyidentified and eliminated. For example, Mr. F,who was concerned about his hands becomingcontaminated, successfully resisted the urge towash his hands, but soon after response pre-vention was implemented, he started to rub hishands together vigorously to “decontaminate”them. When such a substitute ritual is identi-fied, it also needs to be addressed in treatmentwith ritual prevention. Therapists must notonly remain alert to such shifts in ritualistic be-haviors but also alert patients to the possibilityof such shifts.

History of Main Complaintand Treatment History

Many individuals with OCD are unable to givea detailed account of the onset of their symp-toms, because the symptoms began subtly,many years ago. Nevertheless, therapistsshould attempt to collect as much informationas possible about the onset and course of thedisorder. Such information may provide cluesabout aspects of the fear network and variablesassociated with the maintenance of symptoms,and may help to anticipate difficulties that mayarise during treatment (e.g., old obsessions orrituals that may resurface as more prominentones diminish).

Many such individuals also have an exten-sive history of psychological and pharmacolog-ical treatments, and it is important to make adetailed inquiry about the outcome of previoustreatments. If the patient has been treated withEX/RP, the therapist should assess whether thetreatment was implemented appropriately andthe patient was compliant with treatment de-mands. Knowledge that a patient experienceddifficulty complying with response preventioninstructions, or that previous therapy failed toprovide adequate exposure experiences or re-

sponse prevention instructions, is importantfor designing the behavioral program. Otherfactors that may have prevented successful out-come or caused relapse, such as job stress,death in the family, or pregnancy, should bediscussed. At the same time, a prior failedcourse of EX/RP should not necessarily beviewed as prognostic, especially if the patientrecognizes why the therapy was less successfulin the past. One of our patients, who had failedmultiple trials of less intensive EX/RP, came toour center with the knowledge that his non-compliance with exposure exercises betweenweekly sessions greatly reduced the effects oftreatment. He also noted that the slow progresshe observed in these previous therapies demor-alized him and caused further disengagementfrom the treatment. When offered a choice ofdaily versus twice-weekly sessions, he opted forthe daily treatment, noting that the more inten-sive approach might decrease the chance ofsimilar lapses. He has now successfully com-pleted the intensive regimen.

In our clinic we have observed that a sub-stantial majority of our outpatients have beentreated, or are currently being treated, with se-rotonergic medications. Some seek EX/RP toaugment the partial gains they have achievedwith the medication. Others wish to discon-tinue the medication because it was ineffective,because of side effects, or because they do notwant to continue taking medicine indefinitely.Assessment of the patient’s treatment goals isnecessary for planning his or her treatmentprogram.

Social Functioning

Obsessive–compulsive symptoms may severelydisrupt the daily functioning of patients. Ther-apists should assess the impact of OCD symp-toms on the various areas of functioning.Where appropriate, this information should beused to design suitable exposure exercises. Forexample, Ms. D experienced difficulties com-pleting assignments at work because she re-peatedly checked each task. Treatment in-cluded exposures to performing tasks at workwithout checking. Even if the client is not cur-rently working, exposures simulating work sit-uations may be necessary if symptoms createddifficulties in previous jobs.

OCD clearly has a deleterious effect on theintimate relationships of many patients. Abouthalf of married individuals seeking treat-


ment for OCD experience marital distress(Emmelkamp et al., 1990; Riggs et al., 1992).Other family and social relationships may alsosuffer as a result of OCD symptoms. The im-pairment in social functioning may arise eitherbecause social contact is perceived as threaten-ing (e.g., “I may spread germs to other peo-ple”) or because so much of the patient’s timeand energy is invested in performing rituals andplanning ways to avoid distressing situations.Again, information about the relation of socialdysfunction to OCD symptoms may lead thetherapist to include specific exposures aimed atameliorating these social difficulties.

The assessment of social functioning shouldalso include an evaluation of what role, if any,other people play in the patient’s compulsiverituals. If the patient relies on others for reas-surance or compliance with rituals (e.g., familymembers must remove their shoes before enter-ing the house), the therapist should instructfamily members how to respond appropriatelywhen asked to participate in the patient’s ritu-als. A careful analysis of the relationship iscalled for before specific instructions are givento significant others. Moreover, if family mem-bers tend to criticize the patient when obses-sional distress arises, it is important to addressthese negative exchanges in treatment. We haveoften addressed this issue with a combinationof empathic discussion of the frustration expe-rienced by the family member and role playingof more effective responses.

Mood State

Although some patients with serious depres-sion and OCD may benefit from behavioraltherapy for OCD (Foa et al., 1992), researchsuggests that severe depression may limit theextent to which the OCD symptoms are re-duced and the maintenance of those gains (e.g.,Abramowitz et al., 2000). Therefore, it is im-portant to assess the mood state of the patientprior to beginning behavioral therapy. Patientswith severe depression should be treated withantidepressant medication or cognitive therapyto reduce the depressive symptoms prior to im-plementing behavioral therapy for the OCD.Treatment with serotonergic antidepressantsmay reduce OCD symptoms, as well as depres-sion. Because the effects of such medication onOCD symptoms may not be evident until 3months after treatment begins, the therapistneeds to use his or her clinical judgment to de-

cide whether to begin EX/RP when the depres-sion decreases or to wait until the effects of themedication on OCD symptoms can be as-sessed.

Choice of Treatment

How should a therapist determine the mostsuitable treatment for a given patient? As dis-cussed earlier, exposure and response preven-tion, as well as serotonergic medications, havedemonstrated efficacy for OCD. The therapistand patient are faced with the choice of EX/RP,pharmacotherapy, or a combination of the two.Neither treatment is effective with all patients,and no consistent predictors of who will bene-fit most from which treatment modality havebeen identified. Therefore, unless the patienthas been particularly successful or unsuccessfulwith some previous course of treatment, the de-cision should be based on factors such as avail-ability of treatment, amount of time the patientis able or willing to invest in treatment, and hisor her motivation and willingness to tolerateside effects.

The intensive treatment requires a consider-able investment of time over a period of severalweeks. Many patients are unable, or unwilling,to devote 4 to 5 hours a day to treatment.These patients should be advised to try phar-macological treatment, which does not requirethe same extensive time commitment. Recentinvestigations of the effects of a twice-weeklyEX/RP regimen compared to intensive treat-ment suggested comparable outcomes atfollow-up (Abramowitz et al., 2003; Storch,Geffken, et al., 2007); thus, in our center, weroutinely offer either program to patients con-sidering EX/RP. Some patients may be unwill-ing (sometimes expressed as “I can’t do that”)to experience the temporary discomfort causedby EX/RP. These patients, too, may be advisedto try medications. The need to develop “readi-ness programs” designed to prepare such pa-tients to accept EX/RP treatment is often citedin light of the relatively high refusal rate amongpatients offered EX/RP. Such programs may in-clude testimonials from previously treated pa-tients, cognitive strategies designed to help thepatient calculate objective risks more accu-rately, psychoeducation about OCD and EX/RP, and a review of the outcome literature forvarious treatments (Tolin, Maltby, Diefenbach,Hannan, & Worhunsky, 2004). Manualizingsuch programs and investigating their effective-


ness in increasing the acceptance rate and theefficacy of EX/RP for patients who enter suchprograms await controlled studies.

Patients who are concerned about the poten-tial (or have already experienced) side effects ofmedications or their unknown long-term effectsoften prefer EX/RP. Other patients are con-cerned with the prospect of entering an “end-less” treatment because, according to presentknowledge, relapse occurs when medication iswithdrawn (Pato et al., 1988; Thoren, Asberg,Chronholm, Journestedt, & Traskman, 1980).This concern is particularly relevant for womenwho plan to bear children and would need towithdraw from the medication during preg-nancy. EX/RP should be recommended to thesepatients because its effects are more enduring.

As discussed earlier, the long-term effects ofcombining EX/RP and medication are unclear;therefore, it is premature to recommend treat-ment programs that combine the two therapies.However, some patients who present for treat-ment are already on antidepressant medication.Because these medications were found not tointerfere with the effectiveness of EX/RP(Franklin et al., 2000), it is recommended thatpatients continue to take the medication if theyhave experienced some improvement in eitherobsessive–compulsive symptoms or depression.However, if the patient has not experienced im-provement with medication, withdrawal of themedication before or during EX/RP should beconsidered. Special consideration should begiven to patients with severe depression con-current with their OCD. It is recommendedthat these patients be treated with antidepres-sants or with cognitive therapy for the depres-sion prior to entering intensive EX/RP for theOCD, given recent findings of somewhat atten-uated outcome for severely depressed patients(Abramowitz et al., 2000).

INTENSIVE EX/RP

The intensive treatment program comprisesfour phases: (1) information gathering, (2) in-tensive EX/RP, (3) a home visit, and (4) a main-tenance and relapse prevention phase.

Information Gatheringand Treatment Planning

The first stage of information gathering con-sists of a thorough diagnostic evaluation to

determine that the patient’s main psychopath-ology is OCD. The second step is to assesswhether the patient is appropriate for EX/RP.We recommend that individuals who areabusing drugs or alcohol should be treatedfor the substance abuse prior to intensivetreatment for OCD. Patients who have cleardelusions and hallucinations are also poorcandidates for intensive treatment. Individualswith severe MDD should be treated for de-pression before beginning treatment forOCD. The patient’s motivation to complywith the demands of intensive treatmentshould be carefully evaluated. It is importantto describe the treatment program in enoughdetail that the patient is not surprised whentreatment begins. If the patient does not ex-press strong motivation and commitment totreatment, it might be preferable to delay im-plementation of intensive treatment or to of-fer alternative treatments, such as medication.As noted earlier, a study of less intensive EX/RP for patients who appear otherwise moti-vated, yet cannot accommodate the daily reg-imen into their schedules, suggested compara-ble outcome compared to intensive treatment;future research with much larger samples isneeded to determine whether patient factorspredict differential outcome to either treat-ment schedule.

Once a patient is judged to be appropriatefor intensive treatment, information gatheringfor treatment planning begins. This phase typi-cally comprises 4–6 hours of contact with thepatient over a period of 2–3 days. During thisphase, the therapist collects information aboutthe patient’s obsessive–compulsive symptoms,general history, and the history of treatment forOCD, as described earlier. During these ses-sions, the therapist discusses the rationale fortreatment, describes the program in detail,teaches patients to monitor their rituals, anddevelops a treatment plan.

First Information-Gathering Session

It is very important to discuss the rationale fortreatment and to describe the treatment pro-gram in detail. The program requires that thepatient abandon his or her obsessive–compulsive habits, therefore temporarily expe-riencing substantial discomfort. If patients donot understand why they are asked to sufferthis short-term distress or are not convincedthat treatment will work, they are unlikely to


comply with treatment instructions. The treat-ment rationale is explained as follows:

“You have a set of habits that, as you know,are called obsessive–compulsive symptoms.These are habits of thinking, feeling, and act-ing that are extremely unpleasant, wasteful,and difficult to get rid of on your own.Usually, these habits involve thoughts, im-ages, or impulses that habitually come toyour mind, even though you don’t wantthem. Along with these thoughts you haveunwanted feelings of extreme distress oranxiety and strong urges to do something toreduce the distress. To try to get rid of theanxiety, people get into the habit of engagingin various special thoughts or actions, whichwe call ‘rituals.’

“Unfortunately, as you know, the ritualsdo not work all that well, and the distress de-creases for a short time only, then comesback again. Eventually, you may find your-self doing more and more ritualizing to try toreduce anxiety, but even then the relief istemporary and you have to do the ritual allover again. Gradually, you find your-self spending so much time and energyritualizing—which does not work that wellanyway—that other areas of your life are se-riously disrupted.

“The treatment we are about to begin iscalled exposure and response prevention. Itis designed to break two types of associa-tions. The first association is between sensa-tions of anxiety and the objects, situations,or thoughts that produce this distress. [Thetherapist uses information collected as ex-amples, e.g., ‘Every time you touch anythingassociated with urine you feel anxious, dis-tressed, or contaminated.’] The second asso-ciation we want to break is that between car-rying out ritualistic behavior and the feelingof less anxiety or less distress. In otherwords, after you carry out [specifies theidentified rituals], you temporarily feel lessdistress. Therefore, you continue to engagein this behavior frequently. The treatment weoffer breaks the automatic bond between thefeelings of discomfort/anxiety/contamina-tion of [specifies the obsession] and your rit-uals. It will also train you not to ritualizewhen you are anxious.”

After presenting the treatment rationale, thetherapist should begin to collect information

about the patient’s OCD symptoms. The ratio-nale for information gathering and adescription of the treatment is presented as fol-lows:

“In the next two sessions, I will ask you spe-cific questions about the various situationsand thoughts that generate discomfort oranxiety in you. We will order them accordingto the degree of distress they generate in youon a scale from 0 to 100, where 0 means Noanxiety and 100 means Maximum anxiety orpanic. The exposure treatment program in-volves confronting you with situations andthoughts that you avoid because they gener-ate anxiety and urges to carry out ritualisticbehavior. Why do we want to expose you toplaces and objects that will make you un-comfortable, situations that you have at-tempted to avoid even at much cost? Weknow that when people are exposed to situa-tions that they fear, anxiety gradually de-clines. Through exposure, then, the asso-ciation between anxiety and [specifies theobsession] weaken, because you are repeat-edly exposed to these situations, so that thepreviously evoked anxiety decreases withtime.

“For many people with OCD the obses-sions occur within their imagination andrarely take place in reality. This makes it im-possible to practice exposure by actuallyconfronting those situations for prolongedperiods. For example, if a person fears thather home will burn down, we certainly donot wish to have her house catch on fire inorder to practice exposure. Similarly, some-one who fears that he has run over a personwho is now lying in the road cannot in real-ity be exposed to such a situation.

“If confrontation with the feared situationis necessary to reduce obsessions, how canyou improve without directly confrontingthe situation? You can confront these fearsthrough imagery, in which you visualize thecircumstances that you fear will happen. Inimagery practice, you create in your minddetailed pictures of the terrible consequencesthat you are afraid will occur if you do notengage in the ritualistic behavior. Duringprolonged exposure to these images, the dis-tress level associated with them gradually de-creases.

“When people with OCD encounter theirfeared situations or their obsessional


thoughts, they become anxious or distressedand feel compelled to perform the ritualisticbehavior as a way to reduce their distress.Exposure practices can cause this same dis-tress and urge to ritualize. Usually, perform-ing rituals strengthens the pattern of distressand rituals. Therefore in treatment, ritualprevention is practiced to break the habitof ritualizing. This requires that you stopritualizing, even though you are still havingurges to do so. By facing your fears withoutresorting to compulsions, you gradually be-come less anxious. Behavior therapists callthis process ‘habituation.’ Therefore, duringthe 3 weeks of intensive exposure, the associ-ation between relief from anxiety and carry-ing out [specifies the patient’s rituals] willbecome weaker, because you will not be al-lowed to engage in such behaviors; there-fore, you will find out that your anxiety de-creases even if you do not resort to theseactivities.”

The initial information-gathering session isalso used to begin training the patient to moni-tor his or her rituals accurately. Accurate re-ports of the frequency and duration of ritualis-tic behavior are important to evaluate theprogress of treatment and to demonstrate thereality of changes to the patient. In some cases,the monitoring also serves an active role intreatment. Patients begin to recognize that ritu-als do not truly occur “all day long” and theact of monitoring the rituals may decrease theirfrequency and duration.

“It is very important for the treatment pro-gram that we have an accurate picture of theextent to which you engage in obsessivethinking and compulsive behavior. Having aclear picture of how much of your time istaken up by your problem will help us tomonitor your progress and adjust the treat-ment program accordingly. Therefore, dur-ing this week while I am still collecting infor-mation to form a treatment program, Iwould like you to record your symptoms ev-ery day. It is not easy to report accurately onhow much you engage in your obsessive–compulsive behavior; therefore, we willspend some time now and in the next sessiongoing over some rules for how to recordyour symptoms. Here are some monitoringforms on which you will record yourthoughts and rituals.”

The therapist should specify which ritual(s)the patient is to record, go over the instructionscarefully with the patient, and practice fillingout the form with the patient using an “imagi-nary day” of his or her life. The following rulesare helpful in monitoring rituals:

1. Use your watch to monitor the time youspend on your rituals.

2. Do not guess the time of ritualizing; be ex-act.

3. Write the time immediately on your moni-toring form.

4. Do not save the recording to the end of theday or the beginning of the next day.

5. Write a short sentence to describe the trig-ger for ritualizing.

Prior to beginning treatment, the patientidentifies an individual (e.g., parent, spouse, orclose friend) who can serve as a support personduring the intensive treatment program. Thepatient is instructed to rely on this person forsupport during exposures, and the support per-son is asked to help monitor compliance withresponse prevention instructions. If the patientexperiences difficulty resisting the urge to ritu-alize, then the support person is contacted forsupport. Because the support person is in-volved in the therapy, the therapist allocatestime during the information-gathering phase todescribe the treatment and discuss its rationalewith him or her.

The therapist makes an effort to ensure thatthe support person and the patient mutuallyagree that the support person will offer con-structive criticism and observations. In makingthese suggestions, the support person should besensitive to any difficulties that have arisen inthe past. For example, Mr. B, who served as hiswife’s primary source of reassurance, also criti-cized her severely when he “caught” her per-forming her handwashing ritual. To preventthese responses from hampering treatment, tohelp the husband supervise his wife’s responseprevention, the therapist spent time with thecouple negotiating appropriate, uncritical re-sponses to the wife’s requests for reassurance.

The support person is in regular (at leasttwice weekly) contact with the therapist and isnot only informed about the specific home-work exposures that the patient has to accom-plish but also relays his or her observationsabout the patient’s behavior outside the ther-apy session. In addition, with the consent of the


patient, the support person should contact thetherapist if major treatment violations occur(e.g., refusing to do homework or engaging inritualistic behavior).

Second Information-Gathering Session

At the beginning of the second information-gathering session, the therapist devotes time tothe patient’s self-monitoring form, which in-cludes examining the descriptions of situationsthat trigger ritualistic behavior and offeringconstructive comments when necessary. Thetherapist reminds the patient to use shortphrases or sentences to describe the trigger situ-ations, assesses the accuracy of the patient’stime estimates, and emphasizes the need for ac-curate measurements.

Generating a Treatment Plan

The bulk of the second information-gatheringsession is allotted to gathering detailed infor-mation about the patient’s symptoms and,based on what is learned about the symptoms,developing a treatment with the patient. It isimportant to explain to the patient how the ex-posure exercises that comprise his or her treat-ment will reduce the OCD symptoms. For, ex-ample, the patient with religious obsessions istold that the imaginal exposure to burning inhell in excruciating detail is designed to reducehis obsessional distress when a less elaborateimage of burning in hell comes into his mind. Itis important that patients understand the ratio-nale underlying the central concept in EX/RP,that confronting obsession-evoking stimuliduring treatment increases their suffering in theshort run but will reduce it in the long run. Weoften tell patients that the difficulties they ex-perience during the first week of exposure ses-sions are likely to diminish with proper imple-mentation of EX/RP.

Describing Homework

At the end of the second information-gatheringsession, the therapist describes the homeworkassignments included in the treatment pro-gram. The homework, which usually requires2–3 hours, in addition to the 2-hour treatmentsession, consists of additional exposure exer-cises to be done between treatment sessions atthe patient’s home or elsewhere (e.g., a shop-ping mall or a relative’s home). We suggest that

the patient monitor his or her SUDS level every10 minutes during the homework exposures. Insome cases, when it is impossible for the pa-tient to maintain an exposure for 45–60 min-utes, the therapist works with the patient to de-velop a plan that allows the exposure to beprolonged. For example, instead of asking thepatient to spend 45 minutes sitting in the rest-room of a local restaurant, the therapist mightsuggest that he or she contaminate a handker-chief on the toilet seat and carry this “contami-nation rag” in a pocket.

Treatment Period

The treatment program at our center typicallycomprises fifteen 2-hour treatment sessionsconducted daily for 3 weeks. Clinical observa-tion suggests that massed sessions producebetter results than do sessions spread out overtime; therefore, we recommend a minimum ofthree sessions per week. Each session beginswith a 10- to 15-minute discussion of home-work assignments and the previous day’s ritualmonitoring. The next 90 minutes are dividedinto 45 minutes each of imaginal and in vivoexposure. The final 15 minutes are spent dis-cussing the homework assignment for the fol-lowing day. This format may be adjusted whennecessary. For example, if an in vivo exposurerequires that the therapist and patient travel toa local shopping mall to contaminate children’sclothing, the entire session is devoted to this ac-tivity. Some patients have difficulty engagingemotionally in imaginal exposures (i.e., the im-ages fail to elicit distress). In these cases, treat-ment should focus exclusively on in vivo exer-cises.

We recommend that the therapist discuss theplan for that session with the patient in the be-ginning of the session. Barring any unusual cir-cumstances (e.g., patient’s stated objection toproceeding with the planned exposure), it isimportant to limit these discussions to no morethan 15 minutes. Patients with OCD are usu-ally very fearful of engaging in exposure tasks,and elaborate discussion of the task at handmay serve as a form of avoidance of goingahead with the exposure. These preexposurediscussions are also fertile ground for assur-ance seeking (i.e., the patient asking the thera-pist if he or she is certain that the proposed ex-ercise is safe). The therapist should answersuch questions carefully, avoiding either ex-treme (i.e., neither providing compulsive reas-


surance nor of conveying to the patient that theproposed exposure is objectively dangerous).

Imaginal exposure exercises are typicallyconducted prior to in vivo exercises in each ses-sion, often as a prelude to the scheduled in vivoexercise. During imaginal exposure the patientis seated in a comfortable chair and is given thefollowing instructions:

“Today you will be imagining [describesscene]. I’ll ask you to close your eyes so thatyou won’t be distracted. Please try to picturethis scene as fully and vividly as possible, notlike your being told a story, but as if youwere experiencing it now, right here. Everyfew minutes I will ask you to rate your anxi-ety level on a scale from 0 to 100. Please an-swer quickly and try not to leave the image.”

The imaginal exposure sessions are audio-taped, and the patient is asked to repeat the ex-posure by listening to the tape as part of thatday’s homework.

The situations included in in vivo exposurevary greatly from patient to patient (particu-larly with patients with prominent checking rit-uals). Below are some examples of instructionsthat might be offered to patients during in vivoexposure exercises.

For patients with prominent washing rituals:

“Today, you will be touching [specifiesitem(s)]. This means that I will ask you totouch it with your whole hand, not just thefingers, and then to touch it to your face,hair, and clothing, all over yourself, so youfeel that no part of you has avoided contami-nation. Then I’ll ask you to sit and hold itand repeatedly touch it to your face, hair,and clothes during the rest of the session. Iknow that this is likely to make you upset,but remember the anxiety will eventually de-crease. I also want you to go ahead and letyourself worry about the harm you areafraid will occur—for example, disease—since you won’t he washing or cleaning afterthis exposure. I am sorry that this treatmenthas to be difficult and cause so much dis-comfort, but I’m sure you can do it. You’llfind it gets easier as time goes on. OK, here itis, go ahead and touch it.”

The therapist should give the patient the ob-ject to hold, ask him or her to touch it, andthen ask the patient to touch the object or the

“contaminated” hands directly to his or herface, hair, and clothing. Every 10 minutes thepatient should be asked, “What is your level ofanxiety or discomfort from 0 to 100 right nowas you focus on what you’re touching?” Thiscan be shortened to “What is your SUDS?”once the patient understands the question.

For patients with prominent checking ritu-als:

“Now, I’d like you to [e.g., write out yourchecks to pay your monthly bills withoutlooking at them after you’ve finished; justput them in the envelope and then we willmail them right away without checking evenonce after you’ve done it]. Then we will goon and do [e.g., drive on a bumpy road with-out looking in the rearview mirror] in thesame way. While doing this, I would like youto worry about what harm might occur be-cause you aren’t checking your actions, butdon’t let the thoughts interfere with actuallydoing those activities.”

Patients should be reminded of the specificinstructions for response prevention on the firstday of treatment and periodically during treat-ment. We have found that giving patients aprinted copy of the rules for response preven-tion can help them to understand and remem-ber the rules. If the rules as outlined for the pa-tient do not adequately cover the type ofritual(s) the patient exhibits, the therapistshould provide a written set of instructionsmodeled after these forms.

During the last few sessions of treatment, thepatient should be introduced to rules of “nor-mal” washing, cleaning, or checking. Responseprevention requirements should be relaxed toenable the patient to return to what is consid-ered a normal routine.

Home Visit

It is important to ensure that the patient’sgains from the treatment program generalizeto the home environment. Usually homeworkassignments function to produce this general-ization, but we have found that visits by thetherapist to the patient’s home can be quitehelpful, especially in cases where the patientis not able to return home daily during theintensive treatment phase (e.g., patients whoare from out of town or are hospitalized).The home visit also offers the therapist and


patient an opportunity to discuss guidelinesfor “normal” behavior. The therapist shoulddiscuss the plans for these visits with the pa-tient and his or her family before the treat-ment ends. It is also important to note that,in some cases, the majority of the treatmentsessions need to be conducted at the patientshome, such as when treating a hoarder. Thefrequency of home visiting during the coretreatment should be determined based onwhether the patient’s OCD symptoms arereadily “transportable” to situations outsidethe home, or whether they are specific to thehome. For patients with prominent washingrituals, with “safe” rooms and areas in theirhouses, contamination of these areas is im-perative and also quite difficult; it is often ad-visable that the therapist assist directly withthese home-based exposures when it is ques-tionable whether the patient can contaminatethese “sanctuaries” successfully on his or herown.

Typically, the home visit comprises 4-hoursessions held on each of 2 days at the end of thetreatment program. The bulk of the time inthese sessions is used to conduct additional ex-posures to obsessive stimuli in and around thepatient’s home or workplace. For example, thetherapist might accompany the patient as he/she contaminates objects around the house orat the local grocery store. Similarly, the patientmight be asked to turn the stove on and offwithout checking and leave the house with thetherapist. Most patients, particularly thosewho were able to return home during treat-ment, will report little or no discomfort whendoing these exposures, because they representrepetition of homework assignments. In somecases, though, the therapist will discover areasthat the patient has not contaminated, or someareas at home that continue to generate distressdespite previous exposures. The home visitshould focus on exposure to situations or ob-jects that remain problematic.

Maintenance Period

In addition to prescribing continued self-exposure tasks to help the patient maintaintherapy gains, the therapist may wish to sched-ule regular maintenance sessions. These ses-sions may be used to plan additional expo-sures, to refine guidelines for normal behavior,and to address issues that arise as the patientadjusts to life without OCD.

There is some evidence that patients benefitfrom continued contact with the therapist fol-lowing the intensive therapy sessions. In onestudy, 12 weekly supportive therapy sessions(no exposure exercises) appeared to reduce thenumber of relapses in a sample of individualswith OCD treated with 3 weeks of intensiveEX/RP (Foa et al., 1992). In another study, fol-lowing the intensive treatment with 1 week ofdaily cognitive-behavioral sessions followed byeight brief (10-minute) weekly telephone con-tacts resulted in better long-term outcome thanfollowing intensive treatment with 1 week oftreatment with free association (Hiss et al.,1994).

Therapeutic Setting

It is advisable for patients to remain in theirnormal environments during intensive treat-ment. This is particularly important for pa-tients whose fears are cued mainly by stimuli intheir home environment. The hospital may bean artificially protected setting, particularly forpatients with prominent checking rituals, whomay not feel responsible for their surroundingsand as a result do not experience their usualurges to check. If patients live too far away tocommute for daily sessions, we recommendthat they rent an apartment or hotel room nearthe clinic. When this is not possible, hospital-ization should be considered. Hospitalization isrecommended for patients deemed to be at riskfor suicide or psychotic breakdown, and forthose who need close supervision but lack asupport system sufficient to aid them duringtreatment.

If a patient is employed and his or her OCDsymptoms are work related, he or she shouldbe encouraged to continue working, so that rel-evant exposures can he included in treatment.However, since treatment requires 5 to 6 hoursper day, the patient may opt to work half-daysduring the intensive treatment.

When the patient’s symptoms are unrelatedto work, he or she may decide not to continueworking during intensive treatment. Because ofthe time-consuming nature of the treatment,we often suggest that patients take some timeoff from work. If it is not possible for the pa-tient to take 3 full weeks off from work, thetherapist might suggest that the patient workhalf-days or take time off from work during thefirst and second weeks of the treatment pro-gram.


Therapist Variables

Intensive treatment with exposure to feared sit-uations and response prevention of ritualisticbehavior provoke considerable stress for pa-tients. Their willingness to undergo such “tor-ture” attests to their strong motivation to ridthemselves of the OCD symptoms. The inten-sive treatment regimen requires that the thera-pist maintain a delicate balance between pres-suring the patient to engage in the treatmentand empathizing with his or her distress. Clini-cal observations and findings from a study byRabavilas and colleagues (1979) suggest that arespectful, understanding, encouraging, ex-plicit, and challenging therapist is more likelyto achieve a successful outcome than a permis-sive, tolerant therapist. Notably, patients ofwell-supervised, nonexpert EX/RP therapistsappear to fare well with EX/RP (Franklin,Abramowitz, Furr, Kalsy, & Riggs, 2003;Valderhaug et al., 2007).

During treatment, patients’ behavior mayrange from extreme cooperation and willingnessto participate in exposures to blatant manipula-tion and refusal to follow the therapist’s instruc-tions. An individual patient may fluctuate de-pending on what exposure is conducted during aparticular session. To a great extent, the “art” ofconducting behavioral therapy for OCD in-volves knowing when to push, when to confront,and when to be more flexible. Such decisions re-quire that the therapist carefully observe the pa-tient’s reactions and make a judgment based onhis or her experience. As much as possible, thetherapist should display an attitude that coun-teracts the harshness of the treatment program,while maintaining the rules for therapy estab-lished at the beginning of the program. The ther-apist should assure the patient that he or shewould not use force to implement exposure andthat no exposure will be planned without the pa-tient’s consent. If the patient cannot trust that thetherapist will adhere to these essential guide-lines, the treatment is likely to be compromised.We also assure the patients that family memberswill be asked not to present unplanned expo-sures to the patient (e.g., taking out the garbage)without discussing it.

Patient Variables

A primary factor that influences a patient’s po-tential for benefiting from intensive behavioraltreatment is the level of his or her motivation.

Because EX/RP causes high distress, patientsneed to be highly motivated to undertake thetreatment. Often the level of motivation is re-lated to the severity of the patient’s symptoms.When symptoms are sufficiently intolerable,patients are more likely to tolerate considerablediscomfort for a short period to gain relieffrom their symptoms in the long run. Tolin andcolleagues (2004) have also discussed the im-portance of motivational readiness in EX/RP(Tolin et al., 2004), and have suggested specifi-cally how best to prepare patients for the oftengrueling treatment regimen.

Sometimes individuals are pressured into en-tering therapy by their families, and they agreeto participate in treatment only to appease aspouse or a parent. These patients are unlikelyto follow the therapist’s instructions strictly;therefore, they are less likely to make lastinggains in therapy. In light of these observations,we do not recommend that patients enter intoEX/RP if they are not committed to follow suchinstructions; alternative treatment strategiesare typically recommended in such circum-stances.

It is important that the therapist clearly ex-plain to the patient that 1 month of therapy, al-beit intensive, is unlikely to eliminate all OCDsymptoms. Rather, patients should expect thattheir anxiety and the urges to ritualize will di-minish and become more manageable. An ex-pectation of becoming symptom free at the endof treatment may lead to disappointment andcan potentiate relapse, because maintenance oftreatment gains usually requires continued ef-fort over time following the intensive treat-ment. Thus, in the initial interview we tell pa-tients that we do not have a “cure” for OCD;rather, we have a treatment that is likely to helpthem substantially reduce their symptoms inboth the short and the long run.

It is also important to explain to patientsthat EX/RP treatment is not a panacea for all oftheir psychological and interpersonal prob-lems. This treatment is aimed specifically at re-ducing the patient’s obsessions and urges to rit-ualize. Problems that existed prior to treatment(e.g., marital discord or depression) are likelyto remain, although they may be somewhat al-leviated after treatment.

As mentioned earlier, patients with severedepression and/or an extremely strong belief inthe reality of the obsessive fear may not benefitfrom EX/RP. An additional factor that has beenidentified as a potential hindrance to the


cognitive-behavioral and pharmacologicaltreatment of OCD is concurrent schizotypalpersonality disorder (Jenike, Baer, Minichiello,Schwartz, & Carey, 1986). Although somequestions have been raised about the methodused to diagnose schizotypy (see Stanley,Turner, & Borden, 1990), therapists should bealerted to the probability that schizotypal pa-tients may respond poorly to treatment forOCD.

CASE STUDY

In this section we demonstrate through verba-tim material the process of gathering infor-mation relevant to treatment, planning thetreatment program, and conducting exposuresessions.

Case Description

“June,” a 26-year-old married woman whohad just completed her bachelor’s degree innursing, sought treatment for a severe washingand cleaning problem. She was extremely agi-tated in the first interview and described herselfas “crying a whole lot” during the previous 6weeks. She arrived in the company of her hus-band of 6 months and her sister-in-law, whomshe considered a good friend. Previous treat-ment by systematic desensitization, antidepres-sants, tranquilizers, and cognitive restructuringhad proven ineffective. June had been unable toseek employment as a nurse due to her symp-toms.

This information was collected at June’s ini-tial evaluation for participation in EX/RP treat-ment. After ascertaining the absence of psy-chosis, drug and alcohol abuse, and organicdisorders, June was assigned a therapist.

Information Gathering

Current Symptoms

First, the therapist sought information fromJune about the obsessional content, includingexternal and internal fear cues, and beliefsabout consequences, and information aboutpassive avoidance patterns and types of ritu-als. Because rituals are the most concretesymptom, it is often convenient to begin theinquiry by asking for a description of thisbehavior.

THERAPIST: I understand from Dr. F that youare having a lot of difficulty with washingand cleaning. Can you tell me more aboutthe problem?

JUNE: I can’t seem to control it at all recently. Iwash too much. My showers are taking along time, and my husband is very upsetwith me. He and my sister-in-law are tryingto help, but I can’t stop it. I’m upset all thetime and I’ve been crying a whole lot lately(on the verge of tears). Nothing seems tohelp.

THERAPIST: I see. You look upset right now.Please try to explain what your washing hasbeen like in the past few days, so I can under-stand. How much washing have you beendoing?

JUNE: Much too much. My showers use up allthe hot water. And I have to wash my hands,it seems like, all the time. I never feel cleanenough.

THERAPIST: About how long does a showertake? How many minutes or hours wouldyou say?

JUNE: About 45 minutes, I guess. I try to get outsooner. Sometimes I ask Kenny to make mestop.

THERAPIST: And how often do you take one?

JUNE: Usually only twice, once in the morningand once at night before bed, but sometimes,if I’m really upset about something, I couldtake an extra one.

THERAPIST: And what about washing yourhands? How much time does that take?

JUNE: You mean how many times do I wash?

THERAPIST: How long does it take each timeyou wash your hands, and how often do youwash your hands in a day?

JUNE: Umm, maybe 20 times a day. It probablytakes me five minutes each time, maybemore sometimes. I always have the feelingthey’re not really clean, like maybe I touchedthem to the side of the sink after I rinsed andthen I think they’re dirty again.

The therapist now had some basic informa-tion about the most prominent rituals. Somefurther questioning clarified whether othercompulsions were also in evidence.

THERAPIST: Do you do anything else to makeyour self feel clean?


JUNE: Yes, I alcohol things. I wipe with alcohol,like the car seat before I sit down.

THERAPIST: Do you wipe yourself with alcohol?

JUNE: No, only things that I think are dirty.

THERAPIST: Can you tell me how much you dothat?

JUNE: I use about a bottle of alcohol a week.

Here the therapist had to choose whether toinquire about what objects June cleans or toask about possible additional rituals. The ther-apist chose to continue the inquiry about ritu-alistic actions, and to turn to the subject of“contaminants” as soon as the inquiry wascompleted.

THERAPIST: OK, can you think of any otherthings that you do to clean yourself, or otherthings around you that you feel are dirty?

JUNE: That’s all I can think of right now.

THERAPIST: What about other kinds of what wecall “compulsive” type of activities? Do youhave to check or repeat things over andover?

JUNE: No, except when I wash, if I don’t feel it’senough. Then I wash again.

THERAPIST: No other repetitive actions besideswashing?

Since this patient did not appear to havemultiple types of ritualistic behaviors, the ther-apist turned to the obsessional content. Exter-nal cues are usually solicited first.

THERAPIST: What are the things that make youfeel you want to wash? For instance, why doyou wipe the car seat with alcohol?

JUNE: I think that maybe I got dog dirt on itwhen I got in from before, or Kenny mighthave.

THERAPIST: From your shoes?

JUNE: Yes, I also worry about the hem of mydress touching the seat. I’ve been worryingthat my shoe could kick my skirt hem orwhen I step up a step, like to go in a building,the dress could touch the step.

THERAPIST: A dress like this? June was wearinga dress that came to just below her knee.[The likelihood that it could have touched acurb or sole of her shoe was very slim.]

JUNE: Yes.

THERAPIST: Has your skirt ever had dog dirt onit?

JUNE: I don’t think so, but in my mind I thinkthat maybe it could have gotten some on it. Isuppose it would be hard for that to happen,wouldn’t it?

Thoughts that highly improbable eventsmight have occurred are common in OCD.Such distortions may be the result of intenseanxiety. Doubts about “safety” often lead torequests for reassurance or to rituals. Reassur-ing June that her dress is unlikely to be soiledwould have been countertherapeutic, because itperpetuates the neurotic fears. Rather, the ther-apist inquired further about the obsessionalcontent.

THERAPIST: Is dog “dirt” the most upsettingthing that you worry about?

JUNE: Probably. Yes, I think so, but bathroomgerms are pretty bad too.

THERAPIST: What sort of germs?

JUNE: From toilets. You know, when you go tothe bathroom.

THERAPIST: Urine and feces?

JUNE: Yes, urine doesn’t bother me as much asthe other.

THERAPIST: Why?

JUNE: Because I learned in nursing school thatit’s almost sterile. I had a hard time in thecourse about microbiology, because it upsetme to try to learn about bacteria and micro-organisms. They make it sound like there areall kinds of germs everywhere that are realdangerous. I didn’t learn it very well; I triedto avoid thinking about it.

June’s concerns with both dog dirt and bath-room germs suggested that her fear structureincludes apprehension about potential illness.The therapist questioned her to better under-stand the nature of the feared consequences ofcontamination.

THERAPIST: Are you afraid of diseases thatcould come from feces?

JUNE: Yes, I guess so. The thing of it is, though,I know other people don’t worry about itlike I do. To them, you know, they just go tothe bathroom and wash their hands anddon’t even think about it. But I can’t get it


out of my head that maybe I didn’t get cleanenough.

THERAPIST: If you didn’t wash enough, wouldyou get sick or would you cause someoneelse to get sick?

JUNE: Mostly I worry that I’ll get sick, butsometimes I worry about Kenny too.

THERAPIST: Do you worry about a particularkind of disease?

JUNE: I’m not sure. Some kind of illness.

It is not uncommon for patients who fearharm that may ensue from not ritualizing to beunable to identify a specific feared conse-quence. Patients with prominent checking ritu-als often fear they will forget or throw outsomething important, but they do not alwaysknow exactly what this will be. Repeaters mayfear that something bad will happen to a lovedone but often cannot specify what particulardisaster will befall them. However, many in-dividuals with OCD do fear specific conse-quences (e.g., blindness or leukemia). At thispoint, the therapist may either choose to com-plete the inquiry about external threat cues orpursue the investigation about the feared con-sequences and the belief that such harm is in-deed likely to occur. The latter course was se-lected here.

THERAPIST: Let’s say that you did actuallytouch dog feces or human feces, and youweren’t aware of it, so you didn’t wash to re-move it. What is the likelihood that you orKenny would really get seriously ill?

JUNE: Well, I feel like it really could happen.

THERAPIST: I understand that when it happensand you become very distressed, it feels likeyou will actually become sick, but if I askyou to judge objectively, right now, howlikely is it that you will get sick from touch-ing feces and not washing? For example, ifyou were to touch feces 10 times, how manytimes would you get sick?

JUNE: Oh, I know it’s pretty unlikely, but some-times it seems so real.

THERAPIST: Can you put a number on it?What’s the percent chance that if youtouched a small amount of feces and didn’twash that you’d get sick?

JUNE: I’d say low, less than 25%.

THERAPIST: That means that one time in everyfour, you’d get sick.

JUNE: No, that’s not right. I guess it’s really lessthan 1%.

From this dialogue it is clear that June didnot strongly believe that her feared disasterswould actually occur, although her initial esti-mate of the likelihood was high. A person withpoor insight regarding the senselessness of hisor her OCD symptoms would have assignedhigher probabilities (usually over 80%) andwould insist on the accuracy of his or her esti-mate even in the face of persistent questioning.Note also that this exchange is an example ofthe informal cognitive restructuring accompa-nying EX/RP that we discussed earlier. Thetherapist may need to repeat this discussionduring subsequent exposure sessions whenJune, highly anxious about confronting con-taminants, readjusts her likelihood estimateswhen anxious. Strength of belief can change ina given patient but is stronger when the patientperceives threat.

THERAPIST: OK. Now, besides disease, whatelse could happen if you got feces on you?

JUNE: I suppose I’m also afraid of what otherpeople might think if I got dog feces on myshoe or on my dress. Somebody would see itor smell it and think it was really disgusting,and I was a dirty person. I think I’m afraidthey would think I’m not a good person.

The therapist then questioned June furtherabout this feared consequence, inquiring aboutthe possibility of others evaluating her charac-ter negatively because she had feces on herdress. The material regarding feared conse-quences was collected for later inclusion in theimaginal exposure scenes. To conclude the in-quiry about the nature of the obsessions, thetherapist further elucidated the external fearedstimuli.

THERAPIST: Besides dog and human feces andtoilets, what else can “contaminate” you? Isit OK if I use the word “contaminated” todescribe how you feel if you handle thesethings?

JUNE: Yes, it’s like I can feel it on my skin, evenif I can’t see it. Umm, I also get upset if I see“bird doo” on my car.


THERAPIST: Bird droppings? The whitish spots?

JUNE: Yeah, I have to hold my skirt close to meso that I don’t touch any of these spots withmy clothes.

THERAPIST: OK, bird doo, what else?

JUNE: Dead animals, like on the roadside. I feellike the germs, or whatever it is, get on thetires from the pavement and get on the car.Even if I don’t run over it. Like it’s spreadaround the street near it.

THERAPIST: What do you do if you see a deadanimal?

JUNE: I swerve wide around it. Once I parkedthe car and as I got out, I saw this dead catright behind the car. I had to wash all myclothes and take a shower right away. It wasreally a mess that day.

THERAPIST: It sounds like that was very difficultfor you. Is there anything else besides deadanimals that contaminates you?

JUNE: I can’t think of any. There are lots ofplaces I avoid now, but that’s because ofwhat we just talked about.

The therapist questions June further aboutother items that are likely to be contaminated,because of their potential relationship to theones she has already noted.

THERAPIST: What about trash or garbage?

JUNE: Yeah, that bothers me. And I also avoidgutters on the street.

THERAPIST: What’s in the gutter that upsetsyou?

JUNE: Dead animals, I guess. And then the rainspreads the germs down the street. Also rot-ten garbage. It’s really dirty. Sometimes thegutters are really disgusting.

THERAPIST: Um hmm. Are you afraid youcould get sick from dead animals and gar-bage?

JUNE: Yes, it’s like the toilets or dog dirt.

To prepare for an exposure program inwhich objects are presented hierarchically withrespect to their ability to provoke discomfort,June was asked to rank her major contami-nants. Here she also provided informationabout avoidance behaviors associated with hercontaminants.

THERAPIST: Now, let’s make a list of the mainthings that upset you. I’m going to ask youhow distressed you would be on a 0- to 100-point scale if you touched the thing I’ll name.Zero indicates no distress at all and 100means you’d be extremely upset, the mostyou’ve ever felt.

JUNE: OK.

THERAPIST: What if you touched dog dirt?

JUNE: And I could wash as much as I wanted?

THERAPIST: No, let’s say you couldn’t wash fora while.

JUNE: 100.

THERAPIST: A dead animal.

JUNE: Also 100.

THERAPIST: Bird doo on your car.

JUNE: That depends on whether it is wet or dry.

THERAPIST: Tell me for both.

JUNE: 100 wet and 95 dry.

THERAPIST: Street gutter.

JUNE: 95.

THERAPIST: Garbage in your sink at home.

JUNE: Not too bad. Only 50. But, the trash canoutdoors would be 90.

THERAPIST: Why the difference?

JUNE: Because the inside of the trash can isdirty from lots of old garbage.

THERAPIST: I see. What about a public toiletseat?

JUNE: That’s bad. 95.

THERAPIST: Car tires?

JUNE: Usually 90. But if I just passed a dead an-imal, they’d be 99.

THERAPIST: What about a doorknob to a publicbath room?

JUNE: The outside knob is low, like 40. But theinside knob is 80, because people touch itright after they’ve used the bathroom, andI’ve seen that some don’t wash their hands.

THERAPIST: I understand. How about grass in apark where dogs are around?

JUNE: If I did walk in the grass, it would beabout 80 or 85, but I don’t usually do it. Ialso have a lot of trouble on sidewalks. Youknow, the brown spots on the concrete. Iguess most of it is just rust or other dirt, but Ithink maybe it could be dog dirt.


THERAPIST: How much does that bother you?

JUNE: To step on a brown spot? About 90. I al-ways walk around them.

The therapist should continue in this manneruntil a list of 10–20 items is formed. Moreitems may be necessary for patients with multi-ple obsessional fears or rituals. The items areordered from low- to high-level fear in prepara-tion for treatment by exposure. Items equiva-lent with regard to their level of disturbance aregrouped together. Moreover, it is important toprobe the rationale for one stimulus differingfrom another, because it provides further infor-mation about the patient’s particular “OCDlogic.” This information is highly relevant forthe construction of the exposure hierarchy andfor the informal cognitive discussions aboutrisk assessment, responsibility, and so forth.

Considerable information about avoidancepatterns and rituals emerged from the previousinterview about external threat cues. More de-tails may be obtained by asking the patient toprovide a step-by-step description of a typicalday’s activities from the time he/she awakensuntil he/she goes to sleep. Usually, patients arenot entirely accurate when describing theircompulsive behaviors during the interview, be-cause, as one patient told us, they have not“thought of their OCD in that way before.”Thus, the self-monitoring tasks assist patientsin raising their awareness about the OCD pat-terns and provide the therapist with more accu-rate data about rituals and avoidant behaviors.

We were particularly concerned by June’sbathroom routines, her shower, use of the toi-let, handling of towels and dirty clothes, anddressing and putting on shoes. Additional in-formation about avoidance patterns may be as-certained by inquiring about other routineactivities, such as shopping, eating out, house-cleaning, preparing meals, working, and so on.The following dialogue exemplifies the degreeof detail desired.

THERAPIST: June, for us to plan your treatmentcarefully, I need to know what you avoid inyour daily routine. Why don’t you start bydescribing what you do first when you wakeup.

JUNE: I go to use the bathroom first.

THERAPIST: Nightgown on or off?

JUNE: I take off my nightgown because I don’t

want it to touch the toilet. That way it’sclean at night after I shower.

THERAPIST: Go on.

JUNE: I go to the toilet. I suppose I use a lot oftoilet paper because I don’t want to get any-thing on my hand. Then I have to shower af-ter a bowel movement.

THERAPIST: How do you get ready to shower?

JUNE: I have to put a new towel on the rod nearthe shower. I don’t like it to touch anythingbefore I use it. Oh, and I put my slippers fac-ing the door, near the shower, so I can putthem on without stepping on the bathroomfloor when I get out of the shower. Then I getinto the shower.

THERAPIST: You said you shower for 45 min-utes. Why does it take so long?

JUNE: I have to wash myself in a special orderand I count how many times I wash eachpart. Like I wash my arm four times. That’swhy it takes so long.

THERAPIST: What is the order you use?

JUNE: First I wash my hands, then my face andhair and then I go from the top down.

THERAPIST: What about the genital and analarea? [This area should disturb this patientmost, because she fears contamination fromfecal “germs.”]

JUNE: Oh yes, those are last, after my feet.

Such a detailed description helps the thera-pist to anticipate possible avoidance by the pa-tient during treatment and to plan specificexposure instructions. Supervision of normalwashing behavior at the end of treatment willaddress June’s tendency to count and to orderher washing. During the initial session of infor-mation gathering, June was instructed to self-monitor the frequency and duration of hercompulsions.

THERAPIST: Between now and our next session,I’d like you to record all the washing andcleaning that you do, including wipingthings with alcohol. You can use this form(hands her a self-monitoring of rituals form).Please write down every time you wash, howlong you washed, what made you wash, andhow anxious you were before you washed.This kind of record will help us identifyany sources of contamination you’ve for-gotten to mention, and we can also use


it to measure your progress during treat-ment.

JUNE: Do you want me to write in each spacefor each half hour?

THERAPIST: No, only when you wash or use al-cohol.

JUNE: OK.

History of Symptoms and Treatment History

After assessing the patient’s current symptoms,the therapist sought information about the on-set of the problem, with particular reference tothe presence of specific stressors at the time andwhether these stressors are still present.

THERAPIST: How long have you been washinglike this?

JUNE: It started about 2 years ago in my firstyear of nursing school. It wasn’t real badright away. It started with the city. I had togo into the city to classes, and the cityseemed real dirty.

THERAPIST: Did nursing have something to dowith it?

JUNE: Maybe. I was under a lot of tension. Ihad to quit working as a secretary and it waspretty hard without an income and a lot ofschool bills. My mother and dad weren’tmuch help. And then we started to learn allthe sterilizing techniques, and I already toldyou about the course in microbiology.

THERAPIST: Did it gradually get worse?

JUNE: Mostly, but I did notice that it was a lotworse after a rotation on surgery, where Iwas really worried about germs contaminat-ing the instruments. That’s when I started towash more than usual.

THERAPIST: Did you seek help at that time?

JUNE: I was already seeing Dr. W at the univer-sity, and he tried to help.

THERAPIST: You were already in treatment withhim? For what reason?

JUNE: He was helping me with an eating prob-lem. I had anorexia. I’d been seeing him forabout a year when the washing started.

THERAPIST: Anorexia? Did treatment help?

JUNE: Yes. I was down to 85 pounds and I’m uparound 105 now. He mostly asked me to in-crease my weight every week and he did“cognitive therapy,” I think its called.

THERAPIST: I see. What about the washingproblem?

JUNE: He tried the same type of therapy, but itdidn’t work for that. That’s why I’m here.My sister-in-law heard about it, and Dr. Wsaid I should come.

THERAPIST: What about drugs? Were you evergiven medication for this problem?

JUNE: Yes, I tried Anafranil [clomipramine] fora while and it helped a little, but it made medizzy and sleepy, so I decided to stop takingit. Also, I heard that you can’t take the medi-cation when you are pregnant, and Kennyand I want to have a baby soon. Before that,I took Xanax [alprazolam]; it calmed medown but didn’t stop the washing.

THERAPIST: Have you tried any other treat-ments?

JUNE: Only for the anorexia. I went to anothercounseling center at the university for abouta year, but that didn’t really help at all.

June’s history was unusual only in the rela-tively recent onset of her symptoms. Typically,patients in our clinic present a much longer du-ration of symptoms, with the mean around 8years. Other centers in England and Hollandreport similar figures. June’s treatment historyof trying various psychotherapeutic and phar-macological treatments prior to seeking EX/RPwas quite typical. Since previous failure withnonbehavioral treatments has not been foundto influence outcome with exposure and re-sponse prevention, the clinician should not bediscouraged by such a history. However, be-cause of a possible skeptical attitude about thevalue of treatment, the therapist should pro-vide the patient with a clear rationale for EX/RP treatment along the lines discussed earlierand demonstrated below.

THERAPIST: Before I continue to collect moreinformation about your problem, let me tellyou about our treatment.

JUNE: Well, Dr. F told me something about it,but I’m still not sure what this treatment isgoing to be like.

THERAPIST: The treatment is called exposureand ritual prevention. I’ll be asking you toconfront situations and things that frightenyou or make you feel contaminated. We willdo this gradually, working up to the hardestthings. For example, we may begin with the


outside door handles of bathrooms andwork our way up to toilet seats and birddoo. We’ll do this together, and I’ll be thereto help you. The sessions will last 1½ or2 hours, and we’ll meet every weekday.In addition, I’ll assign you homework todo similar things between the therapy ses-sions.

JUNE: You mean I have to touch them, even dogdirt?

THERAPIST: Yes, to get over these kind of fears,people must learn to confront what they’reafraid of and stay with it until the discomfortdecreases.

JUNE: Even if I did, it would probably take me ayear to get used to it.

THERAPIST: Remember, you didn’t always feellike this about dog dirt. When you wereyounger, did you ever step in dog dirt andjust wipe it off on the grass and go on play-ing?

JUNE: Yeah, I forget that. It seems such a longtime ago. I used to not think twice about thisstuff.

THERAPIST: To get you back to how you used tofeel, we need to expose you directly to whatyou’re afraid of. Now, there’s a second partto treatment. I’m also going to ask you notto wash for 3 days at a stretch. No hand-washing or showering for 3 days. Then youcan take a shower, but you will have to limitit to 10 minutes. After the shower, you willhave to contaminate yourself again, thenwait another 3 days for your next shower.

JUNE: I can’t believe it! I’ll never be able to dothat. If I could, I wouldn’t be here. How canI not wash? Every day I resolve to stop, but Ialways give in. You mean I wouldn’t be ableto wash after I use the bathroom or before Ieat? Other people wash after they use thetoilet. Why can’t I just wash less, like normalpeople do?

THERAPIST: Other people don’t have OCD. Re-member for you, washing makes you feel less“contaminated” and less anxious. Right?

JUNE: Yes.

THERAPIST: If you wash, even briefly, wheneveryou feel “contaminated,” you never get achance to learn that the feeling of contami-nation would go away by itself withoutwashing. If you are really very anxious, itmight take a while, even several hours, be-

fore you feel better, but it will eventuallyhappen. On the other hand, if you wash,even briefly, every few hours, it will reinforceyour idea that you have to wash to feelbetter.

JUNE: But why 3 days? Couldn’t I shower oncea day like other people?

THERAPIST: For the same reason. You’d still feelrelief, even if you waited 24 hours betweenwashings. And that would strengthen yourbelief that you need to “decontaminate” bywashing yourself. You must learn to use soapand water to feel clean and fresh but not to“decontaminate” yourself.

JUNE: I think I understand. I know I showernow to get the things I’m afraid of off mybody. I used to shower just to get sweat anddirt off, and feel nice. I’m still not sure Icould stand it though, not washing for thatlong.

THERAPIST: The treatment is very demanding.Before we start the treatment program youwill need to make a commitment to yourselfthat even though you will feel very uncom-fortable and even quite upset at times, youwon’t wash. I’ll try to help you as much as Ican by planning the treatment so you knowwhat to expect each day and by supportingyou whenever you need it. Someone willhave to be available to help supervise andsupport you any time you need it. Betweensessions you can always call me here or athome if a problem comes up. I know thetreatment won’t be easy for you, but I’m sureyou can do it if you make up your mind.

At this point, a firm commitment should notbe requested. Rather, the patient should bemade aware of what will be required of him orher so that he or she can adjust to these expec-tations and plan activities during the treatmentperiod accordingly. The patient should makethe arrangements necessary to attend dailytreatment sessions for 3 to 4 weeks. As we dis-cussed earlier, two to three sessions per weekmay be sufficient for patients with less severesymptoms. It is important that the therapist notminimize the difficulty of the treatment regi-men, so that the patient is prepared to struggleand enters treatment with a readiness to mobi-lize inner resources and emotional supportfrom family and friends.

The history of the patient is usually taken inthe first session. Because collecting histories of


individuals with OCD does not differ from col-lecting histories of other psychiatric patients,details are not provided here.

Treatment Planning

The therapist began the second session bybriefly reviewing the patient’s self-monitoringof rituals form. The remainder of the ses-sion was devoted to developing a treatmentplan.

THERAPIST: OK, now I want to discuss our planfor each day during the first week of therapy.We need to expose you both in imaginationand in reality to the things that bother you,which we talked about in our first sessions.As I said already we’ll also limit your wash-ing. The scenes you imagine will focus on theharm that you fear will occur if you do notwash. The actual exposures will focus onconfronting the things that contaminate you.Restricting your washing will teach you howto live without rituals. In imaginal exposure,you will picture yourself touching somethingyou’re afraid of, like toilet seats, and notwashing and then becoming ill. We can haveyou imagine going to a doctor who can’t fig-ure out what’s wrong and can’t fix it. That’sthe sort of fear you have, right?

JUNE: Yes, that and Kenny getting sick, and itbeing my fault.

THERAPIST: OK, so in some scenes you’ll besick, and in others Kenny will get sick.Should I add that other people blame you fornot being careful? Is this what you’re afraidof?

JUNE: Yes, especially my mother.

THERAPIST: OK. We’ll have her criticize you fornot being careful enough. Can you think ofanything else we should add to the image?

JUNE: No, that’s about it.

THERAPIST: We can compose the scenes in detailafter we plan the actual exposures. Let’s re-view the list of things you avoid or are afraidto touch and make sure that we have listedthem in the right order. Then we’ll decidewhat to work on each day. OK?

JUNE: OK.

June reviewed the list, which included itemssuch as trash cans, kitchen floor, bathroomfloor, public hallway carpet, plant dirt, pud-

dles, car tires, dried dog “dirt,” and bird“doo.” Changes were made as needed.

THERAPIST: Good. Now let’s plan the treat-ment. On the first day we should start withthings that you rated below a 60. Thatwould include touching this carpet, door-knobs that are not inside bathrooms, bookson my shelves, light switches, and stair rail-ings. On the second day, we’ll do the 60- to70-level items, like faucets, bare floors, dirtylaundry, and the things on Kenny’s desk.[The therapist continued to detail Sessions 3to 5, increasing the level of difficulty eachday.] In the second week, we will repeat theworse situations like gutters, tires, public toi-lets, bird doo, and dog dirt, and we’ll alsofind a dead animal to walk near and touchthe street next to it.

On rare occasions, direct confrontation witha feared object (e.g., pesticides or other chemi-cals) may have some likelihood of producingactual harm. In such cases, the therapist’s judg-ment should be exercised to find a middleground between total avoidance and endanger-ment. With chemicals, for example, patientsare exposed to small quantities that are objec-tively nonharmful. In June’s case the therapistdecided that direct contact with a dead animalwas not called for, and that stepping on the ani-mal’s fur with her shoe and then touching theshoe sole constituted sufficient exposure. Ingeneral, the therapist must weigh the level ofobsessional distress that will be evoked by agiven exposure with the objective risks entailedin completing that exposure. Patients withOCD have difficulty assessing such risks realis-tically; thus, it is the responsibility of the thera-pist to evaluate whether exposure is warranted.For example, patients with fears of contractingHIV would certainly be highly distressed ifasked to handle a dirty hypodermic needlefound in a city gutter, but because exposure tosuch stimuli is objectively risky, they should notbe included on the treatment hierarchy.

THERAPIST: How does this plan sound?

JUNE: The first week is OK, but I’m reallyscared about the second week. I’m not sureI’ll be ready to do the bathrooms and dogdirt by then.

THERAPIST: Many people feel this way at thebeginning, but by the end of the first week,


you won’t be as frightened as you are nowabout touching tires or public toilets. Re-member, I will be here to help you, because itwill probably be difficult in the beginning.

JUNE: Yes, I know it. I feel like I don’t reallyhave a choice anyhow. This washing is crazyand I’m disgusted with myself. I suppose I’mas ready as I’ll ever be.

THERAPIST: Good. Now remember, I’ll ask youto keep working on these things for 2 to 3hours at home after each session, but youwill already have done them with me, so Idon’t think it will be too hard. I take it thatyou talked to Kenny about assisting us withsupervising, since I saw him out in the wait-ing room.

JUNE: Yes, he said that’s fine. He wanted toknow what he should do.

THERAPIST: Let’s call him in. Did you talk toyour sister-in-law about being availablewhen Kenny is at work during the day?

JUNE: Yes, she was really good about it, but shecouldn’t come today because of the kids.

THERAPIST: If it’s difficult for her to come, Icould talk to her on the phone. Why don’tyou go get Kenny now?

Treatment

June was seen for 15 treatment sessions, heldevery weekday for 3 weeks. During the fourthweek the therapist visited her twice at her homefor 4 hours each time. During these visits,under the therapist’s supervision, June contam-inated her entire house and exposed herself toobjects at home and in her neighborhoodthat provoked distress. Thereafter, once-weeklyfollow-up sessions were instituted to ensuremaintenance of gains and to address any otherissues of concern to her.

As discussed earlier, treatment begins withexposure to moderately difficult items on thehierarchy and progresses to the most disturbingones by the beginning of the second week. Themost distressing items are repeated during theremainder of the second and third weeks. Thefollowing sequence, which occurred on thesixth day of treatment, exemplifies this process.

THERAPIST: How was your weekend?

JUNE: Not that great. I suppose it was as goodas I could expect. I took my shower Sunday

night and I was so nervous about finishing intime I don’t even know if I washed right.

THERAPIST: Most people feel the same way. Re-member though, you aren’t supposed towash “right,” just to wash. Did Kenny timeit?

JUNE: Yes, he called out the minutes like yousaid, “5, 7, 9,” and then “stop.”

THERAPIST: You stopped when he said to?

JUNE: Yes, but it still wasn’t easy.

THERAPIST: I know. I’m really pleased that youwere careful to follow the rules.

JUNE: I have pretty much decided that this is mychance to get better, so I’m trying my best.

THERAPIST: Good. I am glad you feel so posi-tive. How was the homework?

JUNE: I touched the floor and the soles of myshoes and the cement. It is all written on thedaily sheet there. On Saturday, I went to mysister’s, so I could play with the kids like wesaid. They stepped on me when I lay on thefloor and I tried to touch their bottoms when Iheld them. On Sunday, Kenny and I went tothe park. I didn’t sit in the grass but I did walkaround and touched my shoes afterward.

THERAPIST: The soles?

JUNE: Yeah. We also went downtown and Ithrew some things in the trash cans andpushed them down, and tried to touch thesides. It’s sort of hard because I felt conspicu-ous, but I did it anyway.

THERAPIST: That sounds really good. I’m gladto hear it. How about your doormat and go-ing into the garden?

JUNE: I did the doormat and I stood in the gar-den, but I couldn’t touch the dirt. The neigh-bor’s dog always runs all over. I know Ishould have touched it, but I just couldn’t getup the courage.

THERAPIST: Well, you did do many other things.Let’s plan to go outside today and do it to-gether, so it will be easier for you to walk inthe garden when you go home.

JUNE: OK.

June was very compliant with the treatmentregimen. Some patients occasionally lapse onresponse prevention, particularly during thefirst week of the treatment program. The thera-pist should reinforce the patient for partial


compliance but emphasize the need to complyfully with treatment instructions. With regardto exposure homework, it is not uncommon forpatients to neglect completing some assign-ments. Again, they should be reinforced forwhat they have achieved and encouraged tocomplete all of the assignments.

THERAPIST: How are you and Kenny doing?

JUNE: He got mad on Sunday night after theshower because I started to ask him how heshowered and if I was clean enough. I think Inagged him too much, so he lost his temper.We just watched TV, and after a while wetalked a bit and he sort of apologized for get-ting mad. But I understand; I ask too manyquestions. Otherwise, the rest of the week-end was OK.

THERAPIST: Well, it’s unfortunate that Kennygot mad, but it’s good that he didn’t answeryour questions. He’s not supposed to reas-sure you about cleanliness.

JUNE: I think he has a hard time knowing whento answer me and when not to. I am not realsure either. Could you talk to him beforeWednesday when I shower again?

THERAPIST: That’s a good idea. I’ll call him af-ter we’re done with today’s session. Now, to-day we’ll start with the scene about you driv-ing your car to an appointment with me, andyou get a flat tire and have to change it. Thecars splash water in the puddle near you, andit lands on the car and on you. Then you no-tice a dead animal when you walk behind thecar, and it’s right behind you. You really feelcontaminated. You walk to the gas stationnearby to see if they can fix the tire and youhave to urinate so badly that you have to usetheir rest room. They agree to fix the tire ifyou remove it and bring it to them, becauseotherwise they are too busy. Of course, thatmeans you will have to handle the tire that iscontaminated by the dead animal. We’ll addsome bird doo on the street and on the side-walk too. Then, later you start to feel sick,and you feel like it’s from the dead animal.Sound awful enough?

JUNE: Yeah. Ugh. That one is really bad. Do Ihave to? Never mind, I know the answer.

THERAPIST: OK. I want you to close your eyesnow and imagine that you are driving yourcar on West Avenue.

Note that the therapist checked the patient’sassignment from the previous day to verify thatshe completed it and did not engage in avoid-ance and rituals. This provided an opportunityto reinforce efforts at self-exposure. It is impor-tant to keep track of completion of homework,because patients do not always volunteer infor-mation about omissions. They will, however,admit failure to comply if directly asked andare likely to carry out the next assignment if re-inforced adequately.

With regard to the conflict between June andKenny, it is our experience that, like Kenny,most family members are quite willing to help.Difficulty may, however, arise when they areunable to help without becoming upset,thereby increasing the patient’s tension. Pro-viding them with an opportunity to ventilatetheir frustration by contacting the therapist,who also may coach them in alternative reac-tions, may reduce familial tension.

That same sessions also included imaginalexposure to do a scenario planned in advance.Since that scenario had already been discussedin detail with the patient, it posed no surprisesfor her. It is presented for up to 1 hour, or untila substantial decrease in anxiety is evident.Next, the patient is confronted in vivo with sit-uations like those included in the fantasizedscene.

THERAPIST: It’s time to do the real thing now. Ilooked for a dead animal by the side of theroad yesterday and I found one about a mileaway. I think we should go there.

JUNE: Yuck, that’s terrific. Just for me you hadto find it.

THERAPIST: Today’s our lucky day. You knewwe were going to have to find one today any-how. At least it’s close.

JUNE: Great.

Humor is encouraged and can be quite help-ful if the patient is capable of responding to it.At the same time, it is important that the thera-pist laugh with rather than at the patient. Pa-tients and therapists often develop a shorthandlexicon for discussing OCD and its treatmentthat is specific to them and aimed at promotingcompliance with treatment. For example, onepatient–therapist pair began to discuss ex-posure homework as “swallowing the frog,”based on a proverb that the patient introduced.


When the therapist asked the patient if she had,“swallowed the frog” that morning, it con-veyed the difficulty of the exposure tasks sheneeded to do between sessions. It is importantfor the therapist to observe the patient’s inter-personal style to determine whether such ban-ter is likely to promote the therapeutic goals.

THERAPIST: (outside the office) There it is, be-hind the car. Let’s go and touch the curb andstreet next to it. I don’t think that you needto touch it directly, because it’s a bit smelly,but I want you to step next to it, then touchthe sole of your shoe.

JUNE: Yuck! It’s really dead. It’s gross!THERAPIST: Yeah, it is a bit gross, but it’s also

just a dead cat if you think about it plainly.What harm can it cause?

JUNE: I don’t know. Suppose I get germs on myhand?

THERAPIST: What sort of germs?JUNE: Dead cat germs.THERAPIST: What kind are they?JUNE: I don’t know. Just germs.THERAPIST: Like the bathroom germs that

we’ve already handled?JUNE: Sort of. People don’t go around touching

dead cats.THERAPIST: They also don’t go running home to

shower or alcohol the inside of their car. It’stime to get over this. Now, come on over andI’ll do it first. (June follows.) OK. Touch thecurb and the street. Here’s a stone you cancarry with you and a piece of paper fromunder its tail. Go ahead, take it.

JUNE: (looking quite uncomfortable) Ugh!THERAPIST: We’ll both hold them. Now, touch

it to your front and your skirt, and your faceand hair. Like this. That’s good. What’s youranxiety level?

JUNE: Ugh! 99. I’d say 100, but it’s just short ofpanic. If you weren’t here, it’d be 100.

THERAPIST: You know from past experiencethat this will be much easier in a while. Juststay with it and we’ll wait here. You’re doingfine.

JUNE: (A few minutes pass in which she looksvery upset.) Would you do this if it weren’tfor me?

THERAPIST: Yes, if this were my car and I

dropped my keys here, I’d just pick them upand go on.

JUNE: You wouldn’t have to wash them?

THERAPIST: No. Dead animals aren’t delightful,but they’re part of the world we live in.What are the odds that we’ll get ill from this?

JUNE: Very small, I guess. I feel a little bit betterthan at first. It’s about 90 now.

THERAPIST: Good! Just stay with it now.

The session continued for another 45 min-utes or until anxiety decreased substantially.During this period conversation focused gener-ally on the feared situations and the patient’sreaction to them. The therapist inquired aboutJune’s anxiety level approximately every 10minutes. It is important to note that Juneand the therapist have engaged in conversationthroughout the exposure task, discussing issuessuch as habituation, risk, responsibility, andlong-term outcomes. At the same time it is im-perative to refocus the patient on the exposuretask at hand to ensure that he or she remainsengaged with it. Thus, asking for SUDS ratingsserves two purposes: It provides data aboutfear reduction, and it refocuses the patient onthe exposure. However, if the informal discus-sion serves as a distractor, helping the patient“not think about” what he or she is doing, thetherapist should limit such conversations.

THERAPIST: How do you feel now?

JUNE: Well, it is easier, but I sure don’t feelgreat.

THERAPIST: Can you put a number on it?

JUNE: About 55 or 60, I’d say.

THERAPIST: You worked hard today. You mustbe tired. Let’s stop now. I want you to takethis stick and pebble with you so that youcontinue to be contaminated. You can keepthem in your pocket and touch them fre-quently during the day. I want you to con-taminate your office at work and your apart-ment with them. Touch them to everythingaround, including everything in the kitchen,chairs, your bed, and the clothes in yourdresser. Oh, also, I’d like you to drive yourcar past this spot on your way to and fromwork. Can you do that?

JUNE: I suppose so. The trouble is going homewith all of this dirt.


THERAPIST: Why don’t you call Kenny and planto get home after he does, so he can bearound to help you. Remember, you can al-ways call me if you have trouble.

JUNE: Yeah. That’s a good idea. I’ll just leavework after he does. See you tomorrow.

This scenario illustrates the process of invivo exposure. The therapist answered clearlythe questions raised without detouring fromthe essential purpose of the session, exposureto the feared contaminant. After the initial in-crease, for some patients the anxiety may beginto drop relatively quickly and may require lon-ger for others. As noted previously, it is advis-able to continue the exposure until the patientappears visibly more at ease and reports a sub-stantial decrease in anxiety (40 or 50%).

After 10–15 sessions, the patient’s reportedanxiety level is expected to decrease consider-ably. At the 15th session, June reported a maxi-mum discomfort of 70 SUDs (still somewhathigh, although reduced from 99 SUDs), thatlasted for a few minutes. Her minimal anxietywas 35 SUDs. Her average anxiety level duringthis session was 45 SUDs. Ideally, by the end oftreatment the highest level should not exceed50 SUDs and should drop below 20 SUDs atthe end of the session. In June’s case, morefollow-up sessions were required because heranxiety was still quite high.

To facilitate a transition to normal washingand cleaning behavior, the therapist instituted anormal washing regimen during the third weekof treatment. The patient was allowed one 10-minute shower daily and no more than five 30-second handwashes when there was visible dirton her hands or when they were sticky.

When the therapist arrived for a home treat-ment session the next week, the following con-versation ensued:

THERAPIST: How did it go over the weekend?

JUNE: Not too bad. But I got sort of upset Sat-urday. We went to a picnic and there wereseveral piles of dog dirt around. I had on myflip-flops and I wanted to play volleyball.You can’t in flip-flops, so I went barefoot.

THERAPIST: That’s great! I’m glad to hear it.

JUNE: Yeah, but then I got really upset aboutgoing home and carrying it into the apart-ment. I did it. I walked all over barefoot and

with the flip-flops, but I worried about it foranother whole day, till I talked to Kennyabout my thoughts on Sunday around noon.I felt better when he said he wouldn’t worryabout it. It seems like I feel guilty or some-thing, like the house isn’t clean enough. Butlately if he says it is clean, I’ve been able totake his word for it.

THERAPIST: Well, in time you’ll be able to makethis kind of judgment yourself. How aboutyour washing and cleaning?

JUNE: It was all right. I washed for half a min-ute before I ate, because I was dusty fromplaying volleyball. I deliberately didn’t washwhen I got home, because I felt bad and Iknew that if I did, it would be to “decontam-inate” myself. I showered Saturday nightand I did feel relieved, but I knew I should goand walk around barefoot and touch thefloors I’d walked on. So I did that.

THERAPIST: That’s great! It sounds like youhandled it fine. I’m really pleased. Youavoided washing when it would mean re-ducing feelings of contamination, and youexposed yourself when you felt concernedabout germs. That’s excellent. Now, let’s goover the problem situations that still needwork here at home. What things still disturbyou?

JUNE: The basement. I haven’t done much withthe kitty litter box and old shoes that I threwdown there a year ago because they got con-taminated. The closet still has some contami-nated clothes. And I still worry about thebackyard some. Also the porch. Pigeonshave been perching on the roof and there aredroppings on the railing now, so I thoughtI’d wait until you came to do that.

THERAPIST: OK. Let’s start low and work up.Which is easiest?

JUNE: The basement and closets.

THERAPIST: Fine, down we go.

Exposure to contaminants during the homevisit is conducted in the same manner as thatduring treatment sessions. Typically, home ses-sions last longer, from 2 to 4 hours, until all“dirty” items are touched and “clean” placesare contaminated. These visits should be re-peated if the patient expresses considerableconcern about his or her ability to adopt a per-manent regimen of nonavoidance.


Follow-Up Sessions

June was seen weekly for 3 months, until sheexperienced a setback following the develop-ment of a new obsession. She became con-cerned about hitting a pedestrian while driving.Thoughts that she “might have hit someone”intruded, particularly after turning a corner orglancing in the mirror to change lanes. Onceevoked, they persisted for several hours. Toovercome this new problem, the therapist di-rected her to increase her driving and refrainfrom retracing her path or looking in the mir-ror to check for casualties. June was told thatshe could stop her car only if she knew for cer-tain that she hit someone. Thoughts that it“might” have occurred were to be ignored. Toreduce June’s anxiety about having obsessions(e.g., “Oh, my God, here it is again. This is ter-rible”), she was advised to expect occasionalrecurrences of obsessive thoughts. The fre-quency of obsessions about hitting someonedecreased from several each day to once weeklyafter 3 weeks of self-exposure; the associatedanxiety diminished from 95 to 50 SUDs or less.

Of June’s germ-related obsessions, only thatof dog feces partially recurred. Fears of publicbathrooms and dead animals remained low.The therapist felt that June’s fear of dog feceshad received insufficient attention during treat-ment. To address this return of fear, June wasseen three times a week for 1-hour exposuresessions, in which she touched brown spots onthe sidewalk and walked near, and eventuallystepped on, dog feces. Homework included go-ing to parks, walking on sidewalks withoutlooking, stepping on dog feces, and stepping onthe grass where she thought dogs had been.This treatment continued for 4 weeks and wasreduced to twice a week for an additional 3weeks. Thereafter, June came once weekly foranother 6 weeks, during which the therapist as-signed self-exposure and dealt with June’s ev-eryday concerns. News media coverage of her-pes led to a brief concern about public toilets,but this dissipated within a few days.

In the dialogue below, the therapist reviewedwith June her progress at a 9-month follow-up.

THERAPIST: I’d like to know how you feel com-pared to when you first came here 9 monthsago.

JUNE: I’m definitely a lot better. But, I still havesome bad days when I worry a lot aboutsomething, and I get down on myself. But

when I remember how upset I was last sum-mer and all that washing I did, it’s really awhole lot better. Maybe about 80% better.I’m not ready to be a floor nurse yet, but thejob I got after treatment is pretty good fornow. Kenny and I are doing fine, except he’sreal sensitive if I bring up one of my fears. Iwish he’d just listen and say “OK” or some-thing instead of looking worried about me.It’s like he’s afraid I’m going to get upsetagain. It makes it hard for me to talk freely,but sometimes he does handle it fine. I reallycan’t complain. He’s been through a lot too,when I was really a mess last year and beforethat.

THERAPIST: I’m glad to hear you feel so muchbetter. You look a lot more at ease. Youlaugh more now. I don’t know if you recall,but you never did in the beginning.

JUNE: I remember.

THERAPIST: What’s left now, the other 20%?

JUNE: Obsessions, I guess. I can still work onmy fear of driving over someone. Mostly itlasts less than 15 minutes, but now and thenit hangs on through an evening.

THERAPIST: How often?

JUNE: Once every week or two, I think. And Istill have an urge to avoid walking on thegrass in parks. Like I’m hyperalert. I do itpretty often, but I’m self-conscious.

THERAPIST: You mean you have to remindyourself not to avoid dog feces?

JUNE: Yeah. And I tend to see things in blackand white, all good or all bad. I catch myselffeeling guilty for dumb things like eating des-sert after a full meal. I can stop, but it’s likeI’m out to punish myself or think badlyabout what I did. I have to watch out for it.Still, the thoughts are nothing like they usedto be. I can have fun now. And work is prettyabsorbing, so I can go whole days withoutgetting down on myself for something. Will Ialways do that?

THERAPIST: Maybe to some extent. Weknow that you have a tendency to obsess.Most people who have had an obsessive–compulsive problem say that the rituals andurges to do them decrease more quickly thanthe obsessive ideas. You might have disturb-ing thoughts for a while, but you can expectthem to become less frequent if you’re care-ful not to attempt to control them through


rituals or by avoiding things. Can you han-dle that?

JUNE: I suppose so. They’re not a lot of fun, butI feel like I’m living a normal life again. Isuppose everyone has some problems to dealwith.

Rarely do patients report complete remissionof all obsessions. It is unrealistic to lead a pa-tient to expect that 4 weeks of treatment willresult in a total absence of obsessions and ritu-als. Patients should expect some continuedstruggle with obsessions and urges to ritualize.Strategies for coping with such occasional diffi-culties should be rehearsed.

COMPLICATIONS DURINGBEHAVIORAL TREATMENT

Obviously, difficulties may arise during imple-mentation of EX/RP treatment for OCD. Sev-eral of these are described below and possiblesolutions are discussed.

Noncompliance with Response Prevention

Individuals with OCD often report engaging inrituals despite the response prevention in-structions. In most cases these represent brief“slips” that the therapist addresses by reiterat-ing the rationale for the treatment regimen andthe need to follow the response prevention in-structions strictly. The therapist also may offerways the ritual might be “undone” (e.g.,recontaminating or turning the stove on andoff again).

Sometimes the patient’s support person re-ports violations of response prevention to thetherapist. The therapist should discuss the vio-lations with the patient, emphasizing the factthat continued failure to comply with the re-sponse prevention instructions may result intreatment failure. The following is an exampleof how violations of response prevention maybe presented to the patient.

“I understand from your father that on threeoccasions this weekend he saw you checkingthe front door lock five or six times beforeyou left the house. As we agreed in the firstsession, he called to inform me about yourchecking. I am sure you remember that weagreed that you would check the doors only

once, and that if you had a problem, youwould discuss it with your father or me rightaway, so we could help you overcome yoururge to ritualize. Will you explain to mewhat happened?”

If the patient acknowledges the slip and re-sponds with a renewed agreement to follow in-structions, the therapist need not pursue the is-sue further. However, if a second significantinfraction of the response prevention instruc-tions occurs, the therapist should again remindthe patient of the therapy rules and the ratio-nale for these rules, and “troubleshoot” withthe patient how to successfully implement rit-ual prevention. During the course of this dis-cussion, if it becomes evident that the patient isunwilling to consider these recommendationsand remains committed to rituals and avoid-ance as a means to reduce obsessional distress,then the therapist may broach the subject ofdiscontinuing treatment unless the patient isready to comply.

“It seems that right now you aren’t able tostop ritualizing. For treatment to be success-ful, it is essential that you completely stopyour rituals. Every time that you relieve yourdiscomfort by ritualizing, you prevent your-self from learning that anxiety would havedeclined eventually without rituals, and youdon’t permit your obsessional fears to be dis-connected from distress and anxiety. Ex-posing you to feared situations without stop-ping your rituals won’t be helpful. If youcannot follow the no-rituals rule quitestrictly, then we ought to stop treatment nowand wait until you are really prepared to fol-low through with all the requirements. It isvery hard for people to resist the urge to ritu-alize, and it may be that you are just notready yet and will feel more able to do so inthe future. It is much better for us to stoptreatment now than to continue under con-ditions where you are unlikely to benefitfrom treatment. That would only leave youfeeling more hopeless about future prospectsfor improvement.”

As discussed earlier, patients sometimes re-place identified rituals with less obvious avoid-ance patterns. For example, a patient may usehand lotion to “decontaminate” the hands in-stead of the excessive washing that was doneoriginally. If this occurs, the therapist should


immediately instruct the patient to stop thenew ritual. Other examples of replacementwashing rituals include brushing off one’shands or blowing off “germs”; extensivechecks are often replaced with quick glances.Direct questioning of the patient to solicit suchinformation should proceed as follows:

“Now that you’ve stopped your washing ritu-als, do you find yourself doing other thingsto relieve your anxiety? For example, somepeople start to wipe their hands with papertowels or tissues as a substitute for washingwith soap and water. Are you doing anythinglike this?”

If the answer is “yes,” the therapist shouldidentify these new behaviors as rituals and in-struct the patient to resist engaging in them inthe same manner as he or she resists other com-pulsions.

Continued Passive Avoidance

Patients who continue to avoid situations likelyto evoke obsessional distress are also likely toexperience attentuated outcome in EX/RP. Forexample, a patient may put “contaminated”clothing back in the closet as instructed, but indoing so he/she may ensure that the contami-nated clothes do not touch clean garments.Such avoidance reflects an ambivalent attitudetoward treatment and hinders habituation ofanxiety to feared situations. Because such pro-cesses may hinder outcome, the presence ofcontinued and frequent avoidance behaviorcalls for the therapist and patient to reevaluatewhether the patient should continue treatment.

THERAPIST: Jim, let’s make sure that you are do-ing your homework the right way. I knowthat you had a problem putting your dirtyunderwear in with your other dirty clothes.How are you doing with it now?

PATIENT: Well, I was afraid you might ask that.I still haven’t mixed them up. I was tooscared to do it.

THERAPIST: We discussed this several days agoand you were instructed to do it that night. Itwould have been better had you told me thenext day that you weren’t able to. What I’dlike you to do for tomorrow is to bring insome dirty clothes. Bring in the underwearand the other clothes in separate bags, and

we will mix them here in the office. Are thereany other things you have been avoiding thatyou haven’t told me about?

PATIENT: I don’t think so.

THERAPIST: I want you to pay careful attentionto things you are doing, or not doing, andmake a list of anything you are avoiding,particularly things that you are supposed todo for therapy. It is very important that youdon’t protect yourself by avoiding distressingsituations, since if you don’t face these situa-tions, your obsessive–compulsive symptomswon’t get better. Let’s give it another try, butif you can’t bring yourself to confront theseproblematic situations without these littleavoidances, perhaps you would be better offdelaying your treatment to a later time whenyou will be more ready to comply with thetreatment program.

Arguments

Some individuals who carry out the requiredexposure without ritualizing may attempt toengage the therapist in arguments about the as-signments. It is quite tempting to get involvedin arguments with patients over what they willor will not do during treatment. To avoid this,it is important for the therapist and the patientto agree on some ground rules before the inten-sive program begins. Patients must agree to fol-low the treatment plan that they developed inconjunction with the therapist, and to exposethemselves to the distressing situations withoutargument. New, feared situations that are dis-covered should be discussed, and a new expo-sure program should be developed and agreedto, before exposures to the new situations arecarried out. If a patient balks at or attempts toalter a planned exposure, the therapist shouldacknowledge and empathize with the patient’sthe discomfort, inquire about the reasons forthe hesitation, and encourage the patient toproceed in the following manner:

“I’m sorry to see that you are having so muchtrouble sitting on the floor. I know it’s diffi-cult and that you’re frightened, but it won’tdo you any good if we delay the exposure foranother day or let you skip it all together.You really need to touch the floor, so let’s goahead and do it now. We have agreed that to-day is the ‘floor’ day, and I wouldn’t be do-ing you a favor if I allowed you to avoid it.


Remember, though, I am here to support youas much as I can when you become upset.”

In some instances, difficulties may be over-come by first exposing the patient to similaritems that generate a lower level of distress. Forexample, if a patient refuses to touch a toiletseat, then the therapist may ask him or her tofirst touch the bathroom floor or the door to thebathroom stall. Thereafter, the patient mighttouch the walls of the stall and the toilet handlebefore proceeding to the toilet seat itself.

Emotional Overload

Occasionally during treatment a patient willbecome overwhelmed by fear or another emo-tion that is not directly related to his or herOCD symptoms. For example, a patient maybe upset by a recent event (e.g., the death of arelative) or by fears of facing future plans (e.g.,living on one’s own or getting a job). Imple-menting exposure exercises is inadvisable whenthe patient is extremely upset, because it is un-likely that the patient will adequately attend tothe exposure stimulus; therefore, anxiety isunlikely to habituate. Instead, the therapistshould discuss the distressing situation with thepatient and proceed with exposure only whenthe patient is calmer. On rare occasions, expo-sure may be postponed altogether until thenext day’s session. If this becomes a repetitivepattern, it may be advisable to interrupt treat-ment until the crisis is over.

Nonanxious Reactions to Exposures

Occasionally patients respond to exposureswith emotions other than anxiety or distress,such as anger or depression. Clinical observa-tions suggest that anger often serves as a meansfor the patient to avoid the distress or anxietythat is the target of exposure. If this happens,the anger should be viewed as an avoidance.The therapist should refocus the patient on theanxiety-evoking aspects of the situation andpoint out to the patient that the anger onlystands in the way of progress.

Sometimes during imaginal exposure, whenpatients are exposed to the feared conse-quences of their behaviors, they become de-pressed. Such depression and other emotionalreactions may reduce the efficacy of treatment,and the therapist needs to help the patient tofocus on the anxiety-evoking cues. This may be

done by directing the content of the imaginalexposure away from the feared consequencesand toward the external threat cues. In somecases, such redirection does not resolve theproblem, and the patient continues to display adepressive reaction to the exposure. When thishappens, alternative scenarios that do not elicitdepression should be developed.

Emergent Fears and Rituals

As mentioned earlier, sometimes patients de-velop “new” fears or rituals during treatment.Often, the content of these new symptoms isclosely related to the original fears and may betreated by extending to these fears the EX/RPinstructions given earlier in treatment. For ex-ample, following the successful implementa-tion of response prevention for his compulsivehandwashing, Mr. F began to rub his hands to-gether to decontaminate them. The therapistidentified this as another ritual and instructedMr. F to resist the urge to rub his hands to-gether. Next, Mr. F began subtly to rub his fin-gers against the palms of his hands to cleansehis hands and to reduce anxiety. The therapistasked Mr. F to stop this ritual as he had theothers and was again successful.

Some emergent fears may not be as clearlyconnected to the patient’s original fears. For ex-ample, the fear that June developed of hittingsomeone while driving was not obviously re-lated to her fears of contamination. Further as-sessment often results in the discovery of a con-ceptual link between the two reported fears. InJune’s case, her fear of being blamed for caus-ing someone to become ill or die, and her con-cern about being thought of as a “bad person”because she killed someone, or because shesmelled of dog feces, may have been the con-nection between her two identified fears. Insuch cases, it is important for the therapist todevelop exposures that include cues for thismore general fear. June’s therapist might con-duct imaginal exposures that include images ofpeople criticizing June or blaming her for caus-ing someone to die.

Negative Family Reactions

Because family members have typically experi-enced years of frustration with the patient’ssymptoms, it is not surprising that some areimpatient, expecting treatment to progresssmoothly and to result in total symptom remis-


sion. It is not uncommon for family membersto become disappointed or angry when theyperceive that the symptoms are not subsidingquickly enough. In such cases, the therapistshould assure family members that occasionalstrong anxiety reactions are to be expected anddo not reflect failure. The family should be en-couraged to respond calmly and be supportiveshould the patient experience a burst of anxi-ety.

Often, families have developed patterns ofbehavior designed to reduce the patient’s dis-tress. Some family members may continuethese patterns either in an attempt to protectthe patient from upsetting situations or becauseit is difficult to break habits established overyears of accommodating the patient’s requests.For example, Mr. P, who was accustomed toentering his home through the basement, im-mediately removing his clothes, and showeringfor his wife’s sake, was instructed to enterthrough the front door and toss his overcoat onthe couch. Similarly, family members may findthemselves continuing to perform a variety ofhousehold activities that they have come to re-gard as their responsibility because of the pa-tient’s wishes to avoid the distress that theactivity caused. For instance, Mr. P was respon-sible for preparing all the family meals, becausehis wife was distressed by the possibility thatshe might inadvertently contaminate the food.Because such familiar patterns may hinderprogress in treatment, the therapist should askboth the patient and family members aboutsuch habits and prescribe appropriate alterna-tive behaviors that maximize the patient’s ex-posure and minimize avoidance.

Functioning without Symptoms

At the end of treatment, many individuals withOCD find themselves left with a considerablevoid in their daily routines. The fact that theyno longer need to allocate a large portion oftheir day to performing rituals leaves themwondering what to do. The therapist should besensitive to these issues and aid in planningnew social or occupational goals to be achievedfollowing therapy. If needed, the therapistshould conduct additional sessions or refer thepatient to another therapist who will focus onadjustment-related issues. It may also be thecase that behavioral treatments such as accep-tance and commitment therapy (ACT) are di-rectly applicable to this problem given the ex-

plicit focus on functioning; patients with OCDmight be especially vulnerable to the belief thatthey cannot move forward successfully in theirlives unless their obsessions are gone, and ACTis particularly well suited to address these kindsof problems. Preliminary evidence from a caseseries suggests the applicability of ACT toOCD (Twohig, Hayes, & Masuda, 2006);RCTs, direct comparisons to other establishedtreatments, and further examination of the pu-tative mechanisms that underlie symptom re-duction are now needed.

Because they have spent years performingtheir rituals, patients may be unsure aboutwhat constitutes normal behavior. The thera-pist should offer guidelines for appropriatewashing, checking, repeating, or ordering. Ifrituals are still present, the therapist needs togive instructions to continue the response pre-vention of some behaviors to ensure mainte-nance of treatment gains. A patient may alsodevelop a fear that the OCD symptoms will re-turn. The therapist should reassure the patientthat a single washing of his or her hands doesnot signal the beginning of a relapse.

CONCLUSION

In this chapter we have reviewed the literatureon OCD and its treatment, and provided ver-batim dialogue from patient–therapist interac-tions to demonstrate how EX/RP is imple-mented. Our review illustrates clearly thatmuch is already known about CBT andpharmacotherapy for OCD. In our clinicalpractice with adults, we are guided by the em-pirical research summarized in this chapter, al-though not all of our clinical decisions are un-equivocally supported by empirical studies. Forexample, no controlled, direct comparisonstudy has indicated that intensive EX/RP yieldssuperior outcome to less intensive treatment,yet we typically provide intensive treatment toour adult patients with at least moderately se-vere OCD. Although our clinical experiencesuggests that weekly sessions are probably in-sufficient to produce meaningful gains in mostadult patients with OCD, it has yet to be estab-lished whether two or three weekly sessionswould yield results comparable to daily ses-sions both immediately after treatment and atfollow-up. Future research should examine thisimportant issue to establish a “dose–response”curve for EX/RP. Another important issue is


how to best combine EX/RP with medica-tion. Future research will allow us to identifythe optimal treatment course for a particularpatient.

Empirical results and clinical observationsconverge to indicate that psychosocial treat-ment for OCD must involve both exposure andritual prevention instructions, and that failureto conduct exposures to the most anxiety-evoking situations is likely to compromise out-come. With respect to the therapist-assistedversus self-exposure issue, we routinely choosetherapist-assisted exposure in our clinical prac-tice. At present, eliminating therapist assistancewith exposure exercises seems premature, be-cause existing studies have methodologicalproblems such as insufficient sample sizes.With respect to the role of cognitive interven-tions in the treatment of OCD, the EX/RP pro-gram described in this chapter is a “cognitive-behavioral” treatment in that it targets bothcognitions and behaviors; however, we do nottypically include formal cognitive restructur-ing. Future research needs to delineate whichcognitive and behavioral procedures are mosteffective for correcting particular pathologicalemotions. Cognitive procedures may also beutilized in “readiness programs” designed tohelp patients who are highly ambivalent aboutEX/RP realize that the treatment is both tolera-ble and effective. Empirical research to datesuggest that although antidepressant medica-tions for OCD do not interfere with the effi-cacy of CBT, combination treatment is notnecessarily more effective than EX/RP alone.However, the partial symptom reduction typi-cally found in pharmacotherapy studies forOCD may render some patients more willing totolerate the distress associated with EX/RP;thus, premedication may be helpful in promot-ing readiness in such cases.

What factors seem to enhance long-term effi-cacy of EX/RP for OCD? Studies suggest thatpatients with OCD who show great improve-ment immediately after CBT are more likely toretain their gains at follow-up than those whomake only moderate posttreatment gains (e.g.,Simpson et al., 2004). Thus, emphasis on pro-cedures that are likely to lead to maximalshort-term efficacy also serves to yield superiormaintenance of gains. In our clinical experi-ence, understanding of the treatment rationale,active engagement in exposure exercises, strictadherence to ritual prevention instructions,willingness to design and implement exposure

exercises between sessions, and willingness toconfront even the most difficult tasks on thefear hierarchy are all factors associated withpositive treatment outcome. Thus, verbal rein-forcement of patients when they accomplishthese goals, and reinstruction when they donot, are important in promoting lasting im-provement. In addition, relapse preventiontechniques designed specifically for OCD havebeen found effective in promoting maintenanceof gains at follow-up (Hiss et al., 1994). In clin-ical practice we begin discussing relapse pre-vention procedures long before treatment iscompleted, and focus on maintaining gains inthe last few active treatment sessions. Somecontinuing contact with the treating clinician isalso thought to be of benefit; thus, brief follow-up sessions are held in the first few months af-ter the active treatment is completed, with con-tact as needed following the formal follow-upphase. As part of relapse prevention, we oftenask our patients to plan EX/RP exercises forhypothetical obsessions they might encounterin the future (e.g., “If you became obsessed in 6months that touching tree bark would result inyour contracting a terrible illness, what exer-cises should you do?”) to encourage them toproblem-solve around OCD issues for them-selves rather than relying on the therapist’s in-struction. We also emphasize that the occa-sional occurrence of obsessions should not be acause of great alarm, provided that patients im-plement exposure and ritual prevention tocombat these recurring obsessions and urges toritualize. The patients who are accepting of thisreality are often the ones most able to applywhat they have learned in treatment, and thisprocess enables them to keep their OCD symp-toms under control long after treatment hasterminated.

REFERENCES

Abramowitz, J. S., (1996). Variants of exposure and re-sponse prevention in the treatment of obsessive com-pulsive disorder: A meta-analysis. Behavior Therapy,27, 583–600.

Abramowitz, J. S., & Foa, E. B. (2000). Does major de-pressive disorder influence outcome of exposure andresponse prevention for OCD? Behavior Therapy,31, 795–800.

Abramowitz, J. S., Foa, E. B., & Franklin, M. E. (2003).Exposure and ritual prevention for obsessive–compulsive disorder: Effects of intensive versustwice-weekly sessions. Journal of Consulting andClinical Psychology, 71, 394–398.


Abramowitz, J. S., Franklin, M. E., & Foa, E. B. (2002).Empirical status of cognitive-behavioral therapy forobsessive compulsive disorder: A meta-analytic re-view. Romanian Journal of Cognitive and BehaviorPsychotherapies, 2, 89–104.

Abramowitz, J. S., Franklin, M. E., Street, G. P., Kozak,M. J., & Foa, E. B. (2000). Effects of comorbid de-pression on response to treatment for obsessive–compulsive disorder. Behavior Therapy, 31, 517–528.

Allen, J. J., & Tune, G. S. (1975). The Lynfield Obses-sional/Compulsive Questionnaire. Scottish MedicalJournal, 20, 21–24.



Angst, J. (1993). Comorbidity of anxiety, phobia, com-pulsion and depression. International Clinical Psy-chopharmacology, 8(Suppl. 1), 21–25.

Anderson, R. A., & Rees, C. S. (2007). Group versusindividual cognitive-behavioural treatment forobsessive–compulsive disorder: A controlled trial.Behaviour Research and Therapy, 45(1), 123–137.

Antony, M. M., Downie, F., & Swinson, R. P. (1998).Diagnostic issues and epidemiology in obsessive–compulsive disorders. In R. P. Swinson, M. M. An-tony, S. Rachman, & M. A. Richter (Eds.), Obses-sive–compulsive disorder: Theory, research and treat-ment (pp. 3–32). New York: Guilford Press.

Asbahr, F. R., Castillo, A. R., Ito, L. M., Latorre, R. D.,Moreira, M. N., & Lotufo-Neto, F. (2005). Groupcognitive-behavioral therapy versus sertraline forthe treatment of children and adolescents withobsessive–compulsive disorder. Journal of the Ameri-can Academy of Child and Adolescent Psychiatry,44(11), 1128–1136.

Barrett, P., Farrell, L., Dadds, M., & Boulter, N. (20050.Cognitive-behavioral family treatment of childhoodobsessive–compulsive disorder: Long-term follow-upand predictors of outcome. Journal of the AmericanAcademy of Child and Adolescent Psychiatry, 44,1005–1014.

Barrett, P., Healy-Farrell, L., & March, J. S. (2004).Cognitive-behavioral family treatment of childhoodobsessive–compulsive disorder: A controlled trial.Journal of the American Academy of Child and Ado-lescent Psychiatry, 43, 46–62.

Barsky, A. J., & Ahern, D. (2004). Interventions forhypochondriasis in primary care: In reply. Journal ofthe American Medical Association, 292(1), 42–43.

Basoglu, M., Lax, T., Kasvikis, Y., & Marks, I. M.(1988). Predictors of improvement in obsessive–compulsive disorder. Journal of Anxiety Disorders, 2,299–317.

Beck, A. T. (1976). Cognitive therapy and the emotionaldisorders. New York: International UniversitiesPress.

Bellodi, L., Scuito, G., Diaferia, G., Ronchi, P., &Smeraldi, E. (1992). Psychiatric disorders in the fami-lies of patients with obsessive–compulsive disorder.Psychiatry Research, 42(2), 111–120.

Bogels, S. M., & Bodden, D. (2005, November). Familyversus child CBT for childhood anxiety disorders:Short and long-term results of a multicenter study. InS. M. Bogels (Chair), Family-based prevention andtreatment of childhood anxiety disorders. Sympo-sium presented at the annual meeting of the Associa-tion for Behavioral and Cognitive Therapies, Wash-ington, DC.

Brown, T., Campbell, L., Lehman, C., Grisham, J., &Mancill, R. (2001). Current and lifetime comorbidityof the DSM-IV anxiety and mood disorders in a largeclinical sample. Journal of Abnormal Psychology,110(4), 585–599.

Carr, A. T. (1974). Compulsive neurosis: A review of theliterature. Psychological Bulletin, 81, 311–318.

Comings, D. E. (1990). Tourette syndrome and humanbehavior. Duarte, CA: Hope Press.

Constans, J. I., Foa, E. B., Franklin, M. E., & Mathews,A. (1995). Memory for actions in obsessive compul-sives with checking rituals. Behaviour Research andTherapy, 33, 665–671.

Cottraux, J., Mollard, L., Bouvard, M., Marks, L.,Sluys, M., Nury, A. M., et al. (1990). A controlledstudy of fluvoxamine and exposure in obsessive–compulsive disorder. International Clinical Psycho-pharmacology, 5, 17–30.

Cottraux, J., Note, I., Yao, S. N., Lafont, S., Note, B.,Mollard, E., et al. (2001). A randomized controlledtrial of cognitive therapy versus intensive behaviortherapy in obsessive compulsive disorder. Psycho-therapy and Psychosomatics, 70(6), 288–297.

De Araujo, L. A., Ito, L. M., Marks, I. M., & Deale, A.(1995). Does imaginal exposure to the consequencesof not ritualising enhance live exposure for OCD?: Acontrolled study: I. Main outcome. British Journal ofPsychiatry, 167, 65–70.

de Haan, E., Hoogduin, K. A., Buitelaar, J. K., &Keijsers, G. P. (1998). Behavior thearpy versusclomipramine for the treatment of obsessive–compulsive disorder. Journal of the American Acad-emy of Child and Adolescent Psychiatry, 37, 1022–1029.

Denys, D., Tenney, N., van Megen, J. G., de Geus, F., &Westernberg, H. G. (20040. Axis I and II comorbidi-ty in a large sample of patients with obsessive–compulsive disorder. Journal of Affective Disorders,80, 155–162.

DeVeaugh-Geiss, J., Landau, P., & Katz, R. (1989).Treatment of OCD with clomipramine. PsychiatricAnnals, 19, 97–101.

Diniz, J. B., Rosario-Campos, M. C., Shavitt, R. G.,Curi, M., Hounie, A. G., Brotto, S. A., et al.(2004). Impact of age at onset and duration of ill-ness on the expression of comorbidities inobsessive–compulsive disorder. Journal of ClinicalPsychiatry, 65, 22–27.


Dollard, J., & Miller, N. L. (1950). Personality and psy-chotherapy: An analysis in terms of learning, think-ing and culture. New York: McGraw-Hill.

Dougherty, D. D., Rauch, S. L., & Jenike, M. A. (2002).Pharmacological treatments for obsessive compulsivedisorder. In P. E. Nathan & J. M. Gordon (Eds.), Aguide to treatments that work (2nd ed., pp. 387–410). New York: Oxford University Press.

Eisen, J. L., Phillips, K. A., Baer, L., Beer, D. A., Atala,K. D., & Rasmussen, S. A. (1998). The Brown As-sessment of Beliefs Scale: Reliability and validity.American Journal of Psychiatry, 155, 102–108.

Ellis, A. (1962). Reason and emotion in psychotherapy.New York: Lyle Stuart.

Emmelkamp, P. M. G., & Beens, H. (1991). Cognitivetherapy with obsessive–compulsive disorder: A com-parative evaluation. Behaviour Research and Ther-apy, 29, 293–300.

Emmelkamp, P. M. G., de Haan, E., & Hoogduin, C. A.L. (1990). Marital adjustment and obsessive–compulsive disorder. British Journal of Psychiatry,156, 55–60.

Emmelkamp, P. M. G., & van Kraanen, J. (1977).Therapist-controlled exposure in vivo: A comparisonwith obsessive–compulsive patients. Behaviour Re-search and Therapy, 15, 491–495.

Emmelkamp, P. M. G., Visser, S., & Hoekstra, R. J.(1988). Cognitive therapy vs. exposure in vivo in thetreatment of obsessive–compulsives. Cognitive Ther-apy and Research, 12, 103–114.

Fals-Stewart, W., Marks, A. P., & Schafer, J. (1993). Acomparison of behavioral group therapy and individ-ual behavior therapy in treating obsessive compulsivedisorder. Journal of Nervous and Mental Disease,181, 189–193.

Flament, M., Koby, E., Rapoport, J. L., Berg, C., Zahn,T., Cox, C., et al. (1990). Childhood obsessive com-pulsive disorder: A prospective follow-up study. Jour-nal of Child Psychology and Psychiatry, and AlliedDisciplines, 31, 363–380.

Flament, M., Whitaker, A., Rapoport, J. L., Davies, M.,Berg, C., Kalikow, K., et al. (1988). Obsessive com-pulsive disorder in adolescence: An epidemiologicalstudy. Journal of the American Academy of Childand Adolescent Psychiatry, 27, 764–771.

Foa, E. B., Abramowitz, J. S., Franklin, M. E., & Kozak,M. J. (1999). Feared consequences, fixity of belief,and treatment outcome in patients with obsessivecompulsive disorder. Behavior Therapy, 30, 717–724.

Foa, E. B., Franklin, M. E., & Moser, J. (2002). Contextin the clinic: How well do CBT and medicationswork in combination? Biological Psychiatry, 51,989–997.

Foa, E. B., Huppert, J. D., & Cahill, S. P. (2006). Emo-tional processing theory: An update. In B. O.Rothbaum (Ed.), Pathological anxiety: Emotionalprocessing in etiology and treatment (pp. 3–24). NewYork: Guilford Press.

Foa, E. B., Huppert, J. D., Leiberg, S., Langner, R.,

Kichic, R., & Hajcak, G., et al. (2002). TheObsessive–Compulsive Inventory: Development andvalidation of a short version. Psychological Assess-ment, 14(4), 485–495.

Foa, E. B., & Kozak, M. J. (1985). Treatment of anxietydisorders: Implications for psychopathology. In A. H.Tuma & J. D. Maser (Eds.), Anxiety and the anxietydisorders (pp. 421–452). Hillsdale, NJ: Erlbaum.

Foa, E. B., & Kozak, M. J. (1986). Emotional process-ing of fear: Exposure to corrective information. Psy-chological Bulletin, 99, 20–35.

Foa, E. B., & Kozak, M. J. (1996). Psychological treat-ments for obsessive compulsive disorder. In M. R.Mavissakalian & R. P. Prien (Eds.), Long-term treat-ments of anxiety disorders (pp. 285–309). Washing-ton, DC: American Psychiatric Press.

Foa, E. B., Kozak, M. J., Goodman, W. K., Hollander,E., Jenike, M. A., & Rasmussen, S. (1995). DSM-IVfiled trial: Obsessive compulsive disorder. AmericanJournal of Psychiatry, 152, 90–96.

Foa, E. B., Kozak, M. J., Steketee, G., & McCarthy, P.R. (1992). Treatment of depressive and obsessive–compulsive symptoms in OCD by imipramine andbehavior therapy. British Journal of Clinical Psychol-ogy, 31, 279–292.

Foa, E. B., Liebowitz, M. R., Kozak, M. J., Davies, S.O., Campeas, R., Franklin, M. E., et al. (2005).Treatment of obsessive compulsive disorder by expo-sure and ritual prevention, clomipramine, and theircombination: A randomized, placebo-controlledtrial. American Journal of Psychiatry, 162, 151–161.

Foa, E. B., Steketee, G., Grayson, J. B., Turner, R. M., &Latimer, P. (1984). Deliberate exposure and blockingof obsessive–compulsive rituals: Immediate and long-term effects. Behavior Therapy, 15, 450–472.

Foa, E. B., Steketee, G., Turner, R. M., & Fischer, S. C.(1980). Effects of imaginal exposure to feared disas-ters in obsessive–compulsive checkers. Behaviour Re-search and Therapy, 18, 449–455.

Franklin, M. E., Abramowitz, J. S., Bux, D. A., Zoellner,L. A., & Feeny, N. C. (2002). Cognitive-behavioraltherapy with and without medication in the treat-ment of obsessive compulsive disorder. ProfessionalPsychology: Research and Practice, 33, 162–168.

Franklin, M. E., Abramowitz, J. S., Furr, J., Kalsy, S., &Riggs, D. S. (2003). A naturalistic examination oftherapist experience and outcome of exposure andritual prevention for OCD. Psychotherapy Research,13, 153–167.

Franklin, M. E., Abramowitz, J. S., Kozak, M. J., Levitt,J., & Foa, E. B. (2000). Effectiveness of exposure andritual prevention for obsessive compulsive disorder:Randomized compared with non-randomized sam-ples. Journal of Consulting and Clinical Psychology,68, 594–602.

Franklin, M. E., Kozak, M. J., Cashman, L., Coles, M.,Rheingold, A., & Foa, E. B. (1998). Cognitive behav-ioral treatment of pediatric obsessive compulsive dis-order: An open clinical trial. Journal of the American


Academy of Child and Adolescent Psychiatry, 37,412–419.

Freeman, J. B., Choate-Summers, M. L., Moore, P. S.Garcia, A. M., Sapyta, J. J., Leonard, H. L., et al.(2007). Cognitive behavioral treatment of youngchildren with obsessive compulsive disorder. Biologi-cal Psychiatry, 61, 337–343.

Freeman, J. B., Garcia, A. M., Fucci, C., Karitani, M.,Miller, L., & Leonard, H. L. (2003). Family-basedtreatment of early-onset obsessive–compulsive disor-der [Special issue: Obsessive–compulsive disorder].Journal of Child and Adolescent Psychopharmacol-ogy, 13(2), 71–80.

Freeston, M. H., Ladouceur, R., Gagnon, F., Thibodeau,N., Rhéaume, J., Letarte, H., et al. (1997). Cognitive-behavioral treatment of obsessive thoughts: A con-trolled study. Journal of Consulting and Clinical Psy-chology, 65, 405–413.

Gershuny, B. S., Baer, L., Jenike, M. A., Minichiello, W.E., & Wilhelm, S. (2002). Comorbid posttraumaticstress disorder: Impact on treatment outcome forobsessive–compulsive disorder. American Journal ofPsychiatry, 159, 852–854.

Goodman, W. K., Price, L. H., Rasmussen, S. A.,Mazure, C., Delgado, P., Heninger, G. R., et al.(1989b). The Yale–Brown Obsessive–CompulsiveScale: II. Validity. Archives of General Psychiatry, 46,1012–1016.

Goodman, W. K., Price, L. H., Rasmussen, S. A.,Mazure, C., Fleischmann, R. L., Hill, C. L., et al.(1989a). The Yale–Brown Obsessive–CompulsiveScale: I. Development, use, and reliability. Archivesof General Psychiatry, 46, 1006–1011.

Greist, J. H. (1990). Treatment of obsessive–compulsivedisorder: Psychotherapies, drugs, and other somatictreatments. Journal of Clinical Psychiatry, 51, 44–50.

Greist, J. H. (1992). An integrated approach to treat-ment of obsessive compulsive disorder. Journal ofClinical Psychiatry, 53(Suppl.), 38–41.

Greist, J. H., Jefferson, J. W., Kobak, K. A., Katzelnick,D. J., & Serlin, R. C. (1995). Efficacy and tolerabilityof serotonin reuptake inhibitors in obsessive compul-sive disorder: A meta-analysis. Archives of GeneralPsychiatry, 46, 53–60.

Hanna, G. L. (1995). Demographic and clinical featuresof obsessive compulsive disorder in children and ado-lescents. Journal of the American Academy of Childand Adolescent Psychiatry, 34, 19–27.

Hiss, H., Foa, F. B., & Kozak, M. J. (1994). A relapseprevention program for treatment of obsessive com-pulsive disorder. Journal of Consulting and ClinicalPsychology, 62, 801–808.

Hodgson, R. J., Rachman, S., & Marks, L. M. (1972).The treatment of chronic obsessive–compulsive neu-rosis: Follow-up and further findings. Behaviour Re-search and Therapy, 10, 181–189.

Hohagen, F., Winkelmann, G., Rasche-Raeuchle, H.,Hand, I., Konig, A., Munchau, N., et al. (1998).Combination of behaviour therapy with fluvoxamine

in comparison with behaviour therapy and placebo:Results of a multicentre study. British Journal of Psy-chiatry, 173, 71–78.

Hollander, E., & Stein, D. J. (2006). Clinical manual ofimpulse-control disorders. Arlington, VA: AmericanPsychiatric Publishing.

Insel, T. R., & Akiskal, H. (1986). Obsessive–compulsive disorder with psychotic features: A phe-nomenologic analysis. American Journal of Psychia-try, 12, 1527–1533.

Jaycox, L. H., Foa, E. B., & Morral, A. R. (1998). Influ-ence of emotional engagement and habituation onexposure therapy for PTSD. Journal of Consultingand Clinical Psychology, 66, 185–192.

Jenike, M., Baer, L., Minichiello, W., Schwartz, C., &Carey, R. (1986). Concomitant obsessive–compulsivedisorder and schizotypal personality disorder. Ameri-can Journal of Psychiatry, 143, 530–532.

Kazarian, S. S., Evans, D. L., & Lefave, K. (1977). Mod-ification and factorial analysis of the Leyton Obses-sional Inventory. Journal of Clinical Psychology, 33,422–425.

Keijsers, G. P., Hoogduin, C. A., & Schaap, C. P. (1994).Predictors of treatment outcome in the behaviouraltreatment of obsessive–compulsive disorder. BritishJournal of Psychiatry, 165, 781–786.

Kessler, R. C., Demier, O., Frank, R. G., Olfson, M.,Pincus, H. A., Walters, E. E., et al. (2005). Prevalenceand treatment of mental disorders, 1990 to 2003.New England Journal of Medicine, 352(24), 2515–2523.

Koran, L. M. (2000). Quality of life in obsessive–compulsive disorder. Psychiatric Clinics of NorthAmerica, 23, 509–517.

Kozak, M. J., & Foa, E. B. (1994). Obsessions, overval-ued ideas, and delusions in obsessive compulsive dis-order. Behaviour Research and Therapy, 32, 343–353.

Kozak, M. J., Foa, E. B., & Steketee, G. (1988). Processand outcome of exposure treatment with obsessive–compulsives: Psychophysiological indicators of emo-tional processing. Behavior Therapy, 19, 157–169.

Kurlan, R., Como, P. G., Miller, B., Palumbo, D.,Deeley, C., Andersen, E. M., et al. (2002). The behav-ioral spectrum of tic disorders: A community-basedstudy. Neurology, 59, 414–420.

Ladoucer, R., Rhéame, J., Freeston, M. H., Aublet, F.,Jean, K., Lachance, S., et al. (1995). Experimentalmanipulations of responsibility: An analogue test formodels of obsessive–compulsive disorder. BehaviourResearch and Therapy, 33, 937–946.

Lang, P. J. (1979). A bio-informational theory of emo-tional imagery. Psychophysiology, 6, 495–511.

Leckman, J. F., & Chittenden, E. H. (1990). Gilles de laTourette syndrome and some forms of obsessive–compulsive disorder may share a common geneticdiathesis. L’Encephale, 16, 321–323.

Ledley, D. R., Pai, A., & Franklin, M. E. (2007).Treating comorbid presentations: Obsessive compul-


sive disorder, anxiety, and depression. In M. M. An-tony, C. Purdon, & L. Summerfeldt (Eds.), Psycho-logical treatment of OCD: Fundamentals andbeyond (pp. 281–293). Washington, DC: AmericanPsychological Association Press.

Lelliott, P. T., Noshirvani, H. F., Basoglu, M., Marks, I.M., & Monteiro, W. O. (1988). Obsessive–compulsive beliefs and treatment outcome. Psycho-logical Bulletin, 18, 697–702.

Leon, A. C., Portera, L., & Weissman, M. M. (1995).The social costs of anxiety disorders. British Journalof Psychiatry, 166(Suppl.), 19–22.

Lindsay, M., Crino, R., & Andrews, G. (1997). Con-trolled trial of exposure and response prevention inobsessive compulsive disorder. British Journal of Psy-chiatry, 171, 135–139.

Lochner, C., & Stein, D. J. (2003). Heterogeneity ofobsessive–compulsive disorder: A literature review.Harvard Review of Psychiatry, 11(3), 113–132.

Lovell, K., Cox, D., Haddock, G., Jones, C., Raines, D.,Garvey, R., et al. (2006). Telephone administeredcognitive behaviour therapy for treatment of obses-sive compulsive disorder: Randomised controllednon-inferiority trial. British Medical Journal.com,333, 883.

March, J. S., Franklin, M. E., Leonard, H., Garcia, A.,Moore, P., Freeman, J., et al. (2007). Tics moderatethe outcome of treatment with medication but notCBT in pediatric OCD. Biological Psychiatry, 61,344–347.

Marks, I. M., Lelliott, P., Basoglu, M., Noshirvani, H.,Monteiro, W., Cohen, D., et al. (1988). Clomi-pramine self-exposure, and therapist-aided exposurefor obsessive–compulsive rituals. British Journal ofPsychiatry, 152, 522–534.

Marks, I. M., Stern, R. S., Mawson, D., Cobb, J., &McDonald, R. (1980). Clomipramine and exposurefor obsessive–compulsive rituals—I. British Journalof Psychiatry, 136, 1–25.

Masellis, M., Rector, N. A., & Richter, M. A. (2003).Quality of life in OCD: Differential impact of obses-sions, compulsions, and depression comorbidity. Ca-nadian Journal of Psychiatry, 48(2), 72–77.

Mataix-Cols, D., Marks, I. M., Greist, J. H., Kobak, K.A., & Baer, L. (20020. Obsessive–compulsive symp-toms dimensions as predictors of compliance withand response to behaviour therapy: Results from acontrolled trial. Psychotherapy and Psychosomatics,71, 255–262.

Mathews, A. M., Johnston, D. W., Shaw, P. M., &Gelder, M. G. (1974). Process variables and the pre-diction of outcome in behavior therapy. British Jour-nal of Psychiatry, 125, 256–264.

Matsunaga, H., Kiriike, N., Matsui, T., Oya, K., Okino,K., & Stein, D. (2005). Impulsive disorders in Japa-nese adult patients with obsessive–compulsive disor-der. Comprehensive Psychiatry, 46, 43–49.

McLean, P. L., Whittal, M. L., Thordarson, D. S.,Taylor, S., Sochting, I., Koch, W. J., et al. (2001).

Cognitive versus behavior therapy in the grouptreatment of obsessive–compulsive disorder. Journalof Consulting and Clinical Psychology, 69, 205–214.

Mehta, M. (1990). A comparative study of family-basedand patients-based behavioural management inobsessive–compulsive disorder. British Journal ofPsychiatry, 157, 133–135.

Meyer, V. (1966). Modification of expectations in caseswith obsessional rituals. Behaviour Research andTherapy, 4, 273–280.

Meyer, V., & Levy, R. (1973). Modification of behaviorin obsessive–compulsive disorders. In H. E. Adams &P. Unikel (Eds.), Issues and trends in behavior ther-apy (pp. 77–136). Springfield, IL: Charles C.Thomas.

Meyer, V., Levy, R., & Schnurer, A. (1974). A behavioraltreatment of obsessive-compulsive disorders. In H. R.Beech (Ed.), Obsessional states. London: Methuen.

Mowrer, O. H. (1939). A stimulus–response analysis ofanxiety and its role as a reinforcing agent. Psycholog-ical Review, 46, 553–565.

Mowrer, O. H. (1960). Learning theory and behavior.New York: Wiley.

Neziroglu, F., Stevens, K. P., Yaryura-Tobias, J. A., &McKay, D. (2000). Predictive validity of the Over-valued Ideas Scale: Outcome in obsessive–compulsiveand body dysmorphic disorder. Behaviour Researchand Therapy, 39, 745–756.

Ost, L. G. (1989). One-session treatment for specificphobias. Behaviour Research and Therapy, 27, 1–7.

O’Sullivan, G., Noshirvani, H., Marks, I., Monteiro,W., & Lelliott, P. (1991). Six-year follow-up after ex-posure and clomipramine therapy for obsessive–compulsive disorder. Journal of Clinical Archives, 52,150–155.

Pato, M. T., Zohar-Kadouch, R., Zohar, J., & Murphy,D. L. (1988). Return of symptoms after discontinua-tion of clomipramine in patients with obsessive–compulsive disorder. American Journal of Psychiatry,145, 1521–1525.

Pauls, D. L., Towbin, K. E., Leckman, J. F., Zahner, G.E., & Cohen, D. J. (1986). Gilles de la Tourette syn-drome and obsessive–compulsive disorder. Archivesof General Psychiatry, 43, 1180–1182.

Pediatric OCD Treatment Study Team. (2004).Cognitive-behavioral therapy, sertraline, and theircombination for children and adolescents withobsessive–compulsive disorder: The Pediatric OCDTreatment Study (POTS) randomized controlledtrial. Journal of the American Medical Association,292, 1969–1976.

Piacentini, J., Bergman, R. L., Jacobs, C., McCracken, J.T., & Kretchman, J. (2002). Open trial of cognitivebehavior therapy for childhood obsessive–compulsive disorder. Journal of Anxiety Disorders,16, 207–219.

Purdon, C., & Clark, D. A. (2002). The need to controlthoughts. In R. Frost & G. Steketee (Eds.), Cognitive


approaches to obsessions and compulsions: Theory,research, and treatment (pp. 29–43). Oakland, CA:New Harbinger.

Rabavilas, A. D., Boulougouris, J. C., & Perissaki, C.(1979). Therapist qualities related to outcome withexposure in vivo in neurotic patients. Journal ofBehavior Therapy and Experimental Psychiatry, 10,293–299.

Rachman, S. (1998). A cognitive theory of obsessions:Elaborations. Behaviour Research and Therapy, 36,385–401.

Rachman, S., & DeSilva, P. (1978). Abnormal and nor-mal obsessions. Behaviour Research and Therapy,16, 233–248.

Rachman, S., Thordarson, D. S., Shafran, R., & Woody,S. R. (1995). Perceived responsibility: Structure andsignificance. Behaviour Research and Therapy, 33(7),779–784.

Rapoport, J. L., Swedo, S. E., & Leonard, H. L. (1992).Childhood obsessive compulsive disorder (144th An-nual Meeting of the American Psychiatric Associa-tion: Obsessive compulsive disorder: Integrating the-ory and practice [1991, New Orleans, LA]). Journalof Clinical Psychiatry, 53(Suppl. 4), 11–16.

Rasmussen, S. A., & Eisen, J. L. (1989). Clinical fea-tures and phenomenology of obsessive–compulsivedisorder. Psychiatric Annals, 19, 67–73.

Rasmussen, S. A., & Eisen, J. L. (1990). Epidemiologyof obsessive–compulsive disorder. Journal of ClinicalPsychiatry, 51, 10–14.

Rasmussen, S. A., & Tsuang, M. T. (1986). Clinicalcharacteristics and family history in DSM IIIobsessive–compulsive disorder. American Journal ofPsychiatry, 1943, 317–382.

Reed, G. E. (1985). Obsessional experience and com-pulsive behavior: A cognitive structural approach.Orlando, FL: Academic Press.

Rhéaume, J., Freeston, M. H., Dugas, M. J., Letarte, H.,& Ladouceur, R. (1995). Perfectionism, responsibil-ity, and obsessive compulsive symptoms. BehaviourResearch and Therapy, 36, 385–402.

Riggs, D. S., Hiss, H., & Foa, E. B. (1992). Marital dis-tress and the treatment of obsessive–compulsive dis-order. Behavior Therapy, 23, 585–597.

Rothbaum, B. O., & Shahar, F. (2000). Behavioral treat-ment of obsessive–compulsive disorder in a naturalis-tic setting. Cognitive and Behavioral Practice, 7,262–270.

Rowe, M. K., & Craske, M. G. (1998). Effects of an ex-panding-space vs. massed exposure schedule on fearreduction and return of fear. Beahviour Research onTherapy, 36, 701–717.

Rubenstein, C. S., Peynircioglu, Z. F., Chambless, D. L.,& Pigott, T. A. (1993). Memory in sub-clinicalobsessive–compulsive checkers. Behaviour Researchand Therapy, 31(8), 759–765.

Salkovskis, P. M. (1985). Obsessional compulsive prob-lems: A cognitive-behavioral analysis. Behaviour Re-search and Therapy, 23, 571–583.

Sher, K. J., Frost, R. O., & Otto, R. (1983). Cognitive

deficits in compulsive checkers: An exploratorystudy. Behaviour Research and Therapy, 21, 357–364.

Simpson, H. B., Liebowitz, M. R., Foa, E. B., Kozak, M.J., Schmidt, A. B., Rowan, V., et al. (2004). Post-treatment effects of exposure therapy andclomipramine in obsessive–compulsive disorder. De-pression and Anxiety, 19, 225–233.

Stampfl, T. G., & Levis, D. J. (1967). Essentials of im-plosive therapy: A learning-based psychodynamicbehavioral therapy. Journal of Abnormal Psychology,72, 496–503.

Stanley, M. A., Turner, S. M., & Borden, J. W. (1990).Schizotypal features in obsessive–compulsive disor-der. Comprehensive Psychiatry, 31, 511–518.

Steketee, G., Chambless, D. L., & Tran, G. Q. (2001).Effects of Axis I and II comorbidity on behavior ther-apy outcome for obsessive–compulsive disorder andagoraphobia. Comprehensive Psychiatry, 42, 76–86.

Steketee, G., Eisen, J., Dyck, I., Warshaw, M., & Ras-mussen, S. (1999). Predictors of course in obsessive-compulsive disorder. Psychiatry Research, 89, 229–238.

Steketee, G. S., Foa, E. B., & Grayson, J. B. (1982). Re-cent advances in the treatment of obsessive–compulsives. Archives of General Psychiatry, 39,1365–1371.

Storch, E., Geffken, G., Merlo, L., Mann, G., Duke, D.,Munson, M., et al. (2007). Family-based cognitive-behavioral therapy for pediatric obsessive–compulsive disorder: Comparison of intensive andweekly approaches. Journal of American Academy ofChild and Adolescent Psychiatry, 46, 469–478.

Storch, E. A., Murphy, T. K., Geffken, G. R., Soto, O.,Sajid, M., Allen, P., et al. (in press). Development ofthe Children’s Florida Obsessive–Compulsive Inven-tory (C-FOCI). Journal of Behavior Therapy and Ex-perimental Psychiatry.

Swedo, S. E., Leonard, H. L., Garvey, M., Mittleman,B., Allen, A. J., Perlmutter, S., et al. (1998). Pediatricautoimmune neuropsychiatric disorders associatedwith streptococcal infections: Clinical description ofthe first 50 cases. American Journal of Psychiatry,155, 264–271.

Swedo, S. E., Rapoport, J. L., Leonard, H. L., Lenane,M., & Cheslow, D. (1989). Obsessive compulsivedisorder in children and adolescents: Clinical phe-nomenology of 70 consecutive cases. Archives ofGeneral Psychiatry, 46, 335–341.

Thoren, P., Asberg, M., Chronholm, B., Jornestedt, L.,& Traskman, L. (1980). Clomipramine treatment ofobsessive–compulsive disorder: I. A controlled clini-cal trial. Archives of General Psychiatry, 37, 1281–1285.

Tolin, D. F., Maltby, N., Diefenbach, G. J., Hannan, S.E., & Worhunsky, P. (2004). Cognitive-behavioraltherapy for medication nonresponders withobsessive–compulsive disorder: A wait-list-controlledopen trial. Journal of Clinical Psychiatry, 65, 922–931.


Tukel, R., Polat, A., Ozdemir, O., Aksut, D., & Turksov,N. (2002). Comorbid conditions in obsessive–compulsive disorder. Comprehensive Psychiatry, 43,204–209.

Twohig, M. P., Hayes, S. C., & Masuda, A. (2006). In-creasing willingness to experience obsessions: Accep-tance and commitment therapy as a treatment for ob-sessive compulsive disorder. Behavior Therapy, 37,3–13.

Valderhaug, R., Larsson, B., Gotestam, K. G., &Piacentini, J. (2007). An open clinical trial ofcognitive-behaviour therapy in children and adoles-cents with obsessive–compulsive disorder adminis-tered in regular outpatient clinics. Behaviour Re-search and Therapy, 45, 577–589.

Valleni-Basille, L. A., Garrison, C. Z., Jackson, K. L.,Waller, J. L., McKeown, R. E., Addy, C. L., et al.(1994). Frequency of obsessive–compulsive disorderin a community sample of young adolescents. Journalof the American Academy of Child and AdolescentPsychiatry, 33, 782–791.

van Balkom, A. J., de Haan, E., van Oppen, P.,Spinhoven, P., Hoogduin, K. A., Vermeulen, A. W.A., et al. (1998). Cognitive and behavioral therapiesalone and in combination with fluvoxamine in thetreatment of obsessive compulsive disorder. Journalof Nervous and Mental Disease, 186, 492–499.

van Minnen, A., & Hagenaars, M. (2002). Fear activa-tion and habituation patterns as early process predic-tors of response to prolonged exposure treatment inPTSD. Journal of Traumatic Stress, 15, 359–367.

Verdellen, C. W., Keijsers, G. P., Cath, D. C., &Hoogduin, C. A. (2004). Exposure with responseprevention versus habit reversal in Tourette syn-drome: A controlled study. Behaviour Research andTherapy, 42, 501–511.

Vogel, P. A., Stiles, T. C., & Götestam, K. G. (20040.Adding cognitive therapy elements to exposure ther-apy for obsessive–compulsive disorder: A controlledstudy. Behavioural and Cognitive Psychotherapy, 32,275–290.

Warren, R., & Thomas, J. C. (2001). Cognitive-behavior therapy of obsessive–compulsive disorder inprivate practice: An effectiveness study. Journal ofAnxiety Disorders, 15, 277–285.

Warwick, H. M., Clark, D. M., Cobb, A. M., &Salkovskis, P. M. (1996). A controlled trial ofcognitive-behavioural treatment of hypochondriasis.British Journal of Psychiatry, 169(2), 189–195.

Watts, F. N. (1973). Desensitization as an habituationphenomenon: II. Studies of interstimulus intervallength. Psychological Reports, 33, 715–718.

Weissman, M. M., Bland, R. C., Canino, G. J., Green-wald, S., Hwu, H. G., Lee, C. K., et al. (1994). Thecross national epidemiology of obsessive compulsivedisorder: The Cross National Collaborative Group.Journal of Clinical Psychiatry, 55(Suppl.), 5–10.

Whittal, M. L., Thordarson, D. S., & McLean, P. D.(2005). Treatment of obsessive–compulsive disorder:Cognitive behavior therapy vs. exposure and re-sponse prevention. Behaviour Research and Therapy,43, 1559–1576.


C H A P T E R 5

Emotional DisordersA UNIFIED PROTOCOL

LAURA B. ALLENR. KATHRYN MCHUGH

DAVID H. BARLOW

In this chapter we describe the most recently developed treatment protocol from the Cen-ter for Anxiety and Related Disorders at Boston University. In this “unified” protocol, thera-peutic principles common to psychological treatment of the various emotional disordershave been distilled and integrated into a single protocol that is, in theory, applicable to thefull range of emotional disorders. Unlike protocols for specific problems or disorders de-scribed in other chapters, this new approach has not yet benefited from efforts to establishextensive empirical validation. But the components that make up this treatment haveachieved wide and deep empirical support, and what is new about this approach is theinteraction of these components and the systematic manner in which they are appliedacross disorders. The major advantage, of course, in addition to greatly simplifying dis-semination by eliminating numerous overlapping single-disorder protocols, is that this ap-proach also takes into consideration the extensive comorbidity often found among emo-tional disorders. This approach is illustrated in the treatment of “Oscar.”—D. H. B.

When the first edition of this book was pub-lished in 1985 it was the dawn of evidence-based psychological treatments. Descriptionsof only those treatments with sufficient empiri-cal support that were broadly applicable tolarge numbers of individuals with variousforms of psychopathology at that time were in-cluded. Over succeeding editions of this booknew treatment approaches with wide appealand strong empirical support have been added,whereas others have been deleted. In addition,the field has matured to the point where publichealth services around the world have directedthat evidence-based psychological treatments

become an integral part of health care deliverysystems due to their effectiveness, efficiency,and durability (British Psychological Society,2004; Nathan & Gorman, 2002). It is also asign of the maturity of a field to examineclosely the limitations of existing evidence. Ob-viously, a considerable number of patients stilldo not respond as well as would be desirable toour current arsenal of psychological (or drug)treatments, and there is plenty of room for im-provement. Another problem that has becomeapparent, particularly in the context of emo-tional disorders, is that there now exists aplethora of treatment protocols for each of the

216

discrete anxiety and mood disorders. Althoughthese protocols, by and large, have proven use-ful and been well received, it takes a significantamount of training to become sufficiently fa-miliar with the distinct protocols to integratethem into clinical practice. For example, in thearea of depression, a recent National Instituteof Mental Health (NIMH) Task Force specifiedthree priorities for treatment development, in-cluding more efficacious psychological treat-ments and more “user-friendly” protocols(Hollon et al., 2002). Unless these treatmentsbecome more “user-friendly” as recommended,clinicians are less likely to have a sufficient un-derstanding of, or access to, these evidence-based treatments for the emotional disorders.In this chapter, we present, for the first time(the fourth edition) in this book, a unifiedtreatment protocol for the emotional disorders.“Emotional disorders” in our conception in-clude not only the anxiety and mood disordersbut also other classes of disorders in whichemotional dysregulation plays a prominentrole, such as the somatoform disorders, disso-ciative disorders, and, to some extent, the eat-ing disorders. Borderline personality disorderalso may be conceptualized as a disorder of ex-treme emotional dysregulation (Linehan &Dexter-Mazza, Chapter 9, this volume). Al-though readers should be forewarned that thisprotocol is still a “work in progress,” our sci-ence and practice have advanced sufficientlythat there are now several strong arguments fordeveloping such an approach, in addition tothe very practical advantage of substantially re-ducing the number of existing protocols. Wetouch on these arguments here before describ-ing the latest iteration of this protocol in somedetail, in keeping with the long-standing for-mat of this book.

RATIONALE FORA UNIFIED APPROACH

Perhaps the strongest argument for a unifiedtreatment approach to emotional disorders isan emerging body of evidence supporting com-monalities in the etiology of these disordersthat we have summarized recently in the formof a new etiological model referred to as “triplevulnerability” (Barlow, 1991, 2000, 2002;Suárez, Bennett, Goldstein, & Barlow, inpress). Nevertheless, it is important to note thatshared etiological pathways or pathophysio-

logical processes are not yet firmly established,and that emotional disorders and their sub-types may each turn out to be associated withunique underlying pathology (reflecting a truecategorical organization) with identifiabletaxa. Yet we believe that the evidence on sharedpathological processes is sufficiently strong atpresent to provide further justification for de-veloping and testing a unified treatment ap-proach, even if its mechanisms of action, pro-vided they are proven efficacious, do not turnout to be associated with fundamental changesin hypothetical specific pathophysiology.

A second argument focuses on conceptionsof the major emotional disorders that empha-size their commonalities rather than their dif-ferences. The “spectrum” approach is onemanifestation of this conception. For example,high rates of comorbidity suggest considerableoverlap among disorders. The observed effectsof current psychological treatments on comor-bid conditions also point to at least a partialnonspecificity of treatment response. From aphenomenological perspective, emerging re-search on the latent structure of dimensionalfeatures of emotional disorders is revealing ahierarchical structure that can accommodatethese disorders. The sections to follow brieflyreview evidence relevant to these arguments.

Etiology

We have described in some detail an interactingset of vulnerabilities or diatheses relevant to thedevelopment of anxiety, anxiety disorders, andrelated emotional disorders. This “triple vul-nerability” theory encompasses a generalizedbiological vulnerability, a generalized psycho-logical vulnerability, and a specific psychologi-cal vulnerability emerging from early learning(Barlow, 2000, 2002; Suárez et al., in press). Ageneralized biological vulnerability involvesnonspecific genetic contributions to the devel-opment of anxiety and negative affect. Much ofthe research on this generalized biological vul-nerability has focused on temperaments labeled“neuroticism,” “negative affect,” or “behav-ioral inhibition.” Although the relationshipsamong these closely related traits and tempera-ments have yet to be worked out fully, it seemsthat they substantially overlap, and that each(partially) represents a common theme associ-ated with a biological vulnerability to developemotional disorders generally (Barlow, 2000,2002; Campbell-Sills, Liverant, & Brown,

Emotional Disorders 217

2004; Suárez et al., in press). Additionally,early life experiences fostering a sense thatevents, particularly negative events, are un-predictable or uncontrollable and are ac-companied by weak or nonexistent copingmechanisms, contribute to a generalizedpsychological vulnerability, or diathesis, to ex-perience anxiety and related negative affectivestates later on. This sense of uncontrollabilityseems to be at the core of negative affect andderivative states of anxiety (and depression). Ifa generalized biological and a generalized psy-chological vulnerability happen to line up, andare potentiated by the influence of life stress,the likely result is the clinical syndromes ofgeneralized anxiety disorder and/or depressivedisorders. Notice that false alarms (panic at-tacks) may also occur as a function of stressfullife events, facilitated by high levels of baselineanxiety. But these false alarms seem to have adifferent heritability than anxiety/negative af-fect and are not necessarily implicated in a clin-ical disorder. For that to occur, an additionallayer of a more specific psychological vulnera-bility must be considered. In particular, certainlearning experiences seem to focus anxiety onspecific life circumstances, and these circum-stances or events are associated with a height-ened sense of threat or danger. For example,specific early learning experiences seem to de-termine whether individuals may view somaticsensations, intrusive thoughts, or social evalua-tion as specifically dangerous (Barlow, 2002;Bouton, Mineka, & Barlow, 2001); that is, totake one example, individuals with social anxi-ety often have in their background admoni-tions from parents or family always to be ontheir best behavior and to look their best toavoid the dreaded consequence of being “dis-approved” by others. It is this specific psycho-logical vulnerability that, when coordinatedwith the generalized biological and psychologi-cal vulnerabilities mentioned earlier, seems tocontribute to the development of discrete anxi-ety disorders such as social phobia, obsessive–compulsive disorder, panic disorder, and spe-cific phobias. Existing evidence for this “triplevulnerability” model has been reviewed in de-tail elsewhere (Barlow, 2000, 2002; Bouton etal., 2001; Chorpita & Barlow, 1998; Suárez etal., in press). Whereas confirmation of thismodel awaits further research, it is consistentwith the overriding importance of commonfactors in the genesis and presentation of emo-tional disorders.

Latent Structureof the Emotional Disorders

Whereas the Diagnostic and Statistical Manualof Mental Disorders, fourth edition (DSM-IV;American Psychiatric Association, 1994), rep-resents the zenith of a “splitting approach” tonosology in an attempt to achieve high ratesof diagnostic reliability, this achievement mayhave come at the expense of diagnostic validity;that is, the current system may be highlightingcategories that are minor variations of morefundamental, underlying syndromes. Newquantitative approaches using structural equa-tion modeling are capable of examining the fullrange of anxiety and mood disorders withoutthe constraints of existing (possibly) artificialcategories (Brown, Chorpita, & Barlow, 1998;Chorpita, Albano, & Barlow, 1998; Clark,2005; Clark & Watson, 1991; Watson, 2005;Watson, Clark, & Harkness, 1994) as recentlyarticulated in DSM-V research planning com-mittees (Kupfer, First, & Regier, 2002). Indeed,a recent special issue of the Journal of Ab-normal Psychology is devoted to this topic(Krueger, Watson, & Barlow, 2005). We havebeen studying this question for over a decade.For example, Brown and colleagues (1998), us-ing a sample of 350 patients with DSM-IV anx-iety and mood disorders, confirmed a hierar-chical structure to the emotional disorders. Inthis structure, negative affect and positive af-fect were identified as crucial, higher-order fac-tors to the DSM-IV disorder factors, with sig-nificant paths from negative affect to each ofthe five DSM-IV factors. Interestingly, low pos-itive affect emerged with significant paths tothe mood disorders and social phobia only.In this model, autonomic arousal representsthe phenomenon of panic, and this arousalemerges as a lower-order factor with significantpaths to panic disorder and generalized anxietydisorder (where the relationship was negative).From these findings it seems safe to concludethat what is common to emotional disordersoutweighs what is not (DSM-IV factors).We have concluded from this research thatDSM-IV emotional disorder categories arebest considered as useful concepts or constructsthat emerge as “blips” on a general back-ground of neuroticism/behavioral inhibition,but may not be the best way to organizenosology. It is likely that DSM-V will turn to amore dimensional description of these phe-nomena that would reinforce conceptions of


common underlying components (Kruegeret al., 2002, 2005).

Overlap among Disorders

Anxiety and mood disorders are overlappingand strongly related. We have reviewed the ge-netic and neurobiological evidence on the com-monalities among anxiety and depressive statesin some detail (Barlow, 2002, Chapters 3, 6, 7,and 8; Bouton, 2005; Brown, 2007; Suárez etal., in press). For example, most work on ge-netic contributions to anxiety and depressionsupports the early dictum of Ken Kendler(1996; Kendler et al., 1995) “same genes, dif-ferent environment” (e.g., Eley & Brown,2007; Hettema, Neale, & Kendler, 2001;Rutter, Moffit, & Caspi, 2006; Smoller &Tsuang, 1998). Davidson and colleagues havereported robust findings supporting increasedrelative right-hemisphere interior electroen-cephalographic (EEG) activation linked toheightened negative affect and decreased posi-tive affect in individuals who are either de-pressed or anxious (Davidson, 2000; Wiede-mann, Pauli, & Dengler, 1999). Nemeroff andhis colleagues (Cameron, Champagne, & Par-ent, 2005; Heim & Nemeroff, 1999; Sullivan,Kent, & Coplan, 2000) are focusing on whole-brain systems associated with stress, anxiety,and depression, in which corticotropin-releasing factor (CRF) plays an essential role,with implications for common neurobiologicalunderpinnings. Of course, this is work in prog-ress, and many of its nuances and qualifica-tions are beyond the scope of these few briefsummary sentences.

At the diagnostic level, this is most evident inthe high rates of current and lifetime comorbid-ity (e.g., Brown, Campbell, Lehman, Grisham,& Mancill, 2001; Kessler et al., 1996, 1998;Roy-Byrne, Craske, & Stein, 2006; Tsao,Mystowski, Zucker, & Craske, 2002, 2005).Results from Brown, Campbell, and colleagues(2001) indicate that 55% of patients with aprincipal anxiety disorder had at least one ad-ditional anxiety or depressive disorder at thetime of assessment. But if one examines for thepresence of a disorder over the lifetime of thepatient, whether it is present or not at the timeof interview, this rate increases to 76%. To takeone example, 60% of 324 patients diagnosedwith panic disorder with or without agorapho-bia (PDA/PD) were determined to meet criteriafor either an additional anxiety or mood disor-

der, or both. Specifically 47% presented withan additional anxiety disorder and 33% withan additional mood disorder. When lifetime di-agnoses are considered, the percentages rise to77% experiencing any anxiety or mood disor-der, breaking down to 56% for any anxiety dis-order and 60% for any mood disorder. If post-traumatic stress disorder (PTSD) or generalizedanxiety disorder (GAD) were the principal(most severe) diagnoses, comorbidity rateswere highest. Merikangas, Zhang, andAveneoli (2003) followed almost 500 individu-als for 15 years and found that relatively fewpeople suffer from anxiety or depression alone.When a single disorder did occur at one pointin time, the additional mood state would al-most certainly emerge later.

Several possible explanations for these highrates of comorbidity have been reviewed exten-sively elsewhere (Brown & Barlow, 2002).Among these are relatively trivial issues withoverlapping definitional criteria; artifactualreasons, such as differential base rates of occur-rence in our setting; and the possibility that dis-orders are sequentially related, and that thefeatures of one disorder act as risk factors foranother disorder. For example, depressionseems to follow PDA, and PDA seems to followPTSD. But the more intriguing explanation, forour purposes, offered by individuals such asGavin Andrews or Peter Tyrer (Andrews, 1990,1996; Tyrer, 1989; Tyrer et al., 1998) is thatthis pattern of comorbidity argues for the exis-tence of what has been called a “general neu-rotic syndrome.” Andrews and Tyrer have sug-gested that differences in the expression ofemotional disorder symptoms (individual vari-ation in the prominence of social anxiety, panicattacks, anhedonia, etc.) is simply a trivial vari-ation in the manifestation of a broader syn-drome. This, in turn, is consistent with “triplevulnerability” models mentioned earlier, thatanxiety (and mood) disorders emerge fromshared psychosocial and biological/geneticdiatheses. If this is the case, then a unified treat-ment protocol cutting across current diagnosticcategories to address core features of the anxi-ety and mood disorders could be a more parsi-monious, and, perhaps, powerful option.

The Broad Effects of PsychologicalTreatments for Emotional Disorders

Other findings that may support our conten-tion of an emerging “general neurotic syn-


drome” or “negative affect syndrome” includethe observation that specific psychologicaltreatments for a given anxiety disorder producesignificant improvement in additional comor-bid anxiety or mood disorders that are not spe-cifically addressed in treatment (Borkovec,Abel, & Newman, 1995; Brown, Antony, &Barlow, 1995; Tsao, Lewin, & Craske, 1998;Tsao et al., 2002, 2005). In our own center, weexamined the course of additional diagnoses ina sample of 126 patients being treated for PDA.At pretreatment, 26% had an additional diag-nosis of GAD, but the rate of comorbid GADdeclined significantly to 9% after treatment forPDA and remained at this level at a 2-yearfollow-up (Brown et al., 1995). More recently,a multisite study examining second-stage treat-ments for PDA found that patients who wereconsidered treatment responders after receiving11 sessions of cognitive-behavioral therapy(CBT) for PDA showed significant reductionsin both number and frequency of comorbid di-agnoses, whereas those who were considered“nonresponders” showed no reductions incomorbidity (Allen et al., 2007). This couldrepresent either the generalization of elementsof treatment to independent facets of both dis-orders or a way of effectively addressing“core” features of emotional disorders. In bothcases, the efficiency of a unified treatment pro-tocol is suggested.

In summary, the existing literature supportsseveral arguments for stepping back from indi-vidual DSM-IV diagnostic categories and asso-ciated specific psychological protocols, andconsidering a more unified approach based onnew findings on the nature of emotion regula-tion and dysregulation. In addition, currentCBT protocols for emotional disorders havemuch in common and reduce to three broadprinciples of change: altering emotion-basedmisappraisals of salient events; preventingavoidance of negative, emotionally charged in-ternal or external triggers; and modifyingemotion-driven behaviors. A full explanationof this protocol and a description of the appli-cation of the protocol with a patient follows.

TREATMENT VARIABLES

Setting

All assessments and treatments of patients areconducted in the Center for Anxiety and Re-lated Disorders at Boston University. Our clinic

receives over 500 new admissions per year,with many patients being offered treatment af-ter their initial intake assessments. The Centerfor Anxiety and Related Disorders, in additionto housing staff psychologists and a psychia-trist, is also a training center for doctoral stu-dents and psychiatric residents. At any giventime, numerous NIMH-funded treatment andresearch studies are ongoing at the Center. Re-garding diagnostic breakdown of treatment-seeking patients, the most common diagnosisassigned is GAD, followed by social phobia,PDA, specific phobia, obsessive–compulsivedisorder (OCD), and PTSD. A small percentageof patients are assigned “coprincipal diagno-ses,” referring to cases where two separate di-agnoses are judged to be of equal severity.

Individuals seeking assessments and/or treat-ment at the Center are typically given a briefscreening over the telephone or in person to de-termine eligibility for services. Individuals whoappear to be eligible due to reports of emo-tional dysfunction of some type are then sched-uled for an intake assessment. The assessmentis conducted as soon as possible, although it isrequired that patients currently taking psycho-active medications or in therapy be medicationand/or psychotherapy stable for 3 months priorto the date of the assessment (1 month stabili-zation is required for use of benzodiazepines);that is, they cannot have recently begun a treat-ment that could make their assessment unreli-able due to a rapid change in their condition.For individuals tapering off a medication orpsychotherapy, the “wash out” period is 1month.

Prior to the intake evaluation, each patient ismailed a packet of questionnaires to bring tothe assessment, which are scored and inter-preted subsequent to the interview. The ques-tionnaires provide additional self-report datain examining hypothesized underlying symp-toms and factors associated with diagnosticcategories. The intake evaluation includes theadministration of two semistructured inter-views, typically conducted during a single 4-hour appointment. The UCLA Life StressInterview (Hammen, 1991) is a short interviewdesigned to assess for chronic and episodicstressors occurring over the past 6 months. Themajority of the intake appointment comprisesadministration of the Anxiety Disorders Inter-view Schedule for DSM-IV—Lifetime Version(ADIS-IV-L; Di Nardo, Brown, & Barlow,1994). Following completion of the entire as-


sessment process, consensus diagnoses are de-termined during a weekly staff meeting, afterwhich the patient is provided with diagnosticfeedback and treatment recommendations.Based on the information from the interview,the patient may be offered one of several treat-ment options at the Center, or a referral in thecommunity.

Format

Treatment with the unified protocol (UP) istypically conducted in an individual format, al-though it has been successfully conducted in agroup format with patients with mixed princi-pal diagnoses (anxiety and mood disorders).The treatment described in this chapter reflectsthe individual treatment protocol, which al-lows for greater attention to the descriptionand application of treatment components dur-ing each session. However, when administeredin a group setting, patients were easily able tosee the commonalities among their diverse pre-senting complaints, and this understanding of-ten created a strong bond among group mem-bers. Therefore, it may be useful to consideradapting the protocol for group treatment inthe future.

Therapist Variables

Thus far, the UP has been administered by ther-apists of varying degrees of clinical experienceand expertise with the protocol. To our knowl-edge, both junior therapists (even therapistswho have never had experience with cognitive-behavioral treatments) and senior therapists(i.e., therapists with at least 4 or more years oftreatment experience) have been able to adaptto the UP without significant difficulty. Cer-tainly, a background in cognitive-behavioraltechniques may be helpful when utilizing theprotocol (e.g., facilitating cognitive reap-praisal, and designing and conducting expo-sures). Recent evidence suggests some benefitto therapist experience, at least when treatingPDA with a highly structured CBT protocol(Huppert et al., 2001).

The UP is an emotion-focused treatment ap-proach; that is, the focus of every exercise is oneliciting and changing behavioral responses toa variety of emotions and emotional cues. Per-haps one of the most challenging aspects forthe therapist is to be able to create and utilizeemotion-provoking exposures effectively,

which often begin at the first treatment session.Most importantly, the therapist must have asense of when a patient is avoiding the processof experiencing, expressing, or accepting emo-tions, and this is often signaled by very subtlebehavioral cues, such as avoiding eye contact,changing the topic of discussion, arriving latefor session, and not completing (or “overdo-ing”) homework assignments. Each of thesebehaviors represents attempts to control un-comfortable emotions either through directavoidance or overcontrol. It is essential that thetherapist be able to recognize and address suchbehaviors when they occur, thereby allowingthe opportunity for an effective emotion expo-sure and discussion of avoidance to facilitateemotional processing.

A second challenge for the therapist is to beable to tolerate and experience the expressionof emotion by the patient. It is common for lessexperienced therapists to quickly “rationalizeaway” the patient’s emotional reactions; how-ever, this only feeds into the cycle of emotionsand avoidance. At every step, the therapistshould encourage the expression and accep-tance of emotion, while guiding the patient inhow to “examine” it, without letting the emo-tion “take over.” In such cases, modeling froma senior therapist and/or extensive supervisionmay be helpful instruction for less experiencedtherapists in how to allow their patients to be“emotional.”

Client Variables

As previously mentioned, the rate of comorbid-ity among patients with a principal anxiety ormood disorder is approximately 55% (Brown,Campbell, et al., 2001), depending on the prin-cipal diagnosis. In most CBT protocols, comor-bid diagnoses are never a focus of treatment,although successful treatment of a principalanxiety disorder often results in decreases incomorbidity (e.g., Allen et al., 2007; Brown etal., 1995). One advantage of a unified treat-ment approach is that symptoms related tocomorbid diagnoses can be discussed in treat-ment sessions, and may even be the focus of anemotion exposure. For example, a patient withGAD who experiences chronic worry aboutdaily matters may also feel anxious in socialsituations. Therefore, that patient’s in-sessionemotion exposure might consist of a conversa-tion with a stranger or giving a speech to asmall group of people. Contrary to traditional


protocols, it is the experience of any emotionthat is the target of treatment, which may beparticularly beneficial for patients with signifi-cant comorbidity or for those who would liketo address multiple concerns during the courseof treatment.

Concurrent Drug Treatment

Many patients who present for treatment arealso taking some form of psychoactive medica-tion. As described earlier, patients are requiredto have stabilized on the dosage of medicationprior to the intake interview, so that the thera-pist has a clear picture of actual symptoms (asopposed to symptoms that may be caused bythe initial addition or removal of a medica-tion). Whereas concurrent use of medicationssuch as tricyclic antidepressants and selectiveserotonin reuptake inhibitors (SSRIs) does notappear to have a negative impact on treatmentoutcome initially, once the medications are re-moved, evidence suggests that patients who re-ceived medication alone or CBT in addition tomedication are more likely to relapse (e.g.,Barlow, Gorman, Shear, & Woods, 2000;Heimberg et al., 1998; Liebowitz et al., 1999).The exact mechanisms underlying the return ofsymptoms are unclear, although some authorshave suggested that patients may attribute thesuccess of the treatment to the use of medica-tion and, once it is withdrawn, no longer be-lieve they have the ability to manage theirsymptoms. However, in a recent investigationusing the data from a multisite comparativestudy for the treatment of panic disorder, re-sults indicated that patients who received CBTplus pill placebo were equally likely to believethey had received medication as patients whohad actually received the medication (Raffa etal., in press). Yet those in the CBT plus placebocondition were less likely to relapse comparedto those who had received either CBT plusmedication or medication alone. Therefore, itappears unlikely that the higher relapse rates inthe medication conditions were due solely tothis “attribution hypothesis.” One other hy-pothesis to explain greater relapse in patientsreceiving psychopharmacological treatment isthat the medication provides an unintended“protective” effect against increased physiolog-ical arousal and anxiety. However, elicitinganxiety and panic is a core component of CBTprotocols, so patients taking medication (1)may have experienced greater physical sensa-

tions when the medication was removed and/or(2) were never able fully to confront the physi-ological arousal and panic during the course oftreatment. However, these hypotheses have yetto be fully investigated.

One additional consideration is the use ofbenzodiazepines on a prn, or “as needed,” ba-sis during the course of treatment. Followingconceptions outlined in the UP, any strategyused to reduce the intensity of emotions in themoment is considered an avoidance strategy,and ultimately contributes to increasing levelsof anxiety and emotional reactivity. Therefore,the use of benzodiazepines (or other fast-actingmedications) is discouraged, particularly if themedications are carried with the person as a“safety signal.”

ASSESSMENT

Case Study

Oscar is a 28-year-old graduate student in thesciences, studying at a university in the UnitedStates. He is engaged to be married and livingwith his fiancée. Several months prior to refer-ral to the Clinic, he encountered a difficult situ-ation while working under a vent in a chemis-try laboratory when a distinct odor of a toxicgas produced noticeable tingling in his nose.Knowing this chemical well, he was aware thatthe physical consequences of contact with thegas would be delayed for up to 48 hours butcould then result in potentially serious cardio-vascular and pulmonary complications. Oscarimmediately visited the health clinic, where hewas cleared of any obvious toxic reactions, andreturned again 5 days later and, once again,was given a clean bill of health.

Despite these reassurances, Oscar continuedto experience recollections of the event, partic-ularly upon the occurrence of any physical sen-sations resembling those that occurred duringthe event, such as tingling in his nose, slightheadache, or dizziness. As a result, he hadlargely stopped the laboratory work that wasimportant to his progress toward his doctoraldegree and spent a great deal of time monitor-ing his body for physical symptoms and worry-ing about his health. If certain physical sensa-tions occurred, particularly nasal obstructionor tingling, combined with slight headache, hewould call his mother, who was a physician inSouth America and, if these symptoms per-sisted, possibly visit the university health clinic.


Oscar had been prescribed 0.5 mg of clonaze-pam, which he took upon experiencing anxietydue to physical symptoms that lasted morethan several hours. At the time of referral, hewas taking 1 to 2 pills approximately once aweek.

Contact with any substances that might betoxic was also sufficient to trigger Oscar’shealth concerns. For example, while shoppingin a department store he picked up a plate and,after putting it down, noticed a bit of residueon his fingers. Hypothesizing the source of thissubstance as something utilized in the manu-facture and/or shipping of this plate, he experi-enced a surge of anxiety, left the store, and pro-ceeded to the nearest public men’s room in anearby restaurant to wash his hands thor-oughly. This was followed by additional anxi-ety as he continued to notice mild somatic sen-sations in various systems in his body over thenext several days.

In fact, Oscar reported that although the in-cident several months earlier was perhaps themost “traumatic,” he had experienced health-related worries for approximately 5 years, dur-ing which he would focus on various somaticsensations and report them to his mother ortake other steps to ensure that these sensationsdid not represent an incipient illness. Typically,these episodes would occur every 2–3 monthsand were not connected with any specific trig-gering event that Oscar could remember. Con-cern and worry about his health had occupied asignificant portion of most of Oscar’s days forthe past 5 years.

In addition, he reported intrusive thoughtsof varying content, including crashing his carinto an abutment or driving onto the sidewalk;stabbing himself or his fiancée, if he was hold-ing a knife; drinking acid that he might beholding in the laboratory; as well as jumpingoff high places. Objective assessment of hisemotional state via questionnaires resulted inscore of 9 on the Beck Depressive Inventory–II(BDI-II; Beck, Steer, & Brown, 1996); a scoreof 51 on the Penn State Worry Questionnaire(PSWQ; Meyer, Miller, Metzger, & Borkovec,1990), reflecting mild to moderate worry; anda score of 20 on the Anxiety Sensitivity Index(ASI; Peterson & Reiss, 1992), reflecting mildconcern over somatic symptoms. Diagnoses re-vealed a principal diagnosis of hypochondriasiswith a comorbid diagnosis of OCD at a lowerlevel of clinical severity (less than necessary fora DSM-IV diagnosis).

Interviews

A number of structured and semistructuredclinical interviews may be appropriate for diag-nosing Axis I disorders. The Structured ClinicalInterview for DSM-IV (SCID; First, Spitzer,Gibbon, & Williams, 1996) is widely used andassesses for Axis I disorders. Because this inter-view is highly structured and focuses only oncurrent symptom count, it achieves a high de-gree of interrater reliability for many diagno-ses. However, this interview does not includedimensional ratings of frequency and severityof symptoms, so it may be less useful for re-search projects examining treatment outcome.

The Anxiety Disorders Interview Schedulefor DSM-IV—Lifetime Version (ADIS-IV-L; DiNardo et al., 1994), a semistructured, diagnos-tic clinical interview, focuses on DSM-IV diag-noses of anxiety disorders and their accompa-nying mood states, somatoform disorders, andsubstance and alcohol use. The information de-rived from the interview using the ADIS-IV-Lallows clinicians to determine differential diag-noses and gain a clear understanding of thelevel and severity of each diagnosis. Principaland additional diagnoses are assigned a clini-cian severity rating (CSR) on a scale from 0(No symptoms) to 8 (Extremely severe symp-toms), with a rating of 4 or above (Definitelydisturbing/disabling) passing the clinicalthreshold for DSM-IV diagnostic criteria. In-quiries about suicidal ideation are part of thisinterview. This measure has demonstrated ex-cellent to acceptable interrater reliability forthe anxiety and mood disorders (Brown, DiNardo, Lehman, & Campbell, 2001). The fullADIS-IV-L (focusing on current and lifetime di-agnoses) is typically administered at the initialintake. An abbreviated version of the ADISthat focuses only on current symptomatology(Mini-ADIS-IV; Brown, Di Nardo, & Barlow,1994) is used for posttreatment and follow-upassessments. Because the CSR assigned for eachdiagnosis does not follow a normal curve (dueto a “cutoff” score of 4 to indicate diagnoses ofclinical severity), it is typically not used for sta-tistical analyses. However, the CSR is a dimen-sional measure, so it can assess improvement insymptoms over treatment, even if a patient isstill given the diagnosis at posttreatment.

Two additional clinician-rated measures thatprovide a wider range of scores are the (1)Structured Interview Guide for the HamiltonAnxiety Rating Scale (SIGH-A; Shear, Vander


Bilt, & Rucci, 2001) and (2) Structured Inter-view Guide for the Hamilton Depression Rat-ing Scale (SIGH-D; Williams, 1988). TheSIGH-A was developed to create a structuredformat for administering the Hamilton AnxietyRating Scale (HARS; Hamilton, 1959). Re-spondents are asked to indicate the presenceand severity of a number of symptoms thatmay have been present over the past week, in-cluding anxious mood, tension, sleep prob-lems, irritability, and so forth. The rater is alsoasked to rate the interview behavior of the pa-tient. The SIGH-A includes specific instruc-tions on administration and anchor points forassigning severity ratings. This measure dem-onstrated good interrater and test–retest reli-ability. In addition, scores are similar to (al-though consistently higher than) the HARS.

Similar to the SIGH-A, the SIGH-D was de-veloped to provide more specific instructionsfor administration and scoring of the HamiltonDepression Rating Scale (HDRS; Hamilton,1960). Again, patients are questioned aboutthe presence and severity of a range of depres-sive symptoms over the past week, includingdepressed mood, suicidal ideation, fatigue, feel-ings of hopelessness, weight loss, and so forth.The SIGH-D also demonstrated good interraterand test–retest reliability and produces scoressimilar to the HDRS.

Medical Evaluations

Medical evaluations are generally recom-mended prior to assignment of diagnoses andinitiation of treatment to rule out organiccauses for symptoms of emotional disorders.Some conditions such as hypothyroidism,hyperthyroidism, hypoglycemia, mitral valveprolapse, or alcohol or substance withdrawalmay elicit symptoms similar to those associatedwith GAD or PDA. Although the diagnosis ofsuch medical conditions does not preclude theneed for a psychological treatment, it is gener-ally recommended that such conditions be ex-amined by a medical doctor, because an alter-native treatment may be clinically indicated.

Self-Monitoring

Self-monitoring forms are an important part ofthe treatment protocol for several reasons.First, the therapist is able to discuss specific sit-uations or events that occurred over the pastweek and may have contributed to emotional

reactions. Such records can facilitate dis-cussions of concepts presented during thetreatment sessions and help the therapist inte-grate the general treatment components intothe patient’s specific symptoms. Second, someevidence suggests that patients’ retrospectiverecall of past episodes of anxiety may be in-flated, particularly when recalling panic at-tacks (Margraf, Taylor, Ehlers, Roth, & Agras,1987; Rapee, Craske, & Barlow, 1990). Self-monitoring forms allow for a prospective, andpossibly more accurate, account of anxiety epi-sodes, and may therefore be more useful thera-peutically. In addition, consistent with thethemes outlined in the UP, practicing awarenessof emotions in the present moment is believedto be an important component of changing thecycle of emotions. The very nature of self-monitoring requires the patient to disengage,even briefly, from the habitual anxious processto write down concrete thoughts, feelings, be-haviors, and reactions. Developing this habitultimately aids the patient in beginning tochange emotional reactions and resulting be-haviors.

Self-monitoring forms used in the UP includestandardized forms for automatic thoughts,avoidance, interoceptive and situational expo-sure, and emotion (e.g., Weekly Record of Anx-iety and Depression [WRAD]; see Barlow &Craske, 2000; Barlow, Rapee, & Reisner, 2001;Craske, Barlow, & O’Leary, 1992). Figure 5.1illustrates the WRAD used in the UP. This formis given for several weeks at the beginning oftreatment, in the middle of treatment, and inthe final weeks of treatment to allow for com-parison of rates of anxiety and depression ofthe course of the therapy. Once the concept ofemotion-driven behaviors (EDBs) is intro-duced, patients are asked to monitor situationsand change EDBs on the Changing EDBs form(see Figure 5.2).

Several problems may arise with the intro-duction and completion of self-monitoringforms. First, some patients may not be compli-ant with completion of monitoring forms andhomework assignments, which is an importantissue to be addressed in the therapy session.The therapist must identify cognitive mis-appraisals and emotional reasons contributingto the lack of completion of monitoring forms,and help the patient understand that the avoid-ance is part of the cycle of emotions he or she istrying to break. Therapeutic strategies such ascognitive reappraisal may be used to increase


patients’ willingness to complete homeworkforms. In addition, successful completion ofhomework forms may become an emotion ex-posure in and of itself.

One other common problem with the com-pletion of self-monitoring may be the tendencyfor some patients (particularly those with moreobsessive or “perfectionistic” features, such asin GAD and OCD) to “overdo” the homeworkforms; that is, the patient may write an ex-tremely long and involved description of a situ-ation and his or her reactions to it. This is com-mon with patients who feel a need to “unload”every piece of information about the event. Al-though the patient is doing the homework, hisor her overengagement may also be facilitatingthe anxious/worry process. If this becomes evi-dent, the therapist may want to discuss the ten-dency to overengage in the homework as an

emotional avoidance strategy and encouragethe patient to monitor situations and events us-ing one- or two-word descriptions.

Questionnaires

A number of self-report questionnaires areused over the course of treatment. Generalquestionnaires (measures designed to assess arange of symptoms associated with anxiety andmood disorders, as well as symptoms associ-ated with emotion regulation difficulties) areadministered at pretreatment, midtreatment,posttreatment, and follow-up, and are designedto provide a broad picture of a patient’s overallfunctioning and life satisfaction, in addition tointerference from anxiety and depression.Symptom-specific measures are designed totrack symptoms associated with each patient’s


WEEKLY RECORD OF ANXIETY AND DEPRESSION

Each evening, please make the following ratings, using the scale below:

1. Your AVERAGE level of anxiety (taking all things into consideration).2. Your MAXIMUM level of anxiety, experienced at any one point in the day.3. Your AVERAGE level of depression (taking all things into consideration).4. Your AVERAGE level of pleasantness (taking all things into consideration).

Level of Anxiety/Depression/Pleasantness0—————1—————2—————3—————4—————5—————6—————7—————8

None Slight Moderate A lot As muchas I can imagine

Date Average anxiety Maximum anxietyAverage

depressionAverage

pleasantness

FIGURE 5.1. Weekly Record of Anxiety and Depression (WRAD) monitoring form.

226

CH

AN

GIN

GE

DB

s

Situ

atio

n/tr

igge

rE

mot

ion

ED

BN

ew(in

com

patib

le)

resp

onse

Con

sequ

ence

FIG

UR

E5.

2.C

hang

ing

ED

Bs

mon

itor

ing

form

.

particular principal diagnosis and are adminis-tered at each of these assessment points, as wellas on a weekly basis, prior to each session.

General Measures

The Positive and Negative Affect Scale(PANAS; Watson, Clark, & Tellegen, 1988), abrief, reliable, and valid measure of positiveand negative affect, comprises 20 feeling oremotion words. The PANAS assesses core neg-ative affect and deficits in positive affect inthose disorders (e.g., social phobia) with thischaracteristic, and is useful in determiningchanges in positive and negative affect over thecourse of treatment. The BDI-II (Beck et al.,1996) and the Beck Anxiety Inventory (BAI;Beck & Steer, 1990; Steer, Ranieri, Beck, &Clark, 1993) are widely used measures assess-ing current depressive and anxious symptoma-tology, respectively. Both measures contains 21items, each focusing on severity of particularsymptoms occurring over the past week. In aneffort to establish degree of interference fromsymptoms in various domains of living, theWork and Social Adjustment Scale (WSAS;modification of a scale introduced by Hafner& Marks, 1976) includes five items that askparticipants to rate the degree of interferencecaused by their symptoms in work, home man-agement, private leisure, social leisure, andfamily relationships. The WSAS, a descriptivemeasure of subjective interference in variousdomains of living, has been successfully used inprevious studies (e.g., Brown & Barlow, 1995).In addition, the RAND-modified Medical Out-comes Study 36-Item Short Form Health Sur-vey (RAND MOS-SF-36; Hays, Sherbourne,& Mazel, 1993; Ware & Sherbourne,1992) is a well-validated, comprehensive, self-administered instrument that is widely used inmedical and psychiatric settings to provide amultidimensional assessment of mental andphysical health-related status. The SF-36 mea-sures several health-related dimensions, includ-ing physical functioning, bodily pain, role limi-tations due to physical health problems,general mental health, social functioning,energy/fatigue, and general health perceptions.This measure allows for a detailed analysis ofpsychiatric and medical health changes overtreatment. Finally, the Quality of Life Inven-tory (QOLI; Frisch, Cornell, & Villaneuva,1992) consists of 32 items relevant to overalllife satisfaction, including items related to

work, love relationships, friendships, self-regard, standard of living, recreation, commu-nity, home, and so forth.

Emotion Regulation Measures

Measures designed to assess general conceptsrelated to emotion regulation, emotion dysregu-lation, and related variables may also be usefultools to assess treatment outcome. Although anumber of instruments exist, the following areexamples that are commonly used in psycholog-ical research. The abbreviated version of theTrait Meta-Mood Scale (TMMS; Salovey,Mayer, Goldman, Turvey, & Palfai, 1995) is a30-item questionnaire that measures attentionto moods, clarity of emotions, and ability to re-pair emotional experiences. The subscales of theTMMS demonstrate good reliability and valid-ity (Salovey et al., 1995), and have been shownto be consistent with physiological indicators ofstress (Salovey, Stroud, Woolery, & Epel, 2002).The Anxiety Control Questionnaire—Revised(ACQ-R; Brown, White, Forsyth, & Barlow,2004) is a measure of perceived control overemotional reactions and stressful events. Partic-ipants rate how much each statement is typicalof their experience on a 0 (Strongly disagree) to5 (Strongly agree) scale. Also, the Affective Con-trol Scale (ACS; Williams, Chambless, &Ahrens, 1997), a 42-item scale, examines fear oflosing control while experiencing strong emo-tional reactions such as anxiety, depression, an-ger, and positive affective states. These measuresmay be helpful in assessing the degree to whichpatients are better able to manage a variety ofemotional experiences, in addition to the emo-tions that are a target of treatment.

Symptom-Specific Measures

Given the variability in treatment strategies ad-dressed at each session, in addition to life stress-ors that occur during treatment, it is importantto have repeated measures for each patient at ev-ery treatment session. These repeated measurestypically focus on the patient’s principal diagno-sis and related symptoms, although the thera-pist may want to select specific measures to ad-dress additional concerns (e.g., if a patient’sprincipal diagnosis is an anxiety disorder but heor she is also experiencing a significant depres-sive episode, the therapist may also want to ad-minister a measure of depression at each ses-sion).


The Obsessive–Compulsive Inventory—Revised version (OCI-R; Foa et al., 2002), an18-item self-report instrument that measuresthe frequency of and distress caused by a vari-ety of obsessions and compulsions, is used inpatients with a principal diagnosis of OCD.The Panic Disorder Severity Scale—Self-ReportVersion (PDSS-SR; adapted from Shear et al.,1997) is a 7-item measure adapted from theoriginal, clinician-administered version (PDSS;Shear et al., 1997). This questionnaire for usewith individuals already diagnosed with PDAmeasures the frequency, interference of, anddistressed caused by panic attacks. Scoresrange from 0 to 21, with higher scores reflect-ing more severe panic symptomatology. Thismeasure has been used successfully in treat-ment outcome research for patients with PDA/PD (Shear et al., 1997). For patients with aprincipal diagnosis of GAD, the PSWQ (Meyeret al., 1990) is a 16-item measure to assess gen-eral anxiety and worry; scores on the PSWQare significantly higher for those diagnosedwith GAD compared to other anxiety disordersor depression (Brown, Antony, & Barlow,1992; Molina & Borkovec, 1994). The SocialInteraction Anxiety Scale (SIAS; Mattick &Clarke, 1998) measures the extent of an indi-vidual’s anxiety in a variety of social settingsand is primarily used for patients who are be-ing treated for social phobia. The SIAS hasshown good convergent and divergent validity(Heimberg, Mueller, Holt, Hope, & Liebowitz,1992; Mattick & Clarke, 1998) and ade-quately discriminates between individuals diag-nosed with social phobia and other anxiety dis-orders (Brown et al., 1997). Finally, the HealthAnxiety Inventory, short version (HAI;Salkovskis, Rimes, & Warwick, 2002), a mea-sure designed to assess anxiety about one’shealth, is typically used for individuals with aprincipal diagnosis of hypochondriasis. For in-dividuals with a principal diagnosis of a mooddisorder such as major depressive disorder(MDD) or dysthymia, the BDI-II (describedearlier) may be used as a weekly measure. TheBrief Posttraumatic Diagnostic Scale (BPDS),designed to measure PTSD symptoms, can alsobe administered on a weekly basis (Foa,Cashman, Jaycox, & Perry, 1997).

Functional Analysis

Regardless of the diagnosis, a clear functionalanalysis of the patient’s behavior is essential

prior to beginning treatment. Several compo-nents are important to consider whendiagnosing a patient based on a functionalanalysis. These include a close examination ofsymptom topography (including duration ofillness, physical sensations, level of distress andinterference from symptoms), triggers (situa-tions, physical symptoms, places, thoughts,etc.), cognitions (beliefs about symptoms andmisappraisals), behavioral responses to emo-tions (including avoidance of situations, places,people, or triggers, as well as escape behav-iors), and the consequences of behavioral reac-tions (limiting quality of life, reduced “comfortzone,” etc.).

In this case, Oscar presented with anxietyand avoidance of cues that were reminders ofthe lab accident. Many of his symptoms mim-icked those associated with a PTSD reaction,and because Oscar believed his life was in dan-ger at the time of the accident (the gas towhich he was exposed was potentially life-threatening), the therapist considered assigninga diagnosis of PTSD (although Oscar did notmeet full symptom criteria for the diagnosis).Oscar also presented with significant worryand anxiety about his work and several othertopics, which may have been consistent with adiagnosis of GAD. However, after further in-vestigation, it appeared that all the topics ofworry were in some way related to the lab acci-dent, which therefore precluded a diagnosis ofGAD. After exploring Oscar’s lifelong historyof health-related anxiety, as well as his strongbelief that he had been physically hurt by theexposure to the toxic gas (despite reassurancefrom medical personnel that he was physicallyfine) and his engagement in excessivereassurance-seeking from family members, thetherapist assigned a diagnosis of hypochon-driasis.

For Oscar, this functional analysis was par-ticularly important in guiding treatment plan-ning and application of the UP. Beyond diag-nostic considerations, functional analyses arean essential aspect of determining which pro-cedures may be most helpful for patients. Os-car’s extensive avoidance of both the labora-tory, where he was originally exposed totoxic gas, and somatic sensations associatedwith what he believed would indicate toxicpoisoning provided several important con-texts for developing and creating emotion ex-posures. As we have mentioned, one note-worthy difference when comparing the UP


and traditional CBT protocols is that the fo-cus of treatment techniques in the UP is onan individual’s particular emotional symp-toms, whereas CBT protocols apply the samestrategies to all individuals in a specific diag-nostic category. This difference allows the UPtechniques to be tailored individually to eachpatient and further highlights the necessity ofa clear functional analysis rather than reli-ance on diagnostic categories for treatmentplanning.

COMPONENTS OF TREATMENT

The structure of the UP outlines fifteen 45- to50-minute individual treatment sessions, heldon a weekly basis. However the final three ses-sions (Sessions 13, 14, and 15) may be held ev-ery other week to allow patients to consolidategains and to provide somewhat of a “taper” offthe intensive weekly therapy. However, this for-mat for the final sessions is not a requirement,and it may be more beneficial for the patient tocontinue to have weekly sessions if he or shemight potentially have trouble using the treat-ment concepts consistently without the weeklyreinforcement of sessions. Therefore, spacingof the final sessions is determined by the thera-pist, who takes into consideration the patient’sprogress and any anticipated difficulties aftertreatment.

Currently, the UP comprises four basic com-ponents: (1) psychoeducation about emotions,including a review of the functional nature ofemotions and how emotions become disor-dered; (2) alteration of antecedent cognitivemisappraisals, an intensive antecedent-basedemotional regulation procedure that directlyfacilitates the next two steps in treatment; (3)prevention of emotional avoidance; a broad-based effort that goes well beyond traditionalattempts to prevent behavioral avoidance inphobic disorders by targeting cognitive, behav-ioral, and somatic experiential avoidance; and(4) modification of EDBs, which implementsspecific behaviors not associated with the emo-tion that is disordered. This treatment takesplace in the context of provoking emotional ex-pression (emotion exposure) through situa-tional, internal, and somatic (interoceptive)cues, as well as through standard mood induc-tion exercises, and differs from patient to pa-tient only in the situational cues and exercisesutilized. In addition, “exposure” is not concep-

tualized as a mechanism of action. Rather, suc-cessfully provoking emotions creates a settingcondition to implement the essential treatmentcomponents. Emotion exposures begin withgeneral stimuli (e.g., mood induction andinteroceptive provocation) and are later tai-lored to address each patient’s particular con-cerns and symptoms. Although each sessionaddresses a specific component of the protocol,the expectation is that patients then “carrythrough” to future sessions the strategies theylearn (e.g., patients learn cognitive reappraisalstrategies in Session 3 but are expected to con-tinue to use cognitive reappraisal for emotionexposures conducted throughout the remain-der of the treatment).

Psychoeducation

Psychoeducation about the nature and functionof emotions is directly addressed in the first ses-sion. This component is common to most CBTprotocols, although the focus is expanded toinclude the function of many different emo-tions (anger, sadness, etc.), in addition to thefunction of anxiety. Patients are provided withinformation about the cognitive, physiological,and behavioral sequelae of emotional reactionsand how these three components interact. It isimportant that the patient begin to considerthat his or her reactions are functional andserve the purpose of providing informationabout the environment, in addition to protec-tion from harm. This three-component modelis then applied to a recent situation or eventthat the patient experienced, so that he or shemay better understand the aspects of each com-ponent and how they interact.

Another important aspect of psychoeduca-tion is a discussion of negative reinforcement.Specifically, this is a detailed discussion of howthe patient’s behavioral response to an emo-tional episode (usually escape or some form ofemotional avoidance) is problematic, because itreduces the emotion in the short term (i.e., byremoving the person from the emotional stimu-lus) but reinforces the cycle of emotions in thelong term (i.e., by teaching the person that es-cape/avoidance is the only way to manage thesefeelings in the future). It is extremely importantthat the patient understand the connection be-tween behavioral responses and reinforcementof emotion, so that he or she is better able toappreciate the purpose and function of emo-tion exposures introduced in future sessions.


Antecedent Cognitive Reappraisal

Cognitive therapy, initially developed by AaronT. Beck to treat depression (Beck, 1967/1972;Beck, Rush, Shaw, & Emery, 1979), has be-come a fundamental part of psychologicaltreatments. Particularly common in individualswith emotional disorders are tendencies to-ward appraisals and interpretations of externalevents that are characterized by cognitive bi-ases, such as the tendency to overestimate thelikelihood of the occurrence of negative eventsand to underestimate the ability to cope withthese events. Thus, the aim of cognitive therapyis to evaluate objectively the likelihood of thesenegative appraisals and to incorporate more re-alistic, evidence-based appraisals of the out-come of a situation. On the surface, this tech-nique may appear to be a way to suppress orcontrol negative thoughts by “rationalizing”them, and occasionally, this is how cognitivetherapy is incorrectly used, as noted by Hayes,Strosahl, and Wilson (1999). However, thisstrategy can also be conceptualized from anemotion regulation perspective, if reappraisalsof threat and negativity are made prior to theemotionally arousing situation. Data from theemotion regulation literature have demon-strated that use of antecedent cognitive reap-praisals prior to experiencing heightened levelsof emotional arousal can have a salutary effecton the later expression of negative emotions(Gross, 1998; Richards & Gross, 2000;Thayer, 2000). Antecedent cognitive reap-praisal has also been shown to reduce the sub-jective experience of negative emotion (Gross,1998).

The UP has identified two fundamental ante-cedent misappraisals: the probability of a nega-tive event happening (probability overestima-tion) and the consequences if the negative eventdid happen (catastrophizing) (Barlow &Craske, 2000; Craske et al., 1992). Although anumber of other misappraisals are noted inother, traditional CBT protocols, most mis-appraisals can be condensed into one of thetwo appraisals listed here. The patient mustfirst begin to identify misappraisals that occurin anxiety- or emotion-provoking situations.He/she may be able to list a variety of mis-appraisals occurring in a situation, althoughinitially the therapist may choose to focus onjust one appraisal. Generally, the therapist canelicit appraisals by discussing a situation fromeither the self-monitoring forms from the previ-

ous week or a hypothetical situation. If the pa-tient has trouble identifying misappraisals (orany appraisals at all), the therapist uses stan-dard cognitive therapy techniques, such asquestioning what the patient feared wouldhappen in the situation and identifying anythoughts or reaction he or she might have hadat the time. The goal of this process is to helpthe patient recognize that his or her emotionalreactions are due, at least in part, to theappraisals or interpretations made about thesituation. It is also important to note that inter-pretations of situations are based on past expe-rience (e.g., it is very likely that a person whowas teased by another person for giving an in-correct answer in class will appraise similar sit-uations as negative and anxiety-provoking).

One other strategy used in the UP to identifyand illustrate misappraisals is to present an am-biguous picture and ask the patient to interpretwhat is happening. Typically, the interpretationwill in some way be related to the patient’s pri-mary symptoms, and it is useful for the patientto know that he or she is likely to appraise evenneutral situations in a biased way, because hisor her attention is focused on specific cues and/or threats.

The next step is to assist the patient in reap-praising interpretations of situations by examin-ing the evidence for and against each in-terpretation and generating an alternativeappraisal. Again, standard cognitive restructur-ing strategies may be used, although the focus ofcognitive reappraisal is slightly different. First,the patient is not encouraged to “change” his orher thoughts to “more realistic” ones; rather, heor she is asked to allow for other possible inter-pretations that may be more likely based on theevidence, while allowing all possible appraisalsto exist in his or her mind, without attaching toomuch significance to any particular one. Thegoal of antecedent cognitive reappraisal is flexi-bility in thinking and acknowledging that manypossible outcomes may exist.

In addition, it is important that the therapistutilize cognitive strategies to help the patientidentify his or her “core” cognitive misappraisal.This misappraisal, usually a statement aboutthe patient’s self or self-worth (i.e., “I amworthless” or “I am incompetent”), may un-derlie many of his or her surface appraisals invarious situations. Once the core misappraisalhas been identified, the therapist illustrateshow this misappraisal triggers emotional reac-tions in various situations and helps the patient


examine whether the misappraisal is actuallyaccurate.

As mentioned earlier, the emotion-regulatingproperties of cognitive reappraisal are presentonly when reappraisal strategies are used priorto engagement in an emotionally salient task.Patients have a strong tendency to use the reap-praisal techniques at the height of emotionalexperiences, in an attempt to minimize uncom-fortable emotions. However, this is incorrect,and the therapist must be aware when a patientattempts to use reappraisal strategies in thisway. Using reappraisal techniques prior to atask may help the patient be more willing to en-gage in the task and allow for greater emotionprovocation during the task, as opposed to us-ing the techniques at the height of emotion,which is best conceptualized as a form of emo-tional avoidance.

Prevention of Emotional Avoidance

Evidence now points to a key concept in under-standing emotional disorders, attempts todown-regulate or avoid unexpected, excessiveemotions leading to the development of in-creasingly intense emotional experiences—aprocess that is central to depression, anger, andexcitement (mania), in addition to fear(Barlow, 1988; Gross, 2002; Gross &Levenson, 1997). For example, Roemer, Litz,Orsillo, and Wagner (2001) found that veter-ans with PTSD were more likely to withholdnegative and positive emotions compared toveterans without PTSD. Evidence on the delete-rious consequences of avoidance strategies alsoextends to the calming and relaxation tech-niques that were a significant part of earliertreatment protocols (Barlow & Cerny, 1988).When relaxation strategies are used specificallyfor the purpose of reducing uncomfortableemotions in the moment, these techniques canactually become counterproductive. Similar re-sults have been obtained in examinations of theimpact of distraction techniques (Craske,Street, & Barlow, 1989; Craske, Street,Jayaraman, & Barlow, 1991; Kamphuis &Telch, 2000) and safety signals (Salkovskis,Clark, Hackmann, Wells, & Gelder, 1999;Sloan & Telch, 2002; Wells et al., 1995).

The concept of emotional avoidance is likelyto be particularly important for patients whodo not physically avoid or escape situations butstill experience high levels of anxiety withoutmuch relief. The therapist should convey that

even if the person stays physically in the situa-tion, he or she is likely to engage in a number ofother subtle behaviors to prevent full emo-tional arousal. In the UP, emotional avoidancestrategies are behaviors that prevent the full ex-perience of emotion in a situation, as opposedto EDBs, which are behavioral consequences ofthe experience of emotion (described later inthe chapter). Any technique or strategy the pa-tient uses to reduce or to down-regulate emo-tions may be conceptualized as an emotionalavoidance strategy, and it is essential that emo-tional avoidance be eliminated prior to engage-ment in emotion exposures, to allow for fullemotional processing. Therefore, a very de-tailed description of individual emotionalavoidance strategies must be obtained for eachpatient.

We have identified three general categoriesof emotional avoidance strategies: (1) subtlebehavioral avoidance, (2) cognitive avoidance,and (3) safety signals (see Table 5.1 for exam-ples). Subtle behavioral avoidance strategiesmay consist of a number of different behaviors,with certain behaviors occurring more fre-quently with particular disorders. For example,the avoidance of caffeine in panic disorder orsocial phobia (prior to a social engagement) isan attempt to keep physiological symptomsfrom becoming strong in anxiety-provokingsituations. Patients with social phobia may alsoavoid eye contact or wear sunglasses when in-teracting in social settings. Individuals withGAD may excessively plan, prepare, or write“to do” lists in an attempt to control poten-tially negative outcomes. At the same time,procrastination is also a form of subtle behav-ioral avoidance, if a particular task or projectseems too emotionally arousing for the patient.It is very important to conduct a functionalanalysis of the subtle behaviors to determinewhich ones are avoidance strategies, that is, be-haviors that serve to reduce or avoid emotionalexperience and are functional responses to situ-ations. And, notably, what may be an avoid-ance strategy for one person may be a func-tional response for another.

Cognitive avoidance strategies are typicallymore difficult to identify, because they tend tooccur outside of the patient’s awareness. Some ofthese strategies include distraction, “tuningout,” and mentally checking lists or reviewingpast conversations. However, data also suggestthat worry and rumination actually function as away to down-regulate and avoid emotions (e.g.,


Borkovec, 1994). The function of worry is to en-able the person to prepare for a possible threat.However, when someone experiences chronicworry and associated anxiety, his or her atten-tion is focused on the future as opposed to thepresent moment. Research has shown that whenconfronted with an emotionally arousing event,chronic worriers do not experience the full emo-tional impact of that event, because (1) they havebeen “bracing” themselves for something bad tohappen, and (2) their focus has already shifted toother, future negative outcomes (Borkovec,Hazlett-Stevens, & Diaz, 1999). Therefore, theworry functions as a cognitive avoidance strat-egy and only serves as a maladaptive attempt togain control over seemingly uncontrollable fu-ture events. Obsessions, common to individualswith OCD, may also function in a similar man-ner.

Safety signals are most common to patientswith PDA, although may also be present in in-dividuals with GAD, OCD, or other emotionaldisorders. Safety signals include any object that

the patient carries to feel more safe or “com-fortable,” particularly if he or she is entering anemotionally arousing situation. Safety signalscan range from actual medication (such asbenzodiazepines for reducing physiologicalarousal or medications for gastrointestinal dis-tress) to empty medication bottles, to “lucky”objects (e.g., a talisman, teddy bear, “lucky”pen, etc.). Regardless of the object itself, if itsfunction is to help the patient reduce emotionalarousal in the moment, then it also is problem-atic, because it feeds into the cycle of negativereinforcement.

To combat the problems associated withemotional avoidance, the therapist introducesthe concept of “emotional awareness training”to assist the patient in preventing the emotionalavoidance strategies he/she uses regularly.Emotional awareness training comprises sev-eral components, which include staying in thepresent moment, relinquishing behavioralavoidance strategies and safety signals, andfully engaging with emotions as they come and


TABLE 5.1. Examples of Emotional Avoidance Strategies

Emotional Avoidance StrategyDisorder mostusually associated

1. Subtle behavioral avoidance• Avoid eye contact Social phobia• Avoid drinking caffeine PDA• Attempt to control breathing PDA• Avoid exercise/physiological arousal (interoceptive avoidance) PDA/depression• Avoid touching sink/toilet OCD• Procrastination (avoiding emotionally salient tasks) GAD

2. Cognitive avoidance• Distraction (reading a book, watching television) Depression/PDA• “Tune out” during a conversation Social phobia• Reassure self that everything is OK GAD• Try to prevent thoughts from coming into mind OCD• Distraction from reminders of trauma PTSD• Force self to “think positive” Depression• Worry GAD• Rumination Depression• Thought suppression All disorders

3. Safety signals• Carry cell phone PDA/GAD• Carry empty medication bottles PDA• Hold onto “good luck” charms OCD• Carry items that are associated with positive experiences

(e.g., teddy bears, pictures)GAD/depression

• Have mace at all times PTSD• Carry water bottle PDA• Have reading material/prayer books on hand GAD• Carry sunglasses or item to hide face/eyes Social phobia

go. This requires some practice for the patient,and an in-session emotion provocation andpractice of emotional awareness is often help-ful. After identifying the range of strategies, thetherapist asks the patient to begin to monitor(and attempt to prevent) the use of the strate-gies for the remainder of the treatment ses-sions.

Modifying EDBs

As noted by Izard (1971), data and hypothesesfrom emotion theory suggest that the most effi-cient and effective way to change emotions isby changing responses to them; thus, it is con-ceivable that the mechanism of change duringan exposure is to prevent the action tendencyassociated with a particular emotional experi-ence. Over the past several decades, researchhas focused on these action tendencies, andchanging action tendencies has become an im-portant treatment component for anxiety, aswell as for other emotional disorders (Barlow,1988). For example, Beck and colleagues(1979) based a large part of their treatment fordepression on changing the action tendenciesof their patients to behave in a “passive, re-tarded, and apathetic manner” (p. 312). Morerecently, behavioral activation strategies havebecome a central feature of newer treatmentsfor depression (Dimidjian, Martell, Addis, &Herman-Dunn, Chapter 8, this volume;Dimidjian et al., 2006; Jacobson, Martell, &Dimidjian, 2001).

In fact, these strategies go well beyond sim-ply preventing escape from fearful situations.Laughter, humor, and related facial expressionshave been induced during successful paradoxi-cal intention strategies (Frankl, 1960). Thesestrategies have also effectively counteractedfear (Ascher, 1980). Originally, the proposedmechanism of action was thought to be the in-duction of cognitive changes, although it seemsplausible that these strategies may have workedthrough the prevention of behavioral responsesand modification of action tendencies (facialexpressions) previously associated with theemotion. Linehan (1993) adapted the strategyof modifying action tendencies with much suc-cess in her treatment for individuals with bor-derline personality disorder. Hayes’s accep-tance and commitment therapy (ACT; Hayes etal., 1999) emphasizes the importance of behav-iorally active responses to emotions, with thepurpose of implementing a sense of control

over one’s responses to emotions, as opposed todecreasing the occurrence of unwanted internalevents.

To distinguish action tendencies from avoid-ance strategies (described earlier), the UP hastermed these behavioral responses to emo-tions as “emotion-driven behaviors” (EDBs).Whereas the function of emotional avoidancestrategies is to down-regulate or suppress emo-tions, EDBs include a specific set of “reactive”behaviors associated with each emotion. Forexample, the EDB for a panic attack is escape(fight or flight), whereas that for anxiety ishypervigilance. Similarly, the emotion of angerelicits an EDB of attacking/defending, and theEDB for sadness is cognitive, emotional, andphysical slowing and withdrawal (see Table 5.2for examples of EDBs associated with differentemotional disorders). It is important to note,however, that the therapist must focus onchanging EDBs, in addition to preventing emo-tional avoidance. For example, a patient withsocial phobia may be able to maintain eye con-tact and fully engage in social situations, but hemay escape the situation when his anxiety risesto the level of panic. On the contrary, this samepatient may be able to stay in the situation aslong as needed, but he may be distracting him-self the whole time or avoiding conversationswith people. Clearly, both scenarios are prob-lematic, because the patient is preventinghimself from learning that he can experienceemotions at their fullest without leaving the sit-uation. It should also be noted that, in practice,avoidance and EDBs may be difficult to distin-guish, and this distinction can be downplayedwith a given patient, if necessary or clinicallyrelevant.

After the concept of EDBs is introduced,patients begin to track EDBs on the self-monitoring forms and eventually focus onchanging EDBs. This can be a very concrete ex-ercise for the patient, because he or she is in-structed to “do something different than he orshe would typically do.” The Changing EDBsmonitoring form is illustrated in Figure 5.2.

These concepts are extended during the in-session emotion exposures created by the ther-apist and individually tailored to the patient’spresenting symptoms. In-session exposures canrange from an imaginal exposure to a pastemotional event (for PTSD or GAD), a conver-sation with a stranger (for social phobia), goinginto a dirty bathroom (for OCD), or watchinga sad movie or movie clip (for MDD or


dysthymia). The goal of the emotion exposureis to elicit any emotion, so that the patient hasthe opportunity to practice the techniquestaught in treatment (antecedent cognitive reap-praisal, modifying EDBs, and prevention ofemotional avoidance). With this new conceptu-alization, the exposure to the emotion is thenecessary mechanism of action for emotionalchange. Thus, the actual situational context ofthe exposure becomes less important. Thera-pists must be creative in the design of the expo-sure to maximize the emotion provocation forthe patient.

DETAILED DESCRIPTIONOF PROTOCOL

The following session descriptions and accom-panying transcripts are from the case of “Os-car,” described earlier, treated by one of the au-thors (D. H. B).

Session 1

In the first session, the therapist reviews treat-ment logistics, including session length andstructure, the therapist’s role, and the impor-tance of prioritizing treatment. Sessions are ap-proximately 45–50 minutes in duration and oc-cur weekly for the first 12 weeks and 2 weeksapart for Sessions 13–15. The collaborative na-

ture of treatment is emphasized, including theimportance of feedback from the patient. Fol-lowing this introduction is a detailed review ofthe presenting complaint. In this review, thetherapist focuses on emotions experienced andany strategies the patient uses to manage theseemotions. Below, Oscar describes situationalavoidance resulting from the lab accident.

THERAPIST: Let’s review a couple of thesethings. When exactly was it that the incidentoccurred?

OSCAR: It was at the very end of May.

THERAPIST: Of course, it was an extraordinarilydistressing incident. Would you say that overthe last 6 months you’ve noticed any changein the level of distress?

OSCAR: For a few different reasons the incidentdoes not come up in my mind as often as itdid in the beginning. It could be because ithappened a while ago, or it could be that Icurrently do no acid work in these labs. Infact, somebody else does it for me.

THERAPIST: So that’s something that was ar-ranged then?

OSCAR: Yes, my advisors arranged it, by my re-quest. It wasn’t a big deal. There is anotherperson in my group that was happy to do it. Ihelp her with something else. I know thatgiven your recommendations it is probably


TABLE 5.2. Examples of Emotion-Driven Behaviors and Incompatible Behaviors

EDBsDisorder mostusually associated Incompatible behaviors

Calling relatives to check on safety GAD Restricting contact/calling relatives

Perfectionistic behavior at work orhome

GAD Leaving things untidy or unfinished

Checking locks, stove, or otherappliances

OCD Repeatedly locking–unlocking and turningon–off until memory is unclear

Leaving (escaping from) a theater,religious service, or other crowdedarea

PDA Move to the center of the crowd. Smile orproduce nonfearful facial expressions

Social withdrawal Depression Behavioral activation

Leaving (escaping) a social situation Social phobia Staying in situation and approaching people

Verbally/physically attackingsomeone when in an argument

PTSD Remove self from situation and/or practicerelaxation techniques

Hypervigilance All disorders Focus attention on specific task at hand;meditation; relaxation

not the way to proceed, but at the time itseemed like a good idea.

THERAPIST: Oh sure. It may be, and of courseeverything we do will be fully collaborative,no surprises, but it may be that we talkabout the possibility of getting you back intosome of these situations.

OSCAR: I do encounter situations in the lab thatI work in now where I get anxious. It couldbe a small thing like walking past someoneworking with acids, and wondering “Didsome liquid get on me?” You get little itcheshere and there, and then you attribute themto something instead of just dismissing it.

After reviewing the presenting problem, thetherapist discusses the treatment model, with afocus on normative emotional experience andhow emotions can become disordered. Emo-tions are presented as natural reactions to theenvironment that motivate adaptive behavior.The therapist describes that the goal of treat-ment is to help the patient increase awarenessand control of emotions that are excessive. Neg-ative reinforcement is introduced as a way thatcognitions and behaviors can serve to maintainemotional disorders. The therapist describesthis as an effective strategy that decreases emo-tion in the short term but contributes further tothe disorder in the long run. For Oscar, health-related anxiety had become so pervasive thateven minor physical sensations led to fear of se-rious illness or death. In this discussion, thetherapist emphasized the concept that emotionsthemselves do not present a threat or danger,and that any emotion passes with time.

The three-component model is described as aframework for examining emotions by dividingthem into components (cognitions, behaviors,and physiological sensations), thereby makingthem more manageable. The therapist uses anexample relevant to the patient’s experience todemonstrate the contribution of each compo-nent to his or her emotional experience. Thetherapist guides the patient in identifying eachcomponent in this example and presents themodel as the structural basis for treatment.Each component is addressed separately overthe course of treatment as contributing to theoverall emotion and interacting with the othercomponents. A brief introduction to the neces-sity of provoking emotions as part of treatmentis also presented.

THERAPIST: You may know that emotion is agood thing. It helps us to function, and it isvery adaptive evolutionarily. It motivates usto engage in a behavior that would protectus or sustain the species. And it is absolutelyfine as long as it occurs at the right time, butsometimes it occurs at the wrong time, suchas when the danger is not present or shouldbe tolerable. And in that case emotions cankeep you, to some extent, from doing whatyou want to do, such as, in your case, gettinginto the lab.

OSCAR: Right, among other things.

THERAPIST: We deal with these emotions by try-ing to break them down into different com-ponents. Emotions, as you have pointed out,can be very overwhelming. They grab youand can feel like a big blob. You can be over-whelmed by these feelings and the necessityto do what the emotion is telling you to do.And to get away from this big blob, it will bemuch easier to deal with it one piece at atime. Now in order to do that, we are goingto have you monitor your emotions, and wewant you to actually elicit some of theseemotions. We want to train you to be an ex-pert observer of your emotions.

In this session, the therapist provides a ratio-nale for homework assignments, including pro-spective emotion monitoring, which provides ameans of recording ongoing experiencesthroughout treatment and increasing aware-ness of emotions, as well as exercises designedto elicit emotion outside of the treatment ses-sion. General emotion monitoring with theWRAD is employed beginning in Session 1.This session’s exercise includes assigning amood induction CD we put together that in-cludes selections such as the theme from themovie Schindler’s List and the Forrest Gumpsoundtrack to listen to before Session 2. Thepatient is asked to listen to one or two songs ata time and to record emotional reactions to themusic. This exercise introduces the concepts ofprovoking emotion and developing emotionalawareness. Additionally, the patient is providedwith a worksheet on which he or she is asked tobreak down an emotion from the week into thethree-component model presented in session,by recording the thoughts, behaviors, and feel-ings specific to that example. The therapistshould spend sufficient time in the first session


to ensure that the patient understands the ra-tionale for and importance of homework, andhow to complete each of the forms.

Session 2

The therapist begins Session 2 with a review ofthe treatment model presented in Session 1.Any reservations or concerns about the treat-ment should be validated, and the rationale isreinforced in the context of the lack of successthe patient has had with current emotion regu-lation strategies. Homework forms are thendiscussed in detail, and any difficulties in com-pletion (e.g., procrastination) are addressedas needed. After reviewing the WRAD, thetherapist reviews the three-component modelworksheet and addresses any questions thatarise. Particular attention is paid to the pa-tient’s understanding of the model and abilityto identify the distinct components. The thera-pist then reviews the mood induction exercisein detail with the patient, focusing on identify-ing emotions experienced, and determining thepatients level of awareness and ability to in-duce emotion.

THERAPIST: What were your reactions to themood induction CD? Let’s start with the firstsong.

OSCAR: For this one, I remember feeling anger.

THERAPIST: Are you aware of any thoughts youwere having at the time? How did you knowyou were angry?

OSCAR: I’m not sure. It is kind of a feeling of re-lief in a way. It’s kind of like when you’re an-gry and you scream. It’s as if I’m letting outsome anger through the music, like the musicis doing it for me.

THERAPIST: Any thoughts?

OSCAR: Not that I’m aware of.

THERAPIST: A lot of people have difficulty iden-tifying these kinds of things. But the point ofthis and future exercises is to get you to bemore introspective and aware of your emo-tions.

The therapist then introduces cognitive reap-praisal. “Appraisals” are described as impor-tant aspects of the experience of emotion,which often occur automatically. “Cognitiveerrors,” or “thinking traps,” are defined as ap-praisals that are repeatedly and automatically

assigned to situations at the expense of otherpossible interpretations. When these appraisalslimit the flexibility of interpretation, they cancontribute to the cycle of negative emotion.With this background, the therapist guides thepatient in generating examples from his or herexperience. For Oscar, physical sensations werefrequently interpreted as signs of serious medi-cal problems that might lead to death.

An in-session exercise is used to demonstratethe concept of cognitive flexibility. The therapistshows the patient an “ambiguous” picture (weuse a card from the Thematic Apperception Test[TAT]; Morgan & Murray, 1935) to whichmany cognitive appraisals can be assigned. Thetherapist instructs the patient to look at the pic-ture for approximately 30 seconds, then elicitsboth the patient’s primary interpretation and atleast two or three alternative interpretations.Additionally, the patient is asked what specifi-cally may have contributed to the automaticappraisal (e.g., memories of similar situations,specific details in the picture). This exercise isused to demonstrate the way that situations canbe interpreted in many ways, if all available in-formation is considered. Probability overesti-mation, or the tendency to assume a high likeli-hood of the occurrence of a negative event, andcatastrophizing, or assuming that the conse-quences of an event will be beyond the individ-ual’s ability to cope, are presented as two cogni-tive errors common to all emotional disorders.The therapist guides the patient in beginning toidentify these errors in the context of his or herexperience. Below, the therapist discusses bothof these cognitive errors based on the exampledescribed earlier in which Oscar touched a resi-due on a plate at a department store.

THERAPIST: Sitting here now, what is the proba-bility that the residue was the acid?

OSCAR: Maybe, one in a million—very, verysmall.

THERAPIST: One in a million?

OSCAR: No, there are millions of plates. Proba-bly zero. Vanishingly small.

THERAPIST: When you are right in front of it,feeling emotional, what do you think theprobability is?

OSCAR: That’s not something that came intomy mind. Like a number or something?

THERAPIST: Well, that’s typically not the waywe would think about it. But thinking back


on it in your emotional mind, what do youthink you may have estimated as the proba-bility at that moment?

OSCAR: Maybe 25%.

THERAPIST: So you can see the interplay of thetwo: “This residue is the acid, then I’m goingto die.” Both thinking the worst—“I’mgoing to die,” and overestimating theprobability—“this is an acid.” What wewant to do is to make it crystal clear whatyour automatic thoughts are, and second,what the alternatives are, just as you knowthere are many truths to a situation, includ-ing the rational one in the front of your headand the emotional one in the back of yourhead. One of the ways we approach this is totry to weaken the irrational, strengthen therational, and give you multiple perspectives.

General strategies to address cognitive errorsare presented, including identification of think-ing traps and generation of alternative apprais-als, with more specific reappraisal techniquesintroduced in the next session. Over the nextweek, the patient is asked to begin monitoringautomatic appraisals by identifying the emo-tional trigger, the subsequent appraisal, andwhether the appraisal represents probabilityoverestimation or catastrophizing, in additionto monitoring daily emotions with the WRAD.

Session 3

This session begins with a review of reactionsto the previous session and addresses lingeringquestions about automatic appraisals. TheWRAD is reviewed for the remainder of thetreatment as needed, with more focus on themore specific monitoring forms. In reviewingthe cognitive appraisals monitoring form, thetherapist provides guidance on identifyingthinking traps and tries to focus the patient’sattention on the “core” appraisal through theuse of Socratic questioning. The excerpt belowis from a discussion of an example in whichOscar suspected he had been exposed to achemical in the lab.

THERAPIST: So you were worried about beingexposed to this chemical, but what specifi-cally were you thinking about the chemical?

OSCAR: It’s toxic.

THERAPIST: And what does that mean?

OSCAR: It’s dangerous to me.

THERAPIST: And what?

OSCAR: And, I might die.

THERAPIST: Now that is what we want, the realcore of it, what is driving this interpretation.A lot of these anxious thoughts often docome down to dying. So you think, “I’m go-ing to die and I need to start looking for thesigns of impending death”—the scratchythroat and such. Of course, anyone would bescared or anxious with that type of apprais-al.

The therapist then engages the patient in amore detailed discussion of cognitive reap-praisal, focusing on flexible thinking. The im-portance of identifying core appraisals is em-phasized as the level at which reappraisal willbe focused. Countering probability overestima-tion and decatastrophizing strategies are pre-sented as a way to generate alternative apprais-als and to place less focus on the automaticappraisal. Countering probability estimationinvolves drawing on past evidence to examinehow realistic the patient’s estimation is whenexperiencing an emotion. The therapist guidesthe patient in making concrete estimations andcomparing the original estimation to a more re-alistic one. Decatastrophizing involves helpingthe patient to identify his or her ability to copewith the feared situation using past evidenceand specific examples (e.g., similar experiencesin the past). It is important to communicatethat these strategies do not eliminate negativeappraisals, but they provide greater flexibilityand allow the patient to gain perspective on thefeared situations.

Over the next three sessions, the patient isasked to continue to monitor automatic ap-praisals and to note the strategies used to im-prove interpretive flexibility. The patient isasked to generate at least one additional inter-pretation for each automatic appraisal identi-fied. The therapist again emphasizes that thereis no “correct” interpretation; rather, the rangeof possibilities should be considered.

Session 4

In Session 4, the therapist reviews the cognitivereappraisal monitoring form in detail and clari-fies any residual questions or problems. At thistime, the therapist ensures the patient’s abilityto identify automatic appraisals and to gener-


ate alternatives before moving on to the nexttopic of treatment, which focuses on the behav-ioral component of emotion. First, EDBs aredescribed. A description of the adaptive natureof emotion as a strong motivator for behavioris provided, followed by how EDBs can be-come maladaptive and serve to maintain dis-ordered emotional experiences. The therapistalso distinguishes between EDBs and emo-tional avoidance in this session. The patient isasked to generate examples of EDBs from hisor her experience. In the transcript below, Os-car describes personal examples of EDBs, in-cluding seeking reassurance regarding physicalsymptoms from his mother, a physician.

THERAPIST: Given the past examples we’vetalked about, what would you say are someof the behaviors these emotions cause you todo?

OSCAR: One would be to leave where I am, sothat’s different than not going in the firstplace.

THERAPIST: And how do you feel when you getout?

OSCAR: Well, a little better, but then I start wor-rying again.

THERAPIST: But for the moment you feel a littlebetter, and certainly better than you wouldimagine feeling if you had stayed in the situa-tion. What are some others?

OSCAR: Calling my mother?

THERAPIST: Right. Particularly when you call tocheck in about physical symptoms. What ef-fect do you think this has on your emotions?

OSCAR: Well, I’ve been trying to think aboutthat and I’m not sure. Sometimes when I callmy mother, for example, while I’m speakingto her, it certainly makes me a little calmer.Other times it does not. But has it ever madethe problem completely go away? Probablynot.

THERAPIST: So it does make you feel better inthe short term? You feel reassured, you es-cape. Do you do other checks?

OSCAR: Yeah, like scanning my body internally.Or if there are other people around, I’ll askthem if they smell anything.

THERAPIST: What happens is when you callyour mother or engage in these other be-haviors and feel better temporarily, thatstrengthens that relationship. It’s a very pow-

erful concept. So what will you do nexttime? Something that you know will makeyou feel better. But the problem is, that in-creases the probability that the next time alittle trigger occurs, the whole sequence willfollow.

OSCAR: So you’re saying that after a while, thebehavior will be connected with even smallertriggers, which will seem bigger?

The therapist also guides the patient in un-derstanding that EDBs may decrease the emo-tional experience in the short term but main-tain or heighten the emotion in the long term.Though EDBs have helped the patient to atten-uate the emotional experience to some degree,they have not been working as a long-termstrategy to control emotions. Following thisdiscussion, Oscar was able to identify an exam-ple of this concept.

OSCAR: I’m not sure if this is related, but at thebeginning of the semester I was forced to bein the lab 2 days straight and I was reallyanxious, but I forced myself to stay. Now,being in that lab is not as big a deal.

THERAPIST: There, you broke the cycle. Becausethe fact is, those things you’ve been doinghaven’t been working and in fact make itworse. And you yourself found that if youdon’t do those things and stay in the situa-tion, it works.

OSCAR: It feels a lot worse for a while.

THERAPIST: But in the long run it feels better.

OSCAR: Right.

THERAPIST: So things you do in the short run—wrestling with the emotion, trying to get ridof it—haven’t worked.

OSCAR: Does it mean that I should try not to dothese things?

THERAPIST: We want to get to a point that it isOK to have these emotions. Your emotionsare going to happen. With you, they’re notalways happening at the right time, so whatare we going to do? We need to pay a littleless attention to them. And we also want tostop these behaviors that haven’t been work-ing.

Session 4 concludes with a brief introduc-tion to emotion exposure. The therapist de-scribes the rationale for emotion exposures


and introduces an in-session exercise. Fourslides designed to elicit emotion, selectedfrom the International Affective Picture Sys-tem (IAPS) slides (Center for the Study ofEmotion and Attention, 2001), are presentedsequentially to the patient on a computerscreen. The slides we use are: a young childcrying, a woman being held at knifepoint, amutilated body, and a ski-jumper about todescend a ramp. The patient is instructed toattend to each slide for approximately 30 sec-onds and to notice any emotions that he orshe experiences. After the presentation ofeach slide, the therapist asks the patient whatemotions he or she is experiencing and forany reactions to the slide. During this exer-cise, the therapist emphasizes awareness ofthe emotion and minimization of any avoid-ance strategies. Additionally, the therapistdemonstrates that many situations can triggeran emotion, including the slides from the ex-ercise. It is important for the patient to beable to identify these reactions and their trig-gers. In the following sequence, the slide usedwas the photo of the mutilated body.

THERAPIST: (following the presentation of theslide) What is your reaction to this image?

OSCAR: Kind of disturbing.

THERAPIST: OK, let the emotion wash over you,just take it in.

OSCAR: I guess some anxiety, also not wantingto look at it.

THERAPIST: That’s important, you wanted toturn away to . . . ?

OSCAR: To not see it anymore.

THERAPIST: To decrease the emotion?

OSCAR: Yeah.

THERAPIST: Any physical sensations?

OSCAR: I guess the anxiety comes along withmaybe with a feeling of adrenaline rush. Andthe kind of feeling of tension. Like some-thing I want to get out. I feel like there’ssomething building up.

THERAPIST: OK. We can take the picture awaynow.

OSCAR: I feel like I’m almost going to cry at thispoint. (Begins to cry.)

THERAPIST: That’s OK. And, now the emotionprobably feels a little more intense.

OSCAR: Yeah, but the tension loosens up a bit.

THERAPIST: Thoughts going through yourhead?

OSCAR: It’s interesting, I was really more awareof myself than actually thinking about thepicture.

THERAPIST: What about yourself?

OSCAR: It’s almost like my muscles are hurting,from being tense.

THERAPIST: You see, all of these feelings, someof which might even seem a bit strange, arepart of your emotion.

For homework, in addition to continuingwith cognitive reappraisal, the patient is askedto begin designing emotion exposures for thecoming week. At this stage in treatment, gen-eral instructions are given, but the therapist isnot yet helping the patient to design exposures.The patient is encouraged to enter situationsthat may elicit unwanted emotions. The patientis asked to monitor progress and obstacles thatarise. Additionally, the patient is given a formto monitor EDBs over the next week.

Session 5

This session begins with a review of the home-work, with a particular focus on the EDBsmonitoring form. The therapist then reviewsemotion exposures completed in the past week.It is important to validate any concerns aboutthe exposures and to encourage the patient toseek out opportunities to experience emotion.In the excerpt below, Oscar experienced a highlevel of anxiety following identification of astrange smell in a chemistry lab. The day afterthis incident, he entered the same lab and didnot experience the expected anxiety.

THERAPIST: What are some other possibilitiesas to why you didn’t have a spike of anxietythis morning?

OSCAR: I had already taken the clonazepam.

THERAPIST: Anything else?

OSCAR: I didn’t smell anything. I went througha different door.

THERAPIST: Did you expect to have a difficulttime going back into the lab today?

OSCAR: Yes.

THERAPIST: But you didn’t?

OSCAR: Right. Or was the anxiety from thatnot sufficient to make a noticeable difference


from the already high anxiety? So through-out the day, I wonder every time I cough alittle bit.

THERAPIST: Is there the possibility also thathaving survived yesterday with nothingmore than a suspicious scratchy throat, yourthoughts about the danger lessened a bit?

OSCAR: Yeah, maybe a little bit. But I’m stillwaiting for the delayed reaction. But there isthe constant internal discussion.

THERAPIST: Tell me about that.

OSCAR: First, there is no reason for it to smellin that part of the room. That’s one thing.The other thing is, if it were actually thischemical and it were enough to be danger-ous, I should have had a more immediate re-action in terms of much more coughing andwheezing, that type of thing. Then I comeback saying, “But I’m feeling this or that,and what if it’s some other type of chemi-cal?”

THERAPIST: OK, the one good thing about thisincident, although clearly distressing to you,is that it really gives us something to workwith. These thoughts and behaviors thathave been maintaining these emotions havebecome so automatic that in order to changethem, we need you to fully experience theseemotions.

Following the homework review, the thera-pist introduces the concept of emotional avoid-ance, describing how avoidance is harmful inboth maintaining negative emotion and pre-venting the patient from engaging in certain sit-uations. Types of avoidance, including subtlebehavioral (e.g., procrastination), cognitive(e.g., distraction) and safety signals (e.g., carry-ing a medication bottle), are introduced. Thepatient is asked to generate personally relevantexamples of avoidance. As with the descriptionof EDBs, the therapist emphasizes that avoid-ance can work in the short term but is not along-term solution.

THERAPIST: There are also a number of thingsyou do to avoid even getting these emotionalreactions. For example, you didn’t go to thelab on Friday.

OSCAR: A better example is the lab where theincident happened. Even on a day that I’mfeeling good, I avoid it.

THERAPIST: Those are the easy ones to see,

when you flat out don’t go there. There aresome other things that are often so subtlethat we’re not even aware of them. For ex-ample, some people will give up caffeine.

OSCAR: Yeah, I found myself doing that. In thepast few months I have stopped drinkingcoffee. And I try to avoid things that willmake me uncomfortable, like taking thestairs.

THERAPIST: Your tendency right now is to seekto avoid and to try to suppress or make emo-tions less intense. Attempts to control, di-minish, or decrease them have the paradoxi-cal effect of increasing them in the long run.There are a variety of these subtle and lesssubtle ways to avoid and try to get rid ofemotions.

An in-session exercise to demonstrate emo-tional avoidance is then completed. An imageintended to evoke emotion is chosen by thetherapist from the IAPS slides and presented ona computer screen. The patient is instructed tolook at the image for approximately 30 sec-onds and to attempt to not think about the im-age during that time. The patient can use anyavoidance strategy he or she chooses but mustcontinue looking at the image. The therapistthen takes the image away and asks what strat-egies the patient used and how successful theywere. This exercise demonstrates that avoidinga situation or trigger entails thinking about it insome respect, which prevents avoidance frombeing successful.

At the end of the session, a brief “emotionalawareness training” is provided. Awareness isframed as “not avoiding” or as allowing emo-tions to occur without attempts to suppress orcontrol them. An Awareness of Emotions hand-out (see Figure 5.3) is provided and reviewed.The handout includes specific instructions toguide the patient in accepting and understand-ing emotions to interrupt the cycle of negativeemotion. The therapist emphasizes that ratherthan eliminating emotion, awareness and ac-ceptance of emotion will the patient help togain greater control over disordered emotions.

Homework includes monitoring any emo-tional avoidance strategies that the patient en-gages in over the week, continuing to monitorEDBs, and engaging in general emotion expo-sures. The patient is asked this week also tonote the degree of emotional awareness duringeach exposure by rating awareness on a scale


from 0 (No awareness) to 100 (Completeawareness).

Session 6

Following a detailed homework review, Session6 introduces the third component of emotionby focusing on physical sensations. Followingan introduction to the rationale for intero-ceptive exposure, the therapist and patientwork through a list of exercises designedto elicit emotion through physical activation.Some examples include hyperventilating, spin-ning, and running in place. In addition, any ex-ercises that are particularly relevant to the pa-tient may be added. Prior to each exercise, thetherapist demonstrates the exercise, and fol-lowing the patient’s completion of the exerciseasks the patient to rate the intensity, distress,and similarity to physical sensations typicallyexperiences during an emotion, each on a scaleof 0 (Not at all) to 8 (Very much).

THERAPIST: So today we’ll do a number of exer-

cises to induce some of these physicalsensations that are often the trigger for youranxiety and look at the intensity, distress,and whether it’s similar to your experience.So the first one I’m going to ask you to dois the hyperventilation exercise. (followingthe exercise) What sensations are you feel-ing?

OSCAR: My legs are tingling, and I’m feeling alittle dizzy. My throat is dry, for obvious rea-sons.

THERAPIST: Scan your body and notice any sen-sations.

OSCAR: A little tingling in the fingers and in thehead. I feel like my eyes are kind of . . . notjust dizziness, but a little unsteady, almostlike they’re vibrating.

THERAPIST: How intense were these sensationsin totality from 0 to 8, with 8 being the mostintense?

OSCAR: Maybe a 6?

THERAPIST: How distressing?


Components of “Emotional Awareness”

1. Allowing oneself to fully experience emotions as they happen.

2. Focusing on being in the present moment, not “living” in the future or in the past.

3. Allowing emotions to come and go, without trying to push away feelings or trying to hold on tocertain feelings. When you allow yourself to experience a negative emotion, then you are ableto process it and move forward. If you don’t give yourself permission to experience the emotionbut push it away, then it becomes a “tidal wave” of emotion that feels very out of control.

4. Being aware of your emotions does not always “feel good.” However, if will allow you to movepast the experience, instead of getting stuck trying to push it away, which only reinfoces itspower.

5. Awareness does not necessarily mean acceptance of the situation or environment. Instead, weare focusing on acceptance of one’s emotional reaction to it, before possibly acting in a differentway.

6. Having some understanding for one’s own emotional experience. Getting “mad” or “frustrated”with yourself or saying “I’m so stupid for feeling this way” will only make you want to controland resist emotional experiences even more. This treatment focuses on increasing the range ofemotional experience, instead of keeping it restricted. This includes allowing oneself to be sad,frustrated, and anxious, as well as letting it pass when emotions have changed. Occasionally,it may also mean substituting a different behavior, such as smiling or laughing when afraid—or getting up and doing something when all you feel like doing is going to your room andgoing to sleep. But, again, these new actions must be coming from a place of really trying tochange what you are doing, instead of “forcing” or “punishing” yourself into a new wayof behaving.

FIGURE 5.3. Awareness of emotions.

OSCAR: Maybe a 4, because I knew I was doingit myself?

THERAPIST: What about similarity? Any simi-larity to what you experience in the lab orother situations?

OSCAR: I wasn’t able to describe it well, butdefinitely the eye thing.

THERAPIST: How similar would that be to yournaturally occurring symptoms?

OSCAR: Well, I would give it a 4 just becausethe fingers and the legs tingling are not typi-cally things that I experience, but the dizzi-ness and the eye vibration are.

Following this exercise, the therapist and pa-tient choose the most relevant exercises to en-gage in regularly over the next week. The pa-tient is asked to complete the exercise severaltimes a day, until the associated distress de-creases. Additionally, for homework, the pa-tient is asked to complete an Avoidance Hierar-chy (see Figure 5.4) in which situations that thepatient fears and avoids are listed. The thera-pist provides instructions on how to completethe form, which will be used to guide futureemotion exposures. The patient is asked tocontinue with the general emotion exposuresand to monitor EDBs.


AVOIDANCE HIERARCHY

Rate the degree to which you avoid each of the following situations due to the unpleasant feelingsassociated with them. For each, write the applicable number in the space provided.

Do notavoid

Hesitate to enterbut rarely avoid

Sometimesavoid

Usuallyavoid

Alwaysavoid

0 1 2 3 4 5 6 7 8

No distress Slight distress Definite distress Strong distress Extreme distress

Description Avoid Distress

1WORST

2

3

4

5

6

7

8

9

10

FIGURE 5.4. Example of an Avoidance Hierarchy.

Session 7

Session 7 serves as a review of the concepts pre-sented in the first six sessions and a transitionto the second half of treatment, which focuseson emotion exposures. The therapist reviewsany interoceptive exercises done in the pastweek and encourages the patient to continuethese exercises, if they were helpful. Becausemany of Oscar’s symptoms involve physicalsensations these exercises are continued outsideof session. In particular, exercises that inducedrespiratory symptoms (e.g., hyperventilation)were encouraged because of his sensitivity tothese symptoms. The Avoidance Hierarchy isalso reviewed in detail to ensure that it can beused to guide treatment. It is important to havean appropriate range of avoidant situations tostructure exposures, such that a graded expo-sure framework can be utilized. After dis-cussing the homework, the therapist reviewstreatment concepts in the context of the three-component model, and how cognition, be-havior (including avoidance and EDBs), andphysical sensations all contribute to the experi-ence of emotion. At this stage in treatment, thereview of concepts should be more idiographic,reflecting back on the patient’s experience, andcover the concepts most relevant to the individ-ual. After addressing any lingering questions,the therapist provides a rationale for situa-tional emotion exposures.

THERAPIST: You want to get to a point whereyou can invite these emotions to come onand almost stand aside and watch them. Theonly way to do this is to experience the emo-tion, even at extreme levels. So now we’regoing to start purposefully engaging in exer-cises to provoke these emotions, with thegoal of your being able to let an emotionwash over you, stand beside it, and cut outthese emotion-driven behaviors, strippingaway these procedures that prevent youfrom experiencing emotions.

OSCAR: In terms of distress, every time I go intoone of the labs I get some anxiety, but I don’tworry so much about it, because I feel nowthat I’m able to make it go away. SometimesI feel that if I just went into the lab where theincident happened, I could handle it.

THERAPIST: We want to turn this around now.We want you to invite the emotion, not

thinking about whether you can handle it.And we also don’t want you to talk yourselfout of it. The process of experiencing emo-tion is harmless.

The therapist also describes the process ofdesigning emotional exposures by using theAvoidance Hierarchy. Beginning with this ses-sion through the end of treatment, the patientis asked to monitor emotion exposures pro-spectively, including the situation, intensity ofemotion prior to, during, and following theexposure, and any automatic appraisals and re-appraisals identified. Three exposures are as-signed by the therapist for homework, begin-ning at the bottom of the patient’s hierarchy.The therapist and patient work together to de-sign emotion exposures that are more challeng-ing than the general exposures done thus far.Oscar’s three chosen situations were entering alab that contained hazardous chemicals, goingto a challenging class at a gym (which he antici-pated would lead to strong cardiovascular acti-vation), and performing his lab work withoutthe “extra” precautions he had utilized sincethe lab accident.

Sessions 8–12

Sessions 8–12 focus on both processing theemotion exposures completed outside of ses-sion and completing in-session exposures. Thetherapist begins by reviewing assigned expo-sures in detail, paying close attention to anyobstacles or avoidance behaviors that mayhave prevented the patient from completing as-signed exposures. An exposure near the top ofOscar’s hierarchy was to use a public drinkingfountain in the lab. Since the lab incident, hehad been unable to use the fountain for fearthat a dangerous substance might have con-taminated the water. As is evident below, thisproved very difficult for Oscar at first.

OSCAR: So the first one was the drinking foun-tain, which ended up pretty intense. I was re-ally anticipating that it was going to be bad.I drank. In fact, I drank twice and felt fine,and a few hours later the anxiety startedcoming and lasted for about 3 days. It waspretty bad. And I really felt like calling mymother. Sometimes instead of worrying, Iwould try to really focus on the physical sen-


sation and it helped, but it wouldn’t go100% away.

THERAPIST: Remember the goal is not to makeit go 100% away. We are working towardnot trying to make it go away. In not lookingto make it go away, that is where the changecan be made.

Oscar was encouraged to continue this expo-sure outside of session. Below is an excerptfrom the following session.

THERAPIST: How was drinking from the waterfountain?

OSCAR: The anxiety kind of came and went,and it really felt like this time around I wasable to be really aware and pay attention tothe feelings. And then it went away, and bythe time I went to sleep, it had dissipated alot, and it didn’t come back the next day.And since then I’ve been drinking there everyday, and it hasn’t been a problem exceptmaybe for some anticipatory anxiety. But, ofcourse, going from there back to workingwith acids, you know, it feels like a slightlydifferent beast.

THERAPIST: In what way does it feel different?

OSCAR: Because it feels like a real danger. Withthe water fountain there is no danger, butwith the acid if I made a mistake, there is areal danger.

THERAPIST: What is the probability of a mis-take?

OSCAR: One serious enough to be dangerous?Low, well, very low. They are very uncom-mon.

THERAPIST: And in the event of such a mistake,what safety procedures are available?

OSCAR: We have excellent safety precautions inplace for just that reason. So if a mistake likethat is made, there are built in safeties. Un-less I really purposefully did something stu-pid, it couldn’t produce any real danger.

In-session exposures are also employed inthese sessions, which can be either situationalor imaginal. Situational exposures relevant tothe patient are chosen by the therapist and/ordesigned collaboratively with the patient.Imaginal exposures can be utilized when situa-tional exposures are not appropriate for the pa-tient or are prevented by logistics. Because Os-

car’s feared situations were difficult to recreatein session, imaginal exposures were helpful. Inthese types of exposures, the patient describesan emotion exposure in a very detailed manner,trying to recreate the situation in as much de-tail as possible and to experience the emotionas fully as possible. The therapist leaves at least15 minutes after the exposure to process theemotions experienced. Prior to attempting theexposure at the top of his hierarchy (enteringthe lab where the incident took place), the ther-apist guided Oscar through an imaginal expo-sure, allowing him to visualize this situationand to imagine any physical sensations that hemight experience. Throughout these sessions,additional exposures assigned for homeworkshould be based on both the Avoidance Hierar-chy and other potential situations discussed insession. The exposures increase in difficultyover time, and as the sessions progress, the pa-tient begins to take more responsibility for de-signing the exposures and the therapist pro-vides less guidance. Exposures done outside ofthe session are monitored so that they may bediscussed and processed in session.

Sessions 13–14

In Sessions 13 and 14 the therapist revisits thepatient’s continuing need for emotional expo-sures. If he or she determines that the patientwould benefit from additional emotion expo-sures, they can be continued both in sessionand for homework at this time. Additionally,the therapist can focus on any events or emo-tional reactions over the past week that can beprocessed in session. In these sessions, Oscarcontinues to enter difficult situations, proceed-ing to his most feared situations following Ses-sion 13. Obstacles to exposure completion andreactions to exposures are discussed in session.The patient continues prospectively to monitoremotions, emotional exposures, and EDBsthroughout Session 14.

Session 15

In this session the therapist reviews both themajor concepts presented in treatment and thepatient’s progress. The therapist reviews thethree-component model, focusing on cognitivereappraisals, modifying EDBs, and preventingemotional avoidance strategies. The therapistthen discusses the inevitability of future stress-ors and potential reoccurrence of symptoms.


The continuation of informal emotion expo-sures and use of techniques to think more flexi-bly and to modify EDBs are encouraged. Atthis time the therapist provides to the patienthandouts reviewing the treatment concepts andmay at his or her discretion show the patientthe results of process or outcome measures col-lected over the course of treatment.

OUTCOME AND CONCLUSION

Oscar received follow-up assessments at thetermination of treatment and at 3-monthsposttreatment. These included the Mini-ADIS-IV (Brown et al., 1994), the SIGH-A (Shear etal., 2001) and SIGH-D (Williams, 1998), and abattery of self-report questionnaires. Addi-tionally, the HAI (Salkovskis et al., 2002) wasadministered prior to each session to measureprogress. During treatment, Oscar’s HAI scoredecreased from 24 at Session 2 to 17 at Session12. At the 3-month follow-up assessment, Os-car was assigned a diagnosis of anxiety disor-der not otherwise specified at a CSR of 4, justmeeting the DSM-IV diagnostic threshold. Hissymptoms of hypochondriasis and OCD wereboth noted to be at subclinical levels (CSR = 3and 2, respectively). By 3 monthsposttreatment, Oscar was back in the labora-tory full-time and was on track to complete hisgraduate degree as scheduled. At the nextfollow-up, 6 months posttreatment, he was as-signed no clinical diagnoses, with hypo-chondriasis and OCD noted to be in partial re-mission, both with a CSR of 2.

REFERENCES

Allen, L. B., White, K. S., Barlow, D. H., Gorman, J. M.,Shear, M. K., & Woods, S. W. (2007). Effects ofcognitive-behavioral therapy (CBT) for panic disor-der on comorbid depression and anxiety. Manuscriptsubmitted for publication.


Andrews, G. (1990). Classification of neurotic disor-ders. Journal of the Royal Society of Medicine, 83,606–607.

Andrews, G. (1996). Comorbidity in neurotic disorders:The similarities are more important than the differ-ences. In R. M. Rapee (Ed.), Current controversies inthe anxiety disorders (pp. 3–20). New York: GuilfordPress.

Ascher, L. M. (1980). Paradoxical intention. In A.Goldstein & E. B. Foa (Eds.), Handbook of behavior-al interventions: A clinical guide (pp. 266–321). NewYork: Wiley.

Barlow, D. H. (1988). Anxiety and its disorders: The na-ture and treatment of anxiety and panic. New York:Guilford Press.

Barlow, D. H. (1991). Disorders of emotion. Psycholog-ical Inquiry, 2, 58–71.

Barlow, D. H. (2000). Unraveling the mysteries of anxi-ety and its disorders from the perspective of emotiontheory. American Psychologist, 55, 1247–1263.

Barlow, D. H. (2002). Anxiety and its disorders: The na-ture and treatment of anxiety and panic (2nd ed.).New York: Guilford Press.

Barlow, D. H., & Cerny, J. A. (1988). Psychologicaltreatment of panic. New York: Guilford Press.

Barlow, D. H., & Craske, M. G. (2000). Mastery ofyour anxiety and panic: Client workbook for anxietyand panic (3rd ed.). Boulder, CO: Graywind.

Barlow, D. H., Gorman, J. M., Shear, M. K., & Woods,S. W. (2000). Cognitive-behavioral therapy, imi-pramine, or their combination for panic disorder: Arandomized controlled trial. Journal of the AmericanMedical Association, 283, 2529–2536.

Barlow, D. H., Rapee, R. M., & Reisner, L. C. (2001).Mastering stress 2001: A lifestyle approach. Dallas,TX: American Health.

Beck, A. T. (1972). Depression: Causes and treatment.Philadelphia: University of Pennsylvania Press. (Orig-inal work published 1967)

Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G.(1979). Cognitive therapy of depression. New York:Guilford Press.

Beck, A. T., & Steer, R. A. (1990). Beck Self-ConceptTest. Psychological Assessment, 2, 191–197.


Borkovec, T. D. (1994). The nature, functions, and ori-gins of worry. In G. C. L. Davey & F. Tallis (Eds.),Worrying: Perspectives on theory, assessment, andtreatment (pp. 5–34). New York: Wiley.

Borkovec, T. D., Abel, J. L., & Newman, H. (1995). Ef-fects of psychotherapy on comorbid conditions ingeneralized anxiety disorder. Journal of Consultingand Clinical Psychology, 63, 479–483.

Borkovec, T. D., Hazlett-Stevens, H., & Diaz, M. L.(1999). The role of positive beliefs about worry ingeneralized anxiety disorder and its treatment. Clini-cal Psychology and Psychotherapy, 6, 126–138.

Bouton, M.E. (2005). Behavior systems and the contex-tual control of anxiety, fear, and panic. In L. FeldmanBarrett, P. M. Niedenthal, & P. Winkielman (Eds.),Emotion and consciousness (pp. 205–227). NewYork: Guilford Press.

Bouton, M. E., Mineka, S., & Barlow, D. H. (2001). Amodern learning-theory perspective on the etiologyof panic disorder. Psychological Review, 108, 4–32.

British Psychological Society. (2004). English survey of


applied psychologists in health and social care and inthe probation and prison service. Leicester, UK: Au-thor.

Brown, E. J., Turovsky, J., Heimberg, R. G., Juster, H.R., Brown, T. A., & Barlow, D. H. (1997). Validationof the Social Interaction Anxiety Scale and the SocialPhobia Scale across the anxiety disorders. Psycholog-ical Assessment, 9, 21–27.

Brown, T. A. (2007). Temporal course and structural re-lationships among dimensions of temperament andDSM-IV anxiety and mood disorders. Journal of Ab-normal Psychology, 116, 313–328.

Brown, T. A., Antony, M. M., & Barlow, D. H. (1992).Psychometric properties of the Penn State WorryQuestionnaire in a clinical anxiety disorders sample.Behaviour Research and Therapy, 30, 33–38.

Brown, T. A., Antony, M. M., & Barlow, D. H. (1995).Diagnostic comorbidity in panic disorder: Effect ontreatment outcome and course of comorbid diagno-ses following treatment. Journal of Consulting andClinical Psychology, 63, 408–418.

Brown, T. A., & Barlow, D. H. (1995). Long-term out-come in cognitive-behavioral treatment of panic dis-order: Clinical predictors and alternative strategiesfor assessment. Journal of Consulting and ClinicalPsychology, 63, 754–765.

Brown, T. A., & Barlow, D. H. (2002). Classification ofanxiety and mood disorders. In D. H. Barlow (Ed.),Anxiety and its disorders: The nature and treatmentof anxiety and panic (2nd ed., pp. 292–327). NewYork: Guilford Press.

Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham,J. R., & Mancill, R. B. (2001). Current and lifetimecomorbidity of the DSM-IV anxiety and mood disor-ders in a large clinical sample. Journal of AbnormalPsychology, 110, 49–58.

Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998).Structural relationships among dimensions of theDSM-IV anxiety and mood disorders and dimensionsof negative affect, positive affect, and autonomicarousal. Journal of Abnormal Psychology, 107, 179–192.

Brown, T. A., Di Nardo, P. A., & Barlow, D. H. (1994).Anxiety Disorders Interview Schedule for DSM-IV:Treatment follow-up version (Mini-ADIS-IV). SanAntonio, TX: Psychological Corporation.

Brown, T. A., Di Nardo, P. A., Lehman, C. L., & Camp-bell, L. A. (2001). Reliability of DSM-IV anxiety andmood disorders: Implications for the classification ofemotional disorders. Journal of Abnormal Psychol-ogy, 110, 49–58.

Brown, T. A., White, K. S., Forsyth, J. P., & Barlow, D. H.(2004). The structure of perceived emotional control:Psychometric properties of a revised Anxiety ControlQuestionnaire. Behavior Therapy, 35, 75–99.

Cameron, N. M., Champagne, F. A., & Parent, C.(2005). The programming of individual differences indefensive responses and reproductive stratifies in therat through variations in maternal care. Neurosci-ence and Biobehavioral Reviews, 29, 843–865.

Campbell-Sills, L., Liverant, G. I., & Brown, T. A.(2004). Psychometric evaluation of the behavioral in-hibition/behavioral activation scales in a large sampleof outpatients with anxiety and mood disorders. Psy-chological Assessment, 16, 244–254.

Center for the Study of Emotion and Attention (CSEA-NIMH). (2001). The International Affective PictureSystem: Digitized photographs. Gainsville: Centerfor Research in Psychophysiology, University ofFlorida.

Chorpita, B. F., Albano, A. M., & Barlow, D. H. (1998).The structure of negative emotions in a clinical sam-ple of children and adolescents. Journal of AbnormalChild Psychology, 107, 74–85.

Chorpita, B. F., & Barlow, D. H. (1998). The develop-ment of anxiety: The role of control in the early envi-ronment. Psychological Bulletin, 124, 3–21.

Clark, L. A. (2005). Temperament as a unifying basisfor personality and psychopathology. Journal of Ab-normal Psychology, 114, 505–521.

Clark, L. A., & Watson, D. (1991). Tripartite model ofanxiety and depression: Psychometric evidence andtaxonomic implications. Journal of Abnormal Psy-chology, 103, 103–116.

Craske, M. G., Barlow, D. H., & O’Leary, T. (1992).Mastery of your anxiety and worry. San Antonio,TX: Psychological Corporation/Graywind.

Craske, M. G., Street, L., & Barlow, D. H. (1989). In-structions to focus upon or distract from internalcues during exposure treatment for agoraphobicavoidance. Behaviour Research and Therapy, 27,663–672.

Craske, M. G., Street, L. L., Jayaraman, J., & Barlow,D. H. (1991). Attention versus distraction during invivo exposure: Snake and spider phobias. Journal ofAnxiety Disorders, 5, 199–211.

Davidson, R. J. (2000). Anxiety, depression, and emo-tion. New York: Oxford University Press.

Di Nardo, P. A., Brown, T. A., & Barlow, D. H. (1994).Anxiety Disorders Interview Schedule for DSM-IV.Albany, NY: Graywind.

Dimidjian, S., Hollon, S. D., Dobson, K. S., Shmaling,K. B., Kohlenberg, R. J., Addis, M. E., et al. (2006).Randomized trial of behavioral activation, cognitivetherapy, and antidepressant medication in the acutetreatment of adults with major depression. Journal ofConsulting and Clinical Psychology, 74, 658–670.

Eley, T. C., & Brown, T. A. (2007). Phenotypic and ge-netic/environmental structure of anxiety and depres-sive disorder symptoms in adolescence. Manuscriptsubmitted for publication.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J.B. W. (1996). Structured Clinical Interview for DSM-IV Axis I Disorders—Patient Edition (SCID-I/P, Ver-sion 2.0). New York: Biometrics Research Depart-ment, New York State Psychiatric Institute.

Foa, E., Cashman, L., Jaycox, L., & Perry, K. (1997).The validation of a self-report measure of posttrau-matic stress disorder: The Posttraumatic DiagnosticScale. Psychological Assessment, 9, 445–451.


Foa, E. B., Huppert, J. D., Leiberg, S., Langer, R.,Kichic, R., Hajcak, G., et al. (2002). The Obsessive–Compulsive Inventory: Development and validationof a short version. Psychological Assessment, 14,485–495.

Frankl, V. E. (1960). Paradoxical intention: A logo-therapeutic technique. American Journal of Psycho-therapy, 14, 520–535.

Frisch, M. B., Cornell, J., & Villaneuva, M. (1992).Clinical validation of the Quality of Life Inventory: Ameasure of life satisfaction for use in treatment plan-ning and outcome assessment. Psychological Assess-ment, 4, 92–101.

Gross, J. J. (1998). Antecedent- and response-focusedemotion regulation: Divergent consequences for ex-perience, expression, and physiology. Journal of Per-sonality and Social Psychology, 74, 224–237.

Gross, J. J. (2002). Emotion regulation: Affective, cog-nitive, and social consequences. Psychophysiology, 9,281–291.

Gross, J. J., & Levenson, R. W. (1997). The acute effectsof inhibiting negative and positive emotion. Journalof Abnormal Psychology, 106, 95–103.

Hafner, J., & Marks, I. M. (1976). Exposure in vivo ofagoraphobics: Contributions of diazepam, group ex-posure, and anxiety evocation. Psychological Medi-cine, 6, 71–88.

Hamilton, M. (1959). The assessment of anxiety statesby rating. British Journal of Medical Psychology, 32,50–55.

Hamilton, M. (1960). A rating scale for depression.Journal of Neurology, Neurosurgery, and Psychiatry,23, 56–62.

Hammen, C. L. (1991). The generation of stress in thecourse of unipolar depression. Journal of AbnormalPsychology, 100, 555–561.

Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (1999).Acceptance and commitment therapy: An experien-tial approach to behavior change. New York:Guilford Press.

Hays, R. D., Sherbourne, C. D., & Mazel, R. M. (1993).The RAND 36-Item Health Survey 1.0. Health Eco-nomics, 2, 217–227.

Heim, C., & Nemeroff, C. B. (1999). The impact ofearly adverse experiences on brain systems involvedin the pathophysiology of anxiety and affective disor-ders. Biological Psychiatry, 46, 1509–1522.

Heimberg, R. G., Liebowitz, M. R., Hope, D. A.,Schneier, F. R., Holt, C. S., Welkowitz, L. A., et al.(1998). Cognitive behavioral group therapy vsphenelzine therapy for social phobia: 12-week out-come. Archives of General Psychiatry, 55, 1133–1141.

Heimberg, R. G., Mueller, G. P., Holt, C. S., Hope, D.A., & Liebowitz, M. R. (1992). Assessment of anxi-ety in social interaction and being observed by oth-ers: The Social Interaction Anxiety Scale and the So-cial Phobia Scale. Behavior Therapy, 23, 53–73.

Hettema, J. M., Neale, M. C., & Kendler, K. S. (2001).A review and meta-analysis of the genetic epidemiol-

ogy of anxiety disorders. American Journal of Psy-chiatry, 158, 1568–1578.

Hollon, S. D., Munoz, R. F., Barlow, D. H., Beardslee,W. R., Bell, C. C., Guillermo, B., et al. (2002).Psychosocial intervention development for the pre-vention and treatment of depression: Promoting in-novation and increasing access. Biological Psychia-try, 52, 610–630.

Huppert, J. D., Bufka, L. F., Barlow, D. H., Gorman, J.M., Shear, M. K., & Woods, S. W. (2001). Therapists,therapist variables and cognitive-behavioral therapyoutcome in a multicenter trial for panic disorder.Journal of Consulting and Clinical Psychology, 69,747–755.

Izard, C. E. (1971). The face of emotion. New York:Appleton–Century–Crofts.

Jacobson, N. S., Martell, C. R., & Dimidjian, S. (2001).Behavioral activation treatment for depression: Re-turning to contextual roots. Clinical Psychology: Sci-ence and Practice, 8, 255–270.

Kamphuis, J. H., & Telch, M. J. (2000). Effects of dis-traction and guided threat reappraisal on fear reduc-tion during exposure based treatments for specificfears. Behaviour Research and Therapy, 38, 1163–1181.

Kendler, K. S. (1996). Major depression and generalizedanxiety disorder: Same genes, (partly) differentenvironments—revisited. British Journal of Psychia-try, 30(Suppl.), 68–75.

Kendler, K. S., Walters, E. E., Neale, M. C., Kessler, R.C., Heath, A. C., & Eaves, L. J. (1995). The structureof genetic and environmental risk factors for six ma-jor psychiatric disorders in women: Phobia, general-ized anxiety disorder, panic disorder, bulimia, majordepression, and alcoholism. Archives of General Psy-chiatry, 52, 374–382.

Kessler, R. C., Nelson, C. B., McGonagle, K. A., Lui,J., Swartz, M., & Blazer, D. G. (1996). Comorbidi-ty of DSM-III-R major depressive disorder in thegeneral population: Results from the NationalComorbidity Survey. British Journal of Psychiatry,168, 17–30.

Kessler, R. C., Stang, P. E., Wittchen, H. U., Ustan, T. B.,Roy-Byrne, P. P., & Walters, E. E. (1998). Lifetimepanic–depression comorbidity in the NationalComorbidity Survey. Archives of General Psychiatry,55, 801–808.

Krueger, R. F., Watson, D., & Barlow, D. H. (Eds.).(2005). Introduction to the special section: Towardsa dimensionally based taxonomy of psychopatholo-gy. Journal of Abnormal Psychology, 114, 491–493.

Kupfer, D. J., First, M. B., & Regier, D. A. (Eds.).(2002). Research agenda for DSM-V. Washington,DC: American Psychiatric Press.

Liebowitz, M. R., Heimberg, R. G., Schneier, F. R.,Hope, D. A., Davies, S., Holt, C. S., et al. (1999).Cognitive-behavioral group therapy versusphenelzine in social phobia: Long term outcome. De-pression and Anxiety, 10, 89–98.

Linehan, M. M. (1993). Cognitive behavioral treatment


of borderline personality disorder. New York:Guilford Press.

Margraf, J., Taylor, C. B., Ehlers, A., Roth, W. T., &Agras, W. S. (1987). Panic attacks in the natural envi-ronment. Journal of Nervous and Mental Disease,175, 558–565.

Mattick, R. P., & Clarke, J. C. (1998). Development andvalidation of measures of social phobia scrutiny fearand social interaction anxiety. Behaviour Researchand Therapy, 36, 455–470.

Merikangas, K. R., Zhang, H., & Aveneoli, S. (2003).Longitudinal trajectories of depression and anxiety ina prospective community study. Archives of GeneralPsychiatry, 60, 993–1000.

Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec,T. D. (1990). Development and validation of thePenn State Worry Questionnaire. Behaviour Re-search and Therapy, 28, 487–495.

Molina, S., & Borkovec, T. D. (1994). The Penn StateWorry Questionnaire: Psychometric properties andassociated characteristics. In G. C. L. Davey & F.Tallis (Eds.), Worrying: Perspectives on theory, as-sessment and treatment (pp. 265–284). New York:Wiley.

Morgan, C. D., & Murray, H. A. (1935). A method forinvestigating fantasies: The Thematic ApperceptionTest. Archives of Neurology and Psychiatry, 34, 289–306.

Nathan, P. E., & Gorman, J. M. (Eds.). (2002). A guideto treatments that work. London: Oxford UniversityPress.

Peterson, R. A., & Reiss, S. (1987). Anxiety SensitivityIndex manual. Worthington, OH: IDS.

Raffa, S. D., Stoddard, J. A., White, K. S., Barlow, D.H., Gorman, J. M., Shear, M. K., et al. (in press). Re-lapse following combined-treatment discontinuationin a placebo-controlled trial for panic disorder. Jour-nal of Nervous and Mental Disease.

Rapee, R. M., Craske, M. G., & Barlow, D. H. (1990).Subject described features of panic attacks using anew self-monitoring form. Journal of Anxiety Disor-ders, 4, 171–181.

Richards, J. M., & Gross, J. J. (2000). Emotion regula-tion and memory: The cognitive costs of keepingone’s cool. Journal of Personality and Social Psychol-ogy, 79, 410–424.

Roemer, L., Litz, B. T., Orsillo, S. M., & Wagner, A. W.(2001). A preliminary investigation of the role ofstrategic withholding of emotions in PTSD. Journalof Traumatic Stress, 14, 143–150.

Roy-Byrne, P. P, Craske, M. G., & Stein, M. B. (2006).Panic disorder. Lancet, 368, 1023–1032.

Rutter, M., Moffit, T. E., & Caspi, A. (2006). Gene–environment interplay and psychopathology: Multi-ple varieties but real effects. Journal of Child Psy-chology and Psychiatry, 47, 226–261.

Salkovskis, P. M., Clark, D. M., Hackmann, A., Wells,A., & Gelder, M. G. (1999). An experimental investi-gation of the role of safety-seeking behaviors in the

maintenance of panic disorder with agoraphobia. Be-haviour Research and Therapy, 37, 559–574.

Salkovskis, P. M., Rimes, K. A., & Warwick, H. M. C.(2002). The Health Anxiety Inventory: Developmentand validation of scales for the measurement ofhealth anxiety and hypochondriasis. PsychologicalMedicine, 32, 843–853.

Salovey, P., Mayer, J. D., Goldman, S., Turvey, C., &Palfai, T. P. (1995). Emotional attention, clarity, andrepair: Exploring emotional intelligence using theTrait Meta-Mood Scale. In J. Pennebaker (Ed.), Emo-tion, disclosure, and health (pp. 125–153). Washing-ton, DC: American Psychological Association.

Salovey, P., Stroud, L. R., Woolery, A., & Epel, E. S.(2002). Perceived emotional intelligence, stress reac-tivity, and symptom reports: Further explorations us-ing the Trait Meta-Mood Scale. Psychology andHealth, 17, 611–627.

Shear, M. K., Brown, T. A., Barlow, D. H., Money, R.,Sholomskas, D. E., Woods, S. W., et al. (1997).Multicenter Collaborative Panic Disorder SeverityScale. American Journal of Psychiatry, 154, 1571–1575.

Shear, M. K., Vander Bilt, J., & Rucci, P. (2001). Reli-ability and validity of a structured interview guide forthe Hamilton Anxiety Rating Scale (SIGH-A). De-pression and Anxiety, 13, 166–178.

Sloan, T., & Telch, M. J. (2002). The effects of safety-seeking behavior and guided threat reappraisal onfear reduction during exposure: An experimental in-vestigation. Behaviour Research and Therapy, 40,235–251.

Smoller, J. W., & Tsuang, M. T. (1998). Panic and pho-bic anxiety: Defining phenotypes for genetic studies.American Journal of Psychiatry, 155, 1152–1162.

Steer, R. A., Ranieri, W. F., Beck, A. T., & Clark, D. A.(1993). Further evidence for the validity of the BeckAnxiety Inventory with psychiatric disorders. Journalof Anxiety Disorders, 7, 195–205.

Suárez, L., Bennett, S. M., Goldstein, C., & Barlow, D.H. (in press). Understanding anxiety disorders from a“triple vulnerability” framework. In M. M. Antony& M. B. Stein (Eds.), Handbook of anxiety and theanxiety disorders. New York: Oxford UniversityPress.

Sullivan, G. M., Kent, J. M., & Coplan, J. D. (2000).The neurobiology of stress and anxiety. In D. I.Mostofsky & D. H. Barlow (Eds.), The managementof stress and anxiety in medical disorders (pp. 15–35). Needham Heights, MA: Allyn & Bacon.

Thayer, R. E. (2000). Mood regulation and generalarousal systems. Psychological Inquiry, 11, 202–204.

Tsao, J. C. I., Lewin, M. R., & Craske, M. G. (1998).The effects of cognitive-behavior therapy for panicdisorders on comorbid conditions. Journal of Anxi-ety Disorders, 12, 357–371.

Tsao, J. C. I., Mystkowski, J. L., Zucker, B. G., &Craske, M. G. (2002). Effects of cognitive-behaviortherapy for panic disorder on comorbid conditions:


Replication and extension. Behavior Therapy, 33,493–509.

Tsao, J. C. I., Mystkowski, J. L., Zucker, B. G., &Craske, M. G. (2005). Impact of cognitive-behavioral therapy for panic disorder on comorbidi-ty: A controlled investigation. Behaviour Researchand Therapy, 43, 959–970.

Tyrer, P. J. (1989). Classification of neurosis. Chichester,UK: Wiley.

Tyrer, P. J., Seivewright, N., Murphys, S., Ferguson, B.,Kingdon, D., Barczak, B., et al. (1998). TheNottingham study of neurotic disorder: Comparisonof drug and psychological treatments. Lancet, 2,235–240.

Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-Item Short-Form Health Survey (SF-36): I. Concep-tual framework and item selection. Quality of LifeResearch, 1, 235–246.

Watson, D. (2005). Rethinking the mood and anxietydisorders: A quantitative hierarchical model forDSM-V. Journal of Abnormal Psychology, 114, 522–536.

Watson, D., Clark, L. A., & Harkness, A. R. (1994).

Structures of personality and their relevance to psy-chopathology. Journal of Abnormal Psychology, 103,18–31.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Devel-opment and validation of brief measures of positiveand negative affect: The PANAS scales. Journal ofPersonality and Social Psychology, 54, 1063–1070.

Wells, A., Clark, D. M., Salkovskis, P. M., Ludgate, J.,Hackmann, A., & Gelder, M. (1995). Social phobia:The role of in-situation safety behaviors in maintain-ing anxiety and negative beliefs. Behavior Therapy,26, 153–161.

Wiedemann, G., Pauli, P., & Dengler, W. (1999). Frontalbrain asymmetry as a biological substrate of emo-tions in patients with panic disorders. Archives ofGeneral Psychiatry, 56, 78–84.

Williams, J. B. (1988). A structured interview guide forthe Hamilton Depression Rating Scale. Archives ofGeneral Psychiatry, 45, 742–747.

Williams, K. E., Chambless, D. L., & Ahrens, A. (1997).Are emotions frightening?: An extension of the fearof fear construct. Behaviour Research and Therapy,35, 239–248.


C H A P T E R 6

Cognitive Therapy for Depression

JEFFREY E. YOUNGJAYNE L. RYGH

ARTHUR D. WEINBERGERAARON T. BECK

One of the most important developments in psychosocial approaches to emotional prob-lems has been the success of cognitive therapy for depression. Evidence for the powerfulefficacy of this approach has increased steadily over the years, particularly in regard tosuccessful long-term outcome. Employing a variety of well-specified cognitive and behav-ioral techniques, cognitive therapy is also distinguished by the detailed structure of eachsession with its specific agendas, and by the very deliberate and obviously effective thera-peutic style of interacting with the patient through a series of questions. Moreover, the au-thors underscore very clearly the importance of the collaborative relationship betweentherapist and patient and outline specific techniques to achieve this collaborative state sothat patient and therapist become an investigative team. In this chapter, the authors pres-ent a second important phase of treatment that represents an interesting variation of cog-nitive therapy. This phase, called the “schema-focused” phase of treatment, concentrateson identifying and modifying early maladaptive or “core” schemas that developed duringchildhood in severely depressed and treatment-resistant patients. These schemas maymake the patient vulnerable to relapse. Detailed explication of this second phase of treat-ment will be invaluable to experienced cognitive therapists, as well as to those becomingacquainted with cognitive therapy for depression for the first time. Two compelling cases,new to this edition, illustrate each approach.—D. H. B.

OVERVIEW AND RESEARCH

Depression and the Emergenceof Cognitive Therapy

Depression is one of the most common disor-ders encountered by mental health profession-als. Recent research from the U.S. Nation-al Comorbidity Survey Replication (NCS-R)studies and data provided by the National In-stitute of Mental Health (NIMH) indicate thefollowing:

• The lifetime prevalence estimate for aDSM-IV mood disorder is 20.8% (Kessler,Beglund, et al., 2005).

• The 12-month prevalence estimate for aDSM-IV mood disorder is 9.5% (Kessler,Chiu, Demler, Merikangas, & Walters, 2005).

• Major depressive disorder is associated with27.2 lost workdays and bipolar disorder (I orII) with 65.5 lost workdays for each illworker per year (Kessler, Akiskal, et al.2006).

250

• Depression increases the risk of heart attacksand is a frequent and serious complicatingfactor in stroke, diabetes, and cancer(NIMH, 1999).

• Major depressive disorder is the leadingcause of disability in the United States forages 15–44 (NIMH, 2006).

• The associated costs are more than $30 bil-lion per year (NIMH, 1999).

The high risk of relapse (Scott, 2000), high re-source utilization (Howland, 1993), and loss ofhuman capital (Berndt et al., 2000) associatedwith depression reveal the seriousness of theproblem. Current estimates suggest that by2010, depression will be the second most costlyof all illnesses worldwide; in 1990 it wasranked fourth (Keller & Boland, 1998). Asthese reports indicate, depression is wide-spread, debilitating, and costly.

No amount of data can adequately captureor convey the personal pain and suffering expe-rienced in depression. Many depressed peopledo not get professional help (Frank & Thase,1999; Jarrett, 1995) and although the numberseeking help has increased over the last decade,undertreatment has remained a serious prob-lem (Olfson et al., 2002). The social stigma stillattached to people with depression is no doubtone factor, but the obstacles encountered whilelooking for appropriate care can be anotherstumbling block in getting help. Obtaining theright type of help can be at once inhibiting andoverwhelming, especially to those already im-paired:

Americans who do seek treatment for depressivesymptoms must decide where to seek which treat-ment and from what type of practitioner. . . . Theclinician must select a somatic, psychological, orcombination of treatment, at a given dose and/orschedule of appointments. . . . Throughout thisprocedure, the patient decides to what extent he/she will comply with the recommendations, forhow long, against recognized and unrecognizedeconomic, practical, physical, and emotionalcosts. . . . Sadly, the lack of information as well asthe continued social stigma of psychiatric illnessand treatment influence decision-making. Simul-taneously, the decisions occur in an environmentfilled with social, political, and economic debate,and tension among policy makers, third-partypayers, and clinicians, as well as among differenttypes of practitioner guilds. (Jarrett, 1995, p. 435)

When care is provided, it is frequently inade-quate, reflecting a public health crisis (Keller &

Boland, 1998). The need for delivery of treat-ments with proven and rapid efficacy remainsparamount.

One of the major developments in the treat-ment of depression has been the emergence ofcognitive therapy, developed by Aaron T. Beckover the past 40-plus years. His work and thatof his colleagues (Beck, 1967, 1976; Beck,Rush, Shaw, & Emery, 1979) has led to a para-digm shift within psychotherapy (Salkovskis,1996). Due in part to Beck’s development oftestable hypotheses and clinical protocols, cog-nitive therapy has received an enormousamount of professional attention (Hollon,1998; McGinn & Young, 1996; Rehm, 1990).Of all the cognitive-behavioral treatment ap-proaches to depression, Beck’s paradigm (Beck,1967; Beck et al., 1979) has received the great-est amount of empirical study, validation, andclinical application (Barlow & Hofmann,1997; de Oliveira, 1998; Dobson & Pusch,1993; Hollon, 1998; Hollon, Thase, &Markowitz, 2002; Rehm, 1990; Roberts &Hartlage, 1996; Scott, 1996a). There are manyexcellent books for practitioners that teachcognitive therapy procedures (e.g., J. S. Beck,1995).

Along with this attention, however, hascome confusion about what is actually meantby the term “cognitive therapy.” The actualcognitive therapeutic strategies employed in“cognitive” treatments may differ in manyways from one another and from those explic-itly prescribed by Beck and colleagues (1979)in their manual for cognitive therapy of depres-sion. Thus, the reader should be aware thatcommon use of the term “cognitive therapy”does not necessarily imply uniformity in proce-dures. The therapy described by Beck and col-leagues involves the use of both cognitive andbehavioral techniques, and can therefore be ac-curately labeled “cognitive-behavioral”; how-ever, in the literature, both terms have been ap-plied in describing the Beck and colleaguesprocedures, with some articles utilizing theterm “cognitive therapy” (Sacco & Beck, 1995,p. 345).

Research on Treatment of the Acute Phase

Outcome research has found cognitive therapyto be effective with clinical populations in anumber of controlled trials (Hollon & Shelton,2001). Although some early studies(Blackburn, Bishop, Glen, Whalley, & Christie,

Cognitive Therapy for Depression 251

1981; Rush, Beck, Kovacs, & Hollon, 1977)suggested that cognitive therapy may be supe-rior to drug treatment for depression at termi-nation, Meterissian and Bradwejn (1989)noted that psychopharmacological interven-tions often were not adequately implemented.In research where interventions have been ade-quate, cognitive therapy generally has beenshown to be equivalent in efficacy to antide-pressant medications, including tricyclic anti-depressants (TCAs) and selective serotoninreuptake inhibitors (SSRIs) in the treatment ofoutpatients with nonbipolar depression(DeRubeis et al., 2005; Hollon et al., 1992;Murphy, Simmons, Wetzel, & Lustman, 1984)and a monoamine oxidase inhibitor (MAOI) inthe treatment of outpatients with atypical de-pression (Jarrett et al., 1999). Although manystudies have not included pill-placebo condi-tions, the studies by Jarrett and colleagues(1999) and DeRubeis and colleagues (2005)found the active treatments superior to pill-placebos. In a mega-analysis by DeRubeis,Gelfand, Tang, and Simons (1999) of treatmentoutcome in four studies with severely depressedoutpatients, cognitive therapy was equivalentto antidepressant medication (imipramine ornortriptyline).

Only two studies that included a pill-placebohave found cognitive therapy to be less effec-tive than psychopharmacological intervention.The first was the NIMH Treatment of Depres-sion Collaborative Research Program(TDCRP) with moderate to severe depressionin adults. The second was the multisite Treat-ment for Adolescents with Depression Study(TADS) for reduction of depressive symptomsin adolescents.

The NIMH TDCRP was the first majorstudy to include a pill-placebo condition. Theinitial results (Elkin et al., 1989) suggestedlower rates of improvement with cognitive-behavioral therapy (CBT) than did earlier stud-ies. It also appeared that with more severelydepressed patient groups, interpersonal psy-chotherapy and antidepressant drugs might besuperior to CBT. The high visibility and pres-tige of the NIMH TDCRP study generated agreat deal of debate (Hollon, DeRubeis, & Ev-ans, 1996; Wolpe, 1993), because it appearedthat the benefits of CBT in the acute treatmentphase might have been overestimated in previ-ous studies. However, on later examination ofthe data, Elkin, Gibbons, Shea, and Shaw(1996) acknowledged Jacobson and Hollon’s

(1996) observation that the outcome resultsvaried across sites, with cognitive therapy per-forming as well as medication at one of thethree sites with severely depressed clients. Ja-cobson and Hollon noted that the best resultswere obtained at the site with the most experi-enced therapists. Hollon and colleagues (2002)“suspect that the explanation is not that cogni-tive therapy cannot be effective with such pa-tients, but that the therapist’s expertise makes agreater difference the more difficult the depres-sion is to treat” (p. 62). Additionally, a studyby Albon and Jones (2003) raises the questionof the distinctness of the two types of psycho-therapy treatments in the TDCRP. In this study,Albon and Jones, expert therapists in CBT andinterpersonal psychotherapy, developed proto-types of ideal regimens of their own respectivetreatments. Then, actual transcripts of treat-ment sessions from the TDCRP were comparedto these expert prototypes. Albon and Jonesfound that both CBT and interpersonal psy-chotherapy sessions conformed most closely tothe cognitive-behavioral prototype, and thatcloser adherence to the cognitive-behavioralprototype produced more positive correlationswith outcome measures across both types oftreatment.

The multisite TADS study (2004) for reduc-tion of depressive symptoms in adolescentsfound that the combination of medication(fluoxetine) and CBT produced the most posi-tive outcome, medication alone was superior topill-placebo but CBT alone did not signifi-cantly differ from pill-placebo. However, asnoted by Weisz, McCarty, and Valeri (2006,p. 144) “the CBT ES (effect size) generated inTADS is not characteristic of most CBT,” rais-ing questions about the administration of CBTin this study.

Several studies have not found the combina-tion of CBT and drugs to be superior to eithertreatment alone with depressed outpatients(Biggs & Rush, 1999; Evans et al., 1992;Hollon, Shelton, & Loosen, 1991; Scott,1996a; Shaw & Segal, 1999). The increment inefficacy appears to be modest in the acutephase of treatment at best, with increases in ef-ficacy from 10 to 20% (Conte, Plutchik, Wild,& Karasu, 1986). Studies regarding treatmentwith depressed inpatients suggest beneficial re-sults when CBT is combined with medication(Bowers, 1990; Miller, Norman, Keitner,Bishop, & Dow, 1989; Stuart & Bowers, 1995;Wright, 1996). Although cognitive therapy ap-


pears to be a useful adjunct to standard carewith inpatients, it remains unclear whethercognitive therapy alone is sufficient (Hollon etal., 2002).

Research on Relapse Prevention

Even though the vast majority of patients re-cover from an episode of depression, they nev-ertheless remain vulnerable to future depres-sion.1

Recurrence is a major problem for many individu-als suffering from depression: at least 50% of in-dividuals who suffer from one depressive episodewill have another within 10 years. Those experi-encing two episodes have a 90% chance of suffer-ing a third, while individuals with three or morelifetime episodes have relapse rates of 40% within15 weeks of recovery from an episode. (Kupfer,Frank, & Wamhoff, 1996, p. 293)

Other investigators have estimated that 85% ofpatients with unipolar depression are likely toexperience recurrences (Keller & Boland,1998, p. 350). As these numbers clearly show,there is an urgent need for treatments capableof minimizing and preventing relapse.

What we consider a very exciting finding inthe treatment of depression with cognitive ther-apy is the consistent observation that patientstreated with cognitive therapy alone or with acombination of cognitive therapy and medica-tion fare far better in terms of relapse than dopatients treated with medication alone (whenboth treatments are stopped at termination).Despite differences in sample characteristicsand methodologies across studies, cognitivetherapy appears to have important prophylac-tic properties. After a 1-year follow-up, numer-ous studies have reported lower relapse ratesfor patients treated with cognitive therapy thanfor patients treated with antidepressants. Forexamples, Simons, Murphy, Levine, and Wetzel(1986) found relapse rates of 12% with cogni-tive therapy versus 66% with antidepressants;Bowers (1990) found relapse rates of 20% withcognitive therapy versus 80% with antidepres-sants; Shea and colleagues (1992) reported 9%relapse with cognitive therapy versus 28% withantidepressants; Hollon and colleagues (2005)reported rates of 31% relapse with cognitivetherapy versus 76% with antidepressants. Re-sults from the most extensive meta-analysis todate revealed that “on average, only 29.5% ofthe patients treated with cognitive therapy

relapsed versus 60% of those treated with anti-depressants” (Gloaguen, Cottraux, Cucherat,& Blackburn, 1998, p. 68). The prophylacticbenefits of cognitive therapy are all the moresignificant because “there is no evidence thatpharmacotherapy confers any protectionagainst the return of symptoms after treatmenthas been terminated.2 Since the majority of de-pressed individuals will experience multiple ep-isodes, the capacity of an intervention to pre-vent the return of symptoms after treatmentmay be at least as important as its ability totreat the current episode” (Evans et al., 1992,p. 802).

A related concern—and one of the most sa-lient with psychotropic agents—is the presenceof residual symptoms after treatment: “Treat-ment of depression by pharmacological meansis likely to leave a substantial amount of resid-ual symptoms in most patients” (Fava,Rafanelli, Grandi, Conti, & Belluardo, 1998b,p. 820). Inevitably, patients who improve onantidepressants continue to manifest some ofthe symptoms of depression, and, as numerousinvestigators have concluded, unless patientsachieve full recovery, residual symptoms in-crease the risk of relapse (Evans et al., 1992;Fava et al., 1998b; Keller & Boland, 1998).

A group of investigators concerned aboutthe risk of relapse associated with residualsymptoms looked at the lingering symptoms af-ter treatment with fluoxetine (Prozac). Theyfound that

even among subjects who are considered full re-sponders to fluoxetine 20 mg for 5 weeks, morethan 80% had 1 or more residual DSM-III-Rsymptoms of major depressive disorder, morethan 30% had 3 or more symptoms, and 10.2%met formal criteria for either minor or sub-syndromal depression. . . . These findings implythat minimal depressive symptoms are prodromaland increase the risk of developing an initial full-blown episode of major depression. (Nierenberget al., 1999, pp. 224–225)

Cognitive therapy has been found to be ef-fective in reducing both residual symptoms andrelapse after the termination of medication:“Short-term CBT after successful antidepres-sant drug therapy had a substantial effect onrelapse rate after discontinuation of antidepres-sant drugs. Patients who received CBT re-ported a substantially lower relapse rate (25%)during the 2-year follow-up than those as-signed to [clinical management] (80%)” (Fava


et al., 1998b, p. 818). The protective benefitsof cognitive therapy were still noticeable in a 4-year follow-up study, although the benefitsfaded after a 6-year period (Fava, Rafanelli,Grandi, Canestrari, & Morphy, 1998a). An-other study found that only 5% of the “CBTtreated and recovered” group sought addi-tional treatment compared with 39% of the an-tidepressant group (see Williams, 1997). Paykeland colleagues (1999) found significantlylower cumulative relapse rates at 68 weeks inpatients who received 16 sessions of CBT fol-lowing a partial response to pharmacotherapy.Bockting and colleagues (2005) comparedtreatment as usual (TAU, which included con-tinuation of medication) with TAU augmentedwith brief cognitive therapy and found relapsesignificantly reduced in patients with five ormore previous episodes of depression. The re-lapse rates were 72% for TAU versus 46% forTAU augmented with brief cognitive therapy.

A preventive strategy used by psychiatrists todeal with high relapse rates associated with an-tidepressant medication is “continuation medi-cation,” which is long-term (and in many caseslifelong) maintenance treatment (Evans et al.,1992; Fava et al., 1998b; Thase, 1999)—usu-ally at the same dosage provided during theacute phase of treatment. Some researchershave pointed out the tautological nature of thissolution: “Drug treatment results in a higherrelapse rate than cognitive-behavioral therapy;therefore patients should be maintainedon drugs to prevent relapse” (Antonuccio,Danton, & DeNelsky, 1995, p. 578). In anycase, research comparing relapse rates in pa-tients continuing medication over the long termversus those treated and then withdrawn fromcognitive therapy does not suggest a significantadvantage to this practice. For example,DeRubeis and colleagues (2005) found thatboth groups had equivalent relapse rates(40%); Hollon and colleagues (2005) foundthat 31% relapsed when treated and then with-drawn from cognitive therapy compared to47% who relapsed when treated with continu-ation medication.

What is the optimum frequency and dura-tion of sessions for cognitive therapy to be ef-fective, both at termination and at long-termfollow-up? According to Sacco and Beck(1995),

General guidelines suggest 15 to 25 (50-minute)sessions at weekly intervals, with more seriously

depressed clients usually requiring twice-weeklymeetings for the initial 4–5 weeks. To avoid anabrupt termination, a “tapering off” process isrecommended, with the last few sessions occur-ring once every 2 weeks. After termination, someclients may also need a few booster sessions (fouror five are common). (p. 332)

Some writers have noted that longer treatmentmay be necessary for a full and more lasting re-covery (Elkin et al., 1996; Thase, 1992). Re-search by Jarrett and colleagues (2001) sug-gests that relapse rates in high-risk patientswith an early age of onset or unstable remissionmight possibly be reduced further with “con-tinuation-phase cognitive therapy” (C-CT),which consists of 10 sessions (biweekly duringthe first 2 months and once a month for the fol-lowing 6 months) following the acute phase oftreatment. The focus in the continuation phaseis on relapse prevention and the generalizationof skills (across responses, settings, stimuli, andtimes).

Another alternative for relapse preventionis mindfulness-based cognitive therapy(MBCT), developed by Teasdale, Segal, andWilliams (1995). MBCT draws from the ac-ceptance and meditation strategies from dia-lectical behavior therapy for borderline per-sonality (Linehan, 1993a, 1993b). “MBCTaims at developing participant’s awareness of,and changing their relationship to, unwantedthoughts, feelings, and body sensations, sothat participants no longer avoid them or re-act to them in an automatic way but ratherrespond to them in an intentional and skilfulmanner” (Ma & Teasdale, 2004, p. 32).Teasdale and colleagues (Ma & Teasdale,2004; Teasdale, 1997a, 1997b; Teasdale etal., 1995, 2002) have argued that the primarymechanism of therapeutic change in cognitivetherapy is in distancing or decentering fromcognition rather than changing the content ofthought. The two studies conducted onMBCT found that TAU in comparison withTAU followed by MBCT significantly reducedrelapse in patients with three or more epi-sodes of depression: 66% relapsed with TAUversus 37% with TAU followed by MBCT(Teasdale et al., 2000), 78% relapsed withTAU versus 36% with TAU followed byMBCT (Ma & Teasdale, 2004). The potentialfor lower costs associated with this treatment,in that it can be implemented in a group for-mat, make it an attractive alternative.


Research Related to Chronicversus Nonchronic Depression

Although much progress has been made in thetreatment of depression with cognitive therapyand/or psychopharmacological interventions, itis important to note that fully a third of pa-tients in some studies do not respond to eithertreatment (Blackburn & More, 1997), and amuch higher percentage do not achieve lastingimprovements (Evans et al., 1992; Shaw &Segal, 1999). McCullough (2003) has pro-posed, on the basis of existing research, thatthere are qualitative differences betweenchronic and nonchronic forms of depressionthat require different treatment strategies. Ac-cording to McCullough, research “shows thatthe chronic disorders, when compared withacute/episodic major depression (both singleand recurrent episodes with interepisode fullrecovery), differ significantly in terms of age ofonset, clinical course patterns, developmentalhistory, modal Axis II comorbidity profiles,characteristic response-to-treatment rates, pre-dictable relapse percentages, and need for long-term treatment” (p. 243). He notes that thesedifferences largely have not been addressedin the treatment literature and, consequently,therapeutic work by psychologists, psychia-trists, and social workers has been conductedas if this is an undifferentiated population.Young, Klosko, and Weishaar (2003) also havenoted that although traditional CBT is highlyeffective for many patients with Axis I disor-ders, those with Axis II disorders “either wentlargely unhelped or were helped with their AxisI disorders but still experienced significantemotional distress and impaired functioning—that is significant character psychopathology”(p. 271). Fortunately, two promising CBT-based treatments have been developed specifi-cally to address chronic populations. Thesetreatments are McCullough’s (2000) cognitive-behavioral analysis system of psychotherapy(CBASP) and Young’s (1990/1999; Young etal., 2003) schema therapy.

Before briefly introducing these newer treat-ments, we describe the general characteristicsof this population. An early age of onset (usu-ally in midadolescence, before age 20) in theform of dysthymia tends to differentiate be-tween the chronic and nonchronic forms of de-pression in 70–75% of clinical populations(Keller & Boland, 1998; Keller & Hanks,1995; McCullough, 2000). Early life trauma or

adverse family relations (loss of parent in child-hood; sexual, physical, and/or verbal abuse;neglect; and overprotection) is more evident inthose with chronic depression (Chapman et al.,2004; Dube et al., 2001; Heim & Nemeroff,2001; Kendler et al., 1995; Lizardi et al., 1995;Randolph & Dykman, 1998; Sachs-Ericsson,Verona, Joiner, & Preacher, 2006). In patientswith early-onset dysthymia who experiencechronic stress, those with the addition of an ad-verse family history evidence an increase in de-pression severity over time compared to thosewithout such adverse family histories, but withfamilial loadings for dysthymic disorder(Dougherty, Klein, & Davila, 2004). Cluster Cpersonality disorders are associated withchronic depression (Hayden & Klein, 2001).Maladaptive schemas (identified with theSchema Questionnaire (Young, 1990/1999) inthe domains of Impaired Autonomy andOvervigilance significantly differentiate be-tween chronic and nonchronic depression, al-though elevations across all schema domainsare evident when comparing both types of de-pression and never-ill controls (Riso et al.,2003).

McCullough’s (2000) CBASP is an integra-tive, time-limited treatment that containselements of cognitive, behavioral, interper-sonal, and psychodynamic psychotherapies.McCullough (2003) states:

Treatment begins with a cognitive-emotionally re-tarded adult child who brings a negative “snap-shot” view of the world to the session. Thechronic patient functions, at least in the social–interpersonal arena, with the structural mindset ofa 4–6 year old preoperational (Piaget) child. . . .The patient must be taught to function formally,to perceive that his or her behavior has conse-quences, be taught to generate authentic empathy,and to learn to assert himself or herself effectively.Psychotherapy begins with an “adult child” whomust be assisted to mature developmentally in thecognitive–emotive sphere. (pp. 247, 248)

Change is brought about through a contingencyprogram that relies on negative reinforcement.First, contingencies between behaviors and con-sequences are exposed. Then, as the result ofpositive changes in behavior, discomfort anddistress are reduced or eliminated. Three tech-niques are used to bring about change: situa-tional analysis, interpersonal discrimination ex-ercise, and behavioral skill training/rehearsal.Two studies have used CBASP with chronically


depressed outpatients. The first (Keller et al.,2000) compared the effects of CBSAP alone,Serzone (nefazodone) alone, and combinedtreatment after a 12-week acute phase of treat-ment. The overall response rate was 48% forboth monotherapies and 73% for the combinedtreatment. Among the 76% (519 out of 681)completers of the study, 52% responded toCBSAP, 55% responded to Serzone, and 85%responded to combined treatments. The second(Klein et al., 2004), a 1-year follow-up to thisinitial study, examined relapse and depressivesymptoms over time in the CBSAP respondersby comparing the effects of continuationCBSAP therapy (16 sessions over 52 weeks)with assessment only. There were significantlyfewer relapses and depressive symptoms forthose in the continuation CBSAP therapy.

Schema therapy also is an integrative ther-apy (Young et al., 2003), with elements of cog-nitive, behavioral, and emotion-focused ther-apies. The focus of treatment is on therelationship with the therapist, daily life events,and early life trauma or adverse family rela-tionships. The treatment incorporates the no-tion of “limited reparenting” for the purpose ofbringing about corrective emotional experi-ences with respect to the patient’s early unmetbasic human needs for safety, nurturance, au-tonomy, self-expression, and/or limits. Accord-ing to Young, when basic needs are not met,early maladaptive schemas (EMSs) within theindividual (child or adolescent) are likely to de-velop. EMSs “refer to extremely stable and en-during themes that develop and are elaboratedupon throughout the individual’s lifetime andthat are dysfunctional to a significant degree”(Young, 1990/1999, p. 9). Schemas are con-structions of reality (which are deeply heldemotion-based beliefs) that develop as the re-sult of concrete experiences with the environ-ment, particularly those early in life with signif-icant others. Once developed, they are mostlyoutside of awareness and remain dormant untila life event stimulates one or more schemas.Once schema(s) are activated, the individualautomatically processes information (cogni-tive, behavioral, affective, and interpersonal) inways accordant with the schema(s). Young hasidentified 18 EMSs in five hypothesized do-mains (see Figure 6.1 on pages 261–263). Mostof these schemas have been supported by sub-sequent research (Lee, Taylor, & Dunn, 1999;Schmidt, 1994; Schmidt, Joiner, Young, &Telch, 1995). Once the schema(s) have become

activated, Young has also observed specificcharacteristic modes of responding: childmodes, maladaptive coping modes, and/or dys-functional parent modes (see Figures 6.2a–con pages 264–265). A mode is “the set ofschemas or schema operations—adaptive ormaladaptive—that are currently active for anindividual” (Young et al., 2003, p. 271). Theautomatic processing and responses that occurwhen one is caught in a maladaptive mode pre-serve the underlying maladaptive schema(s).The focus of schema therapy is to increase thepatient’s awareness of the schemas and modes,understand the origin (and possible prior adap-tive value) of the modes, link the modes to cur-rent experience, and to modify and/or incorpo-rate new, more adaptive modes. This isaccomplished by accessing the vulnerable childmode through imagery and helping the patientto reconstruct or alter the historical record ofevents by bringing in a newly developed orstrengthened “healthy adult mode” to guide,support, set limits, and/or protect that vulnera-ble child. A recent study by Giesen-Bloo andcolleagues (2006) found surprisingly strongand significant results in favor of schema-focused therapy over transference-focusedtherapy with the chronic disorder of borderlinepersonality. After 3 years of treatment (sessionstwice per week), 45% of the patients inschema-focused therapy (vs. 24% intransference-focused therapy) completely re-covered. One year later, over half (52%) inschema-focused therapy fully recovered (vs.29% in transference-focused therapy) and two-thirds (70%) in schema-focused therapyshowed significant improvement. In addition,patients in schema-focused therapy were signif-icantly less likely to drop out of therapy (27%dropped out of schema-focused therapy vs.50% dropped out of transference-focused ther-apy). Given the high degree of similarities be-tween patients with chronic disorders (adversechildhood histories, early-onset depression,and a multitude of schemas), we believe thatschema-focused therapy is very likely to be aneffective treatment for the chronically de-pressed population as well.

Research on Pharmacological Interventionsand Cognitive Therapy

Despite the findings cited earlier, drugs are stillthe initial and most frequently prescribed formof treatment for unipolar depression in the


United Stares (Antonuccio et al., 1995) and themost commonly used method to maintaintreatment gains (Geddes et al., 2003). Severalfactors likely have contributed to this treat-ment bias. The government’s DepressionGuideline Panel (1993) recommended that twotrials of antidepressants, both unsuccessful,should be completed before recommendingpsychotherapy for depression. Although a laterreport from the Surgeon General (MentalHealth: A report of the Surgeon General, 1997,Chapter 4) summarized research indicating theeffectiveness of CBT for depression, specifictreatment recommendations were not made. Assuch, the only recommendations made by thegovernment have remained unrevised. Addi-tionally, the most recent White House Confer-ence on Mental Health continued to reveala psychopharmacological treatment bias(Saeman, 1999). Treatment guidelines, such asthe one issued by the American Psychiatric As-sociation (2000), also continue to recommendthat antidepressants be used to treat moderateto severe depression. Apart from the contribu-tion of the political environment to this bias isthat of the managed care environment. Man-aged care companies often use medical doctorsas gatekeepers to the type of care a patient re-ceives. Obvious professional allegiance is builtinto this system. Although psychotropic medi-cations clearly provide enormous benefit andrelief to a substantial portion of depressed pa-tients, research does not support this bias to-ward medication. Three important points havenot been adequately considered: premature ter-mination and iatrogenic effects of pharma-cotherapy, patient preference, and cost-effectiveness.

Research has shown that “a sizeable groupof patients either chooses not to continue long-term pharmacotherapy in the absence of anydepressive symptoms, cannot take medicationdue to a medical condition that precludes theuse of antidepressants, or suffer from side ef-fects that are intolerable to them” (Spanier,Frank, McEachran, Grochocinski, & Kupfer,1999, p. 250). In fact, one group of researchersconcluded that “there is much evidence thatantidepressant medications are not benigntreatments. . . . Many antidepressants arecardiotoxic, have dangerous side effects, andare often used in suicide attempts. . . . [Theyalso] result in relatively poorer compliancethan psychotherapy, have a higher dropoutrate, and result in as much as a 60% non-

response rate with some patient populations”(Antonuccio et al., 1995, p. 581). (Of course,psychotherapy may also have unintended andundesirable side effects [Mohr, 1995], but verylittle is known about any negative effects asso-ciated with cognitive therapy.)

Another consideration is the preference ofpatients, the “consumers.” Research suggeststhat consumers seem to have a preference forpsychotherapy over medication for depression:“The results of treatment acceptability studiesshow that, as treatments for depression inadults, psychotherapies are perceived by poten-tial consumers as more acceptable thanpharmacotherapy alone as well as combina-tions of psychotherapy and pharmacotherapy”(Hall & Robertson, 1998, p. 271). This is notsurprising given that most people attributetheir depressions to negative life experiences,with stress and environmental factors playingmajor roles (Antonuccio et al., 1995; Brown,1996; Eifert, Beach, & Wilson, 1998).

A final consideration is the cost-effectivenessof treatment. The research on this topic is sur-prisingly limited (Johnsson & Bebbington,1994; Rosenbaum & Hylan, 1999; Scott &Freeman, 1992). Antonuccio, Thomas, andDanton (1997) conducted a cost-effectivenessanalysis on several outcome studies on depres-sion and found that over a 2-year period thecost of fluoxetine alone was 33% higher thanindividual CBT and that combined treatmentof flouxetine and CBT was 23% higher thanCBT alone. A recent Australian study (Vos,Corry, Haby, Carter, & Andrews, 2005) exam-ined the cost and benefits of CBT and drugs inthe episodic and maintenance treatment of de-pression. They found maintenance treatmentwith SSRIs to be the most expensive option,nearly double that of bibliotherapy, groupCBT, individual CBT, and TCAs. Scott, Palmer,Paykel, Teasdale, and Hayhurst (2003) com-pared the cost-effectiveness of relapse preven-tion in depression by comparing CBT as an ad-junctive treatment to antidepressants andclinical management, and with antidepressantsand clinical management alone, over a 17-month period. They found the addition of CBTwas not only more expensive but also more ef-fective in reducing cumulative relapse rates (re-lapse rates were 29% with adjunctive CBT vs.47% without adjunctive CBT). Importantly,they noted that the costs associated with ad-junctive CBT in their study were a “worst-case” scenario, because the follow-up extended


only 17 months; other research (as noted in theprevious section) has demonstrated mainte-nance of gains and lowered relapse rates up to6 years with CBT. Scott and colleagues arguedthat the incrementally adjusted costs over timeto relapse should be considered in decisionmaking on the value of various treatments.These initial findings suggest that CBT, eitheralone or in combination with drugs, can en-hance cost-effectiveness, particularly whenhigher short-term costs of combined treatmentsare balanced by better outcomes and lowermarginal costs in the long term.

Research Controversy and Criticism

Despite the fact that cognitive therapy is themost empirically tested psychotherapy for de-pression currently available, it has not been un-equivocally embraced. Some have questionedthe validity of the findings on the basis of meth-odological flaws or insufficient data (Gelder,1994; Klein, 1996; Scott, 1996a), whereasothers have attempted to temper the results byattributing them to “experimental alle-giance”—the preference by the researcher forone therapy over others (Gaffan, Tsaousis, &Kemp-Wheeler, 1995; Luborsky et al., 1999).Some behaviorists have tried to dismiss cogni-tive therapy altogether by claiming that it isbased on “cognitivist oversell” and “self-deception” (Wolpe, 1993, p. 143).

Parker, Roy, and Eyers (2003) conducted areview of meta-analyses and major studies, andconcluded that CBT was less effective than sug-gested by adherents. However, Butler, Chap-man, Forman, and Beck (2006, p. 20) notedthat Parker and colleagues (2003)

excluded . . . high quality clinical trials showingcognitive-behavioral therapy to be superior to al-ternative treatments at follow-up (see Shapiro etal., 1994, for a comparison with psychodynamictherapy; Blatt, Zuroff, Bondi, & Sanislow, 2000,for a comparison with antidepressant medica-tion). Since Parker et al. did not report the criteriaby which they selected research studies for theirreview, it is difficult to interpret their conclusions.

In the most extensive and methodologically rig-orous meta-analysis to date, Gloaguen and col-leagues (1998) found that CBT was superior towaiting-list or placebo controls, modestly supe-rior (ES = 0.24) to a miscellaneous group ofpsychotherapies, and equivalent to behaviortherapy. Wampold, Minami, Baskin, and

Tierney (2000) noted that some of the therapiesincluded in the miscellaneous group were notcomprehensive treatments for depression (e.g.,progressive muscular relaxation). After catego-rizing these miscellaneous therapies into estab-lished and nonestablished treatments for de-pression, they found CBT only marginallysuperior (ES = 0.16) to other established treat-ments. As Butler and colleagues (2006) noted,

a heterogeneous collection of miscellaneous thera-pies, as done by Gloaguen et al., can at best pro-vide limited information. Future meta-analysesare needed that provide direct comparisons be-tween CBT and specific alternative therapies. . . .There are relatively few direct comparisons andnot enough to provide the basis for a meaningfulmeta-analysis. (pp. 22–23)

A recent meta-analysis of treatment outcomeresearch on child and adolescent depression(Weisz et al., 2006) found that CBT performedno better than noncognitive psychotherapytreatments, all of which produced effects thatwere “significant but modest in their strength,breadth, and durability” (p. 132). The resultsof this meta-analysis stand in stark contrast totwo prior meta-analyses published on CBTwith depressed youth (Lewinsohn & Clarke,1999; Reinecke et al., 1998), which indicatedlarge treatment effects. The Weisz and col-leagues (2006) meta-analysis included bothpeer-reviewed and non-peer-reviewed studies,and included results from the TADS (2004)study. Excluding the TADS results, the mean ESof the other 23 CBT studies (out of a total of 44studies) was in the medium range (0.48). Weiszand colleagues noted a broad range of ESsacross the studies (including the CBT studies)in their meta-analysis. They stated: “Indeed,five different treatment programs generated ef-fects exceeding 1.0. Thus, some treatments inthe current armamentarium may already havestrong potential”; however, they also notedthat “the strongest potential may not attach tothe most popular treatments. In the currentzeitgeist, treatments that focus on altering un-realistic, negative cognitions, have particularlyprominent status” (p. 144).

There has been some debate as to whetherthe cognitive components of treatment are nec-essary for treating depression. Some research-ers have argued that the behavioral aspects ofcognitive therapy are primarily responsible forimprovement. For example, the results fromGloaguen and colleagues’ (1998) meta-


analysis, based on clinical trials conducted be-tween 1977 and 1996, indicated that cognitivetherapy was equal to behavior therapy. In an-other study, a component analysis conductedby Jacobson and colleagues (1996), behavioralactivation (BA) alone produced as muchchange as cognitive therapy. Additionally,Dimidjian and colleagues (2006) found a morecomprehensive version of BA (Martell, Addis,& Jacobson, 2001) to be equivalent to medica-tion, outperforming cognitive therapy for indi-viduals with moderate to severe depression atboth 8 weeks and 16 weeks of treatment.Dimidjian and colleagues noted:

A. T. Beck and colleagues (1979) have long sug-gested that therapists focus on behavioral strate-gies early in treatment when patients are more de-pressed and return to that emphasis later ifpatients start to worsen. Although the currentdata do not specifically address whether change incognition is a mediator of symptom change, theyprovide strong evidence that behavioral methodsare sufficient to produce symptom change irre-spective of whether improvement is mediated bycognitive change or not. (cf. Bandura, 1977)(p. 667)

It is important to note that length of treatmentin the Dimidjian and colleagues study was onthe short end (15–25 weeks) of the recom-mended length of CBT treatment and did notexamine long-term efficacy. A study by Gortner,Gollan, Dobson, and Jacoboson (1998), whichfollowed up the Jacobson and colleagues re-search, however, found that patients exposed tothe behavioral components of treatment wereno more likely to relapse following terminationthan those treated with both cognitive andbehavioral components. These findings raiseimportant questions for further research regard-ing parsimony of treatment.

Current Status and Future Researchon Treating Depression

There is now a considerable body of researchon antidepressant and cognitive treatments fordepression. Certainly there are still enough in-consistencies in the literature to warrant con-tinued debate and research regarding the rel-ative merits of different treatments fordepression (Agosti & Ocepek, 1997; Gortneret al., 1998; Oei & Free, 1995; Robinson,Berman, & Neimeyer, 1990; Stewart,Garfinkel, Nunes, Donovan, & Klein, 1998;

Thase et al., 1997). Nevertheless, the efficacyof cognitive therapy for depression is clearly areplicable and robust finding.

Our own hope, however, is that we will see amovement away from the current “horse race”approach so prevalent in the research literature(Williams, 1997). Many important, unan-swered questions make it impossible, in ouropinion, for proponents of any one treatmentto make firm recommendations favoring onetreatment for depression over another. In thespirit of Beck’s testable hypotheses and clinicalprotocols, and through more sophisticated re-search studies, we hope that it will be possibleto assess which types of depressed patients willbenefit most from which type of treatment, orcombination of treatments, and in what se-quence.

The remainder of this chapter is devoted todetailing the basic characteristics of cognitivetherapy and demonstrating applications of cog-nitive therapy to depression in clinical practice.

COGNITIVE MODEL OF DEPRESSION

The cognitive model assumes that cognition,behavior, and biochemistry are all importantcomponents of depressive disorders. We do notview them as competing theories of depressionbut as different levels of analysis. Each treat-ment approach has its own “focus of conve-nience.” The pharmacotherapist intervenes atthe biochemical level; the cognitive therapistintervenes at the cognitive, affective, andbehavioral levels. Our experience suggests thatwhen we change depressive cognitions, we si-multaneously change the characteristic mood,the behavior, and, as some evidence suggests(Free, Oei, & Appleton, 1998; Joffe, Segal, &Singer, 1996), the biochemistry of depression.Although the exact mechanism of change re-mains a target of considerable investigation,speculation, and debate (Barber & DeRubeis,1989; Castonguay, Goldfried, Wiser, Raue, &Hayes, 1996; Crews & Harrison, 1995;DeRubeis et al., 1990; DeRubeis & Feeley,1990; Hayes & Strauss, 1998; Oei & Free,1995; Oei & Shuttlewood, 1996; Shea &Elkin, 1996; Sullivan & Conway, 1991;Whisman, 1993), “there are indications thatcognitive therapy works by virtue of changingbeliefs and information-processing proclivitiesand that different aspects of cognition play dif-


ferent roles in the process of change” (Hollonet al., 1996, p. 314).

Our focus in this chapter is on the cognitivedisturbances in depression. Cognitive scienceresearch emphasizes the importance of infor-mation processing in depressive symptomatol-ogy (Ingram & Holle, 1992). According tothese theories, negatively biased cognition is acore process in depression. This process is re-flected in the “cognitive triad of depression”:Depressed patients typically have a negativeview of themselves, of their environment, andof the future. They view themselves as worth-less, inadequate, unlovable, and deficient. De-pressed patients view the environment as over-whelming, as presenting insuperable obstaclesthat cannot be overcome, and that continuallyresult in failure or loss. Moreover, they viewthe future as hopeless; they believe their ownefforts will be insufficient to change the un-satisfying course of their lives. This negativeview of the future often leads to suicidal idea-tion and actual suicide attempts.

Depressed patients consistently distort theirinterpretations of events, so that they maintainnegative views of themselves, the environment,and the future. These distortions represent de-viations from the logical processes of thinkingtypically used by people. For example, a de-pressed woman whose husband comes homelate one night may conclude that he is havingan affair with another woman, even thoughthere is no other evidence supporting this con-clusion. This example illustrates an “arbitraryinference,” reaching a conclusion that is notjustified by the available evidence. Other dis-tortions include all-or-nothing thinking,overgeneralization, selective abstraction, andmagnification (Beck et al., 1979).

According to subsequent developmentswithin the cognitive model, an important pre-disposing factor for many patients with depres-sion is the presence of early schemas (Stein& Young, 1992; Young, 1990/1999).3 Beck(1976) emphasized the importance of schemasin depression and provided the following defi-nition:

A schema is a cognitive structure for screening,coding, and evaluating the stimuli that impinge onthe organism. . . . On the basis of this matrix ofschemas, the individual is able to orient himself inrelation to time and space and to categorize andinterpret experiences in a meaningful way.(p. 233)

Furthermore, Beck, Freeman, and Associates(1990) noted:

In the field of psychopathology, the term“schema” has been applied to structures with ahighly personalized idiosyncratic content that areactivated during disorders such as depression,anxiety, panic attacks, and obsessions, and be-come prepotent. . . . Thus, in clinical depression,for example, the negative schemas are in ascen-dancy, resulting in a systematic negative bias inthe interpretation and recall of experiences as wellas in short-term and long-term predictions,whereas the positive schemas become less accessi-ble. It is easy for depressed patients to see the neg-ative aspects of an event, but difficult to see thepositive. They can recall negative events muchmore readily than positive ones. They weigh theprobabilities of undesirable outcomes moreheavily than positive outcomes. (p. 32)

It is also becoming increasingly recognized that“focusing on core schemas is a key to effectiveshort-term therapy” (Freeman & Davison,1997, p. 8).

Through clinical observation, Young hasidentified a subset of schemas that he terms“early maladaptive schemas.” The 18 EMSs infive hypothesized domains identified by Youngcan be seen in Figure 6.1. According to Young’s(1990/1999) schema approach, children learnto construct reality through early experienceswith the environment, especially with signifi-cant others. EMSs likely develop when the en-vironment does not meet core needs for safety,stability or predictability, love, nurturance andattention, acceptance and praise, empathy, re-alistic limits, and validation of feelings andneeds. Sometimes these early experiences leadchildren to accept attitudes and beliefs thatlater prove maladaptive. For example, a childmay develop the schema that no matter whathe or she does, his or her performance willnever be good enough. These schemas usuallyoccur outside of awareness and may remaindormant until a life event (e.g., being fired froma job) stimulates the schema. Once the schemais activated, the patient categorizes, selects, andencodes information in such a way that the fail-ure schema is maintained. EMSs therefore pre-dispose depressed patients to distort events in acharacteristic fashion, leading to a negativeview of themselves, the environment, and thefuture.

EMSs have several defining characteristicsthat are (1) a priori truths about oneself and/or



DISCONNECTION AND REJECTIONExpectation that one’s needs for security safety, stability, nurturance, empathy, sharing of feelings, acceptance,and respect will not be met in a predictable manner. Typical family origin is detached, cold, rejecting,withholding, lonely, explosive, unpredictable, or abusive.

1. ABANDONMENT/INSTABILITYThe perceived instability or unreliability of those available for support and connection. Involves the sensethat significant others will not be able to continue providing emotional support, connection, strength, orpractical protection because they are emotionally unstable and unpredictable (e.g., angry outbursts).unreliable or erratically present; because they will die imminently; or because they will abandon the patientin favor of someone better.

2. MISTRUST/ABUSEThe expectation that others will hurt, abuse, humiliate, cheat, lie, manipulate, or take advantage. Usuallyinvolves the perception that the harm is intentional or the result of unjustified and extreme negligence. Mayinclude the sense that one always ends up being cheated relative to others or “getting the short end ofthe stick.”

3. EMOTIONAL DEPRIVATIONExpectation that one’s desire for a normal degree of emotional support will not be adequately met byothers. The three major forms of deprivation are:A. Deprivation of nurturance: Absence of attention, affection, warmth, or companionship.B. Deprivation of empathy: Absence of understanding, listening, self-disclosure, or mutual sharing of

feelings from others.C. Deprivation of protection: Absence of strength, direction, or guidance from others.

4. DEFECTIVENESS/SHAMEThe feeling that one is defective, bad, unwanted, inferior, or invalid in important respects; or that onewould be unlovable to significant others it exposed. May Involve hypersensitivity to criticism, rejection, andblame; self-consciousness, comparisons, and insecurity around others; or a sense of shame regardingone’s perceived flaws. These flaws may be private (e.g., selfishness, angry impulses, unacceptable sexualdesires) or public (e.g., undesirable physical appearance, social awkwardness).

5. SOCIAL ISOLATION/ALIENATIONThe feeling that one is isolated from the rest of the world, different from other people, and/or not part ofany group or community.

IMPAIRED AUTONOMY AND PERFORMANCEExpectations about oneself and the environment that interfere with one’s perceived ability to separate, survive,function independently, or perform successfully. Typical family origin is enmeshed, undermining of child’sconfidence, overprotective, or failing to reinforce child for performing competently outside the family.

6. DEPENDENCE/INCOMPETENCEBelief that one is unable to handle one’s everyday responsibilities in a competent manner, withoutconsiderable help from others (e.g., take care of oneself, solve daily problems, exercise good judgment,tackle new tasks, make good decisions). Often presents as helplessness.

7. VULNERABILITY TO HARM OR ILLNESSExaggerated fear that imminent catastrophe will strike at any time and that one will be unable to prevent it.Fears focus on one or more of the following: (A) medical catastrophes (e.g., heart attacks, AIDS); (B)emotional catastrophes (e.g., going crazy); (C) external catastrophes (e.g., elevators collapsing, victimizedby criminals, airplane crashes, earthquakes).

(continued)

FIGURE 6.1. Early maladaptive schemas with associated schema domains (revised November 1998).Copyright 1999 by Jeffrey E. Young, PhD. Reprinted with permission. Unauthorized reproduction with-out written consent of the author is prohibited. For more information, write Cognitive Therapy Centerof New York, 36 West 44th Street, Suite 1007, New York, NY 10036.


8. ENMESHMENT/UNDEVOLOPED SELFExcessive emotional involvement and closeness with one or more significant others (often parents), at theexpense of full individuation or normal social development. Often involves the belief that at least one ofthe enmeshed individuals cannot survive or be happy without the constant support of the other. May alsoinclude feelings of being smothered by, or fused with, others OR insufficient individual identity. Oftenexperienced as a feeling of emptiness and floundering, having no direction, or in extreme casesquestioning one’s existence.

9. FAILUREThe belief that one has failed, will inevitably fail, or is fundamentally inadequate relative to one’s peers, inareas of achievement (school, career, sports, etc.) Often involves beliefs that one is stupid, inept,untalented. ignorant, lower in status, less successful than others, etc.

IMPAIRED LIMITSDeficiency in internal limits, responsibility to others, or long-term goal orientation. Leads to difficulty respectingthe rights of others, cooperating with others making commitments, or setting and meeting realistic personalgoals. Typical family origin is characterized by permissiveness, overindulgence, lack of direction, or a sense ofsuperiority—rather than appropriate confrontation discipline and limits in relation to taking responsibility,cooperating in a reciprocal manner and setting goals. In some cases, child may not have been pushed totolerate normal levels of discomfort, or may not have been given adequate supervision, direction or guidance.

10. ENTITLEMENT/GRANDIOSITYThe belief that one is superior to other people; entitled to special rights and privileges; or not bound by therules of reciprocity that guide normal social interaction. Often involves insistence that one should be ableto do or have whatever one wants regardless of what is realistic, what others consider reasonable, or thecost to others; OR an exaggerated focus on superiority (e.g., being among the most successful, famous,and wealthy) in order to achieve power or control (not primarily for attention or approval). Sometimesincludes excessive competitiveness toward, or domination of others: asserting ones power, forcing one’spoint of view, or controlling the behavior of others in line with one’s own desires—without empathy orconcern for others’ needs or feelings.

11. INSUFFICIENT SELF-CONTROL/SELF-DISCIPLINEPervasive difficulty or refusal to exercise sufficient self-control and frustration tolerance to achieve one’spersonal goals, or to restrain the excessive expression of one’s emotions and impulses. In its milder form,patient presents with an exaggerated emphasis on discomfort avoidance: avoiding pain, conflict,confrontation, responsibility, or overexertion—at the expense of personal fulfillment, commitment, orintegrity.

OTHER-DIRECTEDNESSAn excessive focus on the desires, feelings and responses of others, at the expense of one’s own needs—inorder to gain love and approval, maintain one’s sense of connection or avoid retaliation. Usually involvessuppression and lack of awareness regarding one’s own anger and natural inclinations. Typical family origin Isbased on conditional acceptance: Children must suppress important aspects of themselves in order to gainlove, attention, and approval. In many such families, the parents’ emotional needs and desires—or socialacceptance and status—are valued more than the unique needs and feelings of each child.

12. SUBJUGATIONExcessive surrendering of control to others because one feels coerced—usually to avoid anger, retaliation,or abandonment. The two major forms of subjugation are:A. Subjugation of needs: Suppression of one’s preferences, decisions, and desires.B. Subjugation of emotion: Suppression of emotional expression, especially anger.Usually involves the perception that one’s own desires, opinions, and feelings are not valid or important toothers. Frequently presents as excessive compliance, combined with hypersensitivity to feeling trapped.Generally leads to a buildup of anger, manifested in maladaptive symptoms (e.g., passive–aggressivebehavior, uncontrolled outbursts of temper, psychosomatic symptoms, withdrawal of affection, “acting out,”substance abuse).

(continued)FIGURE 6.1. (continued)


13. SELF-SACRIFICEExcessive focus on voluntarily meeting the needs of others in daily situations, at the expense of one’s owngratification. The most common reasons are to prevent causing pain to others; to avoid guilt from feelingselfish; or to maintain the connection with others perceived as needy. Often results from an acutesensitivity to the pain of others. Sometimes leads to a sense that one’s own needs are not beingadequately met and to resentment of those who are taken care of. (Overlaps with concept ofcodependency.)

14. APPROVAL SEEKING/RECOGNITION SEEKINGExcessive emphasis on gaining approval, recognition or attention from other people or fitting in, at theexpense of developing a secure and true sense of self. One’s sense of esteem is dependent primarily onthe reactions of others rather than on one’s own natural inclinations. Sometimes includes an overemphasison status, appearance, social acceptance, money, or achievements—as a means of gaining approval,admiration, or attention (not primarily for power or control). Frequently results in major life decisions thatare inauthentic or unsatisfying, or in hypersensitivity to rejection.

OVERVIGILANCE AND INHIBITIONExcessive emphasis on suppressing one’s spontaneous feelings, impulses, and choices OR on meeting rigid,internalized rules and expectations about performance and ethical behavior—often at the expense ofhappiness, self-expression, relaxation, close relationships, or health. Typical family origin is grim, demanding,and sometimes punitive: Performance, duty, perfectionism—following rules, hiding emotions, and avoidingmistakes predominate over pleasure, joy, and relaxation. There is usually an undercurrent of pessimism andworry that things could fall apart if one fails to be vigilant and careful at all times.

15. NEGATIVITY/PESSIMISMA pervasive, lifelong focus on the negative aspects of life (pain, death, loss, disappointment, conflict, guilt,resentment, unsolved problems, potential mistakes, betrayal, things that could go wrong, etc.) whileminimizing or neglecting the positive or optimistic aspects. Usually includes an exaggerated expectation—in a wide range of work, financial, or interpersonal situations—that things will eventually go seriouslywrong, or that aspects of one’s life that seem to be going well will ultimately fall apart. Usually involves aninordinate fear of making mistakes that might lead to financial collapse, loss, humiliation or being trappedin a bad situation. Because potential negative outcomes are exaggerated, these patients are frequentlycharacterized by chronic worry, vigilance, complaining, or indecision.

16. EMOTIONAL INHIBITIONThe excessive inhibition of spontaneous action, feeling, or communication—usually to avoid disapproval byothers, feelings of shame, or losing control of one’s impulses The most common areas of inhibition involve(A) inhibition of anger and aggression; (B) inhibition of positive impulses (e.g., joy, affection, sexualexcitement, play); (C) difficulty expressing vulnerability or communicating freely about one’s feelings,needs, etc.; or (D) excessive emphasis on rationality while disregarding emotions.

17. UNRELENTING STANDARDS/HYPERCRITICALNESSThe underlying belief that one must strive to meet very high internalized standards of behavior andperformance usually to avoid criticism. Typically results in feelings of pressure or difficulty slowing downand in hypercriticalness toward oneself and others. Must involve significant impairment in pleasure,relaxation, health, self-esteem sense of accomplishment, or satisfying relationships. Unrelenting standardstypically present as (A) perfectionism, inordinate attention to detail, or an underestimate of how good one’sown performance is relative to the norm; (B) rigid rules and “shoulds” in many areas of life, includingunrealistically high moral, ethical, cultural or religious precepts; or (C) preoccupation with time andefficiency so that more can be accomplished.

18. PUNITIVENESSThe belief that people should be harshly punished for making mistakes. Involves the tendency to be angry,intolerant, punitive, and impatient with those people (including oneself) who do not meet one’sexpectations or standards. Usually includes difficulty forgiving mistakes in oneself or others because of areluctance to consider extenuating circumstances, allow for human imperfection, or empathize withfeelings.

FIGURE 6.1. (continued)

the environment; (2) self-perpetuating and re-sistant to change; (3) dysfunctional; (4) oftentriggered by some environmental change (e.g.,loss of a job or mate); (5) tied to high levels ofaffect when activated; and (6) usually resultfrom an interaction of the child’s innate tem-perament with dysfunctional developmentalexperiences with family members or caretakers(Young, 1990/1999).

When a multitude of EMSs develop, and aredeeply entrenched, or the patient is highlyavoidant or tends to be stuck in overcompen-sating for EMSs, as is likely in patients withchronic depression, the additional concept of“schema modes” or “modes” (Young et al.,2003) becomes therapeutically useful. Young’snotion of a mode is similar to that of an egostate. A mode is defined as “those schemas orschema operations—adaptive or maladaptive—that are currently active for an individual”(p. 271). A dysfunctional mode is activatedwhen specific maladaptive schemas or copingresponses have erupted into distressing emo-tions, avoidance responses, or self-defeatingbehaviors that take over and control an indi-vidual’s functioning at a given point in time. Anindividual also may shift from one dysfunction-al mode into another; as that shift occurs, dif-ferent schemas or coping responses, previously

dormant, become active (Young et al., 2003).Rapid shifts are referred to as “flipping”modes. Young and colleagues (2003) identifiedfour main types of modes: Child modes (Figure6.2a), Maladaptive Coping modes (Figure6.2b), Dysfunctional Parent modes (Figure6.2c), and Healthy Adult mode. The HealthyAdult mode is

the healthy, adult part of the self that serves an“executive” function relative to the other modes.The Healthy Adult helps meet the child’s basicemotional needs. Building and strengthening thepatient’s Healthy Adult to work with the othermodes more effectively is the overarching goal ofmode work in schema therapy. (p. 277)

In the following sections we present infor-mation about the general characteristics andnature of cognitive therapy, a discussion of theprocess of cognitive therapy, then two casesthat illustrate cognitive therapy in action. Thepatient in the first case has a nonchronic formof depression and is treated with standard cog-nitive therapy. The focus of standard cognitivetherapy is on changing depressive thinking.The patient in the second case has chronic de-pression and is treated with schema therapy.The focus of schema therapy, in this case, is on


FIGURE 6.2a. Child modes. From Young, Klosko, and Weishaar (2003). Copyright 2003 by TheGuilford Press. Reprinted by permission.

Child Mode Description Common associated schemas

Vulnerable Child Experiences dysphoric oranxious affect, especially fear,sadness, and helplessness,when “in touch” with associatedschemas.

Abandonment, Mistrust/Abuse,Emotional Deprivation,Defectiveness, Social Isolation,Dependence/Incompetence,Vulnerability to Harm or Illness,Enmeshment/Undeveloped Self,Negativity/Pessimism.

Angry Child Vents anger directly in responseto perceived unmet core needsor unfair treatment related tocore schemas.

Abandonment, Mistrust/Abuse,Emotional Deprivation,Subjugation (or, at times, any ofthe schemas associated with theVulnerable Child).

Impulsive/Undisciplined Child Impulsively acts according toimmediate desires for pleasurewithout regard to limits or others’needs or feelings (not linked tocore needs).

Entitlement, Insufficient Self-Control/Self-Discipline.

Happy Child Feels loved, connected, content,satisfied.

None. Absence of activatedschemas.

identifying and modifying the patient’s under-lying schemas through mode work. Modework strengthens and develops healthy modesand weakens dysfunctional modes, thereby in-creasing control over one’s responding.

CHARACTERISTICS OF THERAPY

Cognitive therapy with adult depressed outpa-tients is usually undertaken in the therapist’soffice. It has most frequently been applied in aone-to-one setting. However, group cognitivetherapy has also been shown to be successfulwith many depressed outpatients (Beutler et al.,1987; Jarrett & Nelson, 1987), although itmay not be as effective as individual treatment(Wierzbicki & Bartlett, 1987). Additionally,computer-assisted cognitive therapy for depres-sion (50-minute vs. 25-minute sessions withcomputer assistance) also has been found effec-tive (Wright et al., 2005). It is not unusual toinvolve spouses, partners, parents, and otherfamily members during treatment. They maybe used, for example, to provide information

that helps patients test the validity of theirthinking with respect to how other familymembers view them. Moreover, couple therapybased on the cognitive model is often very ef-fective in relieving depression related tochronic interpersonal problems (Beck, 1988;O’Leary & Beach, 1990).

In our clinical experience, a number of thera-pist characteristics contribute to effective cog-nitive therapy. First, cognitive therapists ideallyshould demonstrate the “nonspecific” therapyskills identified by other writers (see, e.g.,Truax & Mitchell, 1971): They should be ableto communicate warmth, genuineness, sincer-ity, and openness. Second, the most effectivecognitive therapists seem to be especiallyskilled at seeing events from their patients’ per-spective (accurate empathy). They are able tosuspend their own personal assumptions andbiases while listening to depressed patients de-scribe their reactions and interpretations.Third, skilled cognitive therapists can reasonlogically and plan strategies; they are not“fuzzy” thinkers. In this respect they resemblegood trial lawyers, who can spot the sometimes


Maladaptive Coping Modes Description

Compliant Surrenderer Adopts a coping style of compliance anddependence.

Detached Protector Adopts a coping style of emotional withdrawal,disconnection, isolation, and behavioralavoidance.

Overcompensator Adopts a coping style of counterattack andcontrol. May overcompensate throughsemiadaptive means, such as workaholism.

FIGURE 6.2b. Maladaptive Coping modes. From Young, Klosko, and Weishaar (2003). Copyright 2003by The Guilford Press. Reprinted by permission.

FIGURE 6.2c. Dysfunctional Parent modes. From Young, Klosko, and Weishaar (2003). Copyright 2003by The Guilford Press. Reprinted by permission.

DysfunctionalParent Mode Description Common associated schemas

Punitive/Critical Parent Restricts, criticizes, or punishes theself or others.

Subjugation, Punitiveness,Defectiveness, Mistrust/Abuse (asabuser).

Demanding Parent Sets high expectations and high levelof responsibility toward others;pressures the self or others toachieve them.

Unrelenting Standards, Self-Sacrifice.

subtle flaws in another individual’s reasoningand skillfully elicit a more convincing interpre-tation of the same events. Skilled cognitivetherapists plan strategies several steps ahead,anticipating the desired outcome. Fourth, thebest practitioners of this approach are active.They have to be comfortable taking the lead,providing structure and direction to the ther-apy process.

Although patient characteristics have re-ceived some empirical attention (Eifert et al.,1998; Padesky & Greenberger, 1995; Persons,Burns, & Perloff, 1988; Shea et al., 1990), wedo not yet have adequate knowledge of whichpatient characteristics are related to success incognitive therapy. Our experience suggests thatpatients with major depressive disorder (singleepisode or recurrent)4 and dysthymic disorder(with or without major depressive disorder) re-spond well to the cognitive therapy approachdescribed in this chapter. To the extent that thepatient is diagnosed with Axis II personalitydisorders and/or the patient’s depression ischronic, schema therapy treatment may be sig-nificantly longer in duration and more crucialin obtaining a more complete and lasting posi-tive response to treatment.

Cognitive therapy can serve an importantadjunctive role to pharmacotherapy for bipolardisorders (Ball et al., 2006; Basco & Rush,1996; Colom, Vieta, Martinez, Jorquera, &Gastó, 1998; Craighead, Miklowitz, Vajk, &Frank, 1998; Lam, Hayward, Watkins, Wright,& Sham, 2005; Scott, 1996b), and is effectivein treating patients with severe endogenous de-pression (Thase, Bowler, & Harden, 1991;Whisman, 1993). Preliminary evidence alsosuggests that cognitive therapy is effective inthe treatment of women with postpartum de-pression (Bledsoe & Grote, 2006).

It is advisable to assess patients’ suitabilityfor cognitive therapy (Padesky & Greenberger,1995; Safran & Segal, 1990; Safran, Segal,Vallis, Shaw, & Samstag, 1993). In our experi-ence, certain patient characteristics are predic-tive of a more rapid response: Patients who areappropriately introspective and can reason ab-stractly; who are well organized, good plan-ners, and conscientious about carrying out re-sponsibilities; who are employed; who are notexcessively angry, either at themselves or atother people; who are less dogmatic and rigidin their thinking; who can identify a clear pre-cipitating event for the depressive episode; andwho have close relationships with others often

show faster improvement in depressive symp-toms through cognitive therapy. Age is not anobstacle, because both younger patients (Har-rington, Wood, & Verduyn, 1998; Reinecke etal., 1998) and older adults (Beutler et al., 1987;Floyd, Scogin, McKendree-Smith, Floyd, &Rokke, 2004; Gallagher-Thompson, Hanley-Peterson, & Thompson, 1990; Koder, Brodaty,& Anstey, 1996; Levendusky & Hufford,1997) seem to benefit from cognitive therapy.Many studies have indicated that children andadolescents show significant clinical improve-ment following CBT (Curry, 2001). Studies ofolder patients show that “various forms of cog-nitive and behavioral psychotherapy can be aseffective in treating geriatric depression as de-pressions occurring earlier in life” (Futterman,Thompson, Gallagher-Thompson, & Ferris,1995, p. 511).

COLLABORATION

Basic to cognitive therapy is a collaborative re-lationship between patient and therapist. Whentherapist and patient work together, the learn-ing experience is enhanced for both, and thecooperative spirit that is developed contributesgreatly to the therapeutic process. Equally im-portant, the collaborative approach helps toensure compatible goals for treatment, and toprevent misunderstandings and misinterpreta-tions between patient and therapist. Because ofthe importance of the collaborative relation-ship, we place great emphasis on the interper-sonal skills of the therapist, the process of jointselection of problems to be worked on, regularfeedback, and the investigative process we call“collaborative empiricism.”

Interpersonal Qualities

Because collaboration requires that the patienttrust the therapist, we emphasize those inter-personal qualities that contribute to trust. Asnoted earlier, warmth, accurate empathy, andgenuineness are desirable personal qualities forthe cognitive therapist, as well as for all psy-chotherapists. It is important that the cognitivetherapist not seem to be playing the role oftherapist. The therapist should be able to com-municate both verbally and nonverbally that heor she is sincere, open, concerned, and direct. Itis also important that the therapist not seem tobe withholding impressions or information, or


evading questions. The therapist should becareful not to seem critical or disapproving ofthe patient’s perspective.

Rapport between patient and therapist iscrucial in the treatment of depressed patients.When rapport is optimal, patients perceive thetherapist as someone who is tuned in to theirfeelings and attitudes, who is sympathetic andunderstanding, and with whom they can com-municate without having to articulate feelingsin detail or qualify statements. When the rap-port is good, both patient and therapist feelcomfortable and secure.

A confident, professional manner is also im-portant in cognitive therapy. A therapist shouldconvey relaxed confidence in his or her abilityto help a depressed patient. Such confidencecan help to counteract the patient’s initial hope-lessness about the future. Because the cognitivetherapist must sometimes be directive and im-pose structure, especially in the early stages oftreatment, it is helpful to maintain a clear senseof professionalism.

Joint Determination of Goals for Therapy

The patient and therapist collaboratively worktogether to set therapeutic goals, determine pri-orities among them, and create an agenda foreach session. Problems to be addressed over thecourse of therapy include specific depressivesymptoms (e.g., hopelessness, crying, and diffi-culty concentrating) and external problems(e.g., couple difficulties, career issues, child-rearing concerns). Priorities are then jointly de-termined in accordance with how much dis-tress is generated by a particular problem andhow amenable that particular problem is tochange. During the agenda-setting portion ofeach therapy session (discussed in detail in thenext section), therapist and patient together de-termine the items to be covered in that session.Through this collaborative process, targetproblems are selected on a weekly basis.

The process of problem selection often pres-ents difficulties for the novice cognitive thera-pist, including failure to reach agreement onspecific problems on which to focus, selectionof peripheral concerns, and the tendency tomove from problem to problem instead of per-sistently seeking a satisfactory solution to onlyone problem at a time. Because the problem se-lection process entails both structuring and col-laboration on the part of the therapist, consid-erable skill is necessary.

Regular Feedback

Feedback is especially important in therapywith depressed patients; it is a crucial ingredi-ent in developing and maintaining the collabo-rative therapeutic relationship. First, as part ofthe regular feedback process, it is important toshare the rationale for each intervention mode.This serves to demystify the therapy processand facilitates the patient’s questioning the va-lidity of a particular approach. When the pa-tient understands how a technique or assign-ment is related to the solution of a problem, heor she is more likely to participate conscien-tiously. The cognitive therapist initiates thefeedback component early in therapy by elicit-ing the patient’s thoughts and feelings aboutmany aspects of the therapy, such as the han-dling of a particular problem, the therapist’smanner, and homework assignments. Sincemany patients misconstrue therapists’ state-ments and questions, only through regularfeedback can the therapist ascertain whether heor she and the patient are on the same “wave-length.” Second, the therapist must be alert forverbal and nonverbal clues to a patient’s covertnegative reactions, and address these reactionsin an empathic manner. Sometimes these prob-lems can be resolved through clarification ofthe rationale for a particular technique or as-signment.

The third element of the feedback process isfor the therapist to check regularly to deter-mine whether the patient understands his orher formulations. Patients sometimes agreewith a formulation simply out of compliance,and depressed patients frequently exhibit bothcompliance and reluctance to “talk straight”with their therapists for fear of being rejectedor criticized, or of making a mistake. There-fore, the therapist must make an extra effort toelicit the patient’s feelings or wishes relevant tocompliance (e.g., anxiety about rejection, wishto please) and be alert for verbal and nonverbalclues that the patient may indeed not under-stand the explanations.

As a regular part of the feedback process, atthe close of each session the cognitive therapistprovides a concise summary of what has takenplace, and asks the patient to abstract andwrite down the main points from the session.The patient keeps this summary for review dur-ing the week. In practice, the therapist usescapsule summaries at least three times during astandard therapeutic interview: in preparing


the agenda, in a midpoint recapitulation of thematerial covered up to that point, and in the fi-nal summary of the main points of the inter-view. Patients generally respond favorably tothe elicitation of feedback and presentation ofcapsule summaries. We have observed that thedevelopment of empathy and rapport is facili-tated by these techniques.

Collaborative Empiricism

When the collaborative therapeutic relation-ship has been successfully formed, patient andtherapist act as an investigative team. Thoughwe elaborate on the investigative process later,it is appropriate to introduce it in the contextof the collaborative relationship. As a team, pa-tient and therapist approach the patient’s auto-matic thoughts and schemas in the same man-ner that scientists approach questions: Eachthought or schema becomes a hypothesis to betested, and evidence is gathered that supportsor refutes the hypothesis. Events in the past,circumstances in the present, and possibilitiesin the future are the data that constitute evi-dence, and the conclusion to accept or rejectthe hypothesis is jointly reached as patient andtherapist subject the evidence to logical analy-sis. Experiments may also be devised to test thevalidity of particular cognitions. Cognitivetherapists need not persuade patients of illogi-cality or inconsistency with reality, because pa-tients “discover” their own inconsistencies.This guided discovery process, a widely ac-cepted educational method, is one of the vitalcomponents of cognitive therapy.

THE PROCESSOF COGNITIVE THERAPY

Here we attempt to convey a sense of how cog-nitive therapy sessions are structured and asense of the course of treatment. Detailed dis-cussion of particular techniques follows thissection.

The Initial Sessions

A main therapeutic goal of the first few ses-sions is to produce some symptom relief. Re-ducing the patient’s suffering helps to increaserapport, collaboration, and confidence in thetherapeutic process. Symptom relief, however,should be based on more than rapport, sympa-

thy, and implied promise of “cure.” In the firstfew sessions, the cognitive therapist starts theprocess of defining the patient’s problems anddemonstrating some of the strategies that willbe used in the therapy to deal with those prob-lems.

Problem definition is a primary goal in theearly stages of therapy. The therapist collabor-atively works with the patient to define the spe-cific problems on which they will focus duringtherapy sessions. The cognitive therapist doesthis by obtaining as complete a picture as possi-ble of the patient’s psychological and life situa-tion difficulties. The therapist also seeks detailsconcerning the depth of depression and partic-ular symptomatology. Cognitive therapists areespecially concerned with how patients seetheir problems.

Once the specific problems have been de-fined, patient and therapist establish prioritiesamong them. Decisions are made on the basisof amenability to therapeutic change and cen-trality of the life problem or cognition to thepatient’s emotional distress. To help establishpriorities effectively, the therapist must see therelationships among particular thoughts, par-ticular life situations, and particular distressingemotions.

Another goal of the initial session is to illus-trate the close relationship between cognitionand emotion. When the therapist is able to ob-serve the patient’s mood change (e.g., crying),he or she points out the alteration in affect andasks for the patient’s thoughts just before themood shift. The therapist then labels the nega-tive thought and points out its relationship tothe change in mood. He or she initially gearshomework assignments toward helping the pa-tient see the close connection between cogni-tion and emotion.

A frequent requirement in the early stage oftherapy is to socialize the patient to cognitivetherapy. A patient who has previously under-taken analytically oriented or Rogerian thera-pies may begin cognitive therapy expecting amore insight-oriented, nondirective therapeuticapproach. The cognitive therapist can facilitatethe transition to a more active and structuredapproach by maintaining a problem-orientedstance, which often entails gently interrupting apatient who tends to speculate about thesources of the problems and seeks interpreta-tions from the therapist.

Finally, the therapist must communicate theimportance of self-help homework assignments


during the initial session by stressing that doingthe homework is actually more important thanthe therapy session itself. The therapist also canenhance motivation by explaining that patientswho complete assignments generally improvemore quickly. The nature and implementationof self-help homework assignments are consid-ered in further detail in a later section of thischapter.

The Progress of a Typical Therapy Session

Each session begins with the establishment ofan agenda for that session. This ensures opti-mal use of time in a relatively short-term,problem-solving therapeutic approach. Theagenda generally begins with a short synopsisof the patient’s experiences since the last ses-sion, including discussion of the homework as-signment. The therapist then asks the patientwhat he or she wants to work on during thesession, and often offers topics to be included.

When a short list of problems and topics hasbeen completed, the patient and therapist de-termine the order in which to cover them and,if necessary, the time to be allotted to eachtopic. There are several issues to consider in es-tablishing priorities, including stage of therapy,severity of depression, likelihood of makingprogress in solving the problem, and potentialpervasiveness of the effect of a particular themeor topic. The cognitive therapist is sensitive to apatient’s occasional desire to talk about some-thing that seems important to him or her at themoment, even if such discussion seems not tobe productive in terms of other goals. This kindof flexibility characterizes the collaborativetherapeutic relationship.

After these preliminary matters have beencovered, patient and therapist move on to theone or two problems to be considered duringthe session. The therapist begins the discussionof a problem by asking the patient a series ofquestions designed to clarify the nature of thepatient’s difficulty. In doing so, the therapistseeks to determine whether early maladaptiveschemas, misinterpretations of events, or unre-alistic expectations are involved. The therapistalso seeks to discover whether the patient hadunrealistic expectations, whether the patient’sbehavior was appropriate, and whether all pos-sible solutions to the problem were considered.The patient’s responses suggest to the therapista cognitive-behavioral conceptualization ofwhy the patient is having difficulty in the area

of concern. The therapist by now has discernedthe one or two significant thoughts, schemas,images, or behaviors to be worked on. Whenthis target problem has been selected, the thera-pist chooses the cognitive or behavioral tech-niques to apply and shares their rationale withthe patient. The specific techniques used in cog-nitive therapy are explained in the followingsections of this chapter.

At the close of the session, the therapist asksthe patient for a summary, often in writing, ofthe major conclusions drawn during the ses-sion. The therapist asks for the patient’s reac-tions to the session to ascertain whether any-thing disturbing was said and to forestall anydelayed negative reactions following the inter-view. Finally, the therapist gives a homeworkassignment designed to assist the patient in ap-plying the particular skills and concepts fromthe session to the problem during the followingweek.

Progression of Session Content over Time

Although the structure of cognitive therapy ses-sions does not change during the course oftreatment, the content often changes signifi-cantly. Treatment with both nonchronic andchronic patients begins with a focus on symp-tom reduction, behavioral activation, overcom-ing hopelessness, identifying problems, settingpriorities, socializing the patient to cognitivetherapy, establishing the collaborative relation-ship, demonstrating the relationship betweencognition and emotion, and working towardprogress on the targeted problems. Therapy isinitially centered on the patient’s symptoms,with attention given to behavioral and motiva-tional difficulties. Once the patient shows somesignificant changes in these areas, the emphasisshifts to the content and pattern of the patient’sthinking.

Once the patient is feeling less depressed,therapist and patient explore the patient’s spe-cific thoughts and assumptions about particu-lar situations and/or problems, and underlyingschemas that may be related. Schemas aredeep-seated core beliefs about the self, relation-ships, and life. Schemas develop as the result ofspecific experiences within an individual’s life-time, mostly during childhood and/or adoles-cence. When schemas are identified, they revealrules and formulas that the individual uses to“make sense” of the world. They are the over-riding principles by which the individual orga-


nizes his or her perceptions and/or behavior.Maladaptive schemas often underlie many ofthe patient’s problems. Cognitive therapy aimsto counteract the effects of maladaptiveschemas. If the schemas themselves can bechanged, we believe that the patient will be-come less vulnerable to future depressions.

Over the course of therapy, the patient as-sumes increased responsibility for identifyingproblems, coming up with solutions, and im-plementing the solutions through homeworkassignments. The therapist increasingly as-sumes the role of advisor or consultant as thepatient learns to implement therapeutic tech-niques without constant support. As the pa-tient becomes a more effective problem solver,the frequency of sessions is reduced, and ther-apy is eventually discontinued.

The remainder of this chapter is devoted to adetailed description of traditional cognitivetherapy and schema therapy strategies.

SYMPTOM REDUCTION

Behavioral Techniques

Behavioral techniques are used throughout thecourse of cognitive therapy, but they are gener-ally concentrated in the earlier stages of treat-ment. Behavioral techniques are especially nec-essary for those more severely depressedpatients who are passive, anhedonic, sociallywithdrawn, and unable to concentrate for ex-tended periods of time. By engaging such a pa-tient’s attention and interest, the cognitive ther-apist tries to induce the patient to counteractwithdrawal and become more involved in con-structive activity. From a variety of behavioraltechniques, the therapist selects those that willhelp the patient cope more effectively with situ-ational and interpersonal problems. Throughhomework assignments, the patient imple-ments specific procedures for dealing with con-crete situations or for using time more adap-tively.

The cognitive therapist uses behavioraltechniques, with the goal of modifying auto-matic thoughts. For example, a patient whobelieves “I can’t get anything done any more”can modify this thought after completing aseries of graded tasks designed to increasemastery. The severely depressed patient iscaught in a vicious cycle in which a reducedactivity level leads to a negative self-label,which in turn results in even further discour-agement and consequent inactivity. Interven-

tion with behavioral techniques can changethis self-destructive pattern.

The most commonly used behavioral tech-niques include scheduling activities that includeboth mastery and pleasure exercises, cognitiverehearsal, self-reliance training, role playing,and diversion techniques. The scheduling of ac-tivities is frequently used in the early stages ofcognitive therapy to counteract loss of motiva-tion, hopelessness, and excessive rumination.The therapist uses the Weekly Activity Schedulefor planning activities hour by hour, day by day(see Figure 6.3). Patients maintain an hourly re-cord of the activities in which they engaged.Activity scheduling also helps patients obtainmore pleasure and a greater sense of accom-plishment from activities on a daily basis. Thepatients rate each completed activity (using a 0-to 10-point scale) for both mastery and plea-sure. The ratings usually contradict patients’beliefs that they cannot accomplish or enjoyanything any more. To assist some patients ininitiating mastery and pleasure activities, thetherapist may sometimes find it necessary tosubdivide an activity into segments, rangingfrom the simplest to the most difficult andcomplex aspects of the activity. We call this the“graded task” approach. The subdivision en-ables depressed patients to undertake tasks thatwere initially impossible, thus providing proofof success.

Cognitive rehearsal entails asking a patientto picture or imagine each step involved in theaccomplishment of a particular task. This tech-nique can be especially helpful with those pa-tients who have difficulty carrying out a taskthat requires successive steps for its comple-tion. Sometimes impairment in the ability toconcentrate creates difficulties for the patientin focusing attention on the specific task. Theimagery evoked by the cognitive rehearsal tech-nique helps the patient to focus and helps thetherapist to identify obstacles that make the as-signment difficult for the particular patient.

Some depressed patients rely on others totake care of most of their daily needs. With self-reliance training, patients learn to assume in-creased responsibility for routine activities suchas showering, making their beds, cleaning thehouse, cooking their own meals, and shopping.Self-reliance involves gaining increased controlover emotional reactions.

Role playing has many uses in cognitive ther-apy. First, it may be used to bring out auto-matic thoughts through the enactment of par-ticular interpersonal situations, such as an


encounter with a supervisor at work. Second, itmay also be used, through homework assign-ments, to guide the patient in practicing and at-tending to new cognitive responses in problem-atic social encounters. A third use of roleplaying is to rehearse new behaviors. Thus, roleplaying may be used as part of assertivenesstraining and is often accompanied by modelingand coaching.

Role reversal, a variation of role playing, canbe very effective in helping patients test howother people might view their behavior. This iswell illustrated by a patient who had a “humili-ating experience” while buying some clothes in astore. After playing the role of the clerk, the pa-tient had to conclude that she had insufficientdata for her previous conclusion that she ap-peared clumsy and inept. Through role reversal,patients begin to view themselves less harshly as“self-sympathy” responses are elicited.

Finally, the therapist may introduce variousdistraction techniques to assist the patient inlearning to reduce the intensity of painful af-fects. The patient learns to divert negative

thinking through physical activity, social con-tact, work, play, and visual imagery. Practicewith diversion techniques also helps the patientgain further control over emotional reactivity.

Cognitive Techniques

The specific cognitive techniques providepoints of entry into the patient’s cognitive orga-nization. The cognitive therapist uses tech-niques for eliciting and testing automaticthoughts, and identifying schemas to help boththerapist and patient understand the patient’sconstruction of reality. In applying specific cog-nitive techniques in therapy, it is important thatthe therapist work within the framework of thecognitive model of depression. Each set of tech-niques is discussed in turn.

Eliciting Automatic Thoughts

“Automatic thoughts” are those thoughts thatintervene between outside events and the indi-vidual’s emotional reactions to them. They of-


Note: Grade activities M for mastery and P for pleasure 0–10.

Mon. Tues. Wed. Thurs. Fri. Sat. Sun.M

orni

ng

6–7

7–8

8–9

9–10

10–11

11–12

Afte

rnoo

n

12–1

1–2

2–3

3–4

4–5

5–6

Eve

ning

6–7

7–8

8–9

9–10

10–11

11–12

12–6

FIGURE 6.3. Weekly Activity Schedule.

ten go unnoticed, because they are part of a re-petitive pattern of thinking, and because theyoccur so often and so quickly. People rarelystop to assess their validity, because they are sobelievable, familiar, and habitual. The patientin cognitive therapy must learn to recognizethese automatic thoughts for therapy to pro-ceed effectively. The cognitive therapist and thepatient make a joint effort to discover the par-ticular thoughts that precede emotions such asanger, sadness, and anxiety. The therapist usesquestioning, imagery, and role playing to elicitautomatic thoughts.

The simplest method to uncover automaticthoughts is for therapists to ask patients whatthoughts went through their minds in responseto particular events. This questioning providespatients with a model for introspective explo-ration that they can use on their own, when thetherapist is not present, and after the comple-tion of treatment.

Alternatively, when a patient is able to iden-tify those external events and situations thatevoke a particular emotional response, thetherapist may use imagery by asking the patientto picture the situation in detail. The patient isoften able to identify the automatic thoughtsconnected with actual situations when the im-

age evoked is clear. In this technique, therapistsask patients to relax, close their eyes, and imag-ine themselves in the distressing situation. Pa-tients describe in detail what is happening asthey relive the event.

If a distressing event is an interpersonal one,cognitive therapists also can utilize role play-ing. The therapist plays the role of the otherperson in the encounter, while the patient playshim- or herself. The automatic thoughts canusually be elicited when the patient becomessufficiently engaged in the role play.

In attempting to elicit automatic thoughts,the therapist is careful to notice and point outany mood changes that occur during the ses-sion, and to ask the patient his or her thoughtsjust before the shift in mood. Mood changes in-clude any emotional reaction, such as tears oranger. This technique can be especially usefulwhen the patient is first learning to identify au-tomatic thoughts.

Once patients become familiar with the tech-niques for identifying automatic thoughts, theyare asked to keep a Daily Record of Dysfunc-tional Thoughts (Beck et al., 1979; see Figure6.4), in which they record the emotions and au-tomatic thoughts that occur in upsetting situa-tions between therapy sessions. In later ses-


FIGURE 6.4. Daily Record of Dysfunctional Thoughts.

DATE

SITUATION EMOTION(S) AUTOMATIC THOUGHT(S) RATIONAL RESPOSNE OUTCOMEDescribe:

1. Actual even leading tounpleasant emotion, or

2. Stream of thoughts,daydream, orrecollection, leading tounpleasant emotion.

1. Specify sad/anxious/angry, etc.

2. Rate degreeof emotion,1–100.

1. Write automaticthought(s) that precededemotion(s).

2. Rate belief in automaticthought(s), 0–100%.

1. Write rational responseto automatic thought(s).

2. Rate belief in rationalresponse, 0–100%

1. Rerate beliefin automaticthought(s),0–100%.

2. Specify andratesubsequentemotions,0–100.

Explanation: When you experience an unpleasant emotion, note the situation that seemed to stimulate the emotion. (If the emotionoccurred while you were thinking, daydreaming, etc., please note this.) Then note the automatic thought associated with the emotion.Record the degree to which you believe this thought: 0% = not at all; 100% = completely. In rating degrees of emotion: 1 = a trace;100 = the most intense possible.

sions, patients are taught to develop rationalresponses to their dysfunctional automaticthoughts and to record them in the appropriatecolumn. Therapist and patient generally reviewthe daily record from the preceding week, nearthe beginning of the next therapy session. Insessions with chronic patients, the therapistalso teaches patients to recognize modes. Vari-ous affective exercises are then used to bringabout change in the underlying schemas associ-ated with those modes.

Eliciting automatic thoughts should be dis-tinguished from the interpretation process ofother psychotherapies. In general, cognitivetherapists work only with those automaticthoughts mentioned by patients. Suggestingthoughts to patients may undermine collabora-tion and inhibit patients from learning to con-tinue the process on their own. As a last resort,however, when nondirective strategies fail, acognitive therapist may offer several possibleautomatic thoughts, then ask the patientwhether any of these choices fit.

Even when many efforts to elicit automaticthoughts have been made by the therapist,sometimes the thought remains inaccessible.When this is the case, the cognitive therapisttries to ascertain the particular meaning of theevent that evoked the emotional reaction. Forexample, one patient began to cry whenevershe had an argument with her roommate, whowas a good friend. Efforts to elicit automaticthoughts proved unsuccessful. Only after thetherapist asked a series of questions to deter-mine the meaning of the event did it becomeclear that the patient associated having an ar-gument or fight with ending the relationship.Through this process, therapist and patientwere able to see the meaning that triggered thecrying.

Testing Automatic Thoughtswith Nonchronic Patients

When therapist and patient have managed toisolate a key automatic thought, they approachthe thought as a testable hypothesis. In this“scientific” approach, which is fundamental tocognitive therapy, the patient learns to think ina way that resembles the investigative process.Through the procedures of gathering data,evaluating evidence, and drawing conclusions,patients learn firsthand that their view of real-ity can be quite different from what actuallytakes place. By designing experiments that sub-

ject their automatic thoughts to objectiveanalysis, patients learn how to modify theirthinking, because they learn the process of em-pirical thinking. Patients who learn to thinkthis way during treatment are better able tocontinue the empirical approach after the endof formal therapy.

The cognitive therapist approaches the test-ing of automatic thoughts by asking the patientto list evidence from his or her experience forand against the hypothesis. Sometimes, afterconsidering the evidence, the patient immedi-ately rejects the automatic thought, recognizingthat it is either distorted or actually false.

When previous experience is not sufficient orappropriate to test a hypothesis, the therapistasks the patient to design an experiment forthat purpose. The patient then makes a predic-tion and proceeds to gather data. If the datacontradict the prediction, the patient can rejectthe automatic thought. The outcome of the ex-periment may, of course, confirm the patient’sprediction. It is therefore very important forthe therapist not to assume that the patient’sautomatic thought is distorted.

Some automatic thoughts do not lend them-selves to hypothesis testing through the exami-nation of evidence. In these cases, two optionsare available: The therapist may produce evi-dence from his or her own experience and offerit in the form of a question that reveals the con-tradiction; or the therapist can ask a questiondesigned to uncover a logical error inherent inthe patient’s beliefs. The therapist might say,for example, to a male patient who is sure hecannot survive without a close personal rela-tionship, “You were alone last year, and yougot along fine; what makes you think you can’tmake it now?”

In testing automatic thoughts, it is some-times necessary to refine the patient’s use of aword. This is particularly true for global labelssuch as “bad,” “stupid,” or “selfish.” What isneeded in this case is an operational definitionof the word. To illustrate, a patient at our clinichad the recurring automatic thought, “I’m afailure in math.” Therapist and patient had tonarrow down the meaning of the word beforethey could test the thought. They operational-ized “failure” in math as “being unable toachieve a grade of C after investing as muchtime studying as the average class member.”Now they could examine past evidence and testthe validity of the hypothesis. This process canhelp patients to see the overinclusiveness of


their negative self-assessments and the idiosyn-cratic nature of many automatic thoughts.

Reattribution is another useful technique forhelping the patient to reject an inappropriate,self-blaming thought. It is a common cognitivepattern in depression to ascribe blame orresponsibility for adverse events to oneself.Reattribution can be used when the patient un-realistically attributes adverse occurrences to apersonal deficiency, such as lack of ability or ef-fort. Therapist and patient review the relevantevents and apply logic to the available informa-tion to make a more realistic assignment of re-sponsibility. The aim of reattribution is not toabsolve the patient of all responsibility, but toexamine the many factors that contribute toadverse events. Through this process, patientsgain objectivity, relieve themselves of the bur-den of self-reproach, and can then search forways to solve realistic problems or preventtheir recurrence.

Another strategy involving reattribution isfor therapists to demonstrate to patients thatthey use stricter criteria for assigning responsi-bility to their own unsatisfactory behavior thanthey use in evaluating the behavior of others.Cognitive therapists also use reattribution toshow patients that some of their thinking orbehavior problems can be symptoms of depres-sion (e.g., loss of concentration) and not signsof physical decay.

When a patient is accurate in identifying arealistic life problem or skills deficit, the cogni-tive therapist can use the technique of generat-ing alternatives, in which therapist and patientactively search for alternative solutions. Be-cause a depressed person’s reasoning often be-comes restricted, an effort to reconceptualizethe problem can result in the patient’s seeing aviable solution that he or she may previouslyhave been rejected.

It should be noted that the cognitive tech-niques outlined here all entail the use of ques-tions by the therapist. A common error we ob-serve in novice cognitive therapists is anexhortative style. We have found that thera-pists help patients to change their thinkingmore effectively by using carefully formedquestions. If patients are prompted to worktheir own way through problems and reachtheir own conclusions, they learn an effectiveproblem-solving process. We elaborate on theuse of questioning in cognitive therapy later inthe chapter.

Questioning

As we have stressed throughout this chapter,questioning is a major therapeutic device incognitive therapy. A majority of the therapist’scomments during the therapy session are ques-tions. Single questions can serve several pur-poses at one time, whereas carefully designedseries of questions can help patients consider aparticular issue, decision, or opinion. The cog-nitive therapist seeks through questioning toelicit what patients are thinking, instead of tell-ing patients what he or she believes they arethinking.

In the beginning of therapy, questions areemployed to obtain a full and detailed pictureof a patient’s particular difficulties. They areused to obtain background and diagnosticdata; to evaluate the patient’s stress tolerance,capacity for introspection, coping methods,and so on; to obtain information about the pa-tient’s external situation and interpersonal con-text; and to modify vague complaints by work-ing with the patient to arrive at specific targetproblems on which to work.

As therapy progresses, the therapist usesquestioning to explore approaches to prob-lems, to help the patient to weigh advantagesand disadvantages of possible solutions, to ex-amine the consequences of staying with partic-ular maladaptive behaviors, to elicit automaticthoughts, and to demonstrate EMSs and theirconsequences. In short, the therapist uses ques-tioning in most cognitive therapeutic tech-niques.

Although questioning is itself a powerfulmeans of identifying and changing automaticthoughts and schemas, it is important that thequestions be carefully and skillfully posed. Ifquestions are used to “trap” patients into con-tradicting themselves, patients may come tofeel that they are being attacked or manipu-lated by the therapist. Too many open-endedquestions can leave patients wondering whatthe therapist expects of them. Therapists mustcarefully time and phrase questions to help pa-tients recognize their thoughts and schemas,and to weigh issues objectively.

Self-Help Homework Assignments

Rationale

Regular homework assignments are very im-portant in cognitive therapy. When patients


systematically apply what they have learnedduring therapy sessions to their outside lives,they are more likely to make significant prog-ress in therapy and to be able to maintain theirgains after termination of treatment. Burns andSpangler (2000) found that patients who didthe most homework showed large and signifi-cant decreases in depression in comparisonwith those who were less compliant. Home-work assignments are often the means throughwhich patients gather data, test hypotheses,and begin to modify their thoughts andschemas. In addition, the data providedthrough homework assignments help to shiftthe focus of therapy from the subjective andabstract to more concrete and objective con-cerns. When a patient and therapist review theprevious week’s activities during the agenda-setting portion of the interview, they may do soquickly, and the therapist can draw relation-ships between what takes place in the sessionand specific tasks, thereby avoiding tangentsand side issues. Homework assignments fur-ther the patient’s self-reliance and providemethods for the patient to continue working onproblems after the end of treatment. Cognitivetherapists emphasize the importance of home-work by sharing with patients their rationalefor assigning homework in therapy. They arealso careful to explain the particular benefits tobe derived from each individual assignment.

Assigning and Reviewing Homework

The cognitive therapist designs each assign-ment for the particular patient. The assignmentshould be directly related to the content of thetherapy session, so that the patient understandsits purpose and importance. Each task shouldbe clearly articulated and very specific in na-ture. Near the end of each session, the assign-ment is written in duplicate, with one copy go-ing to the therapist and one to the patient.

Some typical homework assignments includereading a book or article about a specific prob-lem, practicing distraction or relaxation tech-niques, counting automatic thoughts on a wristcounter, rating activities for pleasure and mas-tery on the Weekly Activity Schedule, maintain-ing a Daily Record of Dysfunctional Thoughts,and listening to a tape of the therapy session.

During the therapy session, the therapistsask for the patient’s reactions to homework as-signments, for example, whether the assign-

ment is clear and manageable. To determinepotential impediments, the therapist may askthe patient to imagine taking the steps involvedin the assignment. This technique can be espe-cially helpful during the earlier stages of ther-apy. The patient assumes greater responsibilityfor developing homework assignments as ther-apy progresses through the middle and laterstages.

It is essential that patient and therapist re-view the previous week’s homework during thetherapy session itself. If they do not, the patientmay conclude that the homework assignmentsare not important. During the first part of thetherapy sessions, therapist and patient discussthe previous week’s assignment, and the thera-pist summarizes the results.

Difficulties in Completing Homework

When a patient does not complete homeworkassignments, or does them without conviction,the cognitive therapist elicits automaticthoughts, schemas, or behavioral problemsthat may help both therapist and patient under-stand where the difficulty resides. The therapistdoes not presuppose that the patient is being“resistant” or “passive–aggressive.” When thedifficulties have been successfully identified,therapist and patient work collaboratively tosurmount them. It is, of course, common forpatients to have difficulties in completinghomework, and here we consider some of thetypical problems and ways to counteract them.

When patients do not understand the assign-ment completely, the therapist should explain itmore fully, specifying his or her expectations indetail. Sometimes using the behavioral tech-nique of cognitive rehearsal (described earlier)can be helpful in such situations.

Some patients believe that they are naturallydisorganized and cannot maintain records andfollow through on detailed assignments. Thera-pists can usually help to invalidate such generalbeliefs by asking patients about other circum-stances in which they make lists—for example,when planning a vacation or shopping trip.Therapists can also ask these patients whetherthey could complete the assignment if therewere a substantial reward entailed. This kindof question helps such patients recognize thatself-control is not the problem; rather, they donot believe that the reward is great enough.When patients realize that the problem is an at-


titudinal one, therapist and patient can proceedto enumerate the advantages of completing theassignment.

More severely depressed patients may needassistance to structure their time, so that home-work becomes a regular activity. This can gen-erally be accomplished by setting a specifictime each day for the homework assignment. Ifnecessary, patient and therapist can set up a re-ward or punishment system to make sure thehomework gets done. For example, patientscan reward themselves for doing the assign-ment with a special purchase, or punish them-selves for failing to do it by not watching a fa-vorite television program.

Some patients are afraid of failing the assign-ments or of doing them inadequately. In thesecases, the therapist can explain that self-helpassignments cannot be “failed”: Doing an as-signment partially is more helpful than not do-ing it at all, and mistakes provide valuable in-formation about problems that still need work.In addition, because performance is not evalu-ated, patients cannot lose if they view the activ-ity from a more adaptive perspective.

Sometimes patients believe their problemsare too deeply embedded and complex to be re-solved through homework assignments. Thetherapist can explain to these patients that eventhe most complex undertakings begin with andconsist of small, concrete steps. A writer, forexample, may resolve “writer’s blocks” by tak-ing the attitude, “If I can’t write a book, I canat least write a paragraph.” When enoughparagraphs have been written, the result is abook. A therapist and patient can consider theadvantages and disadvantages of the patient’sbelieving that problems cannot be solved bydoing homework. Or the therapist can ask thepatient to experiment before reaching such aconclusion. In instances in which a patient be-lieves that he or she has not made enough prog-ress and, therefore, that the homework is nothelpful, the therapist can detail the progress thepatient has made or can help the patient seethat it may take more time before substantialchange can be perceived.

When patients seem to resent being given as-signments, the therapist can encourage them todevelop their own assignments. The therapistmight also offer the patients alternative assign-ments from which to choose, making one of thealternatives noncompliance with homeworkassignments. If patients choose noncompliance,

the therapist can help to examine the conse-quences of that choice. Still another strategy isto present patients with a consumer model oftherapy: Patients have a certain goal (overcom-ing depression), and the therapist offers ameans to achieve that goal; patients are free touse or reject the tools, just as they are free tobuy or not to buy in the marketplace.

Some patients believe that they can improvejust as readily without homework. In this case,therapists have two options. First, they can of-fer their own clinical experience, which is sup-ported by existing empirical evidence that mostpatients who do not actively engage with andcomplete therapeutic homework progress moreslowly in therapy. The other option is to set upan experiment for a given period of time, dur-ing which patients do not have to complete as-signments. At the end of the predetermined pe-riod, therapists and patients can evaluate thepatients’ progress during that time interval.Once again, it is important for cognitive thera-pists to keep an open mind: Some patients doindeed effect significant change without for-mally completing homework assignments.

Special Problems

The novice cognitive therapist often errs bystaying with the standard method we have out-lined here, even if it is not working very well.The cognitive therapist should be flexibleenough to adapt to the needs of patients and tothe several special problems that commonlyarise in therapy. We have grouped these specialproblems into two categories: difficulties in thetherapist–patient relationship and problems inwhich the therapy itself seems not to be work-ing.

Therapist–Patient Relationship Difficulties

The first set of problems concerns thetherapist–patient relationship itself. When thetherapist first perceives a patient to be dissatis-fied, oppositional, angry, or hostile, it is imper-ative that he or she present these observationsto the patient in an empathic manner. It is im-portant that the therapist refrain from respond-ing to the patient simply with an increased andrigid adherence to prescribed techniques or thetherapeutic rationale. It also is important thatthe therapist refrain from self-critical state-ments. Research indicates that such responses


have detrimental effects on the therapeutic re-lationship (Castonguay et al., 1996; Henry,Strupp, Butler, Schact, & Binder, 1993; Piperet al., 1999). Instead, preliminary research(Castonguay et al., 2004; Safran, Muran,Samstag, & Stevens, 2002) indicates that thetherapist’s use of metacommunication skills(open discussion about the patient’s negativereaction, exploration of the patient’s experi-ence, recognition and acknowledgement of anytherapist contribution to those negative reac-tions) is significantly more likely to repair andrestore—and even improve—the therapeuticbond.

It is essential for therapists to be aware thatmany interventions can be misinterpreted in anegative way by depressed patients. Therapistsapproach problems of misinterpretation in thesame way that they approach other thoughts:They work with the patient to gather data andsearch for alternative accounts of the evidence.Difficulties in the therapist–patient relationshipcan generally be resolved through dialogue.There are times when a therapist may need totailor behavior to the particular needs of an in-dividual patient. For instance, a therapist maybecome freer with appropriate self-disclosureand personal reactions to meet the needs of apatient who persist in seeing the therapist asimpersonal. Similarly, the therapist can make apoint of checking formulations of the patient’sthoughts more frequently to meet the needs ofa patient who continues to believe that thetherapist does not understand him or her.

It is imperative in situations like these for thetherapist not to assume that the patient is beingstubbornly resistant or irrational. In fact, ther-apeutic reactance (“a motivational state char-acterized by the tendency to restore or re-assertone’s ability to engage in freedoms perceived aslost or threatened” [Arnow et al., 2003, p.1026]) was found to be a positive predictor oftreatment outcome in directive therapy withchronically depressed patients when therapistsresponded in a flexible manner to such patientbehaviors. Cognitive therapists collaboratewith patients to achieve a better understandingof patients’ responses. The reactions them-selves often provide data regarding the kinds ofdistortions patients make in their other socialand personal relationships. Therefore, patients’responses give the therapists the opportunity towork with them on their maladaptive interpre-tations in relationships.

Unsatisfactory Progress

A second set of problems occurs when thetherapy appears not to be working, evenwhen the patient conscientiously completeshomework assignments and the collaborativerelationship seems successful. Sometimesproblems stem from inappropriate expecta-tions on the part of the patient—or unrealis-tic expectations on the part of the therapist—regarding the rapidity and consistency ofchange. When therapy seems not to be pro-gressing as quickly as it “should,” both pa-tient and therapist must remember that upsand downs are to be anticipated in the courseof treatment. It is important for therapists tokeep in mind that some patients simply prog-ress more slowly than others. The therapistor patient, or both, may be minimizing smallchanges that have indeed been taking place.In this case the therapist can emphasize thesmall gains that have been made and remindthe patient that large goals are attainedthrough small steps toward them.

At times, patients’ hopelessness can leadthem to invalidate their gains. Therapistsshould seek to uncover the thoughts and mal-adaptive assumptions that contribute to thepervasive hopelessness. In these cases, thera-pists must work to correct mistaken notionsabout the process of change and about the na-ture of depression before further progress intherapy can occur.

In some cases in which therapy seems not tobe working successfully, it may be that some ofthe therapeutic techniques have not been cor-rectly used. Problems often arise when patientsdo not really believe the rational responses orare not able to remember them in times of emo-tional distress. It is important for a therapist todetermine the amount of a patient’s belief in therational responses and help him or her use thenew responses as closely as possible to the mo-ment when the automatic thoughts occur. Tothe patient who does not fully believe a rationalresponse, the therapist can suggest an experi-mental stance—taking the new belief and “try-ing it on for size.” The patient who cannotthink of answers because of emotional upsetshould be told that states of emotional distressmake reasoning more difficult, and thatthoughts such as “If this doesn’t work, nothingwill” can only aggravate the problem. Patientsshould be assured that they will be able to


think of rational responses more readily withpractice.

Another problem deriving from the misap-plication of cognitive therapy techniques oc-curs when the therapist uses a particular tech-nique inflexibly. It is often necessary for thetherapist to try out several behavioral or cogni-tive techniques before finding an approach towhich a patient responds well. The cognitivetherapist must stay with a particular techniquefor a while to see whether it works, but he orshe must also be willing to try an alternativetechnique when it becomes apparent that thepatient is not improving. To give a specific ex-ample, behavioral homework assignments aresometimes more helpful with particular pa-tients, even though the therapist has every rea-son to predict in advance that cognitive assign-ments will be more effective.

In some instances in which it appears that lit-tle progress is being made in therapy, it turnsout that the therapist has selected a tangentialproblem. The cognitive therapist should bealert to this possibility, especially during theearly stages of therapy. When there appears tobe little or no significant change in depressionlevel, even when the patient seems to havemade considerable progress in a problem area,the therapist should consider the possibilitythat the most distressing problem has not yetbeen uncovered. A typical example of this kindof difficulty is the patient who presents diffi-culty at work as the major problem, when itturns out that couple problems are contribut-ing significantly to the work difficulties. Thereal issue may be withheld by the patient be-cause it seems too threatening.

Finally, cognitive therapy is not for everyone.If the therapist has tried all available ap-proaches to the problem and has consultedwith other cognitive therapists, it may be bestto refer the patient to another therapist with ei-ther the same or a different orientation.

Regardless of why therapy is not progressingsatisfactorily, cognitive therapists should at-tend to their own affect and cognitions. Theymust maintain a disciplined, problem-solvingstance. If the cognitive therapist finds him- orherself unduly influenced by a patient’s despairor begins to notice that his or her own schemasare triggered by therapeutic interactions, he orshe should seek supervision. Hopelessness inpatients or therapists is an obstacle to problemsolving. If therapists can effectively counteracttheir own negative self-assessments and other

dysfunctional thoughts, they will be better ableto concentrate on helping patients find solu-tions to their problems.

Case Study of Denise:Nonchronic Depression

In the case study that follows, we describe thecourse of treatment for a nonchronically de-pressed woman seen at our center. Through thecase study, we illustrate many of the conceptsdescribed earlier in this chapter, including elici-tation of automatic thoughts, the cognitivetriad of depression, collaborative empiricism,structuring a session, and feedback.

Assessment and Presenting Problems

At the initial evaluation, Denise reported thatshe was a 59-year-old widow, who had beenliving alone for the last year. Denise’s husbandhad been diagnosed with brain cancer threeyears prior and died approximately one yearago. She had two grown unmarried children(27 and 25 years old) who were pursuing ca-reers in other parts of the country. Denise hadan undergraduate degree and had worked untilage 30 but stopped after marrying. Denise de-scribed her major problems as depression (overthe last year and a half), difficulty coping withdaily life, and loneliness. She reported oneprior episode of major depression around age25, following the death of her father.

Denise said she had become increasingly so-cially isolated with the onset of her husband’sillness (brain cancer). She reported having hadnormal friendships as a child, teenager, andyoung adult. She and her husband had led a rel-atively quiet life together, mostly focused onraising their children and respective work.When they had free time, they had enjoyed in-tellectual and cultural activities together (muse-ums, lectures, concerts, and fine restaurants).They had a few close friends with whom theysocialized but those friends had retired inFlorida and Arizona during the time of the hus-band’s illness.

Denise was diagnosed with a major depres-sive disorder, recurrent, on Axis I. Her testscores verified the diagnosis of depression.Denise’s Beck Depression Inventory (BDI) scorewas 28, placing her in the moderate to severerange of depression. Her most prominent de-pressive symptoms included loss of pleasure, ir-ritability, social withdrawal, inability to make


decisions, fatigue, guilt, difficulty motivatingherself to perform daily functions, and loneli-ness.

Session I

The session began with Denise describing the“sad feelings” she was having. The therapist al-most immediately started to elicit Denise’s au-tomatic thoughts during these periods.

THERAPIST: What kind of thoughts go throughyour mind when you’ve had these sad feel-ings this past week?

DENISE: Well, I guess I’m thinking what’s thepoint of all this. My life is over. It’s just notthe same. I have thoughts like, “What am Igoing to do? Sometimes I feel mad at him,you know, my husband. How could he leaveme? . . . Isn’t that terrible of me? What’swrong with me? How can I be mad at him?He didn’t want to die a horrible death. Ishould have done more. I should have madehim go to the doctor when he first startedgetting headaches. . . . Oh, what’s the use?

THERAPIST: It sounds like you are feeling quitebad right now. Is that right?

DENISE: Yes.

THERAPIST: Keep telling me what’s goingthrough your mind right now?

DENISE: I can’t change anything. It’s over. Idon’t know. . . . It all seems so bleak andhopeless. What do I have to look forwardto . . . sickness and then death?

THERAPIST: So one of the thoughts is that youcan’t change things, and that it’s not going toget any better?

DENISE: Yes.

THERAPIST: And sometimes you believe thatcompletely?

DENISE: Yeah, I believe it, sometimes.

THERAPIST: Right now do you believe it?

DENISE: I believe it—yes.

THERAPIST: Right now you believe that youcan’t change things and it’s not going to getbetter?

DENISE: Well, there is a glimmer of hope, butit’s mostly . . .

THERAPIST: Is there anything that you kind oflook forward to in terms of your own lifefrom here on?

DENISE: Well, what I look forward to . . . I en-joy seeing my kids, but they are so busy rightnow. My son is a lawyer and my daughter isin medical school. So, they are very busy.They don’t have time to spend with me.

By inquiring about Denise’s automaticthoughts, the therapist began to understand herperspective—that she would go on forever,mostly alone. This illustrates the hopelessnessabout the future that is characteristic of mostdepressed patients. A second advantage to thisline of inquiry is that the therapist introducedDenise to the idea of looking at her ownthoughts, which is central to cognitive therapy.

As the session continued, the therapistprobed Denise’s perspective regarding her dailylife. The therapist chose to focus on her inactiv-ity and withdrawal. This is frequently the firsttherapeutic goal in working with a severely de-pressed patient.

In the sequence that follows, the therapistguided Denise to examine the advantages anddisadvantages of staying in her house all day.

DENISE: Usually I don’t want to leave myhouse. I want to stay there and just keep theshades closed; you know, I don’t want to doanything. I just want to keep everything out,keep everything away from me.

THERAPIST: Now do you feel better when youstay in the house all day trying to shut every-thing out?

DENISE: Sort of . . .

THERAPIST: What do you mean?

DENISE: Well, I can watch TV all day and justlose myself in these silly shows. I feel betterwhen I see other people and their problemson these shows. It makes me feel less lonelyand like my problems aren’t so bad.

THERAPIST: And so how much time do youspend doing that?

DENISE: Now, lately? . . . Most of the time.Staying inside and watching TV feels safe,sort of secure, everything . . . like my loneli-ness, feels more distant.

THERAPIST: Now after you have spent sometime like this, how do you feel about your-self?

DENISE: Afterwards? I usually try not to paymuch attention to how I’m feeling.

THERAPIST: But when you do, how do you feel?


DENISE: I feel bad. I feel bad for wasting theday. I don’t get to things that I need to takecare of . . . like my bills, like cleaning, liketaking a shower. I usually end up feeling kindof pathetic . . . and guilty.

THERAPIST: On the one hand you seem to feelsoothed and on the other hand, afterwards,you’re a bit critical of yourself?

Note that the therapist did not try to debate orexhort Denise to get out of the house or be-come involved with necessary daily tasks.Rather, through questioning, the therapist en-couraged her to examine more closely her as-sumption that she was really better off watch-ing TV all day in her house. This is the processwe call “collaborative empiricism.” By the sec-ond session, Denise had reexamined her hy-pothesis about watching TV and remaining inthe house all day.

DENISE: About watching TV in the house ver-sus getting out, I thought about that theother day. I remember telling you that itmade me feel better to stay there. When Ipaid attention to what I really felt, it didn’tmake me feel better. It just kind of blockedout feeling bad, but I didn’t feel better.

THERAPIST: It is funny then that when youtalked about it, your recollection of the ex-perience was more positive than it actuallywas, but that sometimes happens with peo-ple. It happens to me too. I think that some-thing is good that’s not so hot when I actu-ally check it out.

We now return to the first session. Aftersome probing by the therapist, Denise men-tioned that it sometimes feels like cognitivetherapy “is my last hope.” The therapist usedthis as an opportunity to explore her hopeless-ness and suicidal thinking.

THERAPIST: What was going through yourmind when you said, “This is my last hope”?Did you have some kind of vision in yourmind?

DENISE: Yeah, that if this doesn’t work, I feellike I couldn’t take living like this the rest ofmy life.

THERAPIST: If it doesn’t work out, then what?

DENISE: Well, I don’t really care what happensto me . . .

THERAPIST: Did you have something more con-crete in mind?

DENISE: Well, right this minute I don’t think Icould commit suicide, but if I keep feelingthis way for a long time, maybe I could. Idon’t know, though—I’ve thought about sui-cide before, but I have never really thoughtabout how I would do it. I know certainthings stop me, like my kids. I think it wouldreally hurt them and some other people too,like my mother. My mom is in good healthnow, but she may need me some day. . . .Yeah, those are the two things that stop me,my children and my mother.

THERAPIST: Now those are the reasons for notcommitting suicide. Now what are some ofthe reasons why you might want to, do youthink?

DENISE: Because sometimes it just feels soempty and hopeless. There’s nothing to lookforward to—every day is the same. My life issuch a waste, so why not just end it?

The therapist wanted Denise to feel as free aspossible to discuss suicidal thoughts; thus, hetried hard to understand both the reasons forher hopelessness and the deterrents to suicide.After determining that she had no imminentplans to make an attempt, the therapist saidthat he would work with her to make somechanges. He then asked her to select a smallproblem that they could work on together.

THERAPIST: Now are there any small things thatyou could do that would affect your life rightaway?

DENISE: I don’t know. Well, I guess just callingmy friend Diane in Florida. She called abouta month ago and then again last week. Bothtimes I told her I was busy and would callher back, but I haven’t. I’ve felt so down. Ihave nothing to say to her.

THERAPIST: Well, when she lived in the area,what kinds of things did you talk about?

DENISE: We have kids about the same age, sowe would talk about our kids. We both liketo read and we used to go to a book clubtogether—so we would talk about the bookswe were reading. Both of us liked art. Weused to attend lectures at the museum duringthe week, so we would talk about art and thelectures. We would spend time making plansto do things together in our free time. It al-


ways was very interesting when I spent timeand with her. We had so much in common. Ido miss her.

THERAPIST: It sounds like you used to be in-volved in a number of interesting activities.What about now?

DENISE: After my husband got sick and thenmy friends moved, I just stopped. I haven’tdone any of those things in quite a while.

THERAPIST: What do you think about attendinga lecture series now?

DENISE: I don’t know.

THERAPIST: Well, what do you think about thatidea?

DENISE: It’s an OK idea, but it just seems liketoo much. I don’t think I’ll enjoy it . . . theway I feel . . . I don’t know.

THERAPIST: Would you be willing to test outthat thought that you won’t be able to enjoyit now?

DENISE: I don’t know . . . I guess so.

THERAPIST: Is that a “yes”?

DENISE: Yes, but I don’t see how I’m going toget myself to do it.

THERAPIST: Well, how would you go aboutfinding out about a lecture series?

DENISE: You look online at the museum’swebsite to see what’s available.

THERAPIST: OK. Do you have a computer?

DENISE: Yes.

THERAPIST: Is it working?

DENISE: Yes.

THERAPIST: How do you feel about doing that?

DENISE: I guess I could do that. . . . I’m so pa-thetic, I know what to do. I don’t need youto spell it out for me. Why didn’t I just dothis before?

THERAPIST: Well, you probably had good rea-sons for not doing it before. Probably youwere just so caught up in the hopelessness.

DENISE: I guess so.

THERAPIST: When you are hopeless you tend todeny, as it were, or cut off possible optionsor solutions.

DENISE: Right.

THERAPIST: When you get caught up in hope-lessness then, there is nothing you can do. Isthat what you think?

DENISE: Yeah.

THERAPIST: So, then, rather than be down onyourself because you haven’t looked this uponline before, why don’t we carry you rightthrough?

This excerpt illustrates the process of gradedtasks that is so important in the early stages oftherapy with a depressed patient. The therapistasked the patient a series of questions to breakdown the process of attending a lecture seriesinto smaller steps. Denise realized that she hadknown all along what to do, but, as the thera-pist pointed out, her hopelessness preventedher from seeing the options.

DENISE: Taking this step is going to be hard forme.

THERAPIST: First steps are harder for every-body, but that’s why there is an old expres-sion: “A journey of a thousand miles startswith the first step.”

DENISE: That’s very true.

THERAPIST: It’s the first step that is so very im-portant, and then you can ready yourself forthe second step, and then the third step, andso on. Eventually, you build up some mo-mentum, and each step begins to followmore naturally. But first, all you have to do istake one small step. You don’t have to takegiant steps.

DENISE: Well, yeah, I can see that. I guess I wasthinking every step was just as hard as thefirst. Maybe it will get easier.

In the second session, Denise reported success.

DENISE: I checked online about the lecture se-ries and I surprised myself. One actuallysounded interesting, and I’m thinking that Imight just register for it online. I really didn’tthink any of those feelings were still there.I’m kind of looking forward to that nextstep.

At the end of the first session, the therapisthelped Denise fill out the Weekly ActivitySchedule for the coming week. The activitieswere quite simple, such as getting up and tak-ing a shower, fixing meals, going out shopping,and checking out the lecture series online.Finally, the therapist asked Denise for feedbackabout the session and about her hopelessness.


THERAPIST: Do you have any reactions?

DENISE: I’m still feeling down, but I’m also feel-ing a little better. It’s interesting that just theidea of looking at what lecture’s might beavailable is making me feel a little lighter. Ieven had the thought of calling Diane to talkover the options. . . . Is this a sign of betterthings to come?

THERAPIST: What do you think?

DENISE: Maybe.

Session 2

In the second session, the therapist began bycollaborating with Denise to set an agenda.Denise wanted to discuss the fact that she hadnot been attending to her bills or to her house-work and was still spending a good part of theday alone in front of the TV; the therapist uti-lized this as an opportunity to discuss the issueof activity versus inactivity on the agenda.They then reviewed the previous homework.Denise had carried out all the scheduled activi-ties and had also listed some of her negativethoughts in between sessions. Her BDI scorehad dropped somewhat. (Patients routinely fillout the BDI before each session, so that boththe patient and the therapist can monitor theprogress of treatment.)

Denise then shared her list of negativethoughts with the therapist. One concern wasthat she had expressed angry feelings about herhusband during the first session.

DENISE: I don’t like revealing things about my-self, but you told me to write down mythoughts. So here it is. When I went to bedthe night after of our first session, I thoughtabout what I said to you, you know, aboutbeing angry at my husband. I was thinkingthat you probably think I am this reallyharsh and cold person. I mean, here my hus-band died this horrible death and I have thishard, insensitive reaction. I started thinkingthat now you probably feel really negativelytoward me because of that statement andthat you don’t want to work with me.

THERAPIST: I’m really glad you’re telling methese thoughts. Let me start by asking youwho is having these negative thoughts?

DENISE: You? Well, no. Actually, it’s me.

THERAPIST: Right. Do you think that someone

like me might have another reaction to whatyou said?

DENISE: I don’t know. I mean it is pretty harshbeing angry at someone who had no controlover what was happening.

The therapist then offered Denise an alterna-tive perspective:

THERAPIST: Do you think that someone mightreact to your statements with empathy?

DENISE: How could they?

THERAPIST: I imagine it would be very upsettingand annoying to have lost both your hus-band and your friends—all around the sametime. Even though you love and care aboutall of them, feeling angry is understandable.It sounds like a basic human reaction tosome very difficult life events.

DENISE: Yeah, I guess that does make sense.Thanks.

This illustrates how a cognitive therapist canutilize events during the session to teach a pa-tient to identify automatic thoughts and to con-sider alternative interpretations. In addition,the therapist provided a summary of a keytheme he had identified from listening toDenise’s automatic thoughts about her hus-band and about therapy. The theme was herfear of being harshly judged and potentiallypunished for her statement (punitivenessschema). Cognitive therapists often identifyand begin to correct EMSs during the firstphase of treatment. More intensive work onchanging schemas in a later phase of treatmentmay be required to inoculate against relapse.We elaborate on this process in the next sectionof this chapter. In the segment that follows, thetherapist explained how he arrived at the con-clusion that punitiveness was an importantschema for Denise.

THERAPIST: When you said that you thought Iwould have a negative opinion about youand not want to work with you because yousaid you felt angry at your husband, itsounded as though you were really con-cerned that you would be harshly judged andpunished for your statements.

DENISE: Yes, that’s right.

THERAPIST: I don’t want to make too much outof this at the moment, but you also said that


after your friends had moved, you felt angryat them and judgmental of their decision.Even though you knew that each set offriends had to move for specific financial orhealth reasons and they had been in the pro-cess of completing their moves over severalyears, part of you still felt very angry withthem. You mentioned that you strongly be-lieve that friends should be there for eachother, especially in times of great need, and ifa friend lets another friend down, that rela-tionship should end. Is that right?

DENISE: Right.

THERAPIST: So here, you largely have with-drawn from these important relationshipsand now you’re feeling quite lonely. Thethought of talking to these friends againbrings fears that they will now be angry andpunitive with you for your reaction to them.You’re caught in a no-win situation. Is thatright?

DENISE: Yes, that sounds right.

THERAPIST: So one of the things that can reallygrab hold of you—and make you feelterrible—is this notion that people, includingyourself, should behave in specific ways, andif they or you don’t behave the “right” way,then harsh punishment should result. Is thatcorrect?

DENISE: Yes, that sounds right. But hearing yousay it makes me realize that it doesn’t reallysound right.


DENISE: It’s too extreme. It’s too harsh. Peopleare human and they have limitations andthey make mistakes sometimes.

THERAPIST: It’s good that you are starting tonotice and evaluate these thoughts ratherthan just responding to them automatically.What this tells us is that you have to be alertfor whenever you have the sense that eitheryou or others should be strongly punishedfor not behaving in a specified way. The ideathat people should not be cut a break, evenunder very difficult circumstances, may notwork very well in real life with real people.You mentioned that both friends told youthey felt terrible about leaving you at thistime, and both have called you regularlysince leaving the area. Do you think that ifyou begin to respond to and return theircalls, they might react differently—in the

same way that I reacted differently fromwhat you expected?

DENISE: Yes, that is very likely.

About halfway through the session, the ther-apist asked the patient for feedback thus far:

THERAPIST: Now at this point, is there anythingthat we have discussed today that botheredyou?

DENISE: That bothered me?

THERAPIST: Yeah.

DENISE: I feel like I’m a bit of a freak.

THERAPIST: That is important. Can you . . .

DENISE: Well, I’m trying not to feel that way,but I do.

THERAPIST: Well, if you are, you are. Why don’tyou just let yourself feel like a freak and tellme about it?

DENISE: Well, I’m feeling like I’m just so differ-ent from everyone else. Other people don’tseem to have my problems. They’re still hap-pily married and carrying on with life. I justfeel so different from everyone.

This comment led to identification of a thirdtheme, the social isolation/alienation schema.Denise had been viewing herself as increasinglydifferent for the past couple of years. By thispoint, however, she was beginning to catch onto the idea of answering her thoughts more ra-tionally. After the therapist pointed out thenegative thought in the preceding excerpt, thepatient volunteered:

DENISE: I know what to do with the thought“I’m a freak.”

THERAPIST: What are you going to do with itright this minute?

DENISE: I am going to say to myself, “I’m notso different from other people. Other peoplehave lost their mates. I’m not the only one.I’m just the first one in my group of friends.Eventually, they will all have the same situa-tion as me. It’s just a part of life.” Seeing youfor help doesn’t mean I’m a freak. You prob-ably see lots of people and help them withproblems like mine.

THERAPIST: Right.

The same automatic thoughts arose later inthe session, when Denise noticed the therapist’s


wedding ring. In the extended excerpt below,the therapist helped her set up an experiment totest the thought “I’m so different from him.”

THERAPIST: OK, now let’s just do an experi-ment and see if you yourself can respond tothe automatic thought, and let’s see whathappens to your feeling. See if responding ra-tionally makes you feel worse or better.

DENISE: OK.

THERAPIST: OK. “ ‘I’m so different from him.”What is the rational answer to that? A realis-tic answer?

DENISE: You are wearing a wedding ring andthat is different from me, because I’m alone,without a mate.

THERAPIST: Yes. And?

DENISE: And? . . . I don’t really know muchabout you, other than you’re married. Iguess from what I do know, that informationcould also be viewed as a similarity. We bothhave gotten married and know what it’s liketo be married. I assume that you’ve neverlost a mate, but maybe that is not true. Youmay have lost a mate as well.

THERAPIST: So, is it that you’re different or thatI’m different? Or is it that we just have dif-ferent situations with respect to our mates atthis point in time?

DENISE: We just have different situations rightnow.

The preceding exchange demonstrates theuse of reattribution. At first, Denise interpretedthe therapist’s ring as evidence that they werevery different. As a result of the guided discov-ery approach, she reattributed the difference toone of two factors: Either she or the therapistwas different or that the situation with respectto mates was different for each. At the end ofthe experiment, Denise expressed satisfactionthat she was finally recognizing this tendencyto distort her appraisals.

DENISE: Right now I feel glad. I’m feeling a lit-tle better that at least somebody is pointingthese things out to me. I never realized I wasso judgmental of myself and other peopleand that I’m assuming I’m so different fromeveryone.

THERAPIST: So you feel good that you havemade this observation about yourself?

DENISE: Yes.

After summarizing the main points of thesecond session, the therapist assigned home-work for the coming week: to fill out the DailyRecord of Dysfunctional Thoughts (see Figure6.4) and the Weekly Activity Schedule (withmastery and pleasure ratings; see Figure 6.3).

Session 3

By the beginning of the third session, Denise’smood had visibly improved. She had registeredfor a lecture series at the museum and waslooking forward to attending the first lecture.She also had called her friend Diane with verypositive results. She was catching negative andpunitive thoughts toward others and herselfand was challenging these thoughts. The pri-mary agenda item Denise chose to work on was“how I back away from other people,” an as-pect of her unrelenting standards, punitiveness,and social isolation/alienation schemas.

DENISE: I want to stop withdrawing from peo-ple. I want to be more accepting and engagedwith others.

THERAPIST: What holds you back?

DENISE: I guess I believe that I have to be a bitremoved and strict in relation to others orthey’ll just behave in whatever way theywant. People have to know my rules andabide by them if they want to have a rela-tionship with me.

The therapist continued probing to under-stand why Denise believed she had to have oth-ers adhere to such a strict set of rules to have arelationship. As the discussion progressed, itbecame obvious that, in the abstract, she couldsee that such hard and fast rules were notnecessarily conducive to having a goodrelationship—in fact, such rules sometimes putothers off. But in real-life situations, Denisenever felt she was wrong. The therapist’s nexttask was to help Denise bring her rationalthinking to bear on her distorted thinking inthe context of a concrete event. At the thera-pist’s request, Denise then described a conver-sation with her friend Diane, and how her in-tolerance for Diane’s deviation from her rulescreated distance. Denise had wanted Diane andher husband to come for a visit the followingsummer. Diane, however, told her that their


dog had been quite sick, and that if the dog wasstill alive she could not leave it. Denise thoughtthis was ridiculous. She believed a relationshipwith a pet should never take precedence over ahuman relationship. This occurred whenDenise actually had wanted and hoped to getcloser to Diane again. The therapist helped heruse logic to evaluate her maladaptive schema.

THERAPIST: You had the thought “I’m right toset the record straight with her. She can’t putme in second place to her dog. She can’t dothat without consequence.” It seems likelythat you believed that thought, and that youbelieved the thought was right. And sinceyou believed that thought was right, youthen felt you had to withdraw your affectionfrom her if she didn’t abide by your wishes.

DENISE: Right.

THERAPIST: Now, let’s look at it. Do you thinkthat thought is correct?

DENISE: Well, yes, it’s insulting.

THERAPIST: What’s insulting?

DENISE: She’s putting her dog in a higher prior-ity position.

THERAPIST: Have you ever had a pet?

DENISE: No.

THERAPIST: Do you think that maybe Dianefeels like her dog is a part of her family?

DENISE: I never thought of it that way.

THERAPIST: If you look at the situation fromthat perspective, how do you feel?

DENISE: I feel like I’m being a little insensitive.. . . That’s not right. I’m not allowing for anyother perspective. I’ve never had a pet, so Idon’t really know what it’s like to have a pet.It’s not right for me to be so judgmental ofDiane. I need to be more understanding. Iwasn’t very caring. I am actually behaving ina way that goes directly against my deepestvalues.

THERAPIST: So, according to your own values,was this right?

DENISE: No, it’s not right. I wasn’t respectingher feelings. I was just demanding that sherespect mine. That wasn’t right.

THERAPIST: OK, now this is one of the prob-lems. If you want to get over this sense thatyou should never give in or bend your rulesfor others, one of the things you can do islook for this thought, “I’m right and you

should have a negative consequence for your‘wrong’ decision”—and refer back to thisconversation we are having now and decidefor yourself whether, indeed, you were right.Now, if every time you approach a conflict ina relationship and allow for the possibilitythat you might not fully understand, but re-ally think underneath, “But I know I’mright,” you are going to feel put out, andthen you are not going to want to engagewith that person. Is that right?

DENISE: Yeah, that sounds right.

THERAPIST: So we have to decide here and now.Do you indeed think that you are right tosuspend your initial negative judgment toleave open the possibility of reevaluatingyour reaction to her behavior?

DENISE: Yes.

THERAPIST: Now, the next time you get thethought, “I’m right and I’m going to makesure this other person knows it,” how areyou going to answer that thought?

DENISE: If I’m right? But I’m not necessarilyright. I need to consider the other person’sperspective. I need to try to understand themand then see if what I’m thinking fits.

THERAPIST: Now are you saying that becausethat is the correct answer, or because you re-ally believe it?

DENISE: No, I really believe it.

The therapist followed this discussion with atechnique called “point–counterpoint” to helpDenise practice rational responses to her auto-matic thoughts even more intensively. In thisexcerpt, the therapist expressed Denise’s ownnegative thinking as Denise tried to defend her-self more rationally.

THERAPIST: Now I am going to be like the pros-ecuting attorney, and I’ll say, “Now I under-stand you let your friend violate one of yourrules of friendship. Is that true?”

DENISE: Yes.

THERAPIST: “Now it seems to me that that wasa very bad thing for you to do.”

DENISE: No, it wasn’t.

THERAPIST: “You don’t think it was?”

DENISE: No, I should try to understand her per-spective.

THERAPIST: “Well, you can sit there and say you


should be more understanding, but I thoughtyou said before that you wanted people torespect you.”

DENISE: I do, but I also need to respect others.

THERAPIST: “I know, but now you are sayingthat you are going to let her get away withthis. What’s next?”

DENISE: What’s next can only be a better under-standing of one another. We’ll feel closer.

THERAPIST: “But how can you feel closer if she’snot respecting your rules of friendship?”

DENISE: Maybe my rules are not appropriate inthis situation. I need to learn to be more un-derstanding, flexible, and tolerant of somedeviations from my rules.

THERAPIST: “But then you’ll lose control of thesituation.”

DENISE: No, that’s an exaggeration. I don’tneed to control the whole situation. I canstill decide what makes sense. I am still incontrol of what’s important.

THERAPIST: “How can that be?”

DENISE: Because I can respect myself and re-spect my friend, as well. I don’t have to turneverything into an either–or situation to tryto make her see and do it my way. That justmakes it difficult for her to get along withme, and I’ll lose out on the relationship inthe long run if I keep on insisting that she ei-ther do it my way or we do nothing.

Finally, the therapist returned to the schemaand asked the patient how much she believedthe new perspective.

THERAPIST: If you’re flexible, you’ll lose con-trol. Now do you believe that?

DENISE: No.

THERAPIST: Do you believe it partially?

DENISE: No. In fact, I’m more likely to losecontrol of any possibility of getting what Iwant if I’m so inflexible. It’s like I lose sightof the importance of the relationship when Iget so stuck on thinking that I have to be incontrol and that the other person has to do itmy way.

THERAPIST: OK, so right now, how much doyou believe that?

DENISE: Completely.

THERAPIST: 100%?

DENISE: Yes.

THERAPIST: You are sure 100%, not 90 or80%?

DENISE: No, 100%.

For the remainder of Session 3, Denise andthe therapist reviewed other instances in whichshe noticed that her standards were not flexibleand felt the urge to be punitive when her ruleswere not met. The session ended with a sum-mary of the main issues raised in the first threesessions.

Summary of Initial Sessions

In the first three sessions, the therapist laid thegroundwork for the remainder of treatment. Hebegan immediately by teaching Denise to iden-tify her negative automatic thoughts. By doingthis, the therapist began to understand her feel-ings of hopelessness and to explore her isola-tion. By identifying her thoughts in a variety ofspecific situations, he was able to deduce severalkey schemas that later proved central to Denise’sthinking: (1) unrelenting standards, (2) puni-tiveness, and (3) social isolation/alienation. Allappeared to be contributing to Denise’s socialisolation and depression. The therapist made es-pecially skillful use of Denise’s thoughts duringthe second therapy session to help her see thatshe was distorting evidence about the therapeu-tic interaction and coming to the inaccurateconclusion that the therapist would be judg-mental and punitive with her and withdraw pos-itive feelings for her, in the same way that Denisetends to respond to others.

Beyond identifying thoughts and distortions,the therapist guided Denise to take concretesteps to overcome her inactivity and with-drawal. He asked her to weigh the advantagesand disadvantages of staying in the house allday watching TV; he broke down the task of at-tending a lecture series at the museum intosmall, manageable steps; and he worked withher to develop an activity schedule to followduring the week.

Finally, the therapist employed a variety ofstrategies to demonstrate to Denise that shecould test the validity of her thoughts, developrational responses, and feel better. For exam-ple, during the course of the three sessions thetherapist set up an experiment, used reat-tribution, offered alternative perspectives, andpracticed the point–counterpoint technique.


One final point we want to emphasize is thatthe primary therapeutic mode was questioning.Most of the therapist’s comments were in theform of questions. This helped Denise to evalu-ate her own thoughts outside of the session andprevented her from feeling attacked by thetherapist.

By the end of these initial sessions, Denise re-ported being more optimistic that her life couldchange.

Later Sessions

Denise continued to fill out the Daily Record ofDysfunctional Thoughts and gathered evidencethat she could relax her standards and be moretolerant of others’ viewpoints and foibles. Shediscovered that she felt happier, both with her-self and others, as a result.

The therapist set up several experimentswith Denise to test a series of beliefs: that herfriends would become punitive with her whenshe did not behave perfectly, and that her rela-tionships would become unpleasant and unde-sirable if she relaxed any of her rigid standardsregarding how others should behave in rela-tionships. Through graded tasks, Denise coun-teracted her tendency to withdraw by graduallyapproaching new and sometimes unfamiliarsituations. When she noticed herself imposingher standards of behavior on others, or noticedin herself the urge to become punitive, Denisepracticed more open and accepting behaviors(by asking open-ended questions that reflectedback her understanding of others’ responses,and by inhibiting harsh and judgmental state-ments). She practiced tolerating the discomfortassociated with these new behaviors until theybegan to feel more comfortable and natural.

When Denise terminated therapy, her BDIscore was in the normal range. The symptomreduction phase of treatment was successfullycompleted in 20 sessions.

The next section describes and illustrates acase example of schema-focused therapy forchronic depression.

SCHEMA THERAPYFOR CHRONIC DEPRESSION

Schema therapy, developed by Young (1990/1999; Young et al., 2003), can be used with pa-tients who present with recurrent depressiveepisodes, an early age of onset (before age 20)

of depression, early life trauma or adverse fam-ily relations (loss of parent in childhood,sexual, physical and/or verbal abuse, neglect,and overprotection), comorbid personality dis-order(s), and a large number of EMSs (identi-fied with the Young Schema Questionnaire;Young & Brown, 1990/1994), particularly inthe domains of Impaired Autonomy and Over-vigilance. Young (1990/1999; Young et al.,2003) has written extensively about the schematherapy approach. Young and Klosko (1994)have published a self-help book for patients toguide them.

Beck and colleagues (1990) have noted that

schemas are difficult to alter. They are held firmlyin place by behavioral, cognitive, and affective ele-ments. The therapeutic approach must take a tri-partite approach. To take a strictly cognitive ap-proach and try to argue patients out of theirdistortions will not work. Having the patientsabreact within the session to fantasies or recollec-tions will not be successful by itself. A therapeuticprogram that addresses all three areas is essential.A patient’s cognitive distortions serve as signpoststhat point to the schema. (p. 10)

As a result, schema therapy is significantly dif-ferent from traditional CBT. It places more em-phasis on early developmental patterns and ori-gins, long-term interpersonal difficulties, thepatient–therapist relationship, and emotive orexperiential exercises.

Case Study of Barbara

The second case study demonstrates the use ofschema therapy with a chronically depressedpatient. We shifted to schema therapy becauseBarbara’s depression was not lifting with stan-dard cognitive therapy. Although Barbara hadlearned how to challenge automatic thoughtswith rational responses and had followedthrough on graded behavioral assignments totest her thoughts, she never believed the ratio-nal responses. Barbara had remained con-vinced that she was worthless, useless, andhopeless. Additionally, many of the behavioralassignments presented “catch-22” scenarioswithin her current life situation. It was neces-sary to add various emotive and experientialtechniques that are part of schema therapy toaccess Barbara’s core beliefs. Core beliefs arethe cognitive components of schemas. Schemasare deeply held emotionally based beliefs orEMSs that are unquestioningly experienced as


true and impervious to rational challenges. Thetreatment was concentrated on schema modework. Before presenting the specifics of theschema therapy treatment, we present Barbara.

History and Presenting Problems

The patient “Barbara” was a very physicallyattractive 46-year-old woman who had beenmarried for the past 20 years to George, a func-tional alcoholic working on Wall Street in in-vestment banking. It was the first marriage forboth, but it had been extremely rocky through-out. Barbara had wanted children in themarriage but had been unable to conceive natu-rally. Although fertility treatments and adop-tion had been discussed, Barbara reported thatGeorge resisted following through on these op-tions, because he questioned her ability to bean adequate parent. At the beginning of Georgeand Barbara’s relationship, Barbara reportedextremely strong sexual chemistry. However,she stated that the chemistry had been erraticand often disappeared for long periods over thecourse of their relationship. She said that shehad only felt real happiness with George in thefirst few months of the relationship, when hewas extremely generous and attentive. Her de-cision to marry George was based on a feelingthat they were meant for each other. They ini-tially met at a high-end bar, where Barbara hadworked as a cocktail waitress. Once their rela-tionship was established, however, George con-tinued to spend most of his free time in cocktaillounges without her, as she did not like todrink.

At the time she came in for her first inter-view, Barbara was spending most of her time inbed or watching TV. She rarely left the houseexcept to go shopping. Outings with her momor friends “to lift her spirits” often culminatedin shopping sprees, followed by extended ver-bal berating by George, drunk upon his returnhome from the bar, because of what he per-ceived as her lack of taste, judgment, and intel-ligence regarding purchases. Barbara enteredtreatment because her husband told her thatshe “was driving him crazy with all her ridicu-lous behavior” and that she should “go getfixed.” Barbara acknowledged feeling very de-pressed. She stated that this latest episode ofdepression started after a particularly upsettingfight with her husband over adopting a child.

Barbara reported having mild to moderatedepression most of her life, interspersed with

multiple episodes of major depression. She firstbecame aware of feeling moderately depressedaround age 11. An only child, she initially de-scribed her childhood family as “fine.” Barbaradescribed her mother as extremely attentiveand devoted, a “very good mother who did ev-erything for me.” She stated that her motherlived for her, but also remembered her motheroccasionally falling into depressions. She re-called that when her mother was depressed, shewas emotionally unavailable. An imagery exer-cise brought up a memory of her mother duringone of these periods, in which Barbara, ataround the age of 6, felt very scared and lost.Barbara described her father as a workaholicwho was almost never home. When he didcome home, he was remote, preferring the soli-tude of his study over interacting with her andher mother. If Barbara tried to engage him as achild, he would upbraid her by calling her “anannoying simpleton” and demand that sheleave him alone. An imagery exercise of her fa-ther revealed his intolerant, demeaning, and re-jecting behavior toward Barbara as a youngchild. Barbara described her first major depres-sive episode, which occurred around age 16, af-ter a breakup with her first serious boyfriend.Later episodes were triggered by other break-ups and the fertility problems.

Barbara scored 29 on the BDl, placing herin the moderate to severe range of depression.She also completed the Young Schema Ques-tionnaire (Young & Brown, 1990/1994), andreceived very high scores on EmotionalDeprivation, Defectiveness, Abandonment,Dependence/Incompetence, Entitlement, Fail-ure, Subjugation, Approval Seeking, and Neg-ativity/Pessimism. Her Schema Mode Ques-tionnaire (Young, Atkinson, Arntz, Weishaar,& Weishaar, 2004) indicated that she primar-ily functioned in the following modes: De-tached Protector, Compliant Surrender, Puni-tive Parent, and Vulnerable Child. On theMultimodal Life History Inventory (Lazarus& Lazarus, 1991), a 15-page assessment toolcovering a wide range of issues dealing withfeelings, thoughts, behaviors, and a variety ofother psychotherapeutic issues, Barbara re-ported her main problems as depression, be-ing unhappy with herself, feeling empty andunloved, and feeling unappreciated. She alsolisted the following behaviors as applicable:procrastination, withdrawal, concentrationdifficulties, sleep disturbance, crying, and oc-casional outbursts of temper. She further indi-


cated that she often felt sad, depressed, un-happy, hopeless, useless, and lonely. She en-dorsed the following statements: “I don’tknow what to do with my life,” “Life isempty, a waste,” and “There is nothing tolook forward to.”

Based on the initial interview, Barbara wasdiagnosed with major depressive disorder, re-current episode on Axis I and dependent per-sonality disorder on Axis II.

This section demonstrates schema therapy,or, more specifically, schema mode work withBarbara. The general overview is as follows. Asnoted earlier in this chapter, there are fourmain types of modes: child modes, maladaptivecoping modes, dysfunctional parent modes,and the healthy adult modes (see Figures 6.2a–c, on pages 264–265). To understand the pa-tient’s current problems in terms of modes, firstit is important to identify with the patient theprimary modes in which he or she is function-ing. Then, it is necessary to help the patient rec-ognize and understand the origins of the mal-adaptive and dysfunctional modes (with theirembedded EMSs), then heal and/or strengthenthe healthy modes (child and adult). Once mal-adaptive modes are fully exposed and thehealthy modes are sufficiently strengthened,the patient is assisted in seriously challengingthe early maladaptive schemas with accompa-nying negative cognitions and the dysfunction-al coping styles. Depression lifts as the patientmore firmly establishes the healthy modes ashis or her primary modes. To summarize, thereare seven general steps in mode work: (1) in-crease awareness of modes by identifying andlabeling modes with the patient; (2) explore or-igins of modes in childhood and/or adolescenceand reveal their adaptive value when relevant;(3) link current problems and symptoms tomaladaptive modes; (4) uncover the advan-tages and disadvantages of each mode; (5) useimagery to access the vulnerable child mode;(6) conduct dialogues between modes; and (7)generalize the results from mode work to reallife. The following sections illustrate the thera-peutic process with each step.

Step 1: Increase Awareness of Modes by Identifyingand Labeling Modes with the Patient

This first step helps both the therapist and thepatient conceptualize the problems in terms ofdifferent parts of the self or modes. From theSchema Mode Questionnaire, the therapist al-

ready was aware that Barbara primarily func-tioned in the Detached Protector, CompliantSurrenderer, Punitive/Critical Parent, and Vul-nerable Child modes. The session reveals howthe therapist queries the patient so that she canbegin to recognize and differentiate these partswithin herself. Throughout this section, the pa-tient is encouraged to label modes with termsthat feel right—rather than simply applying ge-neric terms from the Schema Mode Question-naire. Patients are encouraged to find termsthat best capture the thoughts, emotions, and/or behaviors associated with each mode. A pri-mary goal with this first step is to help the pa-tient observe these parts, and decenter him- orherself from these parts. This step begins theprocess of interrupting the automaticity of themodes.

THERAPIST: I’m noticing in our sessions thatsometimes you seem very sad and upset andcritical of yourself, and other times you seema bit distracted from what you’re feeling, likewhen you were telling me about the greatpair of shoes that you found on your shop-ping trip.

BARBARA: Yeah, I guess that’s true. Talkingabout my purchases makes me feel better.

THERAPIST: When you say better, what do youmean?

BARBARA: I feel good.

THERAPIST: Like happy, peaceful, and content?

BARBARA: I feel pleasure.

THERAPIST: In what way is it pleasurable?

BARBARA: I feel pleasure when I look at prettythings with my mom or my friends, and I likebuying those things. It takes my mind off ofeverything else.

THERAPIST: Like you’re temporarily distractedfrom other feelings?

BARBARA: Yeah, that’s right.

THERAPIST: What are those other feelings?

BARBARA: Just feeling real bad. I can’t standthose feelings.

THERAPIST: So, sometimes there is this one partof you that just wants distance from thisother part of you that feels bad?

BARBARA: Yes.

THERAPIST: Are there other things you do be-sides shopping that help you distance frombad feelings?


BARBARA: Well, yeah, I sleep a lot.

THERAPIST: Anything else?

BARBARA: I watch TV, but TV doesn’t alwayswork.

THERAPIST: This part of you that wants dis-tance from feeling bad—what could we callthat part of you?

BARBARA: That part of me? I don’t know. Idon’t know what to call it.

THERAPIST: What does it feel like?

BARBARA: It feels like I’m escaping.

THERAPIST: OK. So should we call that part“the Escapist” [Detached Protector mode]?

BARBARA: I guess so. . . . That sounds right.

THERAPIST: And the part of you that you’ve es-caped from—can you tell me more aboutthat part of you?

BARBARA: That part feels bad . . . really badand terrible.

THERAPIST: Let me hear that part speak aboutthose feelings.

BARBARA: I’m just a bad person (Starts to cry.). . . I’m a useless good for nothing. I feel sohopeless. I don’t know what to do. I can’tfigure anything out. I’m such a zero, such afailure. . . . Do we really have to talk aboutthis?

THERAPIST: Barbara, I know if feels bad to be intouch with this part of yourself, but if youcan hang in there for just a bit, it will helpme to understand why you’re feeling so bad.Do you know what this part of you wants?

BARBARA: I want to feel good.

THERAPIST: What do you think would help youfeel better?

BARBARA: I don’t know. I really don’t know. Ijust want you to fix me. My husband is right,I am a ridiculous person.

THERAPIST: It sounds like there is one part ofyou that feels really bad, like there is some-thing really wrong with you, and then thereis another part that agrees with what yourhusband says about you—that “you’re ridic-ulous.” I want you to hold back the part thatis agreeing with your husband, and I want tohear more about the part of you that feelsbad—the part of you who wants everythingfixed. That part . . . do you feel it?

BARBARA: Yes, I feel it.

THERAPIST: Tell me more. What are the thingsyou want to fix?

BARBARA: I don’t know. I just want to feel goodabout myself, proud of myself, but I justdon’t. I want to have a child, but my hus-band doesn’t think I can handle it. He’sprobably right. I need so much help with justeveryday living. I don’t know how to handleanything.

THERAPIST: So this part of you, this part thatfeels bad and helpless but also wants to feelbetter, what could we call this part?

BARBARA: I don’t know. What do you think?

THERAPIST: Well, what does it feel like?

BARBARA: It feels helpless, bad and helpless.

THERAPIST: How old do you feel when you’rein touch with this part of yourself?

BARBARA: I feel young, very young.

THERAPIST: Do you want to call that part “BadLittle Barbara” [Vulnerable Child mode withassociated EMSs of defectiveness, depend-ence/incompetence, and enmeshment/unde-veloped self]?

BARBARA: Sure.

THERAPIST: OK. And now this other part—theone that is agreeing with your husband andcalling yourself ridiculous . . .

BARBARA: Well, I am ridiculous and silly. I can’tcope with anything.

THERAPIST: Before you agree with that part, Iwant you to just notice how that partsounds. How does that part sound to you?Does it sound critical?

BARBARA: Yes, but I deserve it. I’m so useless.

THERAPIST: It sounds like you are having a hardtime just listening to that part without auto-matically agreeing with it?

BARBARA: Yeah, I guess that’s true.

THERAPIST: So, what do you want to call thatpart?

BARBARA: I don’t know . . . But . . . you’re notgoing to tell me what to call it, are you?

THERAPIST: Right.

BARBARA: OK. I guess “The Critic” [Punitive/Critical Parent mode with associated EMSsof defectiveness and subjugation].

In this part of the session, the therapist hashelped Barbara begin to recognize and label the


Detached Protector, Punitive/Critical Parent,and Vulnerable Child modes. Although notfully illustrated here, the therapist also used asimilar sequence of questions to help Barbaraidentify other modes. From this portion of thesession, it is clear that “the Critic” (or Punitive/Critical Parent mode) is generating a tremen-dous amount of negative affect in Barbara.Barbara’s only apparent way to cope with theonslaught of criticism from this mode is “theEscapist” (Detached Protector mode), in whichshe sleeps away much of her life. Otherwise,Barbara’s primary experience of herself restswith the “Bad Little Barbara” (VulnerableChild mode), where she feels defective, useless,hopeless, and helpless.

Step 2: Explore Origins of Modes in Childhoodand/or Adolescence

This section illustrates how the therapist helpsBarbara recognize the origins of these modes.In addition, the therapist assesses the strengthof a Healthy Parent mode.

THERAPIST: Does “the Critic” sound like any-one else you know or have known in yourlife?

BARBARA: Yes, it sounds like George.

THERAPIST: Anyone else?

BARBARA: Yeah, it also sounds like my Dad . . .just like my Dad.

THERAPIST: In what way?

BARBARA: My Dad used to talk like that to me. . . whenever I tried to get his attention.

THERAPIST: How old were you?

BARBARA: Young, very young . . . 3 or 4 . . . asfar back as I remember.

THERAPIST: Can you close your eyes and try tolet yourself feel like that young child againwith your dad?

BARBARA: (Closes eyes.)

THERAPIST: Tell me what’s happening.

BARBARA: He’s yelling at me because I pulledon his coat jacket.

THERAPIST: Let me hear what he’s saying toyou.

BARBARA: “Stop it, you little pest. You’re sucha simpleton. Can’t you find anything betterto do than pull on my coat? Get out ofhere!”

THERAPIST: And how are you feeling as he isyelling these things at you?

BARBARA: I feel stupid, like I’m a jerk. I’mnothing—a useless pest, an annoyance.

THERAPIST: Does it feel like he’s right, or areyou angry at him?

BARBARA: No. I don’t feel angry. I just feel bad(starts to cry)—I’m just bad.

THERAPIST: So there is part of you agreeingwith him, becoming critical of yourself—likeyour dad—thinking you’re bad.

BARBARA: Yeah.

THERAPIST: And where is your mom?

BARBARA: She’s telling me to shush and leavehim alone. She says he’s tired from workingso hard all day.

THERAPIST: And what are you thinking andfeeling when she says this?

BARBARA: I’m thinking that I’m a terrible per-son.

THERAPIST: So, you’re getting the message thatyou are the problem from both of your par-ents. It feels like both parents are beingcritical of you, and there is a part of you thatbelieves your parents’ criticisms. Is thatright?

BARBARA: Yes, that’s right.

THERAPIST: So there is also this critical part ofyou, this part that has accepted your “Criti-cal Parents” message, that you’re bad and aproblem—a pest.

BARBARA: Yes.

THERAPIST: What is your mom doing?

BARBARA: She whisks me away and tries tomake me feel better by giving me something,like a toy or something to eat. A lot of timesshe takes me out shopping and buys mesomething special.

THERAPIST: And how do you feel when she doesthis?

BARBARA: I feel a little better, but deep down Iknow I’m bad and useless. I still have thatbad and useless feeling inside.

THERAPIST: So there is still the “Bad LittleBarbara” underneath?

BARBARA: Yes. That is it.

THERAPIST: If you could have rewritten thescript for your family, a family with idealparents, what would you have happen?


BARBARA: I have no idea. They weren’t bad.They were doing their best.

THERAPIST: Yes, but what if you had had adaddy who was excited to see you at the endof his work day . . . a daddy who enjoyedcoming home to his family . . . who foundpleasure in talking with you and getting toknow you, and played with you?

BARBARA: You mean . . . a daddy who lovedme?

THERAPIST: Yes. I mean a daddy who was ableto show you his love through all sorts of ac-tions.

BARBARA: Wow . . . that would have been sodifferent.

THERAPIST: Do you feel any angry feelings athim now as you think about how he spoke toyou—this very young child simply trying toget his attention?

BARBARA: No, I was in his way. He workedhard. I had to leave him alone.

THERAPIST: What you just said—does thatsound like anyone you know?

BARBARA: Yes. It sounds like my mother.

THERAPIST: And what do you think of thatnow?

BARBARA: Well. She was just trying to keep thepeace and fill in for him.

THERAPIST: But what about you, LittleBarbara?

BARBARA: It’s sad. I feel sad for me.

THERAPIST: Right. It is sad for you. You are achild doing what children do—children tryto get their parents’ attention, children wantto know they are loved, valued, appreciated.You were no different from any other child,but what is happening here?

BARBARA: (Cries.)

THERAPIST: No one’s calling your dad on hispoor behavior toward you. Everyone’s ac-commodating him and speaking to you as ifyou are the problem, when all that you aredoing is what every young child does. Andyet your father and your mother are re-sponding as if you are the problem.

BARBARA: Yeah. You’re right. Why did they dothat?

THERAPIST: Do you think there was somethingwrong with you, or do you think there was

something wrong with the way they were be-having with you?

BARBARA: The way they were behaving withme. . . . It’s a problem with the way theytreated me.

In the preceding section, the therapist helpsBarbara to realize the origin of “the Critic,”that part of her self that has accepted her “Crit-ical Parent” message that she is bad anda useless pest. The therapist’s questions andcomments also help Barbara realize that theproblem was actually a problem with her par-ents’ behavior toward her rather than an inher-ent defect in herself.

The therapist now turns Barbara’s attentionto the origin of “the Escapist,” that part thatwas seeking relief from these terrible feelingsabout herself as a child. The therapist alsohelps Barbara become more aware that “theEscapist” was only able to provide short-termrelief.

THERAPIST: Now when you had this bad feelingabout yourself as a child, the “Bad Barbara”feeling, how did you cope with that feeling?What did you do?

BARBARA: A lot of the time I did nothing. I justsat on my bed in my room daydreaming,wishing everything was different. I wouldfantasize being a star, a beautiful person thateveryone adored. People would dote on meand give me expensive gifts and clothes thatmade me look even more beautiful. Some-times when my mom would take me out andbuy me things, it kind of felt like she wasmaking this dream come true.

THERAPIST: Is that the “the Escapist” part ofyou trying to help you feel better?

BARBARA: Yes, it definitely is.

THERAPIST: And then what would happen?

BARBARA: I would feel better, especially as I gotolder and I got a lot of attention from theway I looked, but eventually my dad alwayswould start yelling again. I remember himyelling at my mom for “spending too muchmoney” simply to turn me into “a prettyBimbo.”

THERAPIST: So the efforts to feel better eventu-ally backfired.

BARBARA: Right. I never felt very good for long.I never felt good inside.


Step 3: Link Current Problems and Symptomsto Maladaptive Modes

In this section, the therapist asks Barbara ques-tions that help her to recognize how thesemodes that developed in childhood are still op-erating in the present. Barbara begins to con-nect these modes with why she is feeling so de-pressed, and she begins to see the repetitivepatterns in her life.

THERAPIST: Let’s look at what is happeningnow in your life. Do you recognize any rela-tionship between what we have been talkingabout—these different parts of you—andhow you are thinking and feeling aboutyourself and coping with your life now?

BARBARA: Yeah, I’m still trying to feel good—or just not feel—by being “the Escapist.”When I go on these shopping sprees and letmy mom and my friends dress me up, it stilldoesn’t work any better than it ever did. Andsleeping all the time doesn’t work either.


BARBARA: Well, I only feel good for a shortwhile. Then, instead of my dad cominghome, now I’m coming home to George orhe’s coming home to me. George is just ascritical of me and just as unavailable as myDad.

THERAPIST: And then what happens?

BARBARA: I start to feel really bad again, reallydepressed. I feel like I’m not worth spendingtime with, like I’m boring and useless. I’msaying all sorts of bad things about myselfand I agree with everything George saysabout me. I’m my worst “Critic.” In the end,I still feel like the same old “Bad LittleBarbara” inside who never ever really feelsbetter. Then, I just want to go to sleep to getaway from it all. I’m finally starting to see itmore clearly, going around and around thesame old circle—nothing has changed andnothing changes. I’ve been miserable mywhole life.

THERAPIST: But there is one very importantchange.

BARBARA: What is that?

THERAPIST: You’re starting to see and under-stand what’s been happening for a long timeinstead of just being on the treadmill withoutany awareness.

Step 4: Uncover Advantages and Disadvantagesof Each Mode

The therapist now begins to ask Barbara ques-tions regarding the advantages and disadvan-tages of listening and acting on these differentparts of her self. This helps Barbara to gainmore distance from the modes and to reducefurther automatically responding to them.

THERAPIST: When these different parts of your-self or modes developed, they served a pur-pose. Why don’t we talk about each of themand explore their advantages and/or disad-vantages both in the past and present? Let’sstart with “the Escapist.”

BARBARA: Well, that part lets me feel good—orat least not feel bad.

THERAPIST: For how long?

BARBARA: For a while.

THERAPIST: And in the long run? Does that parthelp you feel better?

BARBARA: No, not really. I can’t really escapefeeling bad. I can’t say I feel any better in thelong run.

THERAPIST: So, is there any advantage to thispart of you—“the Escapist”?

BARBARA: I’m mixed about it.

THERAPIST: Sure. I can understand that. If youdon’t know of any other way to feel better,any relief is better than no relief at all.

BARBARA: Yeah.

THERAPIST: But sometimes it is good to letyourself feel bad, because when you let your-self stay in touch with your feelings, you of-ten can begin to recognize what feels betterin the long run . . . like finding activities thatare truly interesting to you or that give you asense of accomplishment and purpose, orthat bring out a feeling of deep enjoyment.

BARBARA: That makes sense, but I have no ideahow to find those things.

THERAPIST: Well, that is something we canwork on together.

BARBARA: That sounds good.

THERAPIST: And what about “the Critic”? Isthere any advantage to listening to “theCritic”?

BARBARA: That part tells me what’s wrong withme.


THERAPIST: And what does “the Critic” say iswrong with you?

BARBARA: That I’m stupid and silly, useless andannoying.

THERAPIST: Do you think “the Critic” is right?

BARBARA: Of course.

THERAPIST: But if you put the words of “theCritic” back in your dad’s mouth and listento him speak to little 3-year-old Barbara asshe excitedly greets him coming home fromwork, what do you think?

BARBARA: When you put it that way, I thinkhe’s a jerk. I mean, what does he think he isdoing to this poor little girl? What’s wrongwith him?

THERAPIST: Right. So is there any advantage tolistening to these critical words?

BARBARA: No . . . no, definitely not.

THERAPIST: Are there any disadvantages of lis-tening to “the Critic”?

BARBARA: Yes. It’s getting clearer now why thatpart is making me feel so bad. I have to stoplistening to that part. When I accept whatmy dad said as the truth, I feel bad, reallybad, about myself.

THERAPIST: You mean it brings up that “BadLittle Barbara” feeling?

BARBARA: Yes.

THERAPIST: And what about her? Is there anyadvantage to listening to her?

BARBARA: I don’t think anyone has ever reallylistened or paid attention to her, and whatshe needs to feel happy. There’s a big advan-tage and no disadvantage to listening to thatpart of me.

THERAPIST: I agree.

Step 5: Use Imageryto Access the Vulnerable Child Mode

The therapist now begins to engage Barbara inthe Vulnerable Child mode. By accessing thismode, therapist and patient can begin to workon the core schemas that are part of LittleBarbara: her feelings, needs, and beliefs.

THERAPIST: I know it’s sometimes unpleasant tolet yourself be in touch with “Bad LittleBarbara,” but would you be willing to gothere, so that we can get to know that part ofyou to find out what you need to truly feel

better and to find out what is getting in theway?

BARBARA: I guess . . . (Closes eyes.)

THERAPIST: Tell me what you are feeling andthinking right now.

BARBARA: It’s the same old feelings—I feel bad. . . and useless. (Visibly begins to look upset,then opens her eyes.)

THERAPIST: Can you go back there to ”LittleBarbara,” can you let yourself feel her?

BARBARA: OK (Closes her eyes again.)

THERAPIST: Where are you? What are you do-ing?

BARBARA: I’m sitting on my bed in my room.My daddy just told me to go away.

THERAPIST: And how are you feeling?

BARBARA: Awful.

THERAPIST: And what do you want?

BARBARA: I want my mommy to come in andmake me feel better.

THERAPIST: Where is she?

BARBARA: She’s in her room, lying on her bed.She doesn’t feel good either.

THERAPIST: If you could bring in a healthymommy to be with you right now, whatwould she say to you? Let me hear her speakto Little Barbara.

BARBARA: You know he’s just tired from work.If you just leave him alone, everything willbe all right.

THERAPIST: Let me hear what Little Barbarathinks of this?

BARBARA: That doesn’t help much. I still feelupset and bad.

THERAPIST: How about if I come in to help?

BARBARA: OK.

THERAPIST: You are such a loving little girl. Somany fathers would relish coming hometo such a greeting from their little girl.Something is wrong with your Daddy. Whydoesn’t he recognize that you’re offering himsomething that is precious—so special? . . .How does little Barbara feel as she hearsthis?

BARBARA: Better.

THERAPIST: Let me talk to your father now.

BARBARA: OK.

THERAPIST: How can you speak to your daugh-


ter that way? She’s done nothing but wel-come you home with joy and love. And lookat how you are responding. Your response toher is completely inappropriate. You are soclosed off and removed from everything.What is going on with you? Why can’t yousee that you have a beautiful, creative, andloving daughter? Barbara does not deservethis type of treatment. I’m not going to standfor this and let you hurt her anymore. . . .How does little Barbara feel as she hearsthis?

BARBARA: A lot better. I wish my mom wouldhave spoken to my dad that way. But youknow, if you really tried to say anything likethat to him, he wouldn’t listen. He never lis-tened to my mom. He’d probably just saysomething mean to you and then shut thedoor.

THERAPIST: OK. Well if that happened, whatwould you want to happen next?

BARBARA: I don’t know.

THERAPIST: Let me step in again and say this tohim through the door. Whether or not youchoose to listen, there is a limit, a limit as tohow long we will stay here for you. If youchoose not to be here for us, we will not stayhere for you.

BARBARA: I don’t think he’ll change.

THERAPIST: Then you and I, together, will leavehim and create a better life.

BARBARA: Is that really possible?

THERAPIST: Do you want to make it happen?

BARBARA: Yes, but what about him? He’ll be solonely.

THERAPIST: Who will be lonely?

BARBARA: He will, unless we stay.

THERAPIST: But if you stay, who will be lonelythen?

BARBARA: Me.

THERAPIST: Do you want to continue to bethere for someone who chooses not to bethere for you?

BARBARA: Well, no—no, that isn’t right.

THERAPIST: So what do you want to do?

BARBARA: I want to leave.

In this part of the session, the therapist hastaken on the “Healthy Parent” mode forBarbara, because she does not have a strong

template for this mode. This is one example ofwhat is meant by “limited reparenting.” In thisrole, the therapist temporarily steps in to sup-port the client in relation to unmet basic needsfor safety; stability or predictability; love, nur-turance, and attention; acceptance and praise;empathy; realistic limits; and validation of feel-ings and needs. The therapist also counters andchallenges any unreasonable messages or be-liefs that such needs should remain unfulfilled.

Step 6: Conduct Dialogues between Modes

Once the patient clearly identifies and is able toevoke the Healthy Adult and the VulnerableChild in imagery, the next step is to have the pa-tient begin dialogues between the variousmodes. For example, the Healthy Adult canhave a dialogue with the Vulnerable Child; theCritical Parent, with the Healthy Adult; and theDetached Protector, with the Vulnerable Child.

THERAPIST: How about having “LittleBarbara” talk with “the Escapist” aboutthings that she really enjoys?

BARBARA: OK . . . but how do we do that?

THERAPIST: First, I want you to sit on this sideof the couch. When you’re on this side of thecouch, I want you to let yourself get in touchwith and speak for “Little Barbara.” Have“Little Barbara” talk about all those thingsthat interest her—those things that help herfeel alive. And then, I want you to get up andsit on the other side of the couch. Whenyou’re on this side of the couch, I want youto let yourself get in touch with and speakfor “the Escapist.” We’ll go back and forthbetween those two parts of your self to seewhat comes out. I’ll help out “LittleBarbara” if she needs me.

BARBARA: [as “Little Barbara”] You know, Ionce made a dress at school, and I noticedthat I really liked sewing. It was a really sim-ple pattern, nothing special, but I had somuch fun making it. I think I might want totake a class to try sewing again.

[as “the Escapist”] Why do you want to dothat? If it doesn’t turn out, you’ll just feelbad about yourself. People might thinkyou’re ridiculous at this age, trying to learnto sew. Why take the chance that you’ll feelbad or stupid? Why don’t you just lie downand forget about it?


[as “Little Barbara”] But I might like it. If Idon’t start trying the things that I think Imight enjoy, I’ll never know what I like orwhat makes me happy . . . If I keep listeningto you, I’ll do nothing but sleep away the restof my life. I want to have a life.

[as “the Escapist”] I don’t know if it’s worththe risk of feeling bad.

[as “Little Barbara”] Even if I feel bad atfirst, it will only be temporary. Either I’ll getgood at sewing or I can find something elsethat I’m good at. I’ll eventually find some-thing if I just keep at it.

[as “the Escapist”] Well OK, suit yourself.

[as “Little Barbara”] I will.

Part of the therapy at this point is focused onhelping Barbara discover what gives her an in-ner sense of joy and accomplishment. Thisquest is complicated by the fact that Barbaraspent so many of her early years dependentupon her mother, not developing any specialskills or talents, and compliantly surrenderingto what others wanted of her. At this point inher life, she is developmentally delayed in manyareas. In the previous segment, however,Barbara has begun to acknowledge and islearning how to tolerate better the short-termdiscomfort of exposing these delays to herselfand others to grow and feel more comfortablein the long term. The final step in therapy is totake these lessons into her everyday experi-ences.

Step 7: Generalize Results from Mode Workto Real Life

Barbara tried taking one sewing class and sur-prised herself by doing quite well. She then de-cided to take two more classes, a more ad-vanced sewing class and another class incostume design. Following these classes, shestarted to volunteer for a local theater group,working on costumes. Barbara received muchattention, praise, and appreciation from herteachers and theater friends. About 1½ yearsinto treatment, Barbara commented to hertherapist, “You know, I think this is the firsttime I have ever really felt good and proud ofmyself.”

As Barbara’s confidence and her network offriends grew, she became less accepting and lesstolerant of George’s neglectful and demeaningbehaviors. She also became increasingly an-

noyed with her mom’s tendencies to excuseGeorge’s behavior for the sake of “being finan-cially comfortable.” This was in stark contrastto Barbara’s unquestioning acceptance of herrelationship with George at the beginning oftreatment. As she began to see parallels be-tween how she felt with George and how shefelt as a child in her parents’ home, Barbara be-gan to question whether this was a good mar-riage for her. She recognized that many ofGeorge’s behaviors were harmful in a similarway that her parent’s behaviors had been harm-ful. Although Barbara continued to want toadopt a child, she began questioning George’sability to be a good parent. The following vi-gnette is from a session 2 years into Barbara’streatment.

BARBARA: I’m noticing that when Georgespeaks to me in such a condescending tone, Ifeel like a little girl with my father again. Idon’t want this. I want to be treated withlove and respect, and as an equal. It isn’tright for me to accept his behavior—beingignored and demeaned by someone who issupposed to love me.

THERAPIST: So what do you want to do?

BARBARA: I think I am ready to confrontGeorge about his drinking and abusivebehavior. I can’t even imagine adopting achild and bringing that child into our homeif he continues to behave this way.

THERAPIST: What do you want to happen?

BARBARA: I want him to stop drinking and starttreating me with love and respect.

THERAPIST: And if he doesn’t? Then what?

BARBARA: I guess I will leave him and seek a di-vorce.

THERAPIST: And how will that be for you—inthe worst case scenario?

BARBARA: In the worst case . . . you know, Ithink I will be OK. I mean, I have my ownlife now. I have my own friends. And I haveyou, if things really get rough.

THERAPIST: I’ll be here for you whenever youneed support. And financially?

BARBARA: Money is not everything. In anycase, there will be a divorce settlement. Yeah,I think I will be OK.

At this point in treatment, Barbara has fullyincorporated the Healthy Adult and has nur-


tured and is protecting the Vulnerable Childwithin. Along with these changes, Barbara’s de-pression went into complete remission. Soonafter this session, Barbara confronted her hus-band. Despite couple therapy, George refusedto make any changes, and Barbara filed for di-vorce.

Barbara did experience one relapse of majordepression as her divorce was finalized. “BadLittle Barbara” and “the Escapist” temporarilyreturned during this period. Barbara, however,was eventually able to empathize with and for-give herself, and to mourn the lost opportunityto have her own family. She also decided to goback to school to become a certified teacher atthe high school level, teaching costume designand drama in the theater department. She even-tually achieved this goal, and also met a singleand emotionally available man who shared herpassion for theater. Her chronic depressionlifted, and Barbara has been relatively symp-tom free since.

CONCLUSION

Evidence demonstrating the efficacy of cogni-tive therapy in the treatment of unipolar andbipolar depression continues to mount. Adoles-cents, adults, and geriatric patients have allbeen shown to benefit from cognitive therapy.

Cognitive therapy helps patients understandthe relationship among their thoughts, behav-iors, and feelings. Cognitions are “put to thetest” by examining evidence, setting up in vivoexperiments, weighing advantages and disad-vantages, trying graded tasks, and employingother intervention strategies. Through this pro-cess, patients begin to view themselves andtheir problems more realistically, to feel better,to change their maladaptive behavior patterns,and to take steps to solve real-life difficulties.These changes take place as a direct resultof carefully planned, self-help homeworkassignments—one of the hallmarks of cognitivetreatment. Cognitive therapy reduces symp-toms by helping patients to identify and modifyautomatic thoughts, and the behaviors associ-ated with them.

Another type of cognitive therapy—calledschema therapy—has been developed to dealwith the deeper psychological structures thatpredispose patients to chronic depression. Af-ter utilizing interventions aimed at initial symp-tom reduction, a great deal of subsequent at-

tention and effort are directed toward identify-ing and modifying the underlying schemas andschema modes that often predispose individu-als to chronic depression. Following a thor-ough assessment, an extensive change compo-nent is developed and implemented. Duringthis treatment, patients come to understandtheir own schemas and schema modes, their de-velopmental origins, and the way they are trig-gered, reinforced, and maintained.

Throughout the treatment, cognitive thera-pists maintain a collaborative alliance withtheir patients. They are very active in structur-ing sessions, yet go to considerable lengths tohelp patients reach conclusions on their own. Atherapist serves as a guide, helping a patientmaneuver through a labyrinth of dysfunctionalcognitions including EMSs and schema modes.As a result, patients attain the necessary psy-chological tools to become more proactive ontheir own behalf and are able to make the nec-essary cognitive, affective, interpersonal, andbehavioral changes necessary to minimize fur-ther episodes of depression.

NOTES

1. Researchers have found it useful to differentiate be-tween “relapse” (the return of symptoms within 6months after termination of treatment) and “recur-rence” (a whole new episode of depression, occur-ring at least 12 months after treatment has ended;Gelder, 1994; see also Overholser, 1998b). Howeverbecause this distinction has not been uniformly in-corporated into the literature (and for clarity of pre-sentation), we have elected to use the term “relapse”to indicate the return of symptoms, regardless of thetime frame.

2. Reported relapse rates for patients treated with med-ication have varied, depending on the definition of“relapse,” the duration of the follow-up period, andthe severity of depression within the patient popula-tion (Williams, 1997). Because of these differences,some of the estimates have ranged from 34 to 92%(Frank, 1996; Overholser, 1998b; Versiani, 1998;Williams, 1997), though lower rates have also beenreported (Keller & Boland, 1998).

3. The book by Young (1990/1999) cited here andthroughout the chapter, Cognitive Therapy for Per-sonality Disorders: A Schema-Focused Approach,refers to the third edition. However, the ideas ex-pressed were developed in 1990, for the original edi-tion. Similarly, the Young Schema Questionnairewas developed in 1990, then reprinted in the thirdedition.

4. In this chapter, diagnoses are generally based on theDiagnostic and Statistical Manual of Mental Disor-


ders, fourth edition (DSM-IV; American PsychiatricAssociation, 1994). However, diagnoses for patientsdescribed in the case materials are based on DSM-III-R (American Psychiatric Association, 1987), thesystem in effect at the time of assessment.

REFERENCES

Agosti, V., & Ocepek, W. K. (1997). The efficacy ofimipramine and psychotherapy in early-onset chronicdepression: A reanalysis of the National Institute ofMental Health Treatment of Depression Collabora-tive Research Program. Journal of Affective Disor-ders, 43(3), 181–186.

Albon, J. S., & Jones, E. E. (2003). Validity of con-trolled clinical trials of psychotherapy: Findings fromthe NIMH Treatment of Depression CollaborativeResearch Program. American Journal of Psychiatry,159(5), 775–783.

American Psychiatric Association. (1987). Diagnosticand statistical manual of mental disorders (3rd ed.,rev.). Washington, DC: Author.


American Psychiatric Association. (2000). Practiceguideline for the treatment of patients with major de-pressive disorder (revision). American Journal of Psy-chiatry, 157(Suppl. 4).

Antonuccio, D. O., Danton, W. G., & DeNelsky, G. Y.(1995). Psychotherapy versus medication for depres-sion: Challenging the conventional wisdom withdata. Professional Psychology: Research and Prac-tice, 26(6), 574–585.

Antonuccio, D. O., Thomas, M., & Danton, W. G.(1997). A cost-effectiveness analysis of cognitivebehavior therapy and fluoxetine (Prozac) in the treat-ment of depression. Behavior Therapy, 28, 187–210.

Arnow, B. A., Manber, R., Blasey, C., Klein, D. N.,Blalock, J. A., Markowitz, J. C., et al. (2003). Thera-peutic reactance as a predictor of outcome in thetreatment of chronic depression. Journal of Con-sulting and Clinical Psychology, 71(6), 1025–1035.

Ball, J. R., Mitchell, P. B., Corry, J. C., Skillecorn, A.,Smith, M., & Malhi, G. S. (2006). A randomizedcontrolled trial of cognitive therapy for bipolar disor-der: Focus on long-term change. Journal of ClinicalPsychiatry, 67, 277–286.

Bandura, A. (1977). Self-efficacy: Toward a unifyingtheory of behavioral change. Psychological Review,84, 181–215.

Barber, J. P., & DeRubeis, R. J. (1989). On secondthought: Where the action is in cognitive therapy fordepression. Cognitive Therapy and Research, 13(5),441–457.

Barlow, D. H., & Hofmann, S. G. (1997). Efficacy anddissemination of psychological treatments. In D. M.Clark & C. G. Fairburn (Eds.), Science and practice

of cognitive behaviour therapy (pp. 95–117). Ox-ford, UK: Oxford University Press.

Basco, M. R., & Rush, A. J. (1996). Cognitive-behavioral therapy for bipolar disorder. New York:Guilford Press.

Beck, A. T. (1967). Depression: Causes and treatment.Philadelphia: University of Pennsylvania Press.

Beck, A. T. (1976). Cognitive therapy and the emotionaldisorders. New York: International UniversitiesPress.

Beck, A. T. (1988). Love is never enough. New York:Harper & Row.

Beck, A. T., Freeman, A., & Associates. (1990). Cogni-tive therapy of personality disorders. New York:Guilford Press.


Beck, J. S. (1995). Cognitive therapy: Basics and be-yond. New York: Guilford Press.

Berndt, E. R., Koran, L. M., Finkelstein, S. N.,Gelenberg, A. J., Kornstein, S. G., Miller, I. M., et al.(2000). Lost human capital from early-onset chronicdepression. American Journal of Psychiatry, 157,940–947.

Beutler, L. E., Scogin, F., Kirkish, P., Schretlen, D.,Corbishley, A., Hamblin, D., et al. (1987). Groupcognitive therapy and alprazolam in the treatment ofdepression in older adults. Journal of Consulting andClinical Psychology, 55(4), 550–556.

Biggs, M. M., & Rush, A. J. (1999). Cognitive andbehavioral therapies alone or combined with antide-pressant medication in the treatment of depression.In D. S. Janowsky (Ed.), Psychotherapy indicationsand outcomes (pp.121–172). Washington, DC:American Psychiatric Press.

Blackburn, I. M., Bishop, S., Glen, A. I. M., Whalley, L.J., & Christie, J. E. (1981). The efficacy of cognitivetherapy in depression: A treatment trial using cogni-tive therapy and pharmacotherapy, each alone and incombination. British Journal of Psychiatry, 139,181–189.

Blackburn, I. M., & Moore, R. G. (1997). Controlledacute and follow-up trial of cognitive therapy andpharmacotherapy in out-patients with recurrent de-pression. British Journal of Psychiatry, 171, 328–334.

Blatt, S. J., Zuroff, D. C., Bondi, C. M., & Sanislow, C.A. (2000). Short- and long-term effects of medicationand psychotherapy in the brief treatment of depres-sion: Further analysis of data from the NIMHTDCRP. Psychotherapy Research, 10(2), 215–234.

Bledsoe, S. E., & Grote, N. K. (2006). Treating depres-sion during pregnancy and the postpartum: A prelim-inary meta-analysis. Research on Social Work Prac-tice, 16(2), 109–120.

Bockting, C. L. H., Schene, A. H., Spinhoven, P., Koeter,M. W. J., Wouters, L. F., Huyser, J., et al. (2005). Pre-venting relapse/recurrence in recurrent depression


with cognitive therapy: A randomized controlledtrial. Journal of Consulting and Clinical Psychology,73, 647–657.

Bowers, W. A. (1990). Treatment of depressed in-patients: Cognitive therapy plus medication, andmedication alone. British Journal of Psychiatry, 156,73–58.

Brown, G. W. (1996). Onset and course of depressivedisorders: Summary of a research programme. In C.Mundt, M. J. Goldstein, K. Hahlweg, & P. Fiedler(Eds.), Interpersonal factors in the origin and courseof affective disorders (pp. 151–167). London:Gaskell/Royal College of Psychiatrists.

Burns, D. D., & Spangler, D. L. (2000). Does psycho-therapy homework lead to improvements in depres-sion in cognitive-behavioral therapy or does im-provement lead to increased homework compliance?Journal of Consulting and Clinical Psychology,68(1), 46–56.

Butler, A. C., Chapman, J. E., Forman, E. M., & Beck,A. T. (2006). The empirical status of cognitive-behavioral therapy: A review of meta-analyses. Clini-cal Psychology Review, 26, 17–31.

Castonguay, L. G., Goldfried, M. R., Wiser, S., Raue, P.J., & Hayes, A. M. (1996). Predicting the effect ofcognitive therapy for depression: A study of uniqueand common factors. Journal of Consulting andClinical Psychology, 64(3), 497–504.

Castonguay, L. G., Schut, A. J., Aikens, D.,Constantino, M. J., Laurenceau, J. P., Bolough, L., etal. (2004). Integrative cognitive therapy: A prelimi-nary investigation. Journal of Psychotherapy Integra-tion, 14(1), 4–20.

Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S.R., Edwards, V. J., & Anda, R. F. (2004). Adversechildhood experiences and the risk of depressive dis-orders in adulthood. Journal of Affective Disorders,82(2), 217–225.

Colom, F., Vieta, E., Martinez, A., Jorquera, A., &Gastó, C. (1998). What is the role of psychotherapyin the treatment of bipolar disorder? Psychotherapyand Psychosomatics, 67, 3–9.

Conte, H. R., Plutchik, R., Wild, K. V., & Karasu, T. B.(1986). Combined psychotherapy and pharmaco-therapy for depression. Archives of General Psychia-try, 43, 471–479.

Craighead, W. E., Miklovitz, D. J., Vajk, F. C., & Frank,E. (1998). Psychological treatments for bipolar disor-der. In P. E. Nathan & J. M. Gorman (Eds.), A guideto treatments that work (pp. 240–248). New York:Oxford University Press.

Crews, W. D., Jr., & Harrison, D. W. (1995). The neuro-psychology of depression and its implications forcognitive therapy. Neuropsychology Review, 5(2),81–123.

Curry, J. F. (2001). Specific psychotherapies for child-hood and adolescent depression. Biological Psychia-try, 49, 1091–1100.

de Oliveira, I. R. (1998). The treatment of unipolar ma-

jor depression: Pharmacotherapy, cognitive behav-iour therapy or both? Journal of Clinical Pharmacyand Therapeutics, 23, 467–475.

Depression Guideline Panel. (1993). Depression is atreatable illness: A patient’s guide (AHCPR Publica-tion No. 93-0553). Rockville, MD: U.S. Departmentof Health and Human Services, Public Health Ser-vice, Agency for Health Care Policy and Research.

DeRubeis, R. J., & Feeley, M. (1990). Determinants ofchange in cognitive therapy for depression. CognitiveTherapy and Research, 14(5), 469–482.

DeRubeis, R. J., Gelfand, L. A., Tang, T. Z., & Simons,A. D. (1999). Medications versus cognitive behaviortherapy for severely depressed outpatients: Mega-analysis of four randomized comparisons. AmericanJournal of Psychiatry, 156(7), 1007–1013.

DeRubeis, R. J., Hollon, S. D., Amsterdam, J. D.,Shelton, R. C., Young, P. R., Salomon, R. M., et al.(2005). Cognitive therapy vs. medications in thetreatment of moderate to severe depression. Archivesof General Psychiatry, 62, 409–416.

DeRubeis, R. J., Hollon, S. D., Grove, W. M., Evans, M.D., Garvey, M. J., & Tuason, V. B. (1990). How doescognitive therapy work?: Cognitive change andsymptom change in cognitive therapy andpharmacotherapy for depression. Journal of Con-sulting and Clinical Psychology, 58(6), 862–869.

Dimidjian, S., Hollon, S. D., Dobson, K. S., Schmaling,K. B., Kohlenberg, R. J., Addis, M. E., et al. (2006).Randomized trial of behavioral activation, cognitivetherapy, and antidepressant medication in the acutetreatment of adults with major depression. Journal ofConsulting and Clinical Psychology, 74(4), 658–670.

Dobson, K. S., & Pusch, D. (1993). Towards a defini-tion of the conceptual and empirical boundaries ofcognitive therapy. Australian Psychologist, 28(3),137–144.

Dougherty, L. R., Klein, D. N., & Davila, J. (2004). Agrowth curve analysis of the course of dysthymic dis-order: The effects of chronic stress and moderationby adverse parent-child relationships and family his-tory. Journal of Consulting and Clinical Psychology,72(6), 1012–1021.

Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P.,Williamson, D. F., & Giles, W. H. (2001). Childhoodabuse, household dysfunction, and the risk of at-tempted suicide throughout the life span: Findingsfrom the adverse childhood experiences study. Jour-nal of the American Medical Association, 286, 3089–3096.

Eifert, G. H., Beach, B. K., & Wilson, P. H. (1998). De-pression: Behavioral principles and implications fortreatment and relapse prevention. In J. J. Plaud & G.H. Eifert (Eds.), From behavior theory to behaviortherapy (pp. 68–97). Boston: Allyn & Bacon.

Elkin, I., Gibbons, R. D., Shea, M. T., & Shaw, B. F.(1996). Science is not a trial (but it can sometimes bea tribulation). Journal of Consulting and ClinicalPsychology, 64(1), 92–103.


Elkin, I., Shea, M. T., Watkins, J. T., Imber, S. D., Sotsky,S. M., Collins, J. F., et al. (1989). National Instituteof Mental Health Treatment of Depression Collabo-rative Research Program: General effectiveness oftreatments. Archives of General Psychiatry, 46, 971–982.

Evans, M. D., Hollon, S. D., DeRubeis, R. J., Piaseki, J.M., Grove, W. M., Garvey, M. J., et al. (1992). Dif-ferential relapse following cognitive therapy andpharmacotherapy for depression. In D. S. Janowsky(Ed.), Psychotherapy indication and outcomes (pp.802–808). Washington, DC: American PsychiatricPress.

Fava, G. A., Rafanelli, C., Grandi, S., Canestrari, R., &Morphy, M. A. (1998a). Six-year outcome for cogni-tive behavioral treatment of residual symptoms inmajor depression. American Journal of Psychiatry,155(10), 1443–1445.

Fava, G. A., Rafanelli, C., Grandi, S., Conti, S., &Belluardo, P. (1998b). Prevention of recurrent de-pression with cognitive behavioral therapy: Prelimi-nary findings. Archives of General Psychiatry, 55(9),816–820.

Floyd, M., Scogin, F., McKendree-Smith, N. L., Floyd,D. L., & Rokke, P. D. (2004). Cognitive therapy fordepression: A comparison of individual psychothera-py and bibliotherapy for depressed older adults.Behavior Modification, 28(2), 297–318.

Frank, E. (1996). Long-term treatment of depression:interpersonal psychotherapy with and without medi-cation. In C. Mundt & M. J. Goldstein (Eds.), Inter-personal factors in the origin and course of affectivedisorders (pp. 303–315). London: Gaskell/RoyalCollege of Psychiatrists.

Frank E., & Thase, M. E. (1999). Natural history andpreventative treatment of recurrent mood disorders.Annual Review of Medicine, 50, 453–468.

Free, M. L., Oei, T. P. S., & Appleton, C. (1998). Biolog-ical and psychological processes in recovery from de-pression during cognitive therapy. Journal of Behav-ior Therapy and Experimental Psychiatry, 29, 213–226.

Freeman, A., & Davison, M. R. (1997). Short-termtherapy for the long-term patient. In L. Vandecreek,S. Knapp, & T. L. Jackson (Eds.), Innovations in clin-ical practice: A source book (pp. 5–24). Sarasota, FL:Professional Resource Press.

Futterman, A., Thompson, L., Gallagher-Thompson,D., & Ferris, R. (1995). Depression in later life: Epi-demiology, assessment, etiology, and treatment. In E.E. Beckham & W. R. Leber (Eds.), Handbook of de-pression (2nd ed., pp. 494–525). New York: GuilfordPress.

Gaffan, E. A., Tsaousis, J., & Kemp-Wheeler, S. M.(1995). Researcher allegiance and meta-analysis: Thecase of cognitive therapy for depression. Journal ofConsulting and Clinical Psychology, 63(6), 966–980.

Gallagher-Thompson, D., Hanley-Peterson, P., &Thompson, L. W. (1990). Maintenance of gains ver-sus relapse following brief psychotherapy for depres-

sion. Journal of Consulting and Clinical Psychology,58(3), 371–374.

Geddes, J. R., Carney, S. M., Davies, C., Furukawa, T.A., Kupfer, D. J., Frank, E., et al. (2003). Relapse pre-vention with antidepressant drug treatment in de-pressive disorders: A systematic review. Lancet, 361,653–661.

Gelder, M. G. (1994). Cognitive therapy for depression.In H. Hippius & C. N. Stefanis (Eds.), Research inmood disorders: An update (Vol. 1, pp. 115–124).Goettingen, Germany: Hogrefe & Huber.

Giesen-Bloo, J., van Dyck, R., Spinhoven, P., vanTilburg, W., Dirksen, C., van Asselt, T., et al. (2006).Outpatient psychotherapy for borderline personalitydisorder: A randomized trial of schema-focused ther-apy vs. transference-focused therapy. Archives ofGeneral Psychiatry, 63, 649–658.

Gloaguen, V., Cottraux, J., Cucherat, M., & Blackburn,I. M. (1998). A meta-analysis of the effects of cogni-tive therapy in depressed patients. Journal of Affec-tive Disorders, 49, 59–72.

Gortner, E. T., Gollan, J. K., Dobson, K. S., & Jacobson,N. S. (1998). Cognitive-behavioral treatment for de-pression: Relapse prevention. Journal of Consultingand Clinical Psychology, 2, 377–384.

Hall, L. H., & Robertson, M. H. (1998). Undergraduateratings of the acceptability of single and combinedtreatments for depression: A comparative analysis.Professional Psychology: Research and Practice, 3,269–272.

Harrington, R., Wood, A., & Verduyn, C. (1998).Clinically depressed adolescents. In P. J. Graham(Ed.), Cognitive-behaviour therapy for children andfamilies (pp. 156–193). New York: Cambridge Uni-versity Press.

Hayden, E. P., & Klein, D. N. (2001). Predicting out-come of dysthymic disorder at a 5-year follow-up:The impact of familial psychopathology, early adver-sity, personality, comorbidity, and chronic stress.American Journal of Psychiatry, 158, 1864–1870.

Hayes, A. M., & Strauss, J. L. (1998). Dynamic systemstheory as a paradigm for the study of change in psy-chotherapy: An application to cognitive therapy fordepression. Journal of Consulting and Clinical Psy-chology, 66(6), 939–947.

Heim, C., & Nemeroff, C. B. (2001). The role of child-hood trauma in the neurobiology of mood and anxi-ety disorders: Preclinical and clinical studies. Biologi-cal Psychiatry, 49, 1023–1039.

Henry, W. P., Strupp, H. H., Butler, S. F., Schacht, T. E.,& Binder, J. L. (1993). The effects of training in time-limited dynamic psychotherapy: Changes in therapistbehavior. Journal of Consulting and Clinical Psychol-ogy, 61, 434–440.

Hollon, S. D. (1998). What is cognitive behaviouraltherapy and does it work? Current Opinion in Neu-robiology, 8, 289–292.

Hollon, S. D., DeRubeis, R. J., & Evans, M. D. (1996).Cognitive therapy in the treatment and prevention ofdepression. In P. M. Salkovskis (Ed.), Frontiers of


cognitive therapy (pp. 293–317). New York:Guilford Press.

Hollon, S. D., DeRubeis, R. J., Evans, M. D., Weimer,M. J., Garvey, M. J., Grove, W. M., et al. (1992).Cognitive therapy and pharmacotherapy: Singly andin combination. Archives of General Psychiatry, 49,774–781.

Hollon, S. D., DeRubeis, R. J., Shelton, R. C., Am-sterdam, J. D., Salomon, R. M., O’Reardon, J. P.,et al. (2005). Prevention of relapse following cogni-tive therapy vs. medications in moderate to severedepression. Archives of General Psychiatry, 62,417–422.

Hollon, S. D., & Shelton, R. C. (2001). Treatmentguidelines for major depressive disorder. BehaviorTherapy, 32, 235–258.

Hollon, S. D., Shelton, R. C., & Loosen, P. T. (1991).Cognitive therapy and pharmacotherapy for depres-sion. Journal of Consulting and Clinical Psychology,59(1), 88–99.

Hollon, S. D., Thase, M. E., & Markowitz, J. C. (2002).Treatment and prevention of depression. Psychologi-cal Science in the Public Interest, 3, 39–77.

Howland, R. H. (1993). Chronic depression. Hospitaland Community Psychiatry, 44, 633–639.

Ingram, R. E., & Holle, C. (1992). Cognitive science ofdepression. In D. J. Stein & J. E. Young (Eds.), Cog-nitive science and clinical disorders (pp. 187–209).San Diego, CA: Academic Press.

Jacobson, N. S., Dobson, K. S., Truax, P. A., Addis, M.E., Koerner, K., Gollan, J. K., et al. (1996). A compo-nent analysis of cognitive-behavioral treatment fordepression. Journal of Consulting and Clinical Psy-chology, 64, 295–304.

Jacobson, N. S., & Hollon, S. D. (1996). Cognitive-behavior therapy versus pharmacotherapy: Now thatthe jury’s returned its verdict, it’s time to present therest of the evidence. Journal of Consulting and Clini-cal Psychology, 64(1), 74–80.

Jarrett, R. B. (1995). Comparing and combining short-term psychotherapy and pharmacotherapy for de-pression. In E. E. Beckham & W. R. Leber (Eds.),Handbook of depression (2nd ed., pp. 435–464).New York: Guilford Press.

Jarrett, R. B., Basco, M. R., Risser, R., Ramanan, J.,Marwill, M., Kraft, D., et al. (1998). Is there a rolefor continuation phase cognitive therapy for de-pressed outpatients? Journal of Consulting and Clini-cal Psychology, 66(6), 1036–1040.

Jarrett, R. B., Kraft, D., Doyle, J., Foster, B. M., Eaves,G. G., & Silver, P. C. (2001). Preventing recurrent de-pression using cognitive therapy with and withoutcontinuation phase. Archives of General Psychiatry,58(4), 381–388.

Jarrett, R. B., & Nelson, R. O. (1987). Mechanisms ofchange in cognitive therapy of depression. BehaviorTherapy, 18, 227–241.

Jarrett, R. B., Schaffer, M., McIntire, D., Witt-Browder,A., Kraft, D., & Risser, R. C. (1999). Treatment ofatypical depression with cognitive therapy or

phenelzine: A double-blind, placebo-controlled trial.Archives of General Psychiatry, 56, 431–437.

Joffe, R., Segal, Z., & Singer, W. (1996). Change in thy-roid hormone levels following response to cognitivetherapy for major depression. American Journal ofPsychiatry, 153(3), 411–413.

Johnsson, B., & Bebbington, P. E. (1994). What pricedepression?: The cost of depression and the cost ef-fectiveness of pharmacological treatment. BritishJournal of Psychiatry, 164, 665–673.

Keller, M. B., & Boland, R. J. (1998). Implications offailing to achieve successful long-term maintenancetreatment of recurrent unipolar major depression. Bi-ological Psychiatry, 44(5), 348–360.

Keller, M. B., & Hanks, D. L. (1995). Course and natu-ral history of chronic depression. In J. H. Kocsis &D. N. Klein (Eds.), Diagnosis and treatment ofchronic depression (pp. 58–72). New York: GuilfordPress.

Keller, M. B., McCullough, J. P., Klein, D. F., Arnow, B.,Dunner, D. L., Gelenberg, A. J., et al. (2000). A com-parison of nefazodone, the cognitive behavioral-analysis system of psychotherapy, and their combina-tion for the treatment of chronic depression. NewEngland Journal of Medicine, 342, 1462–1470.

Kendler, K. S., Kessler, R. C., Walters, E. E., MacLean,C., Neale, M. C., Heath, A. C., et al. (1995). Stressfullife events, genetic liability, and onset of an episode ofmajor depression in women. American Journal ofPsychiatry, 152, 833–842.

Kessler, R. C., Akiskal, H. S., Ames, M., Birnbaum, H.,Greenberg, P. A., Jin, R., et al. (2006). Prevalence andeffects of mood disorders on work performance in anationally representative sample of U.S. workers.American Journal of Psychiatry, 163(9), 1561–1568.

Kessler, R. C., Berglund, P., Demler, O., Jin, R.,Merikangas, K. R., & Walters, E. E. (2005). Lifetimeprevalence and age-of-onset distributions of DSM-IVdisorders in the National Comorbidity Survey Repli-cation. Archives of General Psychiatry, 62(6), 593–602.

Kessler, R. C., Chui, W. T., Demler, O., Merikangas, K.R., & Walters, E. E. (2005). Prevalence, severity, andcomorbidity of 12-month DSM-IV disorders in theNational Comorbidity Survey Replication. Archivesof General Psychiatry, 62(6), 617–627.

Klein, D. F. (1996). Preventing hung juries about ther-apy studies. Journal of Consulting and Clinical Psy-chology, 64, 81–87.

Klein, D. N., Santiago, N. J., Vivian, D., Blalock, J. A.,Kocsis, J. H., Markowitz, J. C., et al. (2004).Cognitive-behavioral analysis system of psychothera-py as a maintenance treatment for chronic depres-sion. Journal of Consulting and Clinical Psychology,72, 681–688.

Koder, D. A., Brodaty, H., & Anstey, K. J. (1996). Cog-nitive therapy for depression in elderly. InternationalJournal of Geriatric Psychiatry, 11(2), 97–107.

Kupfer, D. J., Frank, E., & Wamhoff, J. (1996). Mooddisorders: Update on prevention of recurrence. In C.


Mundt, M. M. Goldstein, K. Hahlweg, & P. Fiedler(Eds.), Interpersonal factors in the origin and courseof affective disorders (pp. 289–302). London:Gaskell.

Lam, D. H., Hayward, P., Watkins, E. R., Wright, K., &Sham, P. (2005). Relapse prevention in patients withbipolar disorder: Cognitive therapy outcome after 2years. American Journal of Psychiatry, 162(2), 324–329.

Lazarus, A. A., & Lazarus, C. N. (1991). MultimodalLife History Inventory (2nd ed.). Champaign, IL: Re-search Press.

Lee, C. W., Taylor, G., & Dunn, J. (1999). Factor struc-ture of the Schema Questionnaire in a large clinicalsample. Cognitive Therapy and Research, 23(4),441–451.

Levendusky, P. G., & Hufford, M. R. (1997). The appli-cation of cognitive-behavior therapy to the treatmentof depression and related disorders in the elderly.Journal of Geriatric Psychiatry, 30(2), 227–238.

Lewinsohn, P. M., & Clarke, G. N. (1999). Psychosocialtreatments for adolescent depression. Clinical Psy-chology Review, 19(3), 329–342.

Linehan, M. M. (1993a). Cognitive-behavioral treat-ment of borderline personality disorder. New York:Guilford Press.

Linehan, M. M. (1993b). Skills training manual fortreating borderline personality disorder. New York:Guilford Press.

Lizardi, H., Klein, D. N., Ouimette, P. C., Riso, L. P.,Anderson, R. L., & Donaldson, S. K. (1995). Reportsof the childhood home environment in early onsetdysthymia and episodic major depression. Journal ofAbnormal Psychology, 104, 132–139.

Luborsky, L., Diguer, L., Seligman, D. A., Rosenthal, R.,Krause, E. D., Johnson, S., et al. (1999). The re-searcher’s own therapy allegiances: A “wild child” incomparisons to treatment efficacy. Clinical Psychol-ogy: Science and Practice, 6(1), 95–106.

Ma, S. H., & Teasdale, J. D. (2004). Mindfulness-basedcognitive therapy for depression: Replication and ex-ploration of differential relapse prevention effects.Journal of Consulting and Clinical Psychology,72(1), 31–40.

Martell, C. R., Addis, M. E., & Jacobson, N. S. (2001).Depression in context: Strategies for guided action.New York: Norton.

McCullough, J. P., Jr. (2000). Treatment for chronic de-pression: Cognitive behavioral analysis system ofpsychotherapy (CBASP). New York: Guilford Press.

McCullough, J. P., Jr. (2003). Treatment for chronic de-pression: Cognitive behavioral-analysis system ofpsychotherapy. Journal of Psychotherapy Integra-tion, 13(3/4), 241–263.

McGinn, L. K., & Young, J. E. (1996). Schema-focusedtherapy. In P. M. Salkovskis (Ed.), Frontiers of cogni-tive therapy (pp. 182–207). New York: GuilfordPress.

Mental health: A report of the Surgeon General. (1997,

September 30). Available at www.surgeongeneral.gov/library/mentalhealth.

Meterissian, G. B., & Bradwejn, J. (1989). Comparativestudies on the efficacy of psychotherapy,pharmacotherapy, and their combination in depres-sion: Was adequate pharmacotherapy provided?Journal of Clinical Psychopharmacology, 9, 334–339.

Miller, I. W., Norman, W. H., Keitner, G. I., Bishop, S.,& Dow, M. G. (1989). Cognitive-behavioral treat-ment of depressed inpatients. Behavior Therapy, 20,25–47.

Mischoulon, D., Fava, M., & Rosenbaum, J.F. (1999).Strategies for augmentation of SSRI treatment: A sur-vey of an academic psycopharmacology practice.Harvard Review of Psychiatry, 6(6), 322–326.

Mohr, D. C. (1995). Negative outcome in psychothera-py: A critical review. Clinical Psychology: Scienceand Practice, 2, 1–27.

Murphy, G. E., Simmons, A. D., Wetzel, R. D., &Lustman, P. J. (1984). Cognitive therapy andpharmacotherapy, singly and together, in the treat-ment of depression. Archives of General Psychiatry,41, 33–41.

National Institute of Mental Health. (2006). The num-bers count: Mental disorders in America (PublicationNo. 06-4584). Retrieved April 6, 2007, fromwww.nimh.nih.gov/publicat/numbers

Nierenberg, A. A., Keefe, B. R., Leslie, V. C., Alpert, J.E., Pava, J. A., Worthington, J. J., III, et al. (1999).Residual symptoms in depressed patients who re-spond acutely to fluoxetine. Journal of Clinical Psy-chiatry, 60(4), 221–225.

Oei, T. P. S., & Free, M. L. (1995). Do cognitive behav-iour therapies validate cognitive models of mood dis-orders?: A review of the empirical evidence. Interna-tional Journal of Psychology, 30(2), 145–180.

Oei, T. P. S., & Shuttlewood, G. J. (1996). Specific andnonspecific factors in psychotherapy: A case of cog-nitive therapy for depression. Clinical PsychologyReview, 16(2), 83–103.

O’Leary, K. D., & Beach, S. R. H. (1990). Marital ther-apy: A viable treatment for depression and maritaldiscord. American Journal of Psychiatry, 147(2),183–186.

Olfson, M., Marcus, S. C., Druss, B., Elinson, L.,Tanielian, T., & Pincus, H. A. (2002). Nationaltrends in the outpatient treatment of depression.Journal of the American Medical Association, 287,203–209.

Overholser, J. C. (1998). Cognitive-behavioral treat-ment of depression, part X: Reducing the risk of re-lapse. Journal of Contemporary Psychotherapy,28(4), 381–396.

Padesky, C. A., with Greenberger, D. (1995). A clini-cian’s guide to mind over mood. New York: GuilfordPress.

Parker, K., Roy, K., & Eyers, K. (2003). Cognitivebehavior therapy for depression?: Choose horses for


courses. American Journal of Psychiatry, 160, 825–834.

Paykel, E. S., Scott, J., Teasdale, J. D., Johnson, A. L.,Garland, A., Moore, R., et al. (1999). Prevention ofrelapse in residual depression by cognitive therapy.Archives of General Psychiatry, 56, 829–835.

Persons, J. B., Burns, D. D., & Perloff, J. M. (1988). Pre-dictors of dropout and outcome in cognitive therapyfor depression in a private practice setting. CognitiveTherapy and Research, 12(6), 557–575.

Piper, W. E., Ogrodniczuk, J. S., Joyce, A. S., McCallum,M., Rosie, J. S., O’Kelly, J. G., et al. (1999). Predic-tion of dropping out in time-limited, interpretive psy-chotherapy. Psychotherapy, 36, 114–122.

Randolph, J. J., & Dykman, B. M. (1998). Perceptionsof parenting and depression-proneness in the off-spring: Dysfunctional attitudes as a mediating mech-anism. Cognitive Depressive Disorders, 21, 401–449.

Rehm, L. P. (1990). Cognitive and behavioral theories.In B. B. Wolman & G. Stricker (Eds.), Depressive dis-orders: Facts, theories, and treatment methods(pp. 64–91). New York: Wiley.

Reinecke, M. A., Ryan, N. E., & DuBois, D. L. (1998).Cognitive-behavioral therapy of depression and de-pressive symptoms during adolescence: A review andmeta-analysis. Journal of the American Academy ofChild and Adolescent Psychiatry, 37(1), 26–34.

Riso, L. P., du Toit, P. L., Blandino, J. A., Penna, S.,Dacey, S., Duin, J. S., et al. (2003). Cognitive aspectsof chronic depression. Journal of Abnormal Psychol-ogy, 112(1), 72–80.

Roberts, J. E., & Hartlage, S. (1996). Cognitive rehabil-itation interventions for depressed patients. In P. W.Corrigan & S. C. Yudofsky (Eds.), Cognitive rehabil-itation for neuropsychiatric disorders (pp. 371–392).Washington, DC: American Psychiatric Press.

Robinson, L. A., Berman, J. S., & Neimeyer, R. A.(1990). Psychotherapy for the treatment of depres-sion: A comprehensive review of controlled outcomeresearch. Psychological Bulletin, 108, 30–49.

Rush, A. J., Beck, A. T., Kovacs, M., & Hollon, S.(1977). Comparative efficacy of cognitive therapyand imipramine in the treatment of depressed outpa-tients. Cognitive Therapy and Research, 1, 17–37.

Sacco, W. P., & Beck, A. T. (1995). Cognitive theory andtherapy. In E. E. Beckham & W. R. Leber, Handbookof depression (2nd ed., pp. 329–351). New York:Guilford Press.

Sachs-Ericsson, N., Verona, E., Joiner, T., & Preacher,K. J. (2006). Parental abuse and mediating role ofself-criticism in adult internalizing disorders. Journalof Affective Disorders, 93(1–3), 71–78.

Saeman, H. (1999). Psychotherapy gets short shrift atWhite House MH meeting. National Psychologist,8(4), 21.

Safran, J. D., Muran, J. C., Samstag, L. W., & Stevens,C. (2002). Repairing alliance ruptures. In J. C.Norcross (Ed.), Psychotherapy relationships that

work: Therapists contributions and responsiveness topatients (pp. 235–254). New York: Oxford Univer-sity Press.

Safran, J. D., & Segal, Z. V. (1990). Interpersonal pro-cesses in cognitive therapy. New York: Basic Books.

Safran, J. D., Segal, Z. V., Vallis, T. M., Shaw, B. F., &Samstag, L. W. (1993). Assessing patient suitabilityfor short-term cognitive therapy with an interper-sonal focus. Cognitive Therapy and Research, 17(1),23–38.

Salkovskis, P. M. (1996). Frontiers of cognitive therapy:The state of the art and beyond. New York: GuilfordPress.

Schmidt, N. B. (1994). The schema questionnaire andthe schema avoidance questionnaire. Behavior Ther-apist, 17(4), 90–92.

Schmidt, N. B., Joiner, T. E., Young, J. E., & Telch, M. J.(1995). The Schema Questionnaire: Investigation ofpsychometric properties and the hierarchical struc-ture of a measure of maladaptive schemata. Cogni-tive Therapy and Research, 19(3), 295–321.

Scott, A. I., & Freeman, C. P. (1992). Edinburgh pri-mary care depression study: Treatment outcome, pa-tient satisfaction, and cost after 16 weeks. BritishMedical Journal, 304, 883–887.

Scott, J. (1996a). Cognitive therapy of affective disor-ders: A review. Journal of Affective Disorders, 37, 1–11.

Scott, J. (1996b). The role of cognitive behaviour ther-apy in bipolar disorders. Behavioural and CognitivePsychotherapy, 24(3), 195–208.

Scott, J. (2000). New evidence in the treatment ofchronic depression. New England Journal of Medi-cine, 342, 1518–1520.

Scott, J., Palmer, S., Paykel, E., Teasdale, J., &Hayhurst, H. (2003). Use of cognitive therapy for re-lapse prevention in chronic depression. British Jour-nal of Psychiatry, 182, 221–227.

Shapiro, D. A., Barkman, M., Rees, A., Hardy, G. E.,Reynolds, S., & Startup, M. (1994). Effects of treat-ment duration and severity of depression on the ef-fectiveness of cognitive-behavioral and psychodynamic-interpersonal psychotherapy. Journal of Consultingand Clinical Psychology, 62, 522–534.

Shaw, B. F., & Segal, Z. V. (1999). Efficacy, indications,and mechanisms of action of cognitive therapy of de-pression. In D. S. Janowsky (Ed.), Psychotherapy in-dications and outcomes (pp. 173–196). Washington,DC: American Psychiatric Press.

Shea, M. T., & Elkin, I. (1996). The NIMH Treatmentof Depression Collaborative Research Program. In C.Mundt, M. J. Goldstein, K. Hahlweg, & P. Fiedler(Eds.), Interpersonal factors in the origin and courseof affective disorders (pp. 316–328). London:Gaskell/Royal College of Psychiatrists.

Shea, M. T., Elkin, I., Imber, S. D., Sotsky, S. M.,Watkins, J. T., Collins, J. F., et al. (1992). Course ofdepressive symptoms over follow-up: Findings fromthe National Institute of Mental Health Treatment of


Depression Collaborative Research Program. Ar-chives of General Psychiatry, 49(10), 782–787.

Shea, M. T., Pilkonis, P. A., Beckham, E., Collins, J. F.,Elkin, I., Sotsky, S. M., et al. (1990). Personality dis-orders and treatment outcome in the NIMH Treat-ment of Depression Collaborative Research Program.American Journal of Psychiatry, 147(6), 711–718.

Simons, A. D., Murphy, G. D., Levine, J. L., & Wetzel,R. D. (1986). Cognitive therapy and pharmaco-therapy for depression: Sustained improvement over1 year. Archives of General Psychiatry, 43, 43–48.

Spanier, C. A., Frank, E., McEachran, A. B.,Grochocinski, V. J., & Kupfer, D. J. (1999). Mainte-nance interpersonal psychotherapy for recurrent de-pression: Biological and clinical correlates and futuredirections. In D. S. Janowsky (Ed.), Psychotherapyindications and outcomes (pp. 249–273). Washing-ton, DC: American Psychiatric Press.

Stein, D. J., & Young, J. E. (1992). Schema approach topersonality disorders. In D. J. Stein & J. E. Young(Eds.), Cognitive science and clinical disorders(pp. 271–288). San Diego, CA: Academic Press.

Stewart, J. W., Garfinkel, R., Nunes, E. V., Donovan, S.,& Klein, D. F. (1998). Atypical features and treat-ment response in the National Institute of MentalHealth Treatment of Depression Collaborative Re-search Program. Journal of Clinical Psychopharma-cology, 18(6), 429–434.

Stuart, S., & Bowers, W. A. (1995). Cognitive therapywith inpatients: Review and meta-analysis. Journalof Cognitive Psychotherapy, 9(2), 85–92.

Sullivan, M. J. L., & Conway, M. (1991). Dysphoriaand valence of attributions for others’ behavior. Cog-nitive Therapy and Research, 15(4), 273–282.

Teasdale, J. D. (1997a). Assessing cognitive mediationof relapse prevention in recurrent mood disorders.Clinical Psychology and Psychotherapy, 4, 145–156.

Teasdale, J. D. (1997b). The relationship between cogni-tion and emotion: The mind-in-place in mood disor-ders. In D. M. Clark & C. G. Fairburn (Eds.), Scienceand practice of cognitive behavior therapy (pp. 67–93). Oxford, UK: Oxford University Press.

Teasdale, J. D., Moore, R. G., Hayhurst, H., Pope, M.,Williams, S., & Segal, Z. V. (2002). Metacognitiveawareness and prevention of relapse in depression:Empirical evidence. Journal of Consulting and Clini-cal Psychology, 70(2), 275–287.

Teasdale, J. D., Segal, Z. V., & Williams, J. M. G.(1995). How does cognitive therapy prevent depres-sive relapse and why should attentional control(mindfulness) training help? Behaviour Research andTherapy, 33, 25–39.

Teasdale, J. D., Segal, Z. V., Williams, J. M. G.,Ridgeway, V. A., Soulsby, J. M., & Lau, M. A.(2000). Prevention of relapse/recurrence in major de-pression by mindfulness-based cognitive therapy.Journal of Consulting and Clinical Psychology,68(4), 615–623.

Thase, M. E. (1992). Long-term treatments of recurrent

depressive disorders. Journal of Clinical Psychiatry,53(Suppl. 9), 32–44.

Thase, M. E. (1999). How should efficacy be evaluatedin randomized clinical trials of treatments for depres-sion? Journal of Clinical Psychiatry, 60(Suppl. 4),23–32.

Thase, M. E., Bowler, K., & Harden, T. (1991). Cogni-tive behavior therapy of endogenous depression: Part2. Preliminary findings in 16 unmedicated inpatients.Behavior Therapy, 22, 469–477.

Thase, M. E., Greenhouse, J. B., Frank, E., Reynolds,C.F., III, Pilkonis, P. A., Hurley, K., et al. (1997).Treatment of major depression with psychotherapyor psychotherapy–pharmacotherapy combinations.Archives of General Psychiatry, 54(11), 1009–1015.

Treatment for Adolescents with Depression Study(TADS) Team, U.S. (2004). Fluoxetine, cognitive-behavioral therapy, and their combination for adoles-cents with depression: Treatment for adolescentswith depression study (TADS) randomized controlledtrial. Journal of the American Medical Association,292(7), 807–820.

Truax, C. B., & Mitchell, K. M. (1971). Research oncertain therapist interpersonal skills in relation toprocess and outcome. In A. E. Bergin & S. L. Garfield(Eds.), Handbook of psychotherapy and behaviorchange: An empirical analysis (pp. 299–344). NewYork: Wiley.

Versiani, M. (1998). Pharmacotherapy of dysthymicand chronic depressive disorders: Overview with fo-cus on moclobemide. Journal of Affective Disorders,51(3), 323–332.

Vos, T., Corry, J., Haby, M. M., Carter, R., & Andrews,G. (2005). Cost-effectiveness of cognitive-behavioural therapy and drug interventions for ma-jor depression. Australian and New Zealand Journalof Psychiatry, 39(8), 683–692.

Vocisano, C., Klein, D., Arnow, B., Rivera, C., Blalock,J. A., Rothbaum, B., et al. (2004). Therapist variablesthat predict symptom change in psychotherapy withchronically depressed outpatients. Psychotherapy:Theory, Research, Practice, Training, 41(3), 255–265.

Wampold, B. E., Minami, T., Baskin, T. W., & Tierney,S. C. (2002). A meta-(re)analysis of the effects of cog-nitive therapy versus “other therapies” for depres-sion. Journal of Affective Disorders, 68, 159–165.

Weisz, J. R., MacCarty, C. A., & Valeri, S. M. (2006).Effects of psychotherapy for depression in childrenand adolescents: A meta-analysis. Psychological Bul-letin, 132, 132–149.

Whisman, M. A. (1993). Mediators and moderators ofchange in cognitive therapy of depression. Psycho-logical Bulletin, 114, 248–265.

Wierzbicki, M., & Bartlett, T. S. (1987). The efficacy ofgroup and individual cognitive therapy for mild de-pression. Cognitive Therapy and Research, 11(3),337–342.

Williams, J. M. G. (1997). Depression. In D. M. Clark


& C. G. Fairburn (Eds.), Science and practice of cog-nitive behaviour therapy (pp. 259–283). Oxford, UK:Oxford University Press.

Wolpe, J. (1993). Commentary: The cognitivist overselland comments on symposium contributions. Journalof Behavior Therapy and Experimental Psychiatry,24(2), 141–147.

Wright, J. H. (1996). Inpatient cognitive therapy. In P.M. Salkovskis (Ed.), Frontiers of cognitive therapy(pp. 208–225). New York: Guilford Press.

Wright, J. H., Wright, A. S., Albano, A. M., Basco, M.R., Goldsmith, L. J., Raffield, T., et al. (2005).Computer-assisted cognitive therapy for depression:Maintaining efficacy while reducing therapist time.American Journal of Psychiatry, 162, 1158–1164.

Young, J. E. (1994). Cognitive therapy for personalitydisorders: A schema-focused approach (rev. ed.).Sarasota, FL: Professional Resource Press.

Young, J. E. (1999). Cognitive therapy for personalitydisorders: A schema-focused approach (3rd ed.).Sarasota, FL: Professional Resource Exchange.(Original work published 1990)

Young, J. E., Atkinson, T., Arntz, A., Weishaar, E., &Weishaar, M. (2004). Schema Mode Questionnaire.New York: Cognitive Therapy Center of New York.

Young, J. E., & Brown, G. (1994). Young Schema Ques-tionnaire (2nd ed.). In J. E. Young (Ed.), Cognitivetherapy for personality disorders: A schema-focusedapproach (pp. 63–76). Sarasota, FL: Professional Re-source Press. (Original work published 1990)

Young, J. E., & Klosko, J. S. (1994). Reinventing yourlife: How to break free of negative life patterns. NewYork: Plume.

Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003).Schema therapy: A practitioner’s guide. New York:Guilford Press.


C H A P T E R 7

Interpersonal Psychotherapyfor Depression

KATHRYN L. BLEIBERGJOHN C. MARKOWITZ

Much evidence has emerged over the past decade supporting the clinical effectiveness ofinterpersonal psychotherapy (IPT) for a variety of problems, particularly depression. Asubstantial advantage of IPT is the relative ease with which clinicians can learn to admin-ister this protocol with integrity. In this chapter, the process of IPT is illustrated in some de-tail in the context of the treatment of “Sara,” who was suffering from a major depressive ep-isode associated with grief following the death of her baby girl in utero at 27 weeks’gestation, some 2 months earlier. The lead author, Kathryn L. Bleiberg, an internationalauthority on training in IPT, was the therapist. Although IPT is relatively easy to compre-hend, the twists and turns encountered in the administration of IPT (or any therapeutic ap-proach) are particularly evident in this chapter. Here the therapist skillfully focuses on re-solving grief, as well as on the patient’s social isolation and conflicts with her husband overemotional reactions to the loss. Also notable about IPT is the finding that the treatment ismore successful when administered with fidelity to the goals of IPT and adherence to theprotocol. The power of a therapist and patient working well together and staying on taskprovides some good evidence for the specific effects of an interpersonal focus to psycho-therapy.—D. H. B.

Interpersonal psychotherapy (IPT) is a time-limited, diagnosis-targeted, pragmatic, empiri-cally supported treatment that was originallydeveloped to treat outpatients with major de-pression. IPT focuses on current or recent lifeevents, interpersonal difficulties, and symp-toms. Through the use of the medical modeland by linking mood symptoms to recent lifeevents, the IPT therapist helps the patient tofeel understood. IPT aids recovery from depres-sion by relieving depressive symptoms and byhelping the patient to develop more effective

strategies for dealing with current interper-sonal problems related to the onset ofsymptoms (Bleiberg & Markowitz, 2007).

IPT’s success as an individual treatment formajor depression has led to its adaptation forsubpopulations of patients with mood disor-ders including depressed older patients(Sholomskas, Chevron, Prusoff, & Berry,1983), depressed adolescents (Mufson, Moreau,& Weissman, 1993; Mufson, Weissman,Moreau, & Garfinkel, 1999), depressed HIV-positive patients (Markowitz, Klerman, Perry,

306

Clougherty, & Mayers, 1992; Markowitz,Kocsis, et al., 1998), patients with antepartum(Spinelli & Endicott, 2003) and postpartumdepression (O’Hara, Stuart, Gorman, &Wenzel, 2000), dysthymic disorder (Marko-witz, 1998), and bipolar disorder (Frank,2005). IPT has been adapted for anxiety disor-ders, including social phobia (Lipsitz, Fyer,Markowitz, & Cherry, 1999) and posttrau-matic stress disorder (Bleiberg & Markowitz,2005), bulimia nervosa (Fairburn, Jones,Peveler, Hope, & O’Connor, 1993; Fairburn etal., 1995), and, recently borderline personalitydisorder (Markowitz, Skodol, & Bleiberg,2006). This chapter describes the principles,characteristics, and techniques of individualIPT for major depression and illustrates how aclinician implements IPT techniques with anactual patient.

MAJOR DEPRESSION

Major depressive disorder (MDD), the mostcommon depressive illness, affects millions ofAmericans each year. The Global Burden ofDisease Study, initiated by the World HealthOrganization (Murray & Lopez, 1996), esti-mated that depression is the fourth leadingcause of disability and will become the secondleading cause worldwide by the year 2020. Epi-demiological studies of mood disorders haveprovided estimates of the prevalence and corre-lates of MDD. The National Comorbidity Sur-vey Replication (NCS-R; Kessler et al., 2003)found a 16.2% lifetime and 6.6% annual prev-alence of MDD in the United States, with themajority of cases of MDD associated with sub-stantial symptom severity and impairment infunctioning. National and international studieshave consistently found a higher prevalence ofMDD in women than in men; women are twiceas likely as men to experience an episode ofMDD. The average age of onset of MDD is be-tween ages 20 and 40 years (Blazer, 2000).

The Diagnostic and Statistical Manual ofMental Disorders, fourth edition, text revised(DSM-IV-TR; American Psychiatric Associa-tion, 2000) defines MDD as a mood disordercharacterized by one or more major depressiveepisodes. A major depressive episode entails aperiod of at least 2 weeks during which a per-son experiences depressed mood or loss of in-terest or pleasure in nearly all activities most ofthe day, nearly every day, accompanied by at

least four additional symptoms of depressionpresent nearly every day. Symptoms of depres-sion include the following:

• Significant weight loss (not related to diet-ing), or weight gain or decrease or increase inappetite.

• Insomnia or hypersomnia.• Psychomotor agitation or retardation severe

enough to be observable by others.• Fatigue or loss of energy.• Feelings of worthlessness, or excessive or in-

appropriate or excessive guilt.• Diminished ability to think or concentrate,

or indecisiveness.• Recurrent thoughts of death, recurrent sui-

cidal ideation without a specific plan, or asuicide attempt or a specific plan to commitsuicide.

To meet full criteria for a major depressive epi-sode, the patient’s symptoms must cause clini-cally significant distress or impairment in so-cial, occupational, or other important areas offunctioning. Depression is associated with so-cial withdrawal and difficulty in social and oc-cupational functioning. The diagnosis cannotbe made if the symptoms meet criteria for amixed episode, in which symptoms of both amanic episode and major depressive episodecoexist. Nor may the symptoms be due to thedirect physiological effects of a substance or ageneral medical condition. Finally, the symp-toms may not be better accounted for by thedeath of a loved one, unless they persist 2months beyond the loss or include markedfunctional impairment or include intense feel-ings of worthlessness, suicidal ideation, psy-chotic symptoms or psychomotor retardation.

THE DEVELOPMENT OF IPT

Klerman, Weissman, and colleagues developedIPT in the 1970s as a treatment arm for apharmacotherapy study of depression. Beingresearchers, they developed a psychotherapybased on research data. Post–World War II re-search on psychosocial life events and the de-velopment of mental disorders had shown rela-tionships between depression and complicatedbereavement, role disputes (i.e., bad relation-ships), role transitions (e.g., a losing or gettinga new job, or any meaningful life change), andinterpersonal deficits. Stressful life events can

Interpersonal Psychotherapy for Depression 307

trigger depressive episodes in vulnerable indi-viduals, and depressive episodes compromiseinterpersonal functioning, making it difficult tomanage stressful life events, and often trigger-ing further negative life events (Bleiberg &Markowitz, 2007).

IPT was also built on the interpersonal the-ory of Adolph Meyer (1957) and Harry StackSullivan (1953), and on the attachment theoryof John Bowlby (1973). The interpersonalistsbroadened the scope of psychiatry by empha-sizing social, cultural, and interpersonal fac-tors. Sullivan stressed the role of interpersonalrelationships in the development of mental ill-ness and the use of interpersonal relationshipsto understand, assess and treat mental illness.Sullivan also indicated that life events and rela-tionships occurring after early childhood influ-enced psychopathology, in contradistinction tothe then current focus on pre-Oedipal lifeevents.

Bowlby (1973) posited that the disruption ofsecure attachments can contribute to the onsetof depression, and that psychiatric disorders re-sult from difficulties in forming and maintain-ing secure attachments. Indeed, social supportsprotect against depression. In psychiatry, theidea that mental illness was influenced by cur-rent life events and not (simply) due to earlychildhood experiences was novel in an era di-vided between psychoanalytic and biologicalapproaches (Klerman, Weissman, Rounsaville,& Chevron, 1984).

PRINCIPLES OF IPT: THE IPT MODELOF DEPRESSION

Two basic principles of IPT explain the pa-tient’s depression and situation. These princi-ples are simple enough to grasp, even for avery depressed patient with poor concentra-tion. First, the IPT therapist defines depressionas a medical illness and explains that the pa-tient has a common illness that comprises a dis-crete and predictable set of symptoms, thusmaking the symptoms seem less overwhelmingand more manageable. The IPT therapist as-sumes that the etiology of depression is com-plex and multidetermined: The etiology maycomprise biology, life experiences, and familyhistory, among other factors. The IPT therapistemphasizes that depression is a medical illnessthat is treatable and not the patient’s fault. Ex-plaining that depression is treatable inspires in

patients the hope that they can feel better.Hopelessness, a potentially deadly depressivesymptom, distorts the generally good prognosisof the illness. Depressed patients often see theirsymptoms and consequent difficulties in func-tioning as reflections of a personal failure,character flaw, or weakness.

Defining depression as a blameless illnesshelps to combat the depressed patient’s guiltand self-criticism. The therapist diagnoses de-pression using DSM-IV or International Classi-fication of Diseases (ICD-10; World HealthOrganization, 1992) criteria and assessessymptoms using rating scales such as the Ham-ilton Depression Rating Scale (HDRS; Hamil-ton, 1960) or the Beck Depression Inventory–II(BDI-II; Beck, Steer, & Brown, 1996). IPT’s useof the medical model to define depression dis-tinguishes it from other psychotherapies andmakes it highly compatible with antidepressantmedication in combination treatment.

The second principle of IPT is that the pa-tient’s depression is connected to a current orrecent life event. Stressful life events can pre-cipitate depressive episodes in vulnerable indi-viduals and, conversely, depression can makeit difficult for individuals to manage stressfullife events. IPT focuses on solving an inter-personal problem in the patient’s life, a prob-lem area—complicated bereavement, a roledispute, a role transition, or interpersonaldeficits—connected to the patient’s currentdepressive episode. By solving an interper-sonal crisis, the patient can improve his orher life situation and simultaneously relievedepressive symptoms.

CHARACTERISTICS OF IPT

Several characteristics define IPT, sometimesdistinguishing it from other psychotherapies.

• IPT is time-limited and focused. Acutetreatment is set at the start at 12 or 16 weeklysessions. Some studies have included continua-tion and maintenance phases, which involve bi-weekly and monthly sessions. IPT’s time limitprovides hope for patients that their symptomsand life situation can rapidly improve; the timelimit encourages patients to stay focused intreatment. It also pressures both patient andtherapist to work hard and efficiently withinthe therapeutic window of opportunity. Thera-pist and patient agree to focus on one, or at


most two, problem areas at the beginning oftreatment.

• IPT is empirically grounded. Because IPThas shown repeated efficacy in research studies,therapists can offer and deliver the treatmentwith confidence and optimism, and patientscan feel hopeful about the treatment they arereceiving.

• IPT is diagnosis-targeted. IPT focuses ona specific diagnosis, its symptoms, and howsymptoms interfere and interact with socialfunctioning. IPT is not intended for all patients,but it has been tested in a series of randomizedcontrolled trials to determine its efficacy.

• IPT has a “here-and-now” focus. IPT fo-cuses on the present and on improving the pa-tient’s situation for the future—not on whathappened in the patient’s past. The IPT thera-pist relates current symptoms and interper-sonal difficulties to recent or current life events.Although past depressive episodes and rela-tionships are reviewed and relationship pat-terns are identified, treatment focuses oncurrent relationships—building social supportsand resolving disputes—and social functioning.

• IPT focuses on interpersonal problems.The IPT therapist may recognize intrapsychicdefenses but does not see the patient’s currentdifficulties as a function of internal conflict. In-stead, the therapist focuses on interpersonal re-lationships and functioning.

• IPT focuses on the interrelationship be-tween mood and current life events. The IPTtherapist emphasizes that stressful life eventscan trigger episodes of major depression and,reciprocally, depression compromises psycho-social functioning. This makes it difficult tomanage life stressors, leading to further nega-tive life events.

• IPT emphasizes eliciting affect. Depressedpatients often have difficulty understanding,identifying, and articulating what they are feel-ing. Whereas patients with depression tend toreport feeling “bad,” they have often difficultidentifying negative feelings more specifically,such as anger, hurt, shame, rejection, disap-pointment. Furthermore, depressed patientswho do recognize such negative affects tend tofeel ashamed of having such “bad” emotions.The IPT therapist helps the patient to betteridentify what she1 is feeling, validates emotionssuch as anger and disappointment as normaland useful interpersonal signals, and helps thepatient to use such emotion as a guide. Patientsin IPT learn to manage their feelings better and

to use them to decide how to behave and whatto say in interpersonal encounters.

IPT COMPARED WITH OTHERPSYCHOTHERAPIES

Although IPT has a distinct rationale and is dis-tinguishable from other psychotherapies (Hill,O’Grady, & Elkin, 1992; Weissman, Marko-witz, & Klerman, 2000), it is an eclectic psy-chotherapy that uses techniques seen in othertreatment approaches. IPT also includes the so-called “common” factors of psychotherapy(Frank, 1971). IPT shares the “here and now,”diagnostic focus and time limit, and activeapproach with cognitive-behavioral therapy(CBT). IPT and CBT both include role playingand skills building. However, IPT is much lessstructured and assigns no formal homework,although the IPT therapist does encourage ac-tivity between sessions to improve mood and asit relates to resolving the interpersonal problemarea. Whereas the CBT therapist defines de-pression as a consequence of dysfunctionalthought patterns and attributes the patient’sdifficulties to them, the IPT therapist empha-sizes that depression is a medical illness and re-lates difficulties to feeling depressed and to re-cent life events. IPT emphasizes eliciting affectrather than (“hot,” affectively charged) auto-matic thoughts. IPT addresses interpersonal is-sues in a manner similar to that in marital ther-apy. Much like a supportive therapist, IPTtherapists provide support and encouragement(Bleiberg & Markowitz, 2007).

IPT has been termed a “psychodynamic”psychotherapy on occasion, but it is not(Markowitz, Svartberg, & Swartz, 1998). IPTemploys the medical model of depressive ill-ness, whereas psychodynamic psychotherapyuses a conflict-based approach. The IPT thera-pist does not address unconscious processes, orexplore or interpret the transference, but in-stead focuses on current relationships outsideof the treatment room. Childhood experiencesare recognized but not emphasized. The IPTtherapist relates current symptoms and inter-personal problems to recent life events, not tochildhood experiences. IPT acknowledges theinfluence of past experiences on present diffi-culties, but only to identify patterns of interper-sonal behavior and to empathize with thepatient’s struggle. Unlike psychodynamic treat-ment, the goal in IPT is to improve symptoms


and social functioning, not to change characteror personality. Like the psychodynamic psy-chotherapist, however, the IPT therapist em-phasizes facilitating affect in the treatmentroom and helping patients become aware offeelings they may not have been aware of previ-ously (Weissman et al., 2000).

THE IPT THERAPIST

The IPT therapist conducts treatment in an of-fice in an academic, hospital, or private prac-tice setting. The therapist typically does notconsult with the patient’s family or friends.

The IPT therapist takes a non-neutral, activestance, providing psychoeducation about de-pression, emphasizing the IPT rationale, teach-ing social skills, and instilling hope. A some-what directive approach is essential foradhering to the time limit in IPT and keepingthe treatment focused on the interpersonalproblem area. Taking an active stance may bechallenging for therapists with psychodynamicorientations who are used to taking a neutralstance and offering little guidance in the ses-sion.

On the other hand, the IPT therapist doesnot want to reinforce the passivity and depend-ency that are characteristic of depression. Atherapist who solves all problems may rein-force the patient’s sense of inadequacy. Hence,the IPT therapist encourages the patient tocome up with ideas, explore options, and testthem out between sessions. Hence, the patientdeserves (and is given) credit for making the lifechanges that lead to improvement in treatment.

The IPT therapist needs to be supportive, en-thusiastic, and optimistic to instill hope thatthe depression will improve and changes willbe forthcoming, and to encourage and inspirethe patient to make changes. The structure ofthe IPT manual (Weissman et al., 2000) and theempirical validation of IPT in randomized con-trolled studies help to bolster therapist confi-dence. The IPT therapist frequently congratu-lates the patient on her progress in treatmentand efforts to make changes.

Given the emphasis on helping the patient toidentify and express feelings, the IPT therapistneeds to feel comfortable encouraging expres-sion of affect and tolerating intense negative af-fects. The therapist can show by example thatfeelings are indeed potent, but they are only

feelings, and can be understood in an interper-sonal context and will pass, if tolerated.

THE IPT PATIENT

Research has demonstrated that a wide rangeof patients are good candidates for IPT for ma-jor depression. IPT has been validated with mi-nor modifications as a treatment for ado-lescent, adult, and geriatric depression. Thepatient should ideally report some recent lifestressor and have some social contacts. Patientswith depression who report no recent lifeevents and lack basic social skills tend to doleast well in IPT. IPT is not intended for pa-tients with delusional depression, which re-quires antidepressant and antipsychotic phar-macotherapy or electroconvulsive therapy.Patients with comorbid moderate or severe per-sonality disorder may be less likely to respondto short-term psychotherapy (Weissman et al.,2000). Accordingly, the recent adaptation ofIPT for borderline personality disorder allowsup to 32 weeks of treatment (Markowitz et al.,2006).

THE FOUR INTERPERSONALPROBLEM AREAS

Treatment with IPT focuses on solving one ofthe following four interpersonal problem areasrelated to the onset or maintenance of the pa-tient’s current depressive episode. The IPT ther-apist follows strategies specific to each problemarea.

Grief

This problem area addresses a patient’s compli-cated bereavement following the death of asignificant other. The therapist facilitates themourning process, encourages catharsis, andultimately helps the patient to form new rela-tionships and find new activities to compensatefor the loss. The therapist explores the relation-ship and the associated feelings the patient hadwith the loved one. Patients with complicatedbereavement often have had a conflicted rela-tionship with the deceased, often reporting un-resolved feelings about and anger toward theperson, and feeling guilty or uncomfortablewith such feelings. Patients may feel guilty


about what they did or did not say to or do forthe person who died. The therapist encouragesexploration of these feelings, and validates andnormalizes negative feelings, allowing the pa-tient to let go of the guilt she experiences.

The release of negative affects in the therapyoffice can help to diminish the intensity of suchfeelings. The therapist also explores positivefeelings the patient had for the person she lostand empathizes with her loss. Finally, the thera-pist helps the patient explore options for form-ing new relationships and finding new activitiesthat help to substitute for the lost relationshipand give the patient a new sense of direction.The loss may provide opportunities for the pa-tient to meet people and engage in activities shemight not otherwise have encountered.

Role Disputes

A role dispute is a conflict with a significantother: a spouse, friend, parent, relative, em-ployer, coworker, or close friend. The therapistand patient explore the relationship, the natureof the dispute, and options to resolve it. Pa-tients with depression tend to put other peo-ple’s needs ahead of their own. They have diffi-culty asserting themselves, confronting others,or getting angry effectively, which makes it dif-ficult for them to manage interpersonal con-flicts. The therapist discusses these depressivetendencies with patients and explains: “It is notyour fault. You can learn how to assert your-self.”

The therapist validates the patient’s feelingsin the relationship: recognizing anger as a natu-ral response to someone bothering the patient,for example. The next question is how to ex-press such feelings. The therapist helps thepatient to conceive ways to communicatethoughts and feelings more effectively and role-plays potential interactions with her. If, afterexploring and attempting to implement optionsto resolve the dispute, the conflict has reachedan impasse, the therapist helps the patient toconsider ways to live with the impasse or endthe relationship.

Role Transitions

A role transition is a change in life status, suchas beginning or ending a relationship, starting anew job or losing one, a geographic move,graduation or retirement, becoming a parent,

or the diagnosis of a medical illness. The thera-pist helps the patient to mourn the loss of theold role, to explore positive and negative as-pects of the new role, and to determine whatpositive aspects of the old role, if any, she canretain. Ultimately, the therapist helps the pa-tient adjust to and gain a sense of mastery overthe new role. Even if a new role is wanted orpositive, the role may be accompanied by un-anticipated loss. For example, getting marriedmay involve having to spending less time withone’s family of origin because of having tospend time with the spouse’s family. Moving toa new, larger house in a new and better neigh-borhood may disrupt relationships with friendsin the old neighborhood. If a new role is unde-sired, a patient may discover unseen benefits intherapy. A patient who has lost a job may cometo see the loss as an opportunity to pursue abetter job.

Interpersonal Deficits

Interpersonal deficits is the least developed ofthe four problem areas. It is used as the focus ofIPT for major depression only when a patientreports no recent life events and, therefore,lacks any of the first three problem areas. Pa-tients in this category tend to be socially iso-lated and to have few social supports and ahistory of difficulty forming and sustaining re-lationships or finding relationships unfulfilling.The goal of treatment in this problem area is toreduce the patient’s isolation. Because the pa-tient lacks current relationships, treatment fo-cuses on patterns in past relationships andstarting to form new ones. The therapist re-views past significant relationships, exploringpositive and negative aspects and identifyingrecurrent problems. The therapist helps the pa-tient to explore options for meeting people andparticipating in activities that she used to enjoy.

Unlike treatment for the other problem ar-eas, treatment for interpersonal deficits also fo-cuses on the relationship with the therapist. Inthe absence of other relationships, and with theunderstanding that the patient is likely to feeluncomfortable in the therapeutic situation, thetherapist encourages the patient to discuss herfeelings about the therapist and work on prob-lems that arise in their relationship. Ideally, therelationship with the therapist can serve asmodel for the patient to form other relation-ships. Given that IPT usually focuses on a life


event, it is not surprising that patients who lacksuch life events have been shown to respondleast well to IPT. It is important to identify pa-tients who have underlying dysthymic disorder,because such patients often describe few recentlife events. There is an IPT protocol adaptedto the treatment of dysthymic patients(Markowitz, 1998; Weissman et al., 2000).

THE PROCESS OF IPT TREATMENT

Acute IPT treatment for major depression com-prises initial, middle, and termination phases.IPT techniques help the patient to pursue thegoals of treatment for each interpersonal prob-lem area. This section leads into a case exampleillustrating the implementation of these tech-niques.

The Initial Phase (Sessions 1–3)

Tasks of the initial sessions of IPT for major de-pression include eliciting the patient’s chiefcomplaint, reviewing her symptoms and estab-lishing a diagnosis, determining the inter-personal context for the current depressiveepisode, forming a therapeutic alliance, andsetting the frame of the treatment, includingthe treatment goals and strategies. Althoughthese tasks are common to other psychothera-peutic interventions, the techniques and pro-cess for accomplishing them are unique to IPT.

Diagnosing Major Depression

To diagnosis MDD, the therapist reviews cur-rent symptoms of depression using eitherDSM-IV or ICD-10 criteria, inquiring abouteach symptom and about any past history ofdepression. The therapist may administer theHDRS (Hamilton, 1960) or the BDI-II (Beck etal., 1996) and repeat these instruments everyfew weeks to monitor the patient’s progress.The therapist rules out other diagnoses, such asbipolar illness, depression due to general medi-cal condition or substance, and other psychiat-ric disorders. If the patient meets criteria forMDD, the therapist tells her, describing explic-itly each of the symptoms she reported. Thetherapist states:

“You have an illness called major depression.The symptoms you describe having had for

the past couple of months—feeling downmost of the time, having difficulty enjoyingthings, feeling like you have to push yourselfto get things done, difficulty sleeping, loss ofappetite, trouble concentrating, and feelingvery self-critical and down on yourself andpessimistic about your future—are all symp-toms of depression. Depression is a treatableillness. It is not your fault that you feel thisway and have had difficulty functioning, anymore than it would be your fault if you hadasthma or high blood pressure, or any othermedical problem. And although you’re feel-ing hopeless, your prognosis with treatmentis quite good. You can feel better.”

In the initial phase and as needed throughoutthe treatment the therapist provides psychoed-ucation about depression and how it affects so-cial functioning. The therapist helps the patientto identify and understand her depressivesymptoms and how better to manage them,and to distinguish between the illness andpremorbid strengths and capabilities. The ther-apist determines the need for medication basedon symptom severity, past response to medica-tions, and patient preference.

The “Sick Role”

The IPT therapist gives the patient the “sickrole” (Parsons, 1951), a temporary role in-tended to help the patient to recognize that shesuffers from an illness that comprises a distinctset of symptoms that compromise functioning.By assuming the sick role, the patient is relievedof self-blame and exempted from responsibili-ties that the depression compromises until sherecovers. The therapist educates the patientabout the way depression can impair socialfunctioning and how to explain the illness tofamily and friends to gain support. The sickrole also gives the patient the responsibility ofworking on improving her symptoms in treat-ment.

The Interpersonal Inventory

In IPT the psychiatric history includes the “in-terpersonal inventory,” a thorough review ofthe patient’s past and current social function-ing, close relationships, relationship patterns,expectations of others within relationships,and the perceived expectations that others have


of the patient. The interpersonal inventoryshould give the therapist a sense of how the pa-tient interacts with other people. It should alsoilluminate how relationships may have contrib-uted to the current depressive episode and, con-versely, how depressive symptoms may be af-fecting current relationships. To glean thisinformation, the IPT therapist asks detailedquestions about the patient’s past and currentrelationships. An inquiry about current rela-tionships might include the following ques-tions:

“Who is in your life currently? . . . Do you havea significant other? . . . Girlfriend/boyfriend?. . . What is your relationship like with him orher? . . . What do you like about him or her?. . . What don’t you like? . . . Do you and yourspouse argue? . . . What kinds of things to doyou argue about? . . . What happens whenyou argue—what do you do/say? . . . Whatdo you expect of him or her? . . . What do youthink he or she expects of you?”

Identification of Interpersonal Problem Areas

The primary goal of the interpersonal inven-tory is to determine which interpersonal issuesare most related to the patient’s current depres-sive episode. The therapist should identify ma-jor interpersonal problem areas that will be-come the focus of the treatment. The therapistinquires about any changes in the patient’s lifethat occurred around the onset of current de-pressive symptoms: “What was going on inyour life when you became depressed?” Thetherapist also explores different areas in the pa-tient’s life: home, work, relationships with sig-nificant others, family members, and friends.The therapist should pick one, or at most two,problem areas on which to focus; too manytreatment foci yield an unfocused treatment.The therapist relates the depression to a prob-lem area and presents this to the patient in aformulation:

“It seems from what you have been telling methat the main problem has been your diffi-culty in getting along with your husband. Al-though the causes of depression are complexand not fully known, we do know that con-flicts with significant others can be relatedto depression. Furthermore, depression canmake it difficult to handle conflicts with

other people. I suggest that we meet for thenext 12 weeks to figure out how you canbetter deal with the problems with your hus-band you’ve described. In IPT, we call thatconflict a ‘role dispute.’ As you solve yourrole dispute, both your life and your depres-sion should improve. Does this make senseto you?”

Only with the patient’s explicit agreement towork on the chosen interpersonal problem areadoes the therapist proceed to the middle phaseof treatment. The patient’s agreement on thefocus allows the therapist to bring her back tothis theme thereafter, maintaining a thematicflow for the treatment.

Explanation of the Treatment Contractand the IPT Approach

In addition to agreeing on a treatment focus,the therapist and patient discuss and come toan understanding about other aspects of thetreatment in the initial sessions. The therapistdiscusses practical aspects of the treatment, in-cluding the time limit, the specific length andfrequency of sessions, termination date, ap-pointment time, fee, and so forth. The therapistexplains that IPT is one of several empiricallysupported treatments for depression and de-scribes the basic IPT principles. The therapistemphasizes the “here and now,” and the socialand interpersonal focus of the treatment, ex-plaining:

“We’ll work together to try to understandhow current stresses and relationships maybe affecting your mood. We will work onhelping you to better manage these stressesand the problems you may be having in rela-tionships. We’ll explore what you want andneed in your relationships and help you fig-ure out how to get what you want andneed.”

The therapist continues: “I’ll be interested inhearing each week about what’s going onin your relationships with other people, howthose interactions make you feel, and also howyour feelings influence what happens withother people.” The therapist encourages thepatient to bring up any discomforts she hasabout the sessions. “If anything bothers you,please bring it up. I do not intend to do any-


thing to make you uncomfortable, but if youfeel that way, please tell me, so that we can ad-dress the issue. It’s just the sort of interpersonaltension that we need to focus on in your out-side life as well.”

The Middle Phase (Sessions 4–9)

Once the treatment contract has been madeand an interpersonal problem area is chosen asthe focus of treatment, the therapist can pro-ceed to the middle phase of sessions, in whichthe goal is to work on resolving the focal prob-lem area.

Each session begins with the question, “Howhave you been feeling since we last met?” Thisquestion elicits an interval history of mood,events, and interpersonal interactions thattranspired between sessions. It also keeps thepatient focused on current mood and situation.The patient is likely to respond by describing amood (“I felt really down”) or an event (“Myhusband and I had a huge fight”). After furtherinquiry, the therapist links the patient’s moodto a recent event, or an event to her mood (“Nowonder you felt depressed given the fight youhad with your husband”).

Inquiring about emotionally laden interper-sonal interactions, the therapist uses communi-cation analysis—the reconstruction and evalu-ation of affectively charged interactions—tohelp the patient understand how she felt in thesituation and what she might have done tocommunicate more effectively. The therapistexplores the patient’s wishes and options, help-ing her to decide what she wants and exploringthe options for achieving it.

“What did you want to happen in that situa-tion? . . . What could you have done to getwhat you wanted in that situation?”

Depressed patients often have difficulty seeingthat they have options, and their tendency todeem their needs less important than others’contributes to difficulty in being self-assertive.The therapist explains this to the patient andoffers empathy, blaming the depressive syn-drome for the patient’s difficulties where ap-propriate. To help the patient choose which op-tions to pursue, the therapist uses decisionanalysis. These techniques are elaborated in thecase example presented later in this chapter.

The therapist role-plays potential interac-tions with the patient to prepare her for inter-

personal interactions in real life. Rather thansupply words for the patient during role play,the therapist instead encourages her to come upwith her own words. This empowers the pa-tient, because it fosters her independence fromthe therapist. The therapist points out that thepatient is able to figure out what to say and doin interpersonal interactions, but that depres-sion can make her feel unable to do so. Thetherapist asks the patient how she felt duringrole play and how she imagines it will feel tosay the words in reality. When a patient com-municates effectively during role play—and, ul-timately, in real life—the therapist reinforcesthe adaptive behavior with congratulations andencouragement. Real-life successes not onlyimprove mood but also inspire patients to tryto assert themselves in subsequent situations.

Depressed patients tend to withdraw sociallyand to lose interest in activities they previouslyfound pleasurable. The therapist encouragesthe patient to resume social activity and to ex-plore options for new activities and opportuni-ties to form new relationships, if applicable.The therapist empathizes with how difficult itmay be for the patient to push herself to engagewith others, but stresses that, once engaged,she will likely feel better. Indeed, the IPT thera-pist asks the patient to take risks—both in as-serting herself with others and in pushingherself to reengage in social activities. The ther-apist explicitly acknowledges that he or she isasking the patient to take risks, yet assures herthat these risks will likely lead to improvementof her mood and life situation.

The Termination Phase (Sessions 10–12)

In the final sessions, the therapist reviews thepatient’s progress in symptom improvement, aswell as the extent to which the patient has re-solved the focal problem area. By reviewingwhy the patient is better, through the actionsthe patient has taken to resolve the interper-sonal focus, the therapist reinforces the pa-tient’s growing self-esteem by pointing out thatthe patient’s actions have led to her gains. Thetherapist congratulates the patient on her prog-ress and hard work, and expresses optimismthat the patient can maintain that progress in-dependent of the therapist.

The therapist addresses nonresponse in pa-tients whose moods and life situations do notimprove, or only partially improve. The thera-pist should explain that is the treatment that


failed—not the patient. The therapist gives thepatient hope by emphasizing that depression istreatable and many other effective treatmentsexist, and by encouraging her to explore alter-native treatments.

The therapist explores the patient’s feelingsabout the treatment and termination. The ther-apist acknowledges not only his or her ownsadness to be ending their relationship but alsohappiness about the patient’s improvement andconfidence that the patient will be able tomaintain the progress she made in treatment.Should symptoms recur, the patient has gainedtools to manage symptoms of depression onher own. Furthermore, the patient can returnto IPT for “booster” sessions as needed.

Using the IPT medical model, the therapistprovides psychoeducation about relapse andrecurrence of major depression and preparesthe patient about potential for relapse. Patientswho have experienced one or more episodes ofmajor depression are unfortunately vulnerableto future episodes. The therapist explains thisand advises that given the link between stress-ful life events and mood, the patient can antici-pate that she may have difficulty with future,stressful life events. Fortunately, the patient canuse the coping skills she gained in treatment toward off a worsening of symptoms. If the pa-tient has improved in IPT but has either signifi-cant residual symptoms or a history of multipleepisodes, therapist and patient may contractfor continuation or maintenance IPT, whichhas also demonstrated efficacy in forestallingrelapse.

CASE STUDY

The following case demonstrates how a clini-cian (K. L. B.) implemented IPT for major de-pression in a 12-week acute treatment and il-lustrates how one works with the problem areaof grief. In IPT, grief (complicated bereave-ment) is considered as a focal problem areawhen the onset of depression is related to thedeath of a significant other and the patientis experiencing an abnormal grief reaction(Weissman et al., 2000). Although cases thatfocus on the grief problem area usually addresscomplicated bereavement related to the deathof a person who has actually lived, the follow-ing case involves complicated bereavement re-lated to a stillbirth. Indeed, the IPT problem ar-eas can apply to a wide range of cases. IPT

treatment goals and techniques specific toworking with the grief problem area are dem-onstrated.

Background Information

Sara, a 35-year-old, married, childless woman,was referred for treatment of depression fol-lowing the death of her baby girl in utero at 27weeks’ gestation. Her doctor explained that abacterial infection was the most likely cause ofthe stillbirth. Sara’s chief complaint was: “I feellike I should be over it.”

At 27 weeks, after not feeling the baby movefor at least several hours, Sara called her doc-tor, who told her to come to the hospital. Thedoctor was unable to find a heartbeat and toldher that he needed to deliver the fetus. Sara re-called feeling shocked, numb, and unable to cryat first. She was given medication to induce la-bor and an epidural, and delivered the babyvaginally. Despite efforts to revive her, the babywas pronounced dead shortly after delivery.Sara said that she wanted to hold the baby andwas given the baby to hold. She held the baby,who was swaddled in a white-and-pink blan-ket. She recalled that she and her husband crieduncontrollably while they took turns holdingthe baby and for a long time after giving thebaby back to the doctor. She remembers thatthe baby was “very cute” and looked like herhusband. She was given pictures of the babyand footprints to take home. Sara and her hus-band decided not to have a funeral or memorialservice for the baby.

Sara reported that since the stillbirth 2months earlier, she had been feeling sad and ir-ritable most of day, nearly every day, and un-able to enjoy things she used to enjoy, such asreading fiction, cooking, going to the movies,and exercise. She worked as a nurse on an inpa-tient medical floor in a New York City hospi-tal, and prior to the stillbirth had very muchenjoyed her work. She now felt unable to enjoyher work because of her mood, and she fearedhaving to talk about her loss with coworkerswho knew she had been pregnant. She was cry-ing frequently, socially withdrawn, had low en-ergy and difficulty concentrating, experienceddecreased appetite, and felt very bad about her-self. She denied ever having thoughts of suicideor feeling that life was not worth living.

Sara reported that she tried not to thinkabout the baby’s death, but she was frequentlybothered by thoughts about the baby, often


wondering what her life would have been likehad she survived. Sara had returned to work 3weeks after the stillbirth, hoping that it wouldserve as a distraction and help her “get over”her loss. She reported feeling very angry at andavoiding other pregnant women, includingclose friends and women with newborns, in ad-dition to other reminders of her pregnancy. Shefelt angry at having had to go through preg-nancy, labor, and delivery without gaining thepleasure of having a child.

Sara was plagued by inappropriate guilt. Shefelt guilty because she feared that she had donesomething to cause her loss, despite the doctortelling her that there was nothing she couldhave done to prevent it. He explained thatwhen bacterial infections cause fetal death,they often cause no symptoms in the motherand go undiagnosed, until they cause seriouscomplications. Nevertheless, Sara felt that sheshould have known about the infection, andshe felt guilty about having waited until age 35to try to conceive. She described feeling like afailure for having had a stillbirth. Sara feltguilty that she had disappointed and upset herhusband by losing the baby, and she did notwant to burden him with her feelings about theloss.

Sara had never sought treatment prior to hercurrent evaluation. She described one prior epi-sode of major depression in her late 20s lasting4–6 weeks, precipitated by a breakup with aboyfriend of several years, but reported feelingmuch worse since losing the baby. She reportedthat her mother had been treated for depres-sion with antidepressant medication with goodresults.

Sara was an attractive woman of averageheight and weight who looked her stated age.She was casually but neatly dressed in jeansand a large sweater. Sara’s movements wereslightly slowed; her speech was fluent. Hermood was depressed and her affect, congruentand tearful. She denied current or past suicidalideation and any history of substance abuse orpsychotic symptoms. She denied current orpast medical conditions, including thyroid dys-function. She reported no known prior preg-nancies, pregnancy losses, or fertility issuesprior to the stillbirth. In fact, she had conceivedafter just a couple of months of trying to getpregnant.

Sara was good candidate for IPT: She metcriteria for major depression and had experi-enced a recent life event that could be easily

linked to the onset of her symptoms. Further-more, IPT could address the interpersonalproblems she was experiencing related to theonset of her symptoms. Sara was also a poten-tial candidate for CBT, pharmacotherapy, or(evidence-based antidepressant) psychotherapycombined with medication. She was uninter-ested in doing written homework and resistantto taking medication, because she was hopingto conceive again in the near future.

IPT Treatment with Sara

Acute Phase (Sessions 1–3)

Treatment with Sara followed the IPT formatfor acute treatment. In the first three sessions,the therapist obtained a thorough psychiatrichistory and set the treatment framework. In thefirst session, she obtained a chief complaintand a history of Sara’s present illness. UsingDSM-IV criteria, the therapist determined thatSara met criteria for major depression, recur-rent. She administered the HDRS to assess theseverity of Sara’s symptoms.

The therapist offered Sara empathy for herpregnancy loss, saying: “I am so sorry. You’vesuffered a terrible loss. No wonder you havebeen feeling so badly and having such a diffi-cult time.” The therapist gave Sara her diagno-sis of major depression, reviewed her specificsymptoms, and gave her the “sick role.”

“The symptoms you’ve described having inthe past couple of months—depressed mood,not being able to enjoy things and your lossof interest in things, feeling very badly aboutyourself and guilty, your difficulty eating andsleeping, and difficulty concentrating—areall symptoms of major depression. Major de-pression is an illness that is treatable. It is notyour fault that you have been feeling thisway.”

The therapist explained that Sara’s HDRSscore of 24 indicated moderately severe de-pression and that she would readminister theHDRS at regular intervals to monitor Sara’sprogress. Given the severity of Sara’s symp-toms, her willingness to participate in psycho-therapy, and her reluctance to take medicationin anticipation of trying to conceive again, thetherapist did not think medication was needed.

The therapist described IPT and the treat-ment rationale:


THERAPIST: I am trained in a psychotherapycalled interpersonal psychotherapy, which Ithink could be helpful to you. Interpersonalpsychotherapy—often referred to as IPT—isa time-limited treatment that focuses on howrecent life events and stresses—such as losinga baby—affect mood, and how mood symp-toms make it difficult to current life eventsand stresses, particularly problems in rela-tionships. Although we will take the first fewsessions to review your history, our sessionswill focus on the here and now, on your cur-rent difficulties and relationships, not on thepast. Does this make sense to you?

SARA: Yes.

THERAPIST: Often, people respond to treatmentwith IPT in 12 weekly sessions. I proposethat we meet once a week for a 50-minutesession for the next 12 weeks. If it’s helpful,at the end of the 12 sessions, we can discusswhether it might be useful to have additionalsessions to work on issues and maintain yourprogress. How does that sound to you?

SARA: It sounds good. I hope I can feel better in12 weeks.

THERAPIST: You can feel better in 12 weeks. IPThas been shown in numerous research stud-ies to be effective in treating symptoms likethe ones you have described.

After the first session Sara felt somewhatmore hopeful but stated that she did not likethe idea that she had a diagnosis of major de-pression. Although she could understand thatthere was relationship between her stillbirthand her mood, Sara still felt that she should befeeling better after 2 months and did not wantto think of herself as depressed, like hermother, and in need of help. Sara stated thatshe was always the “strong one” and was usedto functioning at a very high level. The thera-pist was not surprised by Sara’s initial skepti-cism, because it can take time for patients toaccept the medical model. Furthermore, pa-tients with depression often feel uncomfortableabout seeking help, because they fear burden-ing others. Nevertheless, therapist and patientagreed to work together for 12 weeks, then de-cide whether further sessions were needed.

In obtaining Sara’s psychiatric history, thetherapist conducted an interpersonal inventory,carefully reviewing Sara’s past and current so-cial functioning and close relationships. She

started the inventory by asking about Sara’sfamily.

“Where did you grow up? . . . Who was inyour family? . . . How would you describeyour relationship with your mother? . . .With your brother?”

Sara grew up in Canada with her father andmother, both in their early 60s, and her youn-ger brother, age 33, all of whom still lived nearToronto where she had been raised. She re-ported that her father had worked a lot whileshe was growing up, and although she wasfond of him, she did not feel so close to him.Sara felt closer to her mother and spoke withher weekly but was easily irritated by her. Itbothered Sara that her mother was not as-sertive and was intermittently depressed. Shespoke weekly with her brother, who lived inCanada with his wife and 2-year-old son. Shedescribed her relationship with her brotheras fairly close. She reported speaking to herbrother less often since the stillbirth, becauseshe felt jealous that he had a child. When theydid speak, Sara avoided asking about hernephew.

After exploring Sara’s relationships withfamily members, the therapist asked aboutother important people in her life and askedher about the relationship with her husband.At age 33, Sara met her husband, Steve, whowas 1 year her junior. She described Steve aswarm and charming, and reported that he tookgreat care of her. She felt she did not “deservehim,” because he was “such a good guy.” Shedescribed her previous boyfriends as less emo-tionally available and “not very nice.” BothSara and Steve were originally from Canadabut met in New York City, when they were in-troduced by mutual friends. Sara had moved toNew York in her early 20s, whereas her hus-band had moved there 2 years prior to theirmeeting.

Since the stillbirth, Sara felt distant fromSteve and argued with him about “littlethings.” She reported feeling guilty that she hadlet him down by losing the baby and fearedthat he blamed her for the baby’s death. She didnot want to burden him further by sharing herown distress about the loss. She also felt thatSteve would not be able to understand her fearsabout trying to conceive again.

Sara reported having a few close girlfriendswho lived in the tristate area and, until the


pregnancy loss, spoke to them about once perweek. She had several friends at work, withwhom she chatted almost daily until the loss.She described herself as “independent,” “out-going,” and “not one to lean on other people”prior to becoming depressed. She was the oneto whom her friends turned when they hadproblems. Sara said that before the depression,her friends would describe her as hardworkingand energetic. Sara reported that she rarely ar-gued with friends because she felt “uncomfort-able” with conflict. She avoided confrontingfriends and coworkers when she disagreed orfelt angry with them.

The therapist asked Sara if there was anyoneto whom she had turned for comfort after herloss, because it is important to have someone inwhom to confide after such a terrible loss orany stressful experience. Sara replied that shehad been avoiding her friends and family sincethe loss. She had felt uncomfortable talking tofriends, family, and coworkers about her preg-nancy when she was pregnant, because she didnot like being the center of attention and feltguilty that she did not enjoy the first trimesterof her pregnancy. She felt even more uncom-fortable discussing her pregnancy loss. Herparents and in-laws came to see Sara and herhusband after the loss, but she felt unable totalk with them about what had happened andhow she was feeling. All of her coworkersknew that she had been pregnant, and Sara feltobligated to say something to them about whathad happened. The therapist noted that itsounded like Sara could trust no one with herfeelings about the stillbirth. Sara did not wantto reach out to family or friends, or let themknow how bad she felt; she explained: “I don’twant to bother people with my problems. Idon’t want to be weak.”

The therapist reframed Sara’s difficulty inreaching out to others, using the medical modelto explain how depression affects social func-tioning:

THERAPIST: You are not weak—you are de-pressed—and that’s not your fault. Peoplewith depression tend to minimize their ownneeds and avoid seeking help from theirfriends, as you have been, because they fearbeing a burden. However, it is not only ap-propriate to seek support from others but italso can be really helpful to get support fromothers. In fact, support from others has been

shown to help in recovery from depression.I appreciate your dilemma. Your depres-sion makes you feel uncomfortable seekingsupport, yet support from others has beenshown to reduce depression and protect peo-ple from becoming depressed. Does thismake sense to you?

SARA: Yes, but I also don’t want to hear whatthey have to say. It just makes me more up-set. They don’t understand what I have beenthrough.

THERAPIST: What kinds of things have peoplesaid to you?

Sara replied that it bothered her when peo-ple said things like “You’ll get pregnantagain” or “I know someone who also lost ababy.” These statements made her feel angry.She felt that others could not understandwhat she had experienced. One close friendhad recently given birth to her first child, andSara had avoided calling and seeing her. Shefelt that it was unfair that her friend had ababy when she did not. A coworker had beenpregnant at the same time but had a rela-tively easy pregnancy. Sara felt that her co-worker was not sympathetic to her physicaldiscomfort during pregnancy.

By the end of the first phase of treatment,the therapist had connected Sara’s major de-pressive episode to her interpersonal situationin a formulation centered on an IPT focalproblem area. Sara’s chief complaint reflectedthat she was still grieving the loss of her babyand unable to resume her normal level offunctioning. Her situation was a clear exam-ple of the grief problem area: Sara was suffer-ing from complicated bereavement. While itis normal to grieve for months after losing aloved one, the severity of Sara’s depressivesymptoms—especially the excessive guilt, lowself-esteem, and social isolation—and heravoidance of thoughts, feelings, and remind-ers of the baby and the baby’s death, reflectedan abnormal grief reaction. She had notsought emotional support after the stillbirthand had not really mourned the loss of herbaby. In fact, people often develop compli-cated bereavement when they lack or havenot used their social network to help themmourn the loss of their loved one.

The therapist presented this formulation toSara:


THERAPIST: From what you are telling me, it’sclear that the loss of your baby triggeredyour current depression. You have suffered aterrible loss and you are having troublegrieving. No wonder you are having such ahard time. This is not your fault. Further-more, your loss and your depression have af-fected your relationships with people in yourlife, like your husband, your friends and co-workers, and you’re having difficulty ex-pressing your feelings to them. I suggest thatwe focus our sessions on handling your griefover this terrible event. Grief is one of theproblem areas that IPT has been shown totreat. I suggest we work on helping you tomourn the loss and to improve your relation-ships that have been affected by your loss.How does this sound to you?

SARA: It sounds good.

With Sara’s explicit agreement about thetreatment focus, the therapist began the middlephase of treatment.

Middle Phase (Sessions 4–9)

During the middle phase, therapist and patientworked on resolving Sara’s interpersonal prob-lem area. In IPT, the strategy for working withgrief is to help the patient to tolerate and man-age the affect of loss, and to gather social sup-port to help the patient through mourning. Inaddition, the therapist helps the patient to useexisting social supports, to reestablish interestsand relationships, and to form new relation-ships and explore new activities to compensatefor the loss (Weissman et al., 2000).

The therapist continued providing psychoed-ucation about complicated bereavement andhow depression affects social functioning, andrepeatedly linked Sara’s depression to the iden-tified problem area. She began each sessionwith the opening question: “How have thingsbeen since we last met?” This question elicitedaffect and a history of Sara’s mood and eventsbetween sessions, and kept her focused on hercurrent mood and life events.

To facilitate the mourning process, the thera-pist encouraged Sara to think about the loss. Infact, this process had begun during the initialphase, while the therapist took a history of theevents related to the onset of Sara’s depression.The therapist asked Sara to describe the events

prior, during, and after the baby’s death—oftena source of patient guilt—and explored Sara’sfeelings associated with these events. In helpinga patient mourn the loss of a loved one, the IPTtherapist asks the patient to describe her feel-ings about the death and about the person whodied. The therapist explores what the patientand the deceased did together, what the patientliked and did not like about the person, andwhat the patient wished they had done togetherbut did not have a chance to do. The therapistasks the patient to describe how the deceaseddied and how she learned about the death, andexplores the patient’s related feelings. Giventhat the Sara’s baby died in utero, the therapistmodified this inquiry somewhat by encourag-ing Sara to talk about her experience of beingpregnant, about the baby, and what she imag-ined the baby would be like. The therapistasked Sara what she liked about carrying thebaby, what she did not like, and what she hadhoped to do with the baby.

Sara tearfully described having had mixedfeelings about her pregnancy. She reported thatshe and Steve started trying to conceive 6months after getting married and, to her sur-prise, she got pregnant after 2 months. Whenshe discovered she was pregnant, Sara felt re-ally happy, but scared about becoming a par-ent. She questioned whether she was “ready.”Sara reported that she made a great effort topractice good prenatal care: She ate healthy,pregnancy-safe foods, took prenatal vitamins,and started prenatal yoga classes. Practicinggood prenatal care made her feel good, “as if Iwas already a mom taking care of my baby.”

Sara quickly began to experience terrible fa-tigue and unrelenting nausea, which lasted forthe first 12 weeks of the pregnancy. She de-scribed feeling as if she had been “taken over”by the pregnancy. She complained that sheloved to cook but did not want to cook, be-cause she felt so sick. Despite the nausea, sheensured that she was getting the nutrients sheneeded for the baby. The fatigue and the nauseawere so debilitating that Sara could no longermeet the physical demands of her job as anurse. As a result, she was unable to hide herpregnancy from her coworkers; she told her su-pervisor, who was happy to accommodate Saraby giving her more administrative responsibili-ties in lieu of patient care, until she felt better.Sara resented having to give up clinical workwith patients, which was the part of her job


she enjoyed. She felt self-conscious about hersymptoms and very guilty that her coworkershad to absorb her patient load, despite their be-ing very supportive. It bothered her that othersidealized pregnancy when she found it so un-pleasant. At the same time, she felt guilty andselfish that she had complained about the preg-nancy: Sara felt that she should have just beengrateful she was pregnant.

The therapist empathized with Sara’s dis-comfort during her first trimester and validatedher need to complain:

THERAPIST: The first trimester of pregnancy canbe really difficult and disruptive. Give your-self a break! It can be hard to appreciate be-ing pregnant when you are feeling so terri-ble. It sounds like you did appreciate beingpregnant—you made a great effort to takecare of yourself. You watched your diet care-fully and rearranged your work situation.

SARA: I don’t know . . . I guess that is true.

When the exhaustion and nausea subsided inher second trimester, Sara began to feel moreoptimistic and excited about having a child.Seeing sonograms made the baby seem “morereal” and helped Sara feel connected to thebaby. At Week 16, Sara learned that the babywas a girl. She felt excited and immediately be-gan considering names and envisioning whatthe baby would look like. Sara imagined thatshe would look like a combination of herselfand her husband, with blue eyes and blond,curly hair. She thought the baby would be akind person, like her husband. She imaginedwalking to the park with the baby in a stroller,and playing with her. At Week 20, Sara beganto feel the baby move, which she very much en-joyed. When the baby moved, Sara would stopwhatever she was doing to watch and feel herabdomen. She described feeling the movementsas “some of the happiest moments in my life.”Neither she nor her husband had thought of aname for the baby, but referred to her as“Sweetie” in utero.

For weeks after the stillbirth, Sara struggledwith physical reminders of the baby. After de-livering the baby, she had leaky breasts for afew days and vaginal bleeding for severalweeks. She reported that she still looked preg-nant for weeks after delivering the baby, as heruterus slowly returned to its prepregnancy size.

At the time of her initial evaluation, Sara re-ported that she still had to lose 5 pounds toreturn to her prepregnancy weight. Sara missedbeing pregnant and described feeling “empty”and “alone” without the baby inside her. Shewas eager and ready to be a parent, yet feltscared about conceiving again, because shefeared losing another baby.

A couple of weeks after the stillbirth, Sara’sdoctor determined that an undetected bacterialinfection caused the stillbirth. The doctor ex-plained that there was nothing Sara or her hus-band could have done to prevent the loss, andthat this kind of loss was very rare. Despite herdoctor’s explanation, Sara blamed herself forher baby’s death and feared that her husbandblamed her too, although he repeatedly deniedthis. The therapist explored Sara’s guilt further:

THERAPIST: What could you have done to pre-vent your baby’s death?

SARA: (tearfully) I don’t know. . . . I shouldhave been able to do something.

The therapist offered Sara empathy and sup-port, and related her guilt to depression:

THERAPIST: It would be great if there was some-thing you could have done to prevent thistragedy, but there is generally nothing par-ents can do to prevent a pregnancy loss. Itsounds like you did everything you could—you took very good care of yourself. You arestruggling with inappropriate and excessiveguilt—a symptom of depression. You areblaming yourself for something you didn’tdo. Perhaps when you find yourself feelingguilty, you can try to label this as a symptomof depression.

SARA: Yes. I guess I can try.

Talking about the pregnancy, the baby, andthe baby’s death, and exploring related feelingsenabled Sara to develop a more balanced andrealistic perception of her relationship with thebaby and her role in the baby’s death. She real-ized that she had not taken her pregnancy forgranted. In fact, she had done everything shecould to manage a difficult first trimester andtake care of her baby. In addition, her experi-ence with the pregnancy and the baby madeSara realize that, despite her initial anxiety, shewas ready and excited to become a parent. By


the end of the first month of treatment, Sara’smood was somewhat improved and her HDRSscore had fallen to 18. She was less self-criticaland more hopeful.

An important part of treating grief is facili-tating the expression of affect related to theloss of the loved one. The therapist exploredSara’s feelings as she spoke about the baby andher loss, giving her time to articulate what shewas feeling and to cry. Although IPT therapistsgenerally take an active stance, when facilitat-ing the expression of painful feelings, it is im-portant to allow for silences. By listening si-lently, the therapist showed that she couldtolerate Sara’s painful feelings, and that cathar-sis was an important part of mourning her loss.Sara was able to express feelings that she notonly had been avoiding but also feelings ofwhich she had previously been unaware.

Sara had avoided looking at the pictures andfootprints of the baby from the hospital, whichhad been stored in a box under her bed. Sheand the therapist explored what it would belike for her to look at these items. Sara feared itwould be scary, and that she would feel reallybad. The therapist gently encouraged Sara totake a risk and look, because it might make herfeel better to experience the feelings she hadbeen avoiding:

“Your feelings are not going to hurt you. Youmight actually feel better if you allow your-self to let out some of the feelings you havebeen trying to keep inside. I know I am ask-ing you to take a risk, but you might bepleasantly surprised.”

Between sessions, Sara looked at the picturesand the footprints. The therapist asked what itwas like for her.

SARA: I cried a lot. She was so cute. It wasn’t ashard as I thought it would be. It felt like a re-lease. I was surprised that I felt a little betterafterwards.

THERAPIST: I am so glad you took a risk andlooked. It sounds like it made you feel better.

In fact, every few weeks before the end of treat-ment, Sara looked at the pictures and the foot-prints. She explained that the pictures were sortof comforting, because they made her feel aconnection to her baby.

In addition to encouraging catharsis, thetherapist encouraged Sara to work on her inter-personal interactions, to reconnect with thepeople in her life, and to consider opportunitiesto form new relationships and start new activi-ties to compensate for the loss. The therapistexplained that people with depression tend toisolate themselves and stop engaging in previ-ously pleasurable activities, both of which canperpetuate depression. Sara reported not want-ing to talk to people, because she feared thatshe would have to talk about the loss, or thatthings people said would make her feel worse.In fact, as Sara and the therapist discussed, shecould guide the conversation in a way thatmade her feel comfortable. They explored androle-played options for maintaining control ofsuch conversations. Furthermore, Sara couldtell people what would be helpful to her. Thetherapist explained:

“People with depression often have difficultyasserting their needs. If you communicateyour needs to others—like your husband,friends, coworkers, and your family—youmight improve those relationships and yourmood. The people in your life may not knowwhat you need. If you tell them, you mightnot only get support from them, but youmight enjoy their company again and feelbetter.”

Using communication analysis, the therapistasked Sara to recount arguments and unpleas-ant interactions with others—what she wasfeeling during the interaction, what she said ordid, and what the other person said or did.They explored what Sara wished other peoplewould say or do, and what options she had forasking them to do these things, and role-playedSara asking for what she wanted. Sara reportedthat she hated running into people who knewshe had been pregnant but did not know aboutthe stillbirth. In fact, she avoided going places,because she feared having to answer questionsabout the stillbirth. Sara and the therapist ex-plored these interactions and how Sara couldhandle them more effectively:

THERAPIST: What kinds of things have peopleasked, or what are you afraid they will ask?

SARA: People have asked “How’s your baby?”or “Weren’t you pregnant?”


THERAPIST: How does that make you feel?

SARA: Awful!

THERAPIST: How do you handle it?

SARA: I don’t know. . . . Sometimes I say “Itdidn’t work out” or “My baby died.”

THERAPIST: That sounds good. How does it feelfor you to say that?

SARA: It feels OK, but then they want to knowwhat happened and say stupid things like“At least you know you can get pregnant” or“You can have another one.”

THERAPIST: What would you like them to sayor do?

SARA: I would like them to just say “I’m sorry,”and not ask any questions. I don’t want totalk about what happened.

THERAPIST: How could you convey that?

SARA: I guess I could say, “I’m sorry, but I’drather not talk about it.”

THERAPIST: How does that sound? How did itfeel to say that?

SARA: It felt OK. Don’t you think that is rude tosay that?

THERAPIST: No. You said it politely and it is ap-propriate for you to assert your needs. It isan uncomfortable situation for both you andthe person who asked the question. If youare polite and direct with people, they arelikely to understand. But why not try it andsee?

Sara reported that she had avoided returningcalls from old friends. She explained that shedid not feel comfortable seeing her friends whohad babies, because it would remind her of thebaby she had lost. Sara also did not want tohave to talk about the loss. She did not want totell them how she felt, because she feared hurt-ing their feelings. Sara and the therapist role-played Sara telling her friends about her dis-comfort and explaining that she did not wantto offend them. Role play helped Sara feel pre-pared and less anxious about going to work,walking around her neighborhood, and talkingto old friends. As a result, she gradually start-ing going out more and began returning phonecalls. She returned to the yoga studio, whereshe had taken prenatal yoga, and started takingregular yoga classes, which helped her moodand provided an opportunity to be amongother people. By midtreatment Sara’s HDRS

had fallen to 13, consistent with mild depres-sion.

Since the stillbirth, Sara had been bickeringwith her husband Steve “over stupid things”and felt “distant from him.” The therapistasked her to describe a recent incident. Sarasaid that Steve came home from work and toldher that his friend’s wife had just had a baby.She felt it was insensitive for him to tell herabout other people’s positive pregnancy experi-ences. It bothered her that Steve did not seemas uncomfortable as she was with this informa-tion, and that he no longer seemed as upset asshe about the loss. The interaction made herfeel “alone.” She had responded to him by say-ing, “That’s nice,” then leaving the room andruminating for the rest of the evening about hisinsensitivity.

Sara reported that they often had similar in-teractions. The therapist once again relatedSara’s difficulty asserting herself with her hus-band to depression, and noted that keeping herfeelings inside might actually be making Sarafeel worse. They explored interpersonal op-tions for handling this situation in a way thatmight make Sara feel better. The therapist alsohelped Sara to explore what her husband’s in-tentions might have been in the situation shedescribed. She wondered whether he was tryingto make her feel better, because his friend’s wifehad experienced several miscarriages. Theyrole-played Sara telling her husband how shefelt. Subsequently, when Sara was able to ex-press her feelings to him, she learned that Stevewas, in fact, telling her these stories to give herhope. Furthermore, her husband revealed thathe was still upset about the loss of their babybut did not want to upset her by sharing hisfeelings. Sara was relieved that she and Stevewere “on the same page” and felt good that shewas able to feel close to him again. They subse-quently were able to share more of their mixedfeelings about the pregnancy experience.

Termination Phase (Sessions 10–12)

During the final sessions the therapist and Sarareviewed the progress Sara had made. She re-ported that her mood was much improved. HerHDRS was now a 5, consistent with euthymiaand remission. Sara’s affect was brighter, andshe was less preoccupied with the loss of herbaby: “I still get upset when I think about mybaby, but I don’t get as upset. It doesn’t ruin myentire day. I am actually able to enjoy things


again.” Furthermore, Sara no longer blamedherself for her baby’s death. She felt good abouther ability to communicate her feelings moreeffectively with her husband, friends, and oth-ers, and to enjoy socializing and other activitiesagain.

The therapist congratulated Sara on her hardwork and achievements, and told her howhappy she was that Sara felt so much better.They discussed the potential for relapse andhow Sara could maintain her progress. GivenSara’s history of depression, the therapist ex-plained that Sara was, unfortunately, vulnera-ble to future episodes; however, Sara could an-ticipate that she would be vulnerable in thesetting of stressful life events—role disputes,role transitions, deaths—and use the copingskills she had learned during their work to-gether. Sara anticipated starting treatments forher clotting disorder, trying to conceive again,and, she hoped, getting pregnant for a secondtime—all role transitions. The therapist andSara explored ways Sara could take care of her-self during this potentially stressful time. Theydiscussed Sara’s reaching out to others for sup-port, communicating with her husband abouthow she was feeling, and forgiving herself if shefound herself having a hard time.

In the final session, Sara told the therapistthat she had reread her diary entries from thedays before beginning treatment, and she couldnot believe how far she had come, that herpregnancy loss had forced her to seek treat-ment for depression that she now realized hadbeen a lifelong problem; in retrospect, she hadsuffered numerous episodes of mild to moder-ate depression. Sara admitted that she was ini-tially very resistant to the medical model. De-fining depression as a medical illness ultimatelyrelieved Sara of her shame and guilt about herdifficulty in functioning. Furthermore, beingable to see depression as a set of discrete symp-toms made it seem more manageable. Sara re-ported that she was getting along better withher mother; now that she understood depres-sion, she felt more sympathy for her mother’sstruggle with depression. She was grateful forthe opportunity to learn coping skills that shefelt confident about maintaining. In addition,Sara said that she would not hesitate to seektreatment in the future should she find herselfbecoming depressed again.

The therapist’s frequent encouragement, thetime limit, and the brief duration of IPT helpedkeep Sara motivated. Sara said that she appre-

ciated the opportunity to talk about her feel-ings about her pregnancy, her baby, and herbaby’s death, and that she felt the therapist un-derstood and supported her. She recognizedthat her feelings, while powerful, made sense incontext and had subsided with discussion. Saraconfessed that she appreciated the therapist’s“pushing” her to reconnect with others. Shehad not thought she could handle being withothers but was pleasantly surprised.

Although each patient is unique, Sara’s ther-apy resembled other IPT treatments for majordepression and is a good example of workingwith the problem area of grief. The explorationand normalization of affect, communicationanalysis, exploration of options, use of roleplay, encouragement to take social risks, andother techniques employed in Sara’s treatmentare characteristic of working with interper-sonal difficulties related to any of the four IPTproblem areas.

COMMON PROBLEMSTHAT ARISE DURING TREATMENT

The problems that typically arise during IPTtreatment for major depression are (1) those in-herent to working with depressed patients and(2) those related to the therapeutic frame. Al-though these problems are not unique to IPT,how the therapist views and treats these issuesdistinguishes IPT from other psychotherapies.In keeping with important IPT themes, thetherapist attributes problems to depression,and to the patient’s difficulties handling inter-personal interactions and communicating ef-fectively outside of the treatment. The therapistcontinues to maintain an optimistic, sup-portive, and nonjudgmental stance and avoidstransference interpretations.

For example, patients with major depressionsuperimposed on dysthymic disorder (“doubledepression”) and their therapists are often dis-couraged by the chronicity of their depression.In these cases, the therapist should remainhopeful and optimistic. Some depressed pa-tients feel that their depression is incurable de-spite reassurances from the therapist. In thesecases, the IPT therapist employs the medicalmodel, labeling the hopelessness as a symptomof depression, and emphasizes that patientsneed not feel hopeless since depression is treat-able. Depressed patients often view seekingtreatment as a personal failure. The IPT thera-


pist frames seeking treatment as the appropri-ate and smart way to treat a medical illness,and a positive step toward gaining masteryover their problems (Weissman et al., 2000).

The most serious problem to arise in anytreatment for depression is when a patient ex-presses suicidal ideation. In IPT, as in any treat-ment, the therapist determines whether hospi-talization is necessary, based on the seriousnessof the intent and the availability of the patient’ssocial supports. The therapist is as available aspossible to the patient and schedules extra ses-sions as needed. The therapist assesses the cir-cumstances in which the suicidal ideation de-veloped and what the patient hoped toaccomplish by ending her life, and reviews howthe patient imagines others would react to hersuicide. The therapist helps the patient explorealternative ways of expressing what she in-tended to communicate by committing suicide.Furthermore, the therapist provides psychoed-ucation about suicide, explaining that it is themost deadly symptom of depression, and thattherapist and patient have to work together tokeep the patient alive long enough to getthrough successful treatment. Augmenting thetherapy with medication treatment should beconsidered for severely suicidal patients. Thetherapist remains optimistic that the patient’sdepression will improve, and that she willno longer feel as if life is not worth living(Weissman et al., 2000).

Problems may arise in the relationship withthe therapist. For example, a patient who haspoor social supports may come to view thetherapeutic relationship as a substitute for rela-tionships outside of the therapy. Because IPTfocuses on relationships outside of the treat-ment, the therapist gently refocuses the patientto outside relationships. Whereas the therapistnotes that the patient’s ability to connect withinthe treatment relationship reflects her ability toform intimate relationships, he or she clarifiesthat they are not friends or family and empha-sizes the importance of the patient’s life outsideof the treatment. The therapist helps the pa-tient explore options for connecting with oth-ers that are similar to the way she connectedwith the therapist (Weissman et al., 2000).

Missed and late sessions are also consideredto be symptomatic of depression. The therapistcalls attention and empathizes with the factthat the patient has missed or is late to a ses-sion, and notes that difficulty getting to ses-

sions may reflect functioning that is character-istic of depression. It is likely that the patient islate to appointments outside of the therapy.The therapist can remind the patient of thetime limit of IPT to motivate her to come tosessions. In the event that the patient misses oris late to sessions because she is uncomfortablewith the material being discussed in session, orbecause she has some other negative feelingabout the treatment or the therapist, the thera-pist empathizes with the patient’s feelings andhelps her to explore ways to express these feel-ings directly.

When a patient is silent to the extent that itseems she is refraining from sharing thoughtsand feelings, or when she changes or avoidssubjects, or has problems with self-disclosure,the therapist notes this behavior and exploresthe patient’s feelings related to the behavior.These behaviors are particularly problematic ina time-limited, focused treatment such as IPT,because they make it difficult to work on thechosen problem area. The patient may feeluncomfortable and ashamed to share herthoughts and feelings with the therapist. Thetherapist assures the patient that there is littlethat can surprise him or her, and that the pa-tient does not have to talk about everything.

PREDICTORS OF RESPONSE

Although there is repeated evidence that IPT isan efficacious treatment for major depressionand other disorders, there are fewer data onfactors that predict response to IPT. Data fromcomparative treatment studies involving IPTsuggest some clinical predictors of response. Inthe multisite National Institute of MentalHealth Treatment of Depression CollaborativeResearch Program (NIMH TDCRP; Elkin etal., 1989), 250 outpatients with major depres-sion were randomly assigned to 16 weeks ofimipramine (IMI), IPT, CBT, or placebo. TheTDCRP dataset has been examined for predic-tors of response by several researchers. Sotskyand colleagues (1991) found that TDCRP sub-jects with low-baseline-level social dysfunctionresponded well to IPT, whereas those with in-terpersonal deficits responded less well. Thesefindings support the discussion earlier in thischapter that IPT works least well for patientswith few or no social contacts, who report norecent life events, and whose treatment, there-


fore, focuses on the IPT problem area of inter-personal deficits. High initial symptom severityand impaired functioning predicted superiorresponse to IPT and to IMI compared to CBT(Sotsky et al., 1991; Weissman et al., 2000) Inanother analysis, TDCRP subjects with symp-toms of atypical depression, such as mood re-activity and reversed neurovegetative symp-toms, responded better to IPT and CBT than toIMI or to placebo (Stewart, Garfinkel, Nunes,Donovan, & Klein, 1998).

Barber and Muenz (1996) found that amongpatients who completed treatment in theTDCRP study, IPT was more efficacious thanCBT for patients with obsessive–compulsivepersonality disorder, whereas CBT was betterfor patients with avoidant personality disorder,as measured by the HDRS. However, anotherstudy examining the relationship between per-sonality traits and outcome in the same datasetfound no significant differences among person-ality traits (Blatt, Quinlan, Pilkonis, & Shea,1995; Weissman et al., 2000).

In another study, Thase and colleagues(1997) found that among 91 patients with de-pression, patients who had abnormal electro-encephalographic (EEG) sleep profiles had sig-nificantly poorer response to IPT than didpatients with normal profiles. In this study, un-like the Sotsky and colleagues (1991) study,symptom severity did not significantly predictresponse to IPT (Weissman et al., 2000).

Patients taking medication, or those whowish to augment their psychotherapy withmedication, may respond well to IPT, whichshares the medical model of depression withpharmacotherapy, although no research to datehas supported this hypothesis. Additional po-tential predictors of response are noted in thesection describing the IPT patient Sara earlierin this chapter.

CONCLUSION

In summary, IPT is a time-limited, diagnosis-targeted treatment with demonstrated efficacyfor patients with major depression and othermood disorders. It has been adapted for thetreatment of anxiety disorders, eating disor-ders, and, most recently, personality disorders.Although there is substantial evidence of IPT’sefficacy for the treatment major depression andother mood and psychiatric disorders, further

evidence of predictors of response to IPT iswarranted. This chapter has focused on theoriginal IPT protocol: IPT as an individualtreatment for major depression.

IPT employs the medical model and focuseson current or recent life events, interpersonaldifficulties, and symptoms. IPT emphasizes theinterrelationship between mood and life events:Negative or stressful life events affect moodand, conversely, mood symptoms affect howpeople manage negative or stressful life events.Treatment with IPT focuses on one of fourinterpersonal problem areas—grief, role dis-putes, role transitions, and interpersonal defi-cits. The therapist helps patients recover fromdepression by relieving depressive symptomsand by helping them to resolve the chosen in-terpersonal problem area. The IPT therapisttakes an active, optimistic, and supportivestance.

The case of Sara clearly demonstrates thetechniques used and the process of treatmentwith IPT. IPT is an eclectic treatment that usestechniques employed by other psychotherapies.The combination of IPT’s main principles, tech-niques, strategies, therapist and patient char-acteristics, and its proven clinical efficacydistinguishes it from other antidepressant treat-ments.

For further information about IPT and itsadaptations, readers should consult the Com-prehensive Guide to Interpersonal Psychother-apy by Weissman and colleagues (2000).

NOTE

1. For stylistic simplicity, patients are referred to inthe feminine. Indeed, most depressed patients arewomen.

REFERENCES

American Psychiatric Association. (2000) Diagnosticand statistical manual of mental disorders (4th ed.,text rev.). Washington, DC: Author.

Barber, J. P., & Muenz, L. R. (1996). The role of avoid-ance and obsessiveness in matching patients to cog-nitive and interpersonal psychotherapy: Empiricalfindings from the Treatment for Depression Collab-orative Research Program. Journal of Consulting andClinical Psychology, 64, 951–958.



Blatt, S. J., Quinlan, D. M., Pilkonis, P. A., & Shea, M.T. (1995). Impact of perfectionism and need for ap-proval on the brief treatment of depression: The Na-tional Institute of Mental Health Collaborative Re-search Program revisited. Journal of Consulting andClinical Psychology, 63, 125–32.

Blazer, D. G., II. (2000). Mood disorders epidemiology.In B. J. Sadock & V. A. Sadock (Eds.), Kaplan andSadock’s comprehensive textbook of psychiatry(7th ed., Vol. 1, pp. 1298–1308). Philadelphia:Lippincott/Williams & Wilkins.

Bleiberg, K. L., & Markowitz, J. C. (2005). Interper-sonal psychotherapy for posttraumatic stress disor-der. American Journal of Psychiatry, 162, 181–183.

Bleiberg, K. L., & Markowitz, J. C. (2007). Interper-sonal psychotherapy and depression. In C. Freeman& M. Power (Eds.), Handbook of evidence-basedpsychotherapies: A guide for research and practice(pp. 41–60). London: Oxford University Press.

Bowlby, J. (1973). Attachment and loss. New York: Ba-sic Books.

Elkin, I., Shea, M. T., Watkins, J. T., Imber, S. D., Sotsky,S. M., Collins, J. F., et al. (1989). National Instituteof Mental Health Treatment of Depression Collabo-rative Research Program: General effectiveness oftreatments. Archives of General of Psychiatry, 46,971–982.

Fairburn, C. G., Jones, R., Peveler, R. C., Hope, R. A.,& O’Connor, M. (1993). Psychotherapy and bulimianervosa: Longer-term effects of interpersonal psycho-therapy, behavior therapy, and cognitive behaviortherapy. Archives of General Psychiatry, 50, 419–428.

Fairburn, C. G., Norman, P. A., Welch, S. L., O’Connor,M. E., Doll, H. A., & Peveler, R. C. (1995). A pro-spective study of outcome in bulimia nervosa and thelong-term effects of three psychological treatments.Archives of General Psychiatry, 52, 304–312.

Frank, E. (2005). Treating bipolar disorder: A clinician’sguide to interpersonal and social rhythm therapy.New York: Guilford Press.

Frank, J. (1971). Therapeutic factors in psychotherapy.American Journal of Psychotherapy, 25, 350–361.

Hamilton, M. (1960). A rating scale for depression.Journal of Neurology, Neurosurgery, and Psychiatry,2, 56–62.

Hill, C. E., O’Grady, K. E., & Elkin, I. (1992). Applyingthe Collaborative Study Psychotherapy Rating Scaleto rate therapist adherence in cognitive-behaviortherapy, interpersonal therapy, and clinical manage-ment. Journal of Consulting and Clinical Psychology,60, 73–79.

Kessler, R. C., Berglund, P., Demler, O., Koretz, D.,Merikangas, K. R., Rush, A. J., et al. (2003). The epi-demiology of major depressive disorder: Results ofthe National Comorbidity Survey Replication (NCS-R). Journal of the American Medical Association,289, 3095–3105.

Klerman, G. L., Weissman, M. M., Rounsaville, B. J., &

Chevron, E. S. (1984). Interpersonal psychotherapyof depression. New York: Basic Books.

Lipsitz, J. D., Fyer, A. J., Markowitz, J. C., & Cherry, S.(1999). An open trial of interpersonal psychotherapyfor social phobia. American Journal of Psychiatry,156, 1814–1816.

Markowitz, J. C. (1998). Interpersonal psychotherapyfor dysthymic disorder. Washington, DC: AmericanPsychiatric Press.

Markowitz, J. C., Klerman, G. L., Perry, S. W.,Clougherty, K. F., & Mayers, A. (1992). Interper-sonal therapy of depressed HIV-seropositive pa-tients. Hospital and Community Psychiatry, 43,885–890.

Markowitz, J. C., Kocsis, J. H., Fishman, B., Spielman,L. A., Jacobsberg, L. B., Frances, A. J., et al. (1998).Treatment of HIV-positive patients with depressivesymptoms. Archives of General Psychiatry, 55, 452–457.

Markowitz, J. C., Skodol, A. E., & Bleiberg, K. (2006).Interpersonal psychotherapy for borderline personal-ity disorder: Possible mechanisms of change. Journalof Clinical Psychology, 62, 431–444.

Markowitz, J. C., Svartberg, M., & Swartz, H. A.(1998). Is IPT time-limited psychodynamic psycho-therapy? Journal of Psychotherapy Practice and Re-search, 7, 185–195.

Meyer, A. (1957). Psychobiology: A science of man.Springfield, IL: Thomas.

Mufson, L., Moreau, D., & Weissman, M. M. (1993).Interpersonal psychotherapy for depressed adoles-cents. New York: Guilford Press.

Mufson, L., Weissman, M. M., Moreau, D., &Garfinkel, R. (1999). Efficacy of interpersonal psy-chotherapy for depressed adolescents. Archives ofGeneral Psychiatry, 56, 573–579.

Murray, C. L., & Lopez, A. D. (1996). The global bur-den of disease (Vol. 1). Cambridge, MA: HarvardUniversity Press.

O’Hara, M. W., Stuart, S., Gorman, L. L., & Wenzel, A.(2000). Efficacy of interpersonal psychotherapy forpostpartum depression. Archives of General Psychia-try, 57, 1039–1045.

Parsons, T. (1951). Illness and the role of the physician:A sociological perspective. American Journal ofOrthopsychiatry, 21, 452–460.

Sholomskas, A. J., Chevron, E. S., Prusoff, B. A., &Berry, C. (1983). Short-term interpersonal therapy(IPT) with the depressed elderly: Case reports anddiscussion. American Journal of Psychotherapy, 36,552–566.

Sotsky, S. M., Glass, D. R., Shea, M. T., Pilkonis, P. A.,Collins, J. F., Elkin, I., et al. (1991). Patient predic-tors of response to psychotherapy and pharmaco-therapy: Findings in the NIMH treatment of de-pression collaborative research program. AmericanJournal of Psychiatry, 148, 997–1008.

Spinelli, M., & Endicott, J. (2003). Controlled clinicaltrial of interpersonal psychotherapy versus parenting


education program for depressed pregnant women.American Journal of Psychiatry, 160, 555–562.

Stewart, J. W., Garfinkel, R., Nunes, E. V., Donovan, S.,& Klein, D. F. (1998). Atypical features and treat-ment response in the National Institute of MentalHealth Treatment of Depression Collaborative Re-search Program. Journal of Clinical Psychopharma-cology, 18(6), 429–434.

Sullivan, H. S. (1953). The interpersonal theory of psy-chiatry. New York: Norton.

Thase, M. E., Buysse, D. J., Frank, E., Cherry, C. R.,

Cornes, C. L., Mallinger, A. G., et al. (1997). Whichdepressed patients will respond to interpersonal psy-chotherapy?: The role of abnormal EEG profiles.American Journal of Psychiatry, 154, 502–509.

Weissman, M. M., Markowitz, J. C., & Klerman, G. L.(2000). Comprehensive guide to interpersonal psy-chotherapy. New York: Basic Books.

World Health Organization. (1992). The ICD-10 classi-fication of mental and behavioral disorders: Clinicaldescriptions and diagnostic guidelines. Geneva, Swit-zerland: Author.


C H A P T E R 8

Behavioral Activationfor Depression

SONA DIMIDJIANCHRISTOPHER R. MARTELL

MICHAEL E. ADDISRUTH HERMAN-DUNN

New to this edition, the behavioral activation (BA) approach to treating depression has re-ceived very strong empirical support in the past few years, with results as good or betterthan cognitive therapy and antidepressant medications, even for the most severe cases ofdepression. Seemingly counterintuitive at first blush, this treatment does not focus on thenotion of “just do it” but takes a comprehensive look at contingent relationships in the pa-tient’s life across the full range of behavior, cognition, and affect that may be maintainingthe depression. As such, this very idiographic approach does not prescribe a set numberof sessions to accomplish certain goals but is very adaptable to the individual with depres-sion. In the detailed and very human description of the treatment of “Mark,” readers willalso note a very up-to-date focus on the role of rumination in depression (or worry in anxi-ety) as a fundamentally avoidant technique. The illustration of creative therapeutic strate-gies for activating clients and overcoming rumination will be of interest to all therapiststreating depression.—D. H. B.

Behavioral activation (BA) is a structured,brief psychosocial approach that aims to alle-viate depression and prevent future relapse byfocusing directly on behavior change. BA isbased on the premise that problems in vulner-able individuals’ lives, and their behavioralresponses to such problems, reduce their abil-ity to experience positive reward from theirenvironment. The treatment aims to increaseactivation systematically in ways that helpclients to experience greater contact withsources of reward in their lives and to solve

life problems. The treatment procedures focusdirectly on activation, and on processes thatinhibit activation, such as escape and avoid-ance behaviors and ruminative thinking, toincrease experiences that are pleasurable orproductive and improve life context. We be-lieve that BA is an important new treatmentfor depression for two main reasons. First, itsefficacy is supported by recent empirical re-search; second, it is based on simple and eas-ily grasped underlying principles and utilizesa small set of straightforward procedures.

328

BEHAVIORAL MODELSOF DEPRESSION

Basic Concepts

The central premise in virtually all behavioralmodels is the assumption that depression is as-sociated with particular behavior–environmentrelationships that evolve over time in a person’slife. “Behavior” is a very broad construct inthese models and includes everything from tak-ing a walk to grieving over the loss of a lovedone. Behaviors can be fairly circumscribed(e.g., lying on the couch watching television af-ter dinner) or they can be part of a more gen-eral repertoire (e.g., avoiding asserting one’sown needs and desires in conflictual interac-tions).

“Environment” is also a broad constructthat can best be thought of as the settings inwhich behaviors currently occur, as well asthose in which behaviors have evolved overtime. The temporal nature of environmentsis crucial for understanding behavioral ap-proaches to depression. Often, behavior thatappears to serve no function in a person’s cur-rent environment has served a very importantfunction in the past. Thus, when practitionerswonder why a depressed person is engaging ina particular set of behaviors, such as remainingin a dead-end job, it is often necessary to con-sider how particular repertoires (e.g., avoidingpotential losses) have evolved over time.

Finally, all behavioral models of depressionemphasize the importance of contingent rela-tionships between behaviors and the environ-ments in which they occur. Contingent rela-tionships are “if–then” relationships betweenhuman activities and their (often interpersonal)environmental consequences. For example,practitioners working within a behavioralframework are probably less interested in thefact that a depressed client stays in bed eachmorning worrying, for example, about the fu-ture of a troubled marriage, than they areabout the consequences of this behavior. Whathappens as a result? Does the client becomemore or less depressed? By staying in bed, doesthe client avoid something aversive, such asconfronting a spouse about an issue in the mar-riage, or going to work and facing a pile of un-completed tasks? Understanding contingent re-lationships is a central feature of behavioralmodels of depression and a requisite skill of BAtherapists.

Behavioral Roots of BA

These general behavioral concepts were de-veloped and refined into specific conceptualframeworks and treatments for depression byboth Ferster (1973, 1981) and Lewinsohn andcolleagues (Lewinsohn, 1974; Lewinsohn,Antonuccio, Breckenridge, & Teri, 1984;Lewinsohn, Biglan, & Zeiss, 1976). Ferster’sprimary assumption was that depression is theresult of a learning history in which the actionsof the individual do not result in positive re-ward from the environment, or in which the ac-tions are reinforced because they allow the in-dividual to escape from an aversive condition.Over time, behavior that would typically pro-duce positive consequences ceases to do so. Forexample, for a variety of different reasons, aperson’s efforts to form close relationships withothers might gradually fade away, because theyare not followed by positive reinforcement(e.g., reciprocal efforts from others).

Ferster reasoned that this decrease inresponse-contingent positive reinforcementproduces two further consequences that facili-tate depression. First, when people’s effortsdo not result in reward, they often becomemore focused on responding to their own de-privation than to potential sources of positivereinforcement in the environment. This is theclassic “turning inward” that is often seen indepression, and that makes sense from abehavioral perspective; when individuals learnthat their own behavior is an unreliablepredictor of positive consequences in theirenvironment, they naturally spend less timeattending to contingencies in that environ-ment.

The second consequence of decreased ratesof positive reinforcement that Ferster observedwas a narrowing of individuals’ repertoire ofadaptive behaviors. This makes logical sense aswell, because fewer and fewer behaviors are be-ing maintained by positive reinforcement. Indi-viduals may adopt extremely passive reper-toires (e.g., “doing nothing”), because theyhave learned that their active attempts to be-come engaged in life do not produce positiveconsequences.

Finally, Ferster observed that increases inaversive consequences following behavior typ-ically lead depressed individuals to becomepreoccupied with escape and avoidance. In ef-fect, more energy is expended attempting to

Behavioral Activation for Depression 329

avoid or escape from anticipated aversiveconsequences than in attempting to contactpotential positive reinforcers in the environ-ment.

Lewinsohn’s (1974) early behavioral modelof depression was very compatible with manyof the ideas proposed by Ferster. Lewin-sohn similarly emphasized the importance ofresponse-contingent reinforcement and con-ceptualized that its rate was influenced by threefactors: the number of potentially reinforcingevents for an individual; the availability of rein-forcement in the environment; and the instru-mental behavior of the individual required toelicit the reinforcement. Lewinsohn also identi-fied social avoidance as a core part of hismodel. Importantly, Lewinsohn, Sullivan, andGrosscup (1980) developed the first stand-alone behaviorally oriented treatment for de-pression. The Coping with Depression Course(Lewinsohn et al., 1984) is a structured grouptreatment for depression that includes bothbehavioral and cognitive components. Thebehavioral component consists of self-monitoring, and structuring and schedulingspecific pleasant activities that are intended toreverse the decrease in positive behavior–environment transactions.

Beck, Rush, Shaw, and Emery (1979) also pi-oneered early work on BA. Incorporating BAstrategies as a core component of cognitivetherapy (CT) for depression, Beck and col-leagues formalized and widely disseminatedsome of the principal BA strategies. Althoughthese strategies were utilized within a largermodel that emphasized the importance of cog-nition in the etiology and treatment of depres-sion, the formalization and dissemination ofBA within CT advanced work on behavioralstrategies significantly.

Current Conceptualization

The current version of BA was influenced byFerster, Lewinsohn, and Beck in importantways. With respect to a conceptual model ofdepression, the current conceptualizationdraws heavily on the work of Ferster andLewinsohn in emphasizing the central impor-tance of context and activity in understandingdepression. While acknowledging that genetic,biological, and other, distal factors may becausally related to depression, the currentbehavioral conceptualization focuses on the as-

pects of a person’s life context that may havetriggered depression, and particular ways of re-sponding to this context that may be maintain-ing depression. Specifically, the model assumesthat one reason people get depressed is becausechanges in the context of their lives providelow levels of positive reinforcement and highlevels of aversive control. Lives that are “lessrewarding” can lead to feelings of sadness anddepressed mood. And when people get de-pressed, they often pull away from the world inimportant ways, and the basic routines of theirlives become easily disrupted. Both of theseprocesses can increase depressed mood andmake it difficult to solve problems effectively inone’s life. In fact, these processes are conceptu-alized as “secondary problem behaviors,” be-cause they frequently (1) prevent people fromconnecting with aspects of their lives that mayprovide some improvement in mood and (2)prevent people from solving problems that mayhelp to decrease stress and improve life con-text.

The BA approach to therapy addresses bothof the factors that may be contributing to de-pression: those aspects of one’s life that need tobe changed to reduce depression and the wayswithdrawal from the world may be maintain-ing or increasing depression. BA accomplishesthese aims through guided activation, which re-fers to a series of behavior change strategiesthat the therapist and client develop togetheron the basis of a careful examination of whatactivities will be reinforcing for a given clientand will help to disrupt the relationships thatare maintaining the depression. The point ofBA is not to engage in increased activation atrandom or activities that are “generally”thought to be pleasing or to improve mood(e.g., seeing a movie); in contrast, activationstrategies are highly individualized and “cus-tom tailored.” The role of the BA therapist is toact as a “coach” as the client implements theactivation strategies, providing expert help insetting achievable goals, breaking difficulttasks down into manageable units, trouble-shooting problems that arise, and maintainingmotivation during the process of change. Thespecific treatment strategies also draw heavilyfrom the pioneering work of Lewinsohn andBeck. Moreover, in accordance with the pri-mary emphasis of Ferster’s work, avoidantbehavioral repertoires are a primary clinicaltarget in BA.


EMPIRICAL CONTEXT

The first study to revitalize interest in apurely behavioral approach to treating de-pression was conducted by Jacobson and col-leagues (1996), who proposed a simple butprovocative question: Could the behavioralcomponent of CT account for the efficacythat CT has demonstrated in previous clinicaltrials? Adults with major depression wererandomly assigned to one of three treatmentconditions, including BA only, BA plus inter-ventions designed to modify automaticthoughts, and the full CT package. Resultssuggested that BA was comparable to the fullCT package in both acute efficacy (Jacobsonet al., 1996) and prevention of relapse over a2-year follow-up period (Gortner, Gollan,Dobson, & Jacobson, 1998).

On the basis of these findings, BA was devel-oped into a more fully articulated behavioralintervention that included the behavioral as-pects of CT and incorporated the early behav-ioral work of Ferster and Lewinsohn, as de-scribed earlier. This expanded BA model wasarticulated in published reports (Jacobson,Martell, & Dimidjian, 2001; Martell, Addis, &Jacobson, 2001) and in a patient-oriented self-help manual (Addis & Martell, 2004). BA wasfurther tested in a large, randomized, placebo-controlled clinical trial that compared its acuteand long-term efficacy to both CT and antide-pressant medication (ADM).

Two hundred forty-one patients wererandomly assigned to BA, CT, ADM, or pill-placebo, and severity was used as a strati-fication variable during randomization. Thefindings suggested that BA is a particularlypromising treatment for depression (Dimidjianet al., 2006). Among more severely depressedpatients, its performance was comparable tothe current standard of care, ADM, and dem-onstrated better retention. Both BA and ADMwere superior to CT among more severely de-pressed patients. Moreover, follow-up resultsindicate that BA appears to have promising en-during effects (Dobson et al., 2007). Finally,BA may demonstrate an important cost-effectiveness advantage compared to continu-ing patients on medication.

These results are consistent with a range ofother studies suggesting that activation inter-ventions are particularly important compo-nents of cognitive-behavioral treatments. In a

classic study, Zeiss, Lewinsohn, and Munoz(1979) reported comparable outcomes fordepressed patients who received treatment fo-cused on interpersonal skills, pleasant activi-ties, or cognitive change. In addition, in a studywith depressed older adults, Scogin, Jamison,and Gochneaur (1989) found no differences inoutcome between a cognitive bibliotherapytreatment and a behavioral bibliotherapy treat-ment among mildly and moderately depressedolder adults. The importance of behavioralstrategies has also been reported in other stud-ies across multiple diagnostic categories (e.g.,Borkovec, Newman, Pincus, & Lytle, 2002;Foa, Rothbaum, & Furr, 2003; Gloaguen,Cottraux, Cucherat, & Blackburn, 1998) andacross other activation-oriented approacheswith depressed clients (e.g., Blumenthal et al.,1999; Hopko, Lejuez, LePage, Hopko, &McNeil, 2003). Process-oriented research onCT has similarly emphasized the importance ofthe BA components of CT (Bennett-Levy et al.,2004) and has suggested the possibility of neg-ative outcomes when therapists work onchanging clients’ thoughts about interpersonalrelationships as opposed to changing ac-tual interpersonal relationships (Hayes,Castonguay, & Goldfried, 1996).

In addition, Ferster’s early focus on avoid-ance has also been underscored by contempo-rary behavior therapists. Specifically, Linehan(1993) incorporates the use of “opposite ac-tion” for sadness as a means to target depres-sion in dialectical behavior therapy. And Hayesand colleagues (1996) have emphasized therole of experiential avoidance in the develop-ment of a wide range of psychopathologies andin the conceptualization of acceptanceand commitment therapy (ACT; Hayes,Strosahl, & Wilson, 1999). A study on an earlyprecursor to ACT found significant treatmentbenefits for depressed clients in comparison tostandard CT (Zettle & Rains, 1989).

ASSESSING DIAGNOSTIC, CLINICAL,AND FUNCTIONAL DOMAINS

The application of BA is based on a compre-hensive diagnostic, clinical, and functional as-sessment. Some of these assessment activitiesare completed as a precursor to the initiation oftherapy, and others are ongoing throughout thecourse of therapy.


In our treatment outcome research, we useda number of structured diagnostic interview in-struments to assess all five axes of Diagnosticand Statistical Manual of Mental Disorders,fourth edition (DSM-IV; American PsychiatricAssociation, 1994) diagnosis, including theStructured Clinical Interview for DSM-IV AxisI Disorders (SCID-I; First, Spitzer, Gibbon, &Williams, 1997), and the Structured ClinicalInterview for DSM-IV Axis II Personality Dis-orders (SCID-II; First, Spitzer, Gibbons, Wil-liams, & Benjamin, 1996). In addition, we usedtwo measures to assess depressive severity: theclinician-administered Hamilton DepressionRating Scale (HDRS; Hamilton, 1960) and theclient self-reported Beck Depression Inventory–II (BDI-II; Beck, Steer, & Brown, 1996). In rou-tine clinical practice, we recommend conduct-ing a baseline diagnostic interview (although itcan be administered using a less structured for-mat than a research-based interview), and theongoing use of a measure to assess depressiveseverity is important. Typically, the most easilyadministered measure is the BDI-II (Beck et al.,1996).

Also, a number of self-report instrumentscurrently under development will serve as use-ful assessment measures of patient activity leveland of the reward available in a client’s envi-ronment. The Behavioral Activation for De-pression Scale (BADS; Kanter, Mulick, Busch,Berlin, & Martell, 2007) is a 29-item measurethat assesses clients on four subscales: Activa-tion, Avoidance/Rumination, Work/School Im-pairment, and Social Impairment. Initial psy-chometric studies on nonclinical samples havedemonstrated good internal consistency andsufficient test–retest reliability and constructvalidity. A second measure, the EnvironmentalReward Observation Scale (EROS; Armento &Hopko, 2007), is a 10-item scale developed tomeasure the construct of response-contingentpositive reinforcement. Whereas the BADS fo-cuses on activities in which the individual didor did not engage, the EROS measures in-creased behavior and improvement in positiveaffect as a consequence of rewarding environ-mental experiences. Initial psychometric stud-ies indicate that the EROS has strong internalconsistency and good test–retest reliability. Cli-nicians who want a nomothetic measure ofpleasurable activities to augment theidiographic behavioral assessment (describedin more detail below) can use the PleasantEvents Schedule (Lewinsohn & Graf, 1973).

These measures may prove to be useful clinicaltools that can be administered periodicallythroughout the course of treatment to assesschange in activity and reward.

Because functional capacity is such a keycomponent of BA treatment, it is important togather detailed information about the impactof the client’s depression on functional statusacross multiple life domains, including work,family, social, and so forth. These domains canbe assessed using clinical interview or standardassessment instruments (e.g., the Social Adjust-ment Scale [Weissman & Bothwell, 1976] andthe Medical Outcomes Study 36-Item ShortForm Health Survey [Ware & Sherbourne,1992]).

COURSE OF TREATMENT

BA is a theory-driven as opposed to a protocol-driven treatment and, as such, is highlyidiographic in its application. Treatment doesnot follow a required session-by-session for-mat; however, it does follow a general courseover time. This general course comprises thefollowing activities, which are described ingreater detail below:

• Orienting to treatment• Developing treatment goals• Individualizing activation and engagement

targets• Repeatedly applying and troubleshooting ac-

tivation and engagement strategies• Reviewing and consolidating treatment gains

Orienting to Treatment

BA begins with orienting the client to the treat-ment, a process that is typically the focus of thefirst two sessions. The primary tasks to accom-plish during this initial phase of treatment in-clude discussing the BA model of depressionand primary treatment strategies, and provid-ing information about the structure of treat-ment, and the roles and responsibilities of theclient and therapist.

The presentation of the treatment model in-cludes describing the behavioral approach todepression and the process of change duringtreatment, discussing the specific ways themodel fits with the client’s experiences, and en-couraging and responding to questions andconcerns about the model. The model is pre-


sented both verbally during the first sessionand in a brief, written description that is givento the client at the end of the first session. A de-tailed transcript illustrating the presentation ofthe treatment rationale is included in the casestudy below, and the key points to address aresummarized in Table 8.1.

During the second session, therapist and cli-ent again discuss the treatment model and theclient’s reactions. It is essential that the thera-

pist obtain “buy in” from the client with thebasic elements of the treatment model. Therapyworks best if the client accepts and buys intothe rationale for treatment and the case con-ceptualization (Addis & Carpenter, 2000). Thetherapist is advised to guard against movingtoo quickly if the client does not express agree-ment with the key tenets of the model. Thus,during the initial sessions, it is essential that thetherapist encourage the client to ask questionsand elicit the client’s potential doubts and con-cerns.

For instance, some clients may find it diffi-cult to accept the idea that changing behaviordirectly is an effective way to work with de-pression. Often, clients are strongly committedto a biological explanation for their depression.It is not advisable for therapists to debate thisposition; instead, therapists can explain thatthere are many sources of vulnerability to de-pression, and that one of the effective ways tochange depression is by changing what onedoes. At times, clients may also think that theemphasis on behavioral change means that allthey need to do is exercise more, go to a fewmore movies, or take a few more walks, andtheir depression will remit. If clients under-stand BA in this way, then it is not surprisingwhen they feel that the treatment invalidatestheir degree of distress and the difficulty theyface in overcoming depression. We often find ithelpful to talk to clients about two points.First, behavior can have a powerful effect onour moods and people are often unaware ofthis link, particularly when they are depressed.For example, subtly shifting the way someoneapproaches interactions with family membersmay not rid that person of depression, but itcan reduce the severity of depressed mood,thereby setting the stage for other changes thatcollectively help to reverse the depression. Thesecond point is that behavior change is noteasy; if it were, we would all behave exactly aswe think we should all the time (and we knowthis is not the case!). Changing behavior re-quires knowing what to change, and this cantake some serious and sustained “detectivework” in therapy. The therapist can again em-phasize his or her role as a coach in helping theclient to figure out what to change and how todo it.

In addition to presenting and discussing thetreatment model, it is also important to discussthoroughly the structure of treatment and theroles of therapist and client. It is necessary to


TABLE 8.1. Ten Key Points to Address WhenPresenting the Treatment Model

1. Behavioral activation (BA) is based on theidea that the events in your life and how yourespond to such events influence how you feel.

2. BA assumes that one reason people getdepressed is that their lives are providing toofew rewards and too many problems.

3. Sometimes it is possible to identify easilystressors or problems; other times there are noclearly identifiable stressors, but there is stillnot adequate reward from the environment.

4. When life is less rewarding or stressful, peoplesometimes pull away from the world aroundthem and find that basic routines in their livesbecome disrupted.

5. Pulling away from the world when feelingdown is natural and understandable. Theproblem is that it also can increase depressionand make it hard to solve life problemseffectively.

6. In this treatment, we will work together tohelp you become more active and engaged inyour life.

7. BA is not just about “doing more.” If feelingbetter were that easy, you would already havedone it. My expertise lies in figuring out whatactivities would be most helpful and whatsmall and manageable steps you can take toget started. You can think of me as a coach orconsultant to you in the process of change.

8. Each session will involve developing practicaland doable steps to engage in activities thatimprove mood and to solve specific lifeproblems.

9. Between sessions, you will work onhomework assignments that we developtogether; these assignments are an essentialpart of therapy and will focus on reconnectingor building parts of your life that increasefeelings of pleasure or accomplishment andbring you closer to important life goals.

10. Activating and engaging in specific ways canhelp you experience more reward andeffectively solve life problems. When you areactive, engaged, and solving problemseffectively, it is likely that you will be movingtoward important life goals and feeling better.

emphasize three key elements: between-sessionpractice, collaboration, and the structure ofsessions.

1. It is important early in treatment to high-light the active nature of BA and establish theimportance of between-session practice. Thetherapist sets the tone of therapy and expecta-tions by highlighting that BA is a very action-oriented and problem-focused treatment inwhich the majority of the “work” of therapygoes on between sessions, as the client imple-ments plans that therapist and client devise insession. Subsequent sessions examine what theclient has learned, highlight relevant conse-quences of activities, and identify barriers andtroubleshoot to maximize the likelihood of suc-cess on future attempts. The assignment ofhomework begins in the very first session,when the therapist asks the client to read thebrief pamphlet that reviews the key ideas of theBA approach (see Martell et al., 2001).

2. Therapists should explain that BA is acollaborative approach to therapy, in whichtherapist and client work together in the serviceof the client’s goals.

3. Therapists should explain the structurednature of each treatment session. Although thespecific foci of each session vary across clientsand across time, each session nonetheless fol-lows a similar general outline. Specifically, as incognitive therapy for depression (Beck et al.,1979), each session begins with therapist andclient setting an agenda in a collaborative fash-ion. The aim of the agenda is to organize thetime effectively and to ensure that sessions ad-dress the client’s most important topics and havemaximal likelihood of helping the client reachkey goals. Because the goal of treatment is acti-vation, the more control given to the client insetting the agenda, the better. In addition, giventhe integral role of homework in BA, the major-ity of each session is devoted to reviewing home-work from the previous session and assigninghomework for the next session. The ending ofeach session includes asking the client to reiter-ate a “take-home message” from the session,verifying that the client has a clear understand-ing of the assigned homework, and reviewingthe time for the next scheduled session andmethods to reach the therapist in the interim.

In attending to each of these specific tasks,the therapist must also establish his or her cred-ibility as an expert who can help the client.

Doing so requires conveying a sense of hopeand establishing a strong foundation of collab-oration in which the client and therapist worktogether toward the client’s goals.

Developing Treatment Goals

The ultimate goal of BA is to help clients mod-ify their behavior to increase contact withsources of positive reinforcement in their lives.Typically, this process involves first addressingbasic avoidance patterns and areas of routinedisruption and, second, addressing short- andlong-term goals. A good deal of the therapist’stime in BA is spent helping clients to increaseawareness of their escape and avoidance reper-toires and to replace them with active copingresponses. Often, part of this process also in-volves changing basic routines (e.g., sleep, eat-ing, and social contact). Thus, short-term goalsare set to specify concrete accomplishmentsthat can help clients shift their life situations ina less depressing direction. Common examplesmight include getting the house cleaned, spend-ing time with friends and family, making head-way on a large pile of paperwork on which theclient has procrastinated, exercising more fre-quently, and so on. Often, behaviors necessaryfor progress on short-term goals can be sched-uled and structured in such a way that theysubstitute for maladaptive secondary copingresponses. A therapist works with a client tomake progress on short-term goals regardlessof how the client is feeling. In other words, oneof the therapist’s primary goals is to help theclient change the pattern of having his or herbehavior governed by mood. Rather, the objec-tive is to make progress on goals regardless ofhow one feels at a particular point in time,under the assumption that progress on lifegoals is itself antidepressive. This is a crucialpoint, because clients often judge the success ofa particular behavior change based on how itmakes them feel in the moment. Thus, it is im-portant for therapists to remain mindful of theconsequence of different behaviors specificallywith respect to whether they help clients movetoward or achieve treatment goals.

Once the short-term goals relating to avoid-ance, withdrawal, and routine disruption havebeen addressed, clients are assisted in address-ing larger life circumstances that may be re-lated to depression. Larger life goals are thoseshifts that take time to accomplish but have thepotential to alter a person’s life situation in sub-


stantial ways. Common examples include find-ing a new job, getting out of a distressingrelationship, starting a new relationship, ormoving to a new city or town. BA can be help-ful in teaching and encouraging clients to con-tinue to make progress toward these goals eventhough their actual realization may take sometime to accomplish. Essentially, in BA, clientslearn basic change strategies that can be used toaccomplish short-term and larger life goals.

Individualizing Activationand Engagement Targets

No two people are the same, and no person isexactly the same from situation to situation. Inthe context of treating depression, this meansthat the particular activation strategies thatwork for one person may not work for anotherperson, and they may work for the same personsometimes, but not always. Consider the be-haviors associated with a client meeting afriend for lunch, which might include getting inthe car or walking to a restaurant, talking witha friend about current events, sharing details ofhis or her personal life with the friend, eating,worrying about what the friend thinks of himor her, and so on. Each of these different reper-toires of activity can be associated with positiveor negative shifts in mood depending on a per-son’s particular history and the way events un-fold in the present. Whether these activity–environment transactions shift someone’smood in a positive or negative direction is anempirical question, and one that requires care-ful attention and assessment to answer.

Much of the skill of doing competent BA re-lies on just this kind of careful examination. InBA, this process is referred to as “functionalanalysis,” which is the key to individualizingactivation targets and the heart of BA. Func-tional analysis involves identifying for each cli-ent the variables that maintain the depressionand are most amenable to change. This under-standing forms the basis of the case conceptual-ization and guides the idiographic applicationof specific activation strategies. In general, thetherapist must engage the client in a detailedexamination of the following:

• What is maintaining the depression?• What is getting in the way of engaging in and

enjoying life?• What behaviors are good candidates for

maximizing change?

This process sounds simple, yet in practice, itcan be complicated given that we all frequentlylack awareness of the contingencies that con-trol our behavior. Given this reality, early intreatment, we talk explicitly about the goal(and the challenge!) of identifying these rela-tionships.

A number of steps are involved in identifyingthe contingencies that control behavior. First,therapist and client must clearly and specifi-cally define the behavior of interest; this in-cludes defining the frequency, duration, inten-sity, and setting of the behavior. The behaviorsof greatest interest are those most closely tiedto changes in mood. Often, therapists are un-certain about how to sequence treatment tar-gets; generally, we focus initially on behaviorchange that has the greatest likelihood of suc-cess, which may be based on ease of accom-plishment or level of importance to client goals.For instance, for one client, difficulty complet-ing basic household tasks was maintaining andexacerbating his experience of depression. Itwas important to specify the problems of(1) filling and closing bags of household trash,then leaving them in the pantry room ratherthan taking them to the dumpster behind thehouse for the past 6 weeks; and (2) receivingbills in the mail and placing them unopened ina file drawer for the past 4 months. Treatmentwith this client began with a series of gradedtasks to accomplish the goal of taking out thetrash. This was selected as an initial focus, be-cause the therapist and client decided it wasmore easily accomplished than addressing thebills, which the client experienced as over-whelming. Next, the therapist and clientworked to identify the antecedent and conse-quences of the behavior, with the goal of speci-fying the variables that elicited or reinforcedthe depression.

Understanding basic behavioral principlescan often be of great value in identifying thesecontingent relationships. What does this meanin practical terms? For depressed clients, a num-ber of contingent relationships are often ob-served. First, as we discussed earlier, negative re-inforcement contingencies are often pervasive.Negative reinforcement means that the likeli-hood of a behavior occurring is increased by theremoval of something from the environment,typically an aversive condition. Negative rein-forcement contingencies can be a very adaptivepart of human behavior. For instance, putting ona warm coat to avoid getting cold, stopping at


stop signs to avoid getting in accidents, and be-ing extra nice to a parent after crashing the car toavoid being punished are all examples of waysthat negative reinforcement can serve one well.Unfortunately, however, when individuals be-come depressed, their behavioral repertoires canbecome dominated by escape and avoidance be-haviors that temporarily allow a person to es-cape from painful feelings or difficult interper-sonal situations. In this way, many avoidant andescape behaviors may be understood as second-ary coping responses—efforts to cope with theexperience of depression that, unfortunately,make it worse. For example, a person might at-tempt to escape from feelings of hopelessnessand fatigue by taking a long nap in the middle ofthe afternoon. This might, as a consequence,temporarily remove the person from an aversivecontext, but it can also prevent him or her fromtaking the steps necessary to shift into a less de-pressive context overall (exercising, applying forjobs, cleaning the house, etc.). Substance abuse,excessive sleeping, watching too much televi-sion, and general inactivity are all common ex-amples of secondary coping responses that canbe maintained by negative reinforcement. Thesetypes of negative reinforcement contingenciesare often central in preventing people from com-ing into contact with potentially reinforcing en-vironments that contribute to leading a moreadaptive and engaged life. Careful assessment isrequired to identify whether and how specificnegative reinforcement contingencies are activein a client’s life.

Second, positive reinforcement contingen-cies may also be problematic for clients. Inthese cases, the likelihood of a behavior is in-creased, because it is contingently associatedwith positive consequences. For instance, go-ing to bed early may be positively reinforcedby family members offering empathy andsupport. Some behaviors, such as overeatingand substance abuse, can also provide imme-diate positive reinforcement but detract fromlong-term goals, thereby maintaining the de-pression.

How do we conduct functional analyses inBA? Activity monitoring is the heart of the pro-cess. Through detailed and ongoing activitymonitoring that the client completes betweensessions, therapist and client can work togetherto develop an understanding of the questionslisted earlier. Given the role of activity monitor-ing, it is important early in treatment for the

therapist to set the stage to clearly explain howto do it, and carefully and skillfully reviewmonitoring tasks to reinforce the client’s effortsto begin developing activation and engagementassignments.

Therapists may elect to use a number of dif-ferent formats for activity monitoring, the mostbasic form of which includes recording both anactivity and a mood rating for each hour of theclient’s waking day. Time sampling proceduresmay also be used. In such a procedure, clientand therapist agree on a specified number ofhours when activities will be monitored duringthe week between sessions. Time sampling pro-cedures need to include a variety of situationsin which the client functions during the week.Therefore, times during the early morning, thework day, and the evening and weekend shouldbe set aside for monitoring.

Therapists can provide clients with a simpleweekly Activity Record for monitoring assign-ments (e.g., see Figure 8.1). However, as is of-ten emphasized in BA, the function of thebehavior is more important than its form; thus,clients may eschew monitoring on an ActivityRecord but be comfortable using personal cal-endars, personal digital assistants, or other id-iosyncratic recording methods. Adapting themonitoring assignment to a form that is com-patible with the client’s normal daily routine ishighly encouraged. For instance, some clientswho spend a lot of time in their car may keeptheir records above the visor; others may find ithelpful to keep it taped to the refrigerator orbathroom mirror; still others find it useful tocarry it with them in a pocket, purse, back-pack, or briefcase. In general, therapists willwant to encourage clients to complete entries atregular intervals throughout the day and areadvised to work with clients to develop plansfor recording that take advantage of naturalcues and rhythms in their lives to facilitate suchregular recording.

When a client has completed an activitymonitoring assignment, it is essential for thetherapist to review it in detail. Failing to reviewactivity monitoring records may mean missedopportunities to reinforce the client’s behaviorand to develop the case conceptualization.Much of competent BA lies in skillfully review-ing a completed Activity Record. What doesthe competent therapist attend to when review-ing the records? In general, the therapist willwant to keep the case conceptualization ques-


tions listed earlier in the forefront of his or hermind when reviewing a completed activity log.The therapist reviews the Activity Record tounderstand the client’s activities, routines, andlife context, and to begin to identify patterns

that may be maintaining or exacerbating de-pressed mood.

Specific questions that therapists can use toguide their review of activity schedules arelisted below:


Instructions: Record your activity for each hour of the day, and record a mood rating associated witheach activity. Use the scale below with the anchors that you and your therapist develop to guide yourmood rating. Aim to make entries on your Activity Record at least every 3–4 hours each day.

Monday Tuesday Wednesday Thursday Friday Saturday Sunday

5–6

6–7

7–8

8–9

9–10

10–11

11–12

12–1

1–2

2–3

3–4

4–5

5–6

6–7

7–8

8–9

9–10

10–11

11–12

12–1

1–2

2–3

3–4

4–5

Mood Ratings: 0: Feeling really good; not depressed at all (Examples of associated activities:)

5: Intermediate (Examples of associated activities:)

10: Feeling the worst (Examples of associated activities:)

FIGURE 8.1. Activity Record (hourly).

• What would the client be doing if he or shewere not depressed (working, managingfamily responsibilities, exercising, socializ-ing, engaging in leisure activities, eating,sleeping, etc.)?

• Is the client engaging in a wide variety of ac-tivities, or have his or her activities becomenarrow?

• What is the relationship between specific ac-tivities and mood?

• What is the relationship between specific lifecontexts/problems and mood?

• In what ways are avoidance and withdrawalmaintaining or exacerbating depression?What does the client avoid or from what ishe or she pulling away? In what specificways? Are there routine disruptions?

• Where has contact with reinforcers beenlost?

• Are there deficits in coping skills and strate-gies?

To answer these questions, it is essential tofocus on the parts of the client’s activity andcontext that change and the parts that are con-sistent over time. It is not uncommon for de-pressed clients to report that their mood is al-ways low, no matter what they are doing orwhere they are doing it. They may describe thatthey simply feel “blah” all the time or that, per-haps, their mood changes, but such changesare minor or irrelevant. A central premise ofBA, however, is that variability is everywhere,though sometimes it is difficult to detect.Moreover, the variability is not random. In-stead, variations in behavior and its settingshave a direct effect on a person’s mood and, assuch, provide critical information about centralcontingencies. When clients report that theyfeel depressed “all the time,” it is either becausethey are inaccurately reporting on their moodretrospectively, their behavioral repertoires areextremely narrow (e.g., lying in bed all daylong), or they have not learned to discriminatebetween subtle differences in mood. This lastpoint is a critical one. One of the key tasks fortherapists is to help clients understand thattheir moods are intimately linked to what theyare doing, where they are doing it, and the re-sulting consequences. Treating depression re-quires making a series of strategic changes ineach of these domains, all of which are basedon understanding the basic contingent relation-ships.

Repeatedly Applying and TroubleshootingActivation and Engagement Strategies

Given the idiographic nature of BA, the courseof treatment may look quite different across arange of clients. Despite this diversity, a few,straightforward behavioral methods are fre-quently used. These are discussed below.

Activity Scheduling and Self-Monitoring

The major work of therapy in BA occurs be-tween treatment sessions. It is uncommon forclients to leave a session without some specificactivities with which to experiment during theweek. Thus, activity scheduling and monitor-ing of outcome are standard methods usedthroughout treatment. At the end of each ses-sion, the client should clearly understand thespecific activity assignment and have a clearstrategy for implementing it during the week.

Specifically scheduling the activity is a usefultool for having the client commit to times whenhe or she will do the homework. When the as-signment is listed in writing on a particular dayof the week, at a particular hour, the client hasthe benefit of an external aid to motivatebehavior change (i.e., working from the “out-side-in” as opposed to the “inside-out”). Activ-ity scheduling is also used frequently for clientswith significant routine disruptions. Schedulescan be an aid in the effort to develop and fol-low regular routines for eating, working, sleep-ing, exercising, and maintaining social contact.Depending on the activity, the client may notspecifically schedule the time, but may insteadrecord its completion on a daily log.

A key part of scheduling activities frequentlyinvolves careful attention to contingency man-agement. As social psychologists have longknown, “The correlations between intentionand behavior are modest . . . the weak inten-tion–behavior relation is largely due to peoplehaving good intentions but failing to act onthem” (Gollwitzer, 1999, p. 493). Given thisreality, it is essential for BA therapists to con-sider ways they can help clients structure theirenvironment so as to maximize success withtreatment assignments and goals. One helpfulmethod of contingency management is the useof public commitment to enhance the likeli-hood of completing assignments (Locke &Latham, 2002). Often, we explore whetherfriends, coworkers, or family members are


available to be included in activation plans. Forinstance, one of our clients learned in treatmentthat when his depression was more severe, itwas essential for him to tell his wife each morn-ing the primary tasks he intended to completethat day. For him, “keeping his word” was aneffective reinforcer that helped to increase hisactivation. In addition, we work with clients tostructure their environments in ways that willenhance activation. Thus, for a client whoworks on an exercise plan during the week,putting on her exercise clothes before sheleaves work at the end of the day is an essentialpart of the plan. Finally, clients may also exper-iment with the use of arbitrary reinforcers forspecific behavior change tasks. Although theemphasis of BA is heavily on increasing contactwith natural reinforcers in one’s environment,the selective use of arbitrary reinforcers can attime be helpful. The client we noted earlierplanned a special dinner with his wife at theend of a week of adherence with activationtasks; the client who was working on the exer-cise program bought a new shirt for exercisingafter she started her new program.

Occasionally, therapists may suggest the useof aversive contingencies to help promote be-havior change; for instance, one client found ituseful to agree to call his therapist if he was goingto stay in bed and miss work for the day. We havedescribed to other clients the method of writingdonation checks to their least favored charities,which are then cashed if they do not completescheduled activities (Watson & Tharp, 2002).Often, simply the suggestion of an aversive con-tingency is sufficient to motivate change. For in-stance, the client who agreed to call his therapistdid, in fact, call early one Monday morning;when he was halfway through leaving a message,he stated, “This is ridiculous. Forget it, I am go-ing work.”

When clients do activity scheduling, it isalso essential to build in a monitoring com-ponent such that they record the context andconsequences of activation. This providesboth information about the specific activationassignment, and regular and ongoing practicein noticing contingent relationships betweenactivity and mood. It is essential for the ther-apist to discuss what the client learned fromthe activation and self-monitoring tasks ineach session. In addition, the therapist needsto provide regular feedback about progressand to highlight areas of improvement or

troubleshoot problems that may have arisen.It is essential for the therapist not to shyaway from asking about homework that isincomplete or not done. Attending to thesedomains is exactly the job of the BA thera-pist. Repeated and persistent focus on a smallset of activation tasks often occupies a greatdeal of the course of treatment. Asking aboutwhat prevented the completion of homeworkgives the therapist and client essential infor-mation about important barriers and possibleexamples of avoidance patterns. The purposeof such discussions is not to punish or shamethe client, so it is important to approachhomework review with a direct andnonjudgmental attitude. At the same time,discussing incomplete or partially completedhomework may be experienced as aversive bythe client, which may help to facilitate his orher doing the homework next time (i.e.,through being negatively reinforced by escap-ing from the therapist’s uncomfortable ques-tioning). If this occurs naturally, it is not aproblem and can in fact enhance treatmentprogress. However, the therapist should neveruse shame or criticism, however subtly, to en-hance homework completion.

Graded Task Assignment

Graded task assignment, which is a core partof cognitive therapy for depression (Beck etal., 1979), is also a hallmark of BA and a keypart of most activity scheduling efforts. Thus,it is important for therapists to help clientsbreak down behaviors into specific, achiev-able units to facilitate successful behaviorchange. Knowing how to break down tasksand grade them appropriately in a stepwiseprogression from simple to complex requirestherapists to use a broad array of basic self-management and problem-solving skills. To-ward this end, it is also important for thera-pists to help clients learn the method ofgrading tasks during therapy, so that they canapply this skill to new contexts and tasks af-ter therapy has ended. In explaining the toolof graded task assignment, it is often impor-tant to remind clients that the goal is not toaccomplish all parts of the activity; rather, thegoals are to get started on important tasks,increase activation, and disrupt avoidance.Therapists can also explain to clients thatbreaking tasks down helps to ensure success


with subtasks, and such success experiencescan in turn reinforce and motivate work onsuccessive components of the larger task.

It is very important to break down tasks insuch a way that early success is guaranteed. Ifclients experience difficulty with tasks, thera-pists may want to revisit explicitly whether thetask was broken down and graded sufficiently.Clients also can be asked to envision the stepsinvolved in a given task before attempting itoutside of therapy, and to anticipate any obsta-cles that might arise. If this process suggeststhat certain components may be difficult tomaster, the therapist and client can grade thattask into smaller and more achievable compo-nents.

Avoidance Modification and Problem Solving

This is perhaps one of the most rich and var-ied aspects of BA. As noted earlier, clientsmay be addressing avoidance of concretetasks at work or home, avoidance of painfulemotions such as grief or fear, avoidance ofinterpersonal conflict, and so forth. The spe-cific methods used to address these areas aretied to the specific nature of the avoidance.For instance, a client who avoids the experi-ence of grief over a lost relationship may beassisted in spending time each day reviewingphotos of the former partner, reminiscingabout times they shared, and so forth. A cli-ent who avoids tasks at work may be assistedin breaking down tasks, making specific to-do lists, asking others for help, and so forth.A client who avoids interpersonal conflictmay be assisted in practicing assertive com-munication in role plays in sessions, experi-menting with discussions with friends, bring-ing a family member to a therapy session,and so forth. That said, within this broad do-main, a number of basic strategies may behelpful in planning treatment.

First, it is essential when addressing avoid-ance to start from a collaborative stance withthe client. It is important to understand andcommunicate an understanding of the discom-fort that a client may experience in a particularsituation, which is then followed by some ac-tion on the part of the client to end the aversiveexperience. Therapists can emphasize the waysthat avoidance may serve an adaptive functionin the short term but be problematic in the longterm. Given that many avoidant behaviors may

be under the control of immediate contingen-cies, it is often helpful to highlight repeatedlythe long-term consequences of particular be-haviors.

Second, basic problem-solving methods arefrequently used to address avoidance. Al-though we rarely teach problem-solving stepsin a structured and formal manner, addressingavoidance typically involves figuring out howto approach and solve problems. Thus, it is es-sential for therapists to maintain a problem-solving mindset with respect to avoidance andto work with clients to generate a range of op-tions for possible alternative coping behaviors.Problem-solving strategies include defining andassessing the problem, generating alternativesolutions, managing environmental contingen-cies, and trouble-shooting the solution whenneeded. It should also be noted that all of theother basic strategies are also frequently usedin the service of avoidance modification andproblem solving (e.g., activity scheduling andmonitoring, graded task assignment, etc.).

Third, to help maintain a consistent focus onavoidance, clients may be assisted by severalmnemonic devices. These devices help to orga-nize a method of examining, “What is the func-tion of a behavior; what are its consequences?”We have used the acronym ACTION to iden-tify the general approach that we ask clients totake (Martell et al., 2001):

• Assess—Is this behavior approach or avoid-ance? Will it be likely to make me feel betteror worse?

• Choose—Either choose to continue thisbehavior, even if it makes me feel worse, ortry a new behavior.

• Try—Try the behavior chosen.• Integrate—Any new behavior needs to be

given a fair chance, so integrate new behav-iors into a routine before assessing whether ithas been helpful or not.

• Observe the Results—Pay close attentionand monitor the effects of the new behavior.

• Never Give Up—Remembering that makingchanges can often require repeated effortsand attempts.

When clients demonstrate behavioral pat-terns characterized by avoidance behaviors,therapists may use the acronym of being in aTRAP (Martell et al., 2001):


• Trigger—usually something in the environ-ment (e.g., criticized by employer)

• Response—usually an emotional reaction(e.g., feels shame and sadness)

• Avoidance Pattern—The avoidance behav-ior(s) used to cope with the emotional re-sponse (e.g., leaves work early, complains topartner about displeasure with job)

Therapists then ask the client to get out ofthe TRAP and back on TRAC; that is, underthe same trigger and response conditions theyare asked to experiment with “AlternativeCoping” behaviors.

Engagement Strategies

Research has well documented the frequencyand negative consequences of ruminative be-havior among people with depression (Nolen-Hoeksema, 2000). The BA approach considersruminating as a behavior that frequently pre-vents people from engaging fully with their ac-tivities and environments. BA therapists are onalert for client reports of ruminating and arealso careful to assess whether ruminating is aproblem when clients return to sessions re-porting that an activation assignment “didn’twork.” If, for example, a client reports that sheexperienced little mood improvement whileplaying with her son in the park, the therapistwill want to examine whether she was onlypartially engaged with the task of playing be-cause her mind was focused on why her formerpartner did not want to continue their relation-ship, what that meant about her value as awoman, and so forth.

In addressing rumination in BA, we are lessinterested in the particular content of rumina-tive thoughts than in the context and conse-quences of rumination. For instance, in work-ing with a client who frequently ruminatedabout a job that he regretted having declined,the BA therapist asked him to examine the fol-lowing types of questions: What were you do-ing while you were thinking about the otherjob? How engaged were you with the activityof the moment and with your surroundings?What happened during and after your thinkingabout the other job? In examining these ques-tions, the therapist and client identified that theclient was more likely to ruminate when he wasworking on aversive tasks in his current job,about which he felt significant anxiety; thus,

ruminating was negatively reinforced by dis-tracting the client from anxiety and decreasinghis focus on the aversive tasks. BA therapistsmay ask clients who ruminate frequently to ex-periment with “attention to experience”practice, in which they deliberately focus theirattention on their current activity and sur-roundings. For instance, they may be asked tobring their full awareness to physical sensa-tions (colors, sounds, smells, tastes, physicalmovements, etc.). These strategies are akin tomindfulness practices (Segal, Williams, &Teasdale, 2001) and are also very consistentwith the dialectical behavior strategy of “oppo-site action all the way” (Linehan, 1993).

Reviewing and Consolidating Treatment Gains

As the therapist and client agree that there hasbeen sufficient improvement and terminationseems indicated, remaining sessions should fo-cus on relapse prevention, which largely in-volves reviewing and consolidating gains theclient has made. Toward the end of therapy, itis often wise to focus on anticipating situationsin the client’s life that may trigger depressivefeelings and behaviors, and to generate plansfor coping with such situations. Upcoming lifeevents that can be anticipated as challenges forthe client (e.g., the death of a parent, a careerchange) should be discussed in detail, and theclient can draft a self-help plan to activate inthe face of the stressor. In addition, it is alsovery important to review the basic methodsused in therapy to ensure that the client is leav-ing with a solid understanding of how to applythese as tools in the future. For instance, thera-pists will want to review how to approachmood from a behavioral perspective (e.g., iden-tifying the pattern of antecedents and conse-quences), how to set specific and concretegoals, how to use graded task assignment, andhow to identify and target primary avoidancepatterns (noticing that one is ruminating moreat work, noticing that one is starting to fail toreturn phone calls of friends, etc.).

The Therapy Setting

BA therapy has been most widely investigatedin the context of outpatient individual psy-chotherapy settings. Treatment duration hasranged from 12- to 24-session formats over 16weeks. In our research protocol, sessions were


offered twice per week during the first 8 weeksof treatment and once per week for the remain-ing 8 weeks. This intensity of sessions accom-modated comparisons with other treatmentsincluded in our research designs; however, limi-tations of client resources, insurance reim-bursement, and other factors may make such acourse of therapy impractical in many treat-ment settings. Twice-weekly appointments arealways be recommended in the beginning oftherapy with severely depressed clients. Whendoing BA in settings that are not compatiblewith this level of frequency, we often schedulebrief phone contacts between sessions to rein-force the interpersonal connection to the thera-pist and the importance of completing thebetween-session assignments, and to providean opportunity to address questions or prob-lems with the between-session assignments. Al-though BA is an individual treatment modality,significant others in clients’ lives are often in-cluded in sessions across the course of treat-ment. Decisions about whether to include sig-nificant others in treatment follow from thefunctional analysis; thus, they are determinedon an idiographic basis. Conjoint sessions canbe helpful in assessing patterns that may bemaintaining the client’s depression, providinginformation and education to the family mem-ber about the treatment rationale and ap-proach, and providing opportunities for theclient and family member to practice new inter-personal behaviors with the therapist’s directand immediate feedback.

BA has also been applied in a group format.Peter Lewinsohn and colleagues developed theCoping with Depression Course for adoles-cents (Lewinsohn et al., 1984). This treatmentprogram includes activity scheduling, relax-ation training, assertiveness and social skillstraining, and cognitive restructuring. Admit-tedly, the inclusion of cognitive restructuring inthe Coping with Depression Course places thisprogram in the broader area of CBT ratherthan specific BA, but the emphasis is on sched-uling pleasant events.

Therapist Qualities

In our experience, a number of qualities are im-portant for BA therapists to bring to their clini-cal work. All of the BA strategies require thebasic skill of being a good problem solver.Therapists approach most matters that arise intreatment simply as problems to be solved.

Thus, therapists must be naturally curiousabout factors that maintain problems and beskilled at generating a range of more functionalalternative behaviors. Toward this end, it is im-portant to be comfortable being concrete, spe-cific, and structured in one’s approach to ther-apy. In addition, therapists also need to becomfortable with a direct, matter-of-fact, andnonjudgmental manner of communication.This is essential in reviewing homework assign-ments effectively and troubleshooting prob-lems that may arise. It is important to be skilledin maintaining a persistent focus on activationassignments within and across sessions. Thera-pists must also balance a genuine sense of em-pathy and understanding for the suffering andstruggles of their clients with an optimisticand dogged commitment to the possibility ofchange. Finally, understanding basic behavioralconcepts and principles can help therapistsconceptualize cases according to a behavioralmodel of depression and present a coherentframework to clients.

In our treatment outcome research, we haveused ongoing clinical supervision and therapistconsultation teams to assist therapists in adher-ing to and refining these important therapistqualities. Therapists typically meet togetherweekly for 1–2 hours. The teams help thera-pists increase skills with basic treatment strate-gies (conceptualizing treatment plans, doingfunctional analysis, effectively grading tasks,etc.). In the context of a focus on specific casesand strategies, the team also provides essentialreinforcement for the therapist qualities of em-pathy, nonjudgment, problem-solving curiosityand persistence, and optimism about change.Although we have not tested this hypothesisempirically, we strongly suspect that it wouldbe difficult for many therapists to maintainthese qualities in their work with clients strug-gling in particular with severe, complicated, orchronic depression without the support of aneffective consultation team.

Client Variables

Our treatment outcome research suggests thatdepressed adults can experience acute and en-during clinical benefit from a course of BA.Our current work explores what specific char-acteristics help to predict whether a client willrespond to BA. Our clinical impressions arethat engagement with the basic treatment ratio-nale and willingness to complete homework as-


signments are both important predictors ofoutcome in BA. Moreover, it is important tonote that we used a number of client character-istics as exclusionary criteria in our treatmentresearch. For instance, if a client was acutelysuicidal, such that his or her risk could not bemanaged on an outpatient basis, then wewould refer him or her for more acute and in-tensive treatment. In addition, if a client had acomorbid diagnostic disorder that was moresevere and prominent (more interfering and theprimary focus of the client), and necessi-tated another evidence-based treatment (e.g.,obsessive–compulsive disorder), then we wouldalso refer him or her to a more specifically ap-propriate treatment. We also carefully evalu-ated any potential medical problems that mayhave contributed to depression and referred cli-ents for appropriate concurrent medical treat-ment, if necessary.

CASE STUDY

Background Information

The following section presents the treatment ofMark, a 43-year-old man with a long history ofdepression, who sought treatment after the endof his second marriage. Mark was in treatmentfor 19 sessions across 4 months. The descrip-tion here is presented to illustrate the imple-mentation of core BA principles and strategies.Earlier sessions are described in greater detailto provide the reader with “how-to” informa-tion regarding the primary principles and strat-egies. Later sessions emphasize a thematic fo-cus for which the same types of principles andstrategies are applied. It is important to empha-size at the outset that this case description isnot intended to communicate a prescriptivecourse of treatment, and readers are advisedagainst following the sequence of strategies in alockstep fashion. BA is a highly idiographictreatment in which the choice of specific activa-tion strategies is driven by functional analysis;given this, the reader is encouraged to attend tothe ways the therapist conceptualizes Mark’sdifficulties and implements treatment strategiesover the course of therapy. It is our hope thatthis detailed illustration will inspire readers toapply the basic principles and core strategies ina flexible and idiographic manner.

Mark sought treatment at the urging of hisprimary care physician. His recent episode ofdepression had lasted without remission for 3

years. Mark also had a history of alcoholabuse. His early alcohol abuse had caused sig-nificant problems in Mark’s first marriage,which ended in divorce when he was in hisearly 20s; however, problems with alcoholwere not a cause of current concern. He hadbeen in therapy previously, during his separa-tion and divorce 4 years earlier. However,Mark described it as unstructured and un-focused, and reported that he stopped going af-ter a few sessions. He lived alone, although hehad joint custody of his twin adolescent daugh-ters; he and his ex-wife alternated parenting ev-ery other week.

Mark reported that he had had periods ofdepression for “as long as I can remember.” Inparticular, he recalled his first episode of de-pression at age 12, shortly after his fatherabruptly left and severed all contact with Markand his family. Mark reported that he had be-lieved that his parents were happily marriedand, at that time, blamed himself for his fa-ther’s departure. Mark reported that hismother and older siblings never discussed hisfather. In describing his mood during adoles-cence and adulthood, he reported, “I have peri-ods when I’m able to function OK. I go to workand all that, but I’m never really happy.”Mark’s primary depressive symptoms includeddepressed mood, loss of pleasure in nearly allactivities, excessive guilt, fatigue, difficultyconcentrating, and occasional passive thoughtsof death.

Mark had had a social network that re-volved primarily around his former marriage,but he had been withdrawn from that networksince his separation and divorce. Currently, hespent most of his time alone, with the excep-tion of caring for his daughters. Mark was col-lege educated and worked as an accountant fora local manufacturing company. He also wrotechildren’s stories and, prior to his most recentepisode of depression, was working on a num-ber of stories as a member of a local writer’sgroup.

Case Conceptualizationand Overview of Treatment

Mark’s depression was conceptualized as beingcontrolled by a pattern of interpersonal avoid-ance that was negatively reinforced by reduc-tions in grief and anxiety. Specifically, Markhad trouble fully engaging in his significant re-lationships and, instead, avoided intimacy in


various ways that included ruminating aboutmistakes he made in the past, and failing to ex-press commitment to the relationship and whathe thought or felt about various topics on aregular basis. Over the course of treatment, thetherapist and Mark hypothesized that avoidingclose interpersonal connections in his adult lifekept Mark detached enough that he would notfeel subsequent losses as acutely as those he feltas a child. However, these patterns of avoid-ance also maintained his depression by limitingMark’s experience of reward in many of hiscurrent contexts. Treatment focused initially onincreasing activation and addressing many sec-ondary problems and routine disruptions thathad become established. Although Mark in-creased activation relatively quickly, his mooddid not improve significantly. This led to a pri-mary focus on Mark’s rumination and the waysit functioned to avoid intimacy in his relation-ships, and experimentation with new behaviorsdesigned to move Mark closer to his goal ofhaving a close intimate relationship.

Session 1

Session 1 focused on reviewing the results fromthe assessment process, presenting the treat-ment model, encouraging questions and feed-back, and tailoring the model to Mark’s spe-cific experiences. The review of the assessmentprocess is typically brief; in this part of the ses-sion, the aim of the therapist is to ensure thathe or she has a solid understanding of the cli-ent’s presenting problems, relevant history, andprevious experience with treatment, if any. Thetherapist also reviews the basic diagnostic for-mulation to ensure that the assessment out-come matches the client’s subjective experienceof his or her current problems. Discussion ofthe treatment typically forms the bulk of theearly sessions. The following transcript pro-vides an example of the therapist presenting thetreatment model and responding to frequentlyasked questions about the etiology of depres-sion. Specifically, the therapist puts forth theidea that depression is treated behaviorally, re-gardless of etiology.

THERAPIST: Let me tell you a little about the ba-sic model that guides BA. The first idea isthat there are often things that happen inpeople’s lives that make it hard for them toconnect with the kinds of experiences thatwould normally help them feel good. These

shifts can be clear and easy to detect changeslike major losses or disruptions in life. Andthey can also be smaller things, like the kindof things that just bug you a little but theykeep happening, or you have a bunch ofthem happen all around the same period oftime. The most important part is the ideathat the effect of these events in your life isthat it’s harder to connect with the kinds ofexperiences that could give you a sense ofpleasure or accomplishment in your life, andthat could help you feel better.

MARK: I would say that is true for me. There is-n’t much that helps me feel better. Eventhings that I think should help me feel betterdon’t do much.

THERAPIST: Yes, exactly. What we find oftenhappens is that people can respond to thesechanges by pulling away from their liveseven more. This pulling away can happensometimes in obvious ways, like staying inbed or calling in sick to work, or cancelingsocial engagements, and sometimes in moresubtle ways, like being focused more on yourthinking than on the activities you are en-gaged in. The problem with pulling awaylike this is that it tends to keep people stuckin feeling depressed, and the pulling awaycan become a problem in its own right. So,the ultimate goal of our work together is tofigure out what sorts of experiences may po-tentially have a positive effect on your mood,and then help you activate and engage toconnect with these experiences. And to fig-ure out how to solve the problems that arecreating stress or dissatisfaction in your life.How do you see this fitting with your experi-ence? Do you have questions about what I’vesaid? Parts that fit or don’t fit?

MARK: I understand what you are saying, and Ithink some of it fits, but I guess I don’t un-derstand why I get so depressed. I mean,other people have stressful things in theirlives and they seem to function. Other peo-ple get divorced or have crappy jobs, andthey move on. I mean, come on, I’ve been di-vorced for 4 years now. I think depressionruns in my family. My older brother has beendepressed forever, and I sometimes wonder ifmy dad was depressed when he took off.Sometimes, I can’t really identify anythingthat has happened in my life. I mean, I amnever really happy and then, it’s just like aswitch goes in my brain and I’m back in that


dark hole again. But how does that fit withwhat you are saying?

THERAPIST: That is an excellent question. WhatI mean as I talk about depression is thatsome people are more vulnerable than oth-ers. And, there are many ways that you maybe vulnerable to depression—through genet-ics, biology, or experiences in your history.What this treatment emphasizes is that it’spossible to change depression by makingchanges in what you do.

MARK: That makes sense to me. One part ofwhat you said definitely fits for me—the partabout pulling away more. I definitely dothat. Sometimes I don’t talk to another per-son or get out of bed all weekend. I know itmakes it all worse. But I still do it. I guess Idon’t really understand that part either. Itfits, but it doesn’t make much sense to me.

As the therapist replies to these very com-mon questions, she seeks to normalize avoidantresponses to depression. It is essential for theclient to experience the therapist as someonewho understands and has sincere empathy forhis or her struggle. The therapist must commu-nicate that the client’s behaviors make sense,even though they may not serve the client wellin the long term. In this way, the client is morelikely to experience the therapist as an ally inthe change process as opposed to someone whooversimplifies or “doesn’t get” the challengesof making changes. Additionally, the therapistalso emphasizes to Mark the importance ofguided activity, highlighting her role as an ex-pert and the importance of careful assessment.She emphasizes the difference between assign-ments derived from a list of pleasurable activi-ties and those that are guided by functionalanalysis, a key aspect of the treatment, towhich the therapist will return many times.

THERAPIST: Those are great observations andare really on target with what I have noticedfor a lot of people as well. What many peo-ple experience is that when they start to acti-vate and engage, they can actually feel worseinitially! The troubling fact about pullingaway or avoiding is that it does providesome short-term relief. But in the long term,it sets up a downward spiral and can keepyou trapped in depression.

MARK: That makes a ton of sense to me. I justdon’t want to do anything. Making a meal

makes me tired. I feel irritated by the soundof silverware scratching on plates. It’s kindof crazy, but I just want to crawl in a hole, toturn out a light in my head and make it all goaway. Then, I end up feeling worse when I dostay in bed. I used to drink, too. I knew itwould make it worse, and I don’t do it muchanymore, but it helped in the moment, eventhough I knew it didn’t really help. I guess Ifelt better temporarily and that was enough.

THERAPIST:Yes, exactly. Avoidance is a per-fectly natural response to depression. Whatunfortunately happens, though, is that youare not in touch with all those things thatcan give you pleasure and a sense of accom-plishment, and you are not engaged in solv-ing the problems that create stress in yourlife.

MARK: I know that would help, but it just allfeels so overwhelming. Just the thought ofit . . .

THERAPIST: Yes, I know. That is where I comein. It’s important to emphasize that thistreatment is not just about me saying youshould “do more” in general. Sometimes Itell people that it’s not the Nike approach totherapy, where I tell you each week to “justdo it.” You have probably received feedbacklike that from other people in your life, andyou may even say something similar to your-self.

MARK: Yeah, guilty as charged.

THERAPIST: My assumption is that if this wereeasy to figure out, you would already havedone it. The reason that you are here is thatit’s not so easy, and that is where my exper-tise comes in. A major part of this treatmentis the idea of guided activation. This meansthat you and I will be working together toidentify specific ways in which you can ex-periment with activation. My expertise liesin figuring out, first, where the places arethat would be the most helpful in increasingyour activation and engagement, and sec-ond, what small and manageable steps youcan take to get started. You can think of meas a coach or consultant to you in the pro-cess of change. We will work together, insmall steps, all along the way. How does thatsound?

MARK: The idea of it sounds good. I guess it’sworth trying.

THERAPIST: I’d like to ask you to read a short


pamphlet about this treatment, between nowand the next time that we meet. It will pro-vide you with more information. When wemeet next time, we can talk more about howwe will put the ideas into practice.

With this initial session, the therapist has be-gun to teach the client about the treatmentmodel and is getting the client actively involvedin and on board with the rationale. The thera-pist has oriented the client to their roles intreatment and has given the client his firsthomework assignment (reading the treatmentrationale pamphlet). These critical tasks of thefirst session set the stage for additional discus-sion in Session 2.

Session 2

In Session 2, the therapist carefully followsup on a number of the key orienting tasks,including ensuring that Mark is on boardwith the basic treatment model and explain-ing the structure of the therapy. The therapistattends to these topics in the opening of thesession:

THERAPIST: It’s great to see you today, Mark.

MARK: Thanks. It’s good to be back here.

THERAPIST: That’s great. You know, when I wasthinking about our session last time, I real-ized that there were a couple of points Iwanted to emphasize more. One of the im-portant ones is that this is a very collabora-tive approach to therapy, and one that is alsofairly structured. So each time we meet, wewill start out by setting an agenda for thesession, and we will do this collaboratively.In fact, over time, you will set the agendamore and more, though I may have more tosay about it in the beginning. The idea is thatI’m the expert on how to get over depres-sion, and you are the expert on yourself andyour life, and what things help or don’t help.

MARK: That sounds reasonable to me.

THERAPIST: Great. So in terms of the agenda fortoday, I have a couple of things. I’d like totalk more about the treatment approach andyour reaction, and more about how we putsome of the ideas into practice. Do you haveitems you want to be sure we address today?

MARK: No, that sounds good. I did read thepamphlet, and it really hit home. It was like

they wrote it about me, basically. I thought,“Thank goodness somebody has figured thisout.”

THERAPIST: That’s great. I think one of the coreideas of the model is things happen that tendto trigger depressed mood, and then peopletend to do things, or not do things, thatmake the depression worse. For you, my un-derstanding is that the main trigger was yourdivorce.

MARK: Yeah, I’ve really pulled back on a lot,like not exercising and not doing things withother people or even with my girls. We usedto cook these great dinners together and nowit’s like an effort to get organized to orderpizza. It’s kind of like that everywhere—atwork, too. I’m just managing the minimumand, honestly, a lot of times I’m not even do-ing that.

THERAPIST: I know. It can be very hard to keepdoing the sort of things that will keep youfeeling well. And that is where this therapycomes in. From our session last time andfrom your reading, what is your understand-ing of what we are going to be doing in here,and how I am going to be helpful? If youwere to tell a friend of yours what we weregoing to do in this therapy, what would yousay?

MARK: I guess I would say that we are going topinpoint the activities that give me somepleasure or help me feel like I’m handlingthings well. Then, we will figure out how tohelp me get into the position of being moreinvolved in some of those things.

THERAPIST: Yes, that is a big part of it. Some-times in people’s lives something that is com-pletely beyond their control triggers depres-sion, and then what I call secondary problembehaviors get triggered or made worse.These are the behaviors that involve pullingaway or avoiding, as you were saying, likestopping fun activities with your girls orwithdrawing at work. And in those cases wework on the secondary problem behaviors,and that is the core of the therapy. Othertimes, we also need to address larger prob-lems in your life that may be related to whatmakes you vulnerable to depression. In thosecases, therapy can involve both directly ad-dressing the secondary problem behaviorsand working directly on the problems, afterwe have kind of cleared the path for doing


some problem solving by getting you acti-vated and engaged.

MARK: That sounds like it’s probably the casefor me, because I know I had a lot of prob-lems relating to Diane that were part of ourdivorce, and those are not any better.

THERAPIST: Yes, we will talk more as we goalong about what set off the depression foryou. In a global sense, we know now that itwas the divorce. But, as we start followingyour mood and activities day to day, we willsee the ways that your mood has ups anddowns. We will work together and look atthat carefully, asking what set that off, howyou responded to your mood hitting thatpoint, and whether it would help if you triedsomething different.

The therapist has now stated twice that gen-erally a contextual event triggers depression,while earlier acknowledging that several thingscan contribute to vulnerability. This is a subtlebut important point, because clients sometimesbelieve their depression came “out of the blue”or that it is simply “biological” and not modifi-able by behavioral means. By emphasizing anenvironmental antecedent (e.g., a loss of posi-tive reinforcement), the therapist sets up theidea that rather than depression being com-pletely beyond patients’ control, their depres-sive response makes sense, and more impor-tantly, it is possible to make behavioral changesto regain or establish new reinforcers in theirlives. Moreover, the therapist has continued toemphasize the importance of carefully moni-toring and assessing the relationships amongmood, activity, and context as a key part of de-signing effective behavior change plans. Thetherapist then builds on this foundation as shemoves into the other main focus of Session 2—the initiation of activity monitoring. Here, thetherapist explains to Mark why activity moni-toring is important, begins to teach him how tocomplete an Activity Record (see Figure 8.1),and links it directly to some of his recent expe-riences.

THERAPIST: One of the main tools that we usein this therapy is called an Activity Record.This is one example (hands Mark the re-cord); as you can see, it has blocks for eachhour of the day. I’d like you to use this tostart recording your activity and your mood.It’s basically a way to keep track of how you

are spending your time during the day andhow you are feeling. We want to learn whatyou are doing on an hour-by-hour and day-to-day basis. What things in your life helpyou feel better, and what things make youfeel worse? You and I will review these verycarefully together, focusing on how you arespending your time and how you feel. Some-times the Activity Record tells us right awaywhere changes need to be made, and othertimes we have to look at it over a couple ofweeks.

MARK: OK.

THERAPIST: Is there anything you have been do-ing since you started to feel more depressedthat is different from what you normally do?

MARK: Yes, exercising less, watching more TV,and just the amount of time I spend thinkingabout all this. It’s just crazy.

THERAPIST: It’s hard not to do that, but also it’snot helping you very much. And it’s very dif-ficult, which is why you are here. We canstart to figure this out together. It’s great thatyou are already aware of those patterns, andthose are good examples of looking con-cretely at what you are doing. This therapy isabout increasing your awareness of howyour mood is affected subtly from activity toactivity and increasing those that tend to bemore rewarding.

MARK: So, should I write all this down? Doyou really want me to do this every hour?

THERAPIST: Here’s the guideline that I use: Iwant people to record their activity fre-quently enough that they are not relyingheavily on memory. The problem with mem-ory when you are depressed is that yourawareness can be dulled or biased by the de-pression. So you don’t have to do it everyhour. We have to be realistic about the rest ofyour life! But you may want to experimentwith doing it every 3–4 hours. Sometimes,people like to do it at breakfast, lunch, din-ner, and before bed.

MARK: That might work for me.

THERAPIST: Let’s go over what to write down.You put your activity down for each hourblock and then for each hour block you alsoassign a mood rating from 0 to 10. Let’s lookat today as an example. What were you do-ing in the hours before you came here?

MARK: I was at work.


THERAPIST: Okay, great. What were you doingat work?

MARK: I was teaching a new employee how touse our computer system. It was really frus-trating because she wasn’t picking it up and Ididn’t have much patience.

THERAPIST: That’s great information to record.Why don’t you write down “working–teaching new employee.” Now, I also wantyou to record your mood on a 0- to 10-pointscale of depression, so let’s see if we can getsome anchors here. What would be 0 moodfor you? This would mean that you feel re-ally good, absolutely no depression or feel-ing down at all. What would be 10? Thiswould be when you feel your absolute worst,the worst you could possibly feel. It might behelpful to think of some activities that are a5 or in between, when you’re not feelingyour best but you’re not feeling particularlybad either. Which activities might be associ-ated with each?

The therapist and Mark then worked to-gether to identify activities that were associatedwith the low, middle, and high ends of thescale. This was completed to provide Markwith anchors to use when completing the moni-toring at home. It should also be noted thattherapists may ask clients to rate mastery andpleasure associated with activities (Beck et al.,1979). Mastery and pleasure may be rated in-stead of or in addition to mood. Often we be-gin by asking clients to record mood ratingsgiven that this is an easier starting place formany clients because it requires less discrimina-tion of subjective experience; moreover, themood rating provides essential informationabout the relationships between specific activi-ties and depression. For some clients, it is help-ful to build on this by teaching them how tohow to distinguish between mastery and plea-sure, and the ways in which both can be helpfulin regulating mood. In the case of rating moodor mastery–pleasure, it is important to reviewcarefully the method and scale we want clientsto use.

THERAPIST: Given this scale, what was yourmood rating for the 2 hours of “working”today?

MARK: Probably a 5.

THERAPIST: That’s exactly it. Now, sometimes

what happens is that people don’t fill it outbecause they think, “I wasn’t doing any-thing.” It’s important to realize that even ifyou are not doing an activity, we want toknow that, too.

MARK: What do you mean?

THERAPIST: Well, when people think “activi-ties,” they often think of things like “goingto the store,” “watching a movie,” “pickingup my child from school.” But, we are con-ceptualizing activity more broadly. It mightbe driving by Diana’s house, or having a sig-nificant phone conversation with someone,or even lying in bed, spending time thinkingabout Diana.

MARK: That would be true on a lot of days.

THERAPIST: Yes, and you can write that down.Those are some of the most importantthings. In some ways, the more detail, thebetter. We want to start noticing subtlechanges. We want to build on those timesthat you feel just a little better, and we wantto figure out what the problem is when youfeel worse.

MARK: I think I got it.

THERAPIST: Great! People usually come awayfrom this thinking that it sounds really sim-plistic. And it does. It sounds simple, but inpractice it is not that simple. It can be diffi-cult to do in the beginning, to really look atall your activities and figure out how yourmood is related to them. It takes skill andhard work on both our parts.

The therapist ended the session by askingMark to review his understanding of the home-work assignment, encouraged him to makecontact by phone if any questions arose, andoffered encouragement about the likelihoodthat she could be helpful to him.

Session 3

As noted earlier, one of the necessary compe-tencies of a BA therapist is the ability to reviewan Activity Record and glean information thatwill help to customize activation and engage-ment strategies. Session 3 focused heavily onreviewing Mark’s Activity Record (see Figure8.2) and using the information collected as aspringboard for more detailed assessment ofkey problem behaviors. Again, the therapist’s



FIGURE 8.2. Sample completed Activity Record (hourly) (assigned on Monday and reviewed on Thurs-day).

Instructions: Record your activity for each hour of the day, and record a mood rating associated witheach activity. Use the scale below with the anchors that you and your therapist develop to guide yourmood rating. Aim to make entries on your Activity Record at least every 3–4 hours each day.

Monday Tuesday Wednesday Thursday Friday Saturday Sunday

5-6

6-7 Awake,

thinking in

bed (9)7-8

8-9 At work (7)

9-10 Getting

ready for

work (8)

10-11 At work (7)

11-12

12-1

1-2

2-3

3-4 Working/

teaching

employee (5)

Home (7)

4-5

5-6

6-7 Therapy (5) Making

Dinner (6)

7-8 TV (9) TV (9)

8-9

9-10

10-11

11-12

12-1

1-2

2-3

3-4

4-5

Mood Ratings: 0: Feeling really good; not depressed at all (Examples of associated activities:writing; playing with my kids )

5: Intermediate (Examples of associated activities: doing a work task that is only

moderately interesting but I’m focused and concentrating )

10: Feeling the worst (Examples of associated activities: thinking about how I’ve screwed

everything up )

focus in these early sessions is on increasing ac-tivation in areas that will improve Mark’smood; this work will set the foundation forlater work on avoidance modification andproblem solving.

THERAPIST: Shall we go over your Activity Re-cord?

MARK: Okay. (Hands the therapist the re-cord.)

THERAPIST: Why don’t you walk me through it?(Hands the record back to Mark.)

MARK: I’m not sure if this is what you had inmind. I started the next day after our last ses-sion. I went to work that day, but I was feel-ing so lousy that I left early and came home.I was just kind of fiddling around the houseuntil dinner. I felt really down all day up tothat point; I rated my mood as a 7. I didmake dinner, which was a little better for me.I used to love to cook for Diana and myself,and we would make these big feasts some-times with the girls. Since the divorce,though, sometimes I just grab a bag of chipsor something like that, or on a good night, Imight order a pizza. When I was cooking, Ifelt a little better then, about a 5.

THERAPIST: This is terrific. You did a reallygreat job with this. You completed the re-cord exactly as we talked about—writingdown your activities and also your moodrating—and all of this information is ex-tremely useful. I want to ask you some morequestions about specific parts of the day in aminute, but right now let me just get an over-all sense of things.

Notice how the therapist is careful to rein-force her client’s efforts early in the review pro-cess. Clients are often uncertain about how tocomplete the record, and it is not uncommonfor them to return with partially or improperlycompleted records. In such cases, therapistsmust balance the need to provide correctivefeedback and to reinforce the client’s efforts.Frequent client errors include writing down ac-tivities very globally (e.g., “at work” for 6hours), failing to record mood ratings, or fail-ing to record anything because they did “noth-ing.” In such cases, the therapist should ad-dress these problems in a straightforward andmatter-of-fact manner.

THERAPIST: What happened after you madedinner?

MARK: Well, after dinner, I started watchingTV, and everything kind of tanked fromthere. I sat and watched TV until 2:00 in themorning. I guess it helped in that it kept mymind off of worries about work and just feel-ing lousy about Diana, but I was really de-pressed the whole time. In fact, I rated mymood as a 9.

THERAPIST: That is really important informa-tion. I see that you were also up the nextnight watching TV until 1:00 A.M. Is this trueof a lot of nights for you, or are these twomore like exceptions?

MARK: I wish they were exceptions, but no, it’sbeen more the rule. And then what happensis that I just can’t get up in the morning.Well, I guess I do wake up, but I just lie therein bed. I’ve been getting to work pretty late,and some days I just call in sick.

THERAPIST: So we’ll use this log to pick upthemes of specific activities that can help youfeel good and those that may be contributingto your depression. It seems like there are afew that might be important. I’m thinkingthat the watching TV and going to bed late isone big one, and the other two are cookingand how you are doing at work.

MARK: I think the TV is a really big one.

Notice here that the therapist identified afew broad areas that appear to be related to themaintenance of the client’s depression. The BAtherapist is also alert to disruptions in the cli-ent’s normal routines; in Mark’s case, both eat-ing and sleeping routines appear to have beensignificantly altered. The therapist then workscollaboratively with Mark to target a specificarea for further assessment and problem solv-ing (i.e., nighttime TV watching). At this point,the therapist begins the more explicit process offunctional analysis.

THERAPIST: OK, why don’t we start there? Let’sget clear first about what the problem is, be-cause it doesn’t sound like it’s watching TVin general.

MARK: That’s true. Normally, I would watchsome TV, like I might watch for an hour. But,actually, come to think of it, I was more in-volved in my writing then, too. So, normally, I


might watch TV until about 9:00 P.M. andthen turn off the TV and write for anotherhour. Or, if I had the girls, we might watch ashow together and then turn off the TV andread or play a game, or just hang out together,or maybe I’d be on a phone call or something.

THERAPIST: So, this is different for you. Theproblem, then, is that you don’t turn off theTV at 9:00 P.M. and instead watch it for anadditional 4–5 hours.

MARK: Yes, that’s the problem.

THERAPIST: Are you doing this every night ofthe week or just on work nights?

MARK: I hate to admit it, but it’s pretty muchall nights, not always so late, but prettymuch always later than is good for me.

To this point, the therapist has worked suc-cessfully to define the problem in specific andbehavioral terms. With a clear and mutual un-derstanding of the problem, the therapist andclient can begin to consider the contingenciesthat may be maintaining the problem and whatmay be amenable to change.

THERAPIST: We should probably look at whatgets in the way of turning the TV off, since itdoes not seem to have a great effect on yourmood. If you were to turn off the TV at 9:00P.M. now, what do you think would happen?

MARK: I thought about turning it off last night,but I just didn’t want to think about all ofthis stuff.

THERAPIST: By “all this stuff,” do you mean thedivorce and the pressures at work?

MARK: Yes, both of them.

THERAPIST: So, that is what you are activelyavoiding. And the TV helps you to distract?

MARK: Yes, I just don’t have the mental aware-ness now to start writing. I can’t focus on it,and I’m just not interested.

THERAPIST: I think you have the right idea interms of distraction, but the problem is thatyou are distracting yourself with somethingthat doesn’t give you much pleasure and notmuch accomplishment.

MARK: And meanwhile the house is a mess. Ihaven’t paid my bills in months, and . . .

THERAPIST: I think it might be good to help yousolve the problem of watching TV. It might

be simple to solve, but my guess is that thereis more to understand about it.

Notice how easily the client can becomeoverwhelmed and hopeless in response to themyriad problems in his life. The therapist isalert to this possibility during sessions and iscareful to refocus the client on the problem athand. In addition, the therapist also takes akeen interest in the “minutiae” of the client’sday-to-day behavior, particularly if such behav-ior is related to mood. This detailed level of in-terest is critical. Its intent is twofold: First, suchdiscussions guide the choice of activation tar-gets and specific assignments; and second, it isthe intent that such discussions will teach Markto take a similar interest and begin to noticepatterns that are more and less helpful in work-ing his way out of depression.

THERAPIST: Let’s understand better what hap-pens with the TV. Does it come into yourmind, the thought that you might be betteroff if you turned off the TV?

MARK: Typically, I think, “I should go to bed.”But I know that if I go to bed, I’ll just liethere awake anyway, thinking about whatDiana is doing, thinking about how much Iam going to hate being at work the next day.So, then I think I might as well watch TVand distract myself.

THERAPIST: Is that what happens in bed? Youlie there and ruminate about Diana or thingsyou have done or haven’t done at work?

MARK: Pretty much exactly.

In this transcript, the therapist has effectivelyidentified a number of key relationships. Theseinclude the following: (1) nighttime TV watch-ing is associated with deteriorated mood; (2)nighttime TV watching is associated with poorperformance at work; and (3) nighttime TVwatching is potentially maintained via a pro-cess of negative reinforcement in which nega-tive affect (specifically, grief and anxiety) arereduced when the client is watching TV. Thetherapist has done so in a collaborative andnonjudgmental manner, and the client is onboard. At this point, the therapist explicitly ex-amines her hypothesis with Mark about the re-lationship between TV and mood. On the basisof this understanding, they can then considerpossible activation strategies.


THERAPIST: I’m wondering if part of what ishappening is that watching TV is helpful inthe short run, because it takes your mindaway from these topics that are connected toa lot of potential sadness and also anxietyabout the future.

MARK: Yeah, that’s true.

THERAPIST: But the tough part is that while itworks in the short run, it’s that same viciouscycle in the long run, because watching TVgives you almost no pleasure and it keepsyou from doing activities that you previouslygot a lot out of, and it sets you up for havingproblems at work.

MARK: Yes, exactly. It’s crazy, I know, but it’ssuch an easy way out when I’m just beatenby the day.

THERAPIST: Absolutely! So we have to take thatinto account when we think about makingany changes here. I’m thinking that youcould try going to bed in spite of that, andwe could work on the ruminating. Or, if youare going to be up, you could do things thatare better than the TV. Which do you leantoward?

The therapist attends to the function (dis-traction) of the problem behavior (TV watch-ing), while engaging in problem solving in avery collaborative manner.

MARK: Probably finding other things, betterthan the TV. I used to go to a book groupone night a week. It was made up of otherwriters and I liked a few of the people a lot,so when I was doing that, I was also doingreading in the evenings, too.

THERAPIST: OK, does reading seem more of away you could start getting back into someof your writing, versus jumping in with writ-ing itself?

MARK: Yes, there is no way I could write now. Iwould just be staring at a blank page, feelinglike crap.

THERAPIST: OK, that makes sense. So whatabout starting with this? One option is thatyou could have a limit for yourself of 9:00P.M. for TV and we could work on identify-ing a book that you could read instead.

MARK: It’s a good idea. It’s more a question ofmy doing it.

It is very important that the therapist notgloss over comments such as Mark’s final state-ment. When clients express doubt about howor whether they will implement an activationstrategy, it is essential to attend to this in detail.Additionally, it is helpful for the therapist to beattentive to statements such as “I’ll just have tomake myself do it,” which generally indicatethat the therapist and client have not suffi-ciently identified the contingencies that controlthe behavior. In our experience, use of sheerwillpower is unlikely to meet with great suc-cess, and suggestions of such signal that furtherassessment is required, as the therapist illus-trates.

THERAPIST: So we need to be sure we are get-ting at the real problem, instead of just say-ing, “Oh, you are going to do this,” andleaving it at that. What kind of reader areyou? Are you someone who can get really in-volved in a book?

MARK: I do get really involved. In fact, I’llthink about the book a lot during the day, ifI’m already into it.

THERAPIST: But getting yourself to do it is hard.

MARK: Yes, it’s getting started on things.

THERAPIST: That is great to know. So, we haveto somehow get you involved in the book sothat when 9:00 P.M. rolls around, you are al-ready involved, so it will be easier to turn offthe TV.

MARK: That would make it easier.

THERAPIST: What if you were to buy a book onthe way back from our session and beginreading it in the café of the bookstore.

MARK: Oh, yeah, that is right on the way back.I can do that.

THERAPIST: Mark, I think the trick with all ofthis is to figure out what is going to help youmove toward the things that will be benefi-cial for your mood. And this is what is reallyhard—getting yourself to go back to thethings that you used to enjoy, when you haveno interest in them right now. When you feelgood, you take it for granted that it’s easy todo things like reading, spending time withfriends or your girls, and even writing. Whenyou don’t feel good, you really notice it. Theproblem is that you’re in this vicious cycleagain. The longer you do not do things, the


more badly you feel and the less you want todo. The trick for us is to figure out ways tohelp you start to do some of the things thatwill give you pleasure again.

The therapist acknowledges to the client thatthe new behavior will be hard to initiate be-cause of mood, and that it is necessary to doso anyway. Many times, clients employ an“inside-out” or mood-dependent approach totheir depression; that is, they passively wait fortheir mood to improve before making behav-ioral changes. BA therapists teach clients thatwhen they feel down, they cannot afford towait for a better mood to strike them. The goalis to get active when they feel bad (as hard asthat is). Increasing activation will eventuallyimprove mood, even if not immediately, and itwill interrupt the pattern of secondary prob-lems created by withdrawal and avoidance.Addressing mood-dependent behavior (or talk-ing about an outside-in vs. inside-out ap-proach) is a sensitive point in therapy, in whicha great deal of empathy for the experience offeeling depressed is required. The therapistmust skillfully balance encouragement for ac-tion with validation of the difficulty of activat-ing when depressed. In addition, heavily rein-forcing the client (frequently with the use ofsignificant praise) for any steps taken is essen-tial in supporting the process of change.

MARK: Yeah, I know. A lot of the time I mightknow what I need to do, but I have no ideahow to get myself to do it. I’ve just droppeda lot of stuff, like anything social in the eve-nings. I don’t try to make plans. And, like Isaid, for almost a year I was going to thatwriter’s book group every Thursday. Butthen I said to myself, I’m not writing. Thiswhole divorce is wiping me out. What is thepoint of going? I have nothing to add.” But,it’s true that when I went, I used to get a lotout of it. I’m just not interested now, though.

THERAPIST: Yes, exactly. That’s where you and Iwill work together. I want to come back tothe social connections and the routinesaround writing, but let’s stick with the read-ing and nighttime TV for a bit longer first, ifthat’s OK?

MARK: Yeah, that makes sense.

THERAPIST: Let’s think through this book planagain. Is there anything that might come up

between here and the bookstore that wouldderail that plan?

At this point, the therapist and Mark spendthe remainder of the session discussing particu-lar books he could purchase that would maxi-mize his engagement, and they discuss poten-tial barriers that might arise to derail him fromthe intended plan. They also continued to re-view the Activity Record to identify other keyproblems, including ruminating at work, with-drawing from social networks, and experienc-ing disruptions in routines that previouslybrought him pleasure (e.g., cooking, exercise).In each case, the therapist uses a similarmethod that she used with the problem of TVwatching: defining the problem, identifying theantecedent and consequences, and checkingout hypotheses about how the activity is re-lated to mood with client. In each case, thetherapist also continues to emphasize that sim-ply deciding to “make myself do it” is notlikely to be an effective activation strategy forMark, and that it is essential to tie the activa-tion plan to a clear understanding of the func-tion of the problem behaviors. The therapistemploys a combination of gentle prompting,consistent validation of the difficulty of thetasks and understanding of the temptationof withdrawal and avoidance, and repeateddiscussion of potential barriers to activationplans. Importantly, the therapist also highlightsfor Mark that compliance with the homeworkassignments may not bring immediate relief.

“What will be really good this week is to seewhat effect these things have on your mood.Even if they have just a little bit of a positiveeffect, then we know that we are on the righttrack. And, Mark, they might not have animmediate positive effect on your mood. Itmight be that the act of getting yourself to doit is the success itself, and that you need tokeep doing it for a while before you start tofeel good again. But, it’s my guess that someof this stuff will help your mood a little bit,even in the short run.”

The session concluded with the therapist andMark reviewing the homework assignments,which included purchasing a new book, start-ing to read it in the café, and turning off the TVevery night at 9:00 P.M. and reading. In addi-tion, they agreed that Mark would return a


telephone call of an old friend, Mary, who livedin his neighborhood and had been trying tocontact him recently.

Session 4

Mark returned to Session 4 with little improve-ment in the severity of his depression. He re-ported that he had increased his social contact,but that he was not feeling any better. Markalso had delayed the task of purchasing thenew book and continued to watch TV late atnight. The therapist addressed both of theseproblems in a direct, matter-of-fact way.

MARK: It was a really bad weekend. I did callMary and ended up going to this kind ofcocktail party at the community pool thatshe had organized. I was kind of shockedthat I went, but I thought being outsidewould do me good. I was thinking aboutwhat we were talking about, and I thoughtabout how much I used to love swimming. Iwas actually a lifeguard during summers incollege. But I think I felt worse after I went. Isuppose there were moments that were fun,but I was so frustrated by it all. I just spentthe rest of the weekend holed up in myapartment.

THERAPIST: Would that be good to put on theagenda? Doing things that you used to enjoyand not enjoying them?

MARK: Sure.

THERAPIST: And I want to make sure we checkin about how it went with the book versusthe TV, too. Which do you want to talkabout first—calling Mary and the party orthe TV?

MARK: I guess we can do the TV first. I justbought the book today. On Friday, I was atthat party, so I didn’t get home and in beduntil midnight.

THERAPIST: Did you watch TV then?

MARK: No, I do think I was more tired frombeing outside all night, so I just fell asleepwhen I got back.

THERAPIST: And what about Saturday and lastnight?

MARK: It was kind of par for the course. Istayed up late both nights.

Given the importance of attending consis-tently and regularly to the completion of home-

work, here the therapist assesses what in-terfered with Mark’s full completion of the pre-vious assignment.

THERAPIST: I’m glad you bought the book.What got in the way of getting it sooner? AmI recalling correctly that you were going tobuy it on the way back from the session lastweek?

MARK: Yeah, I was, but when I left someonecalled me from work about needing to meet,so I didn’t have as much time as I thought.But I thought I could do it after work, andthen in the evening, I thought, “I’ll get it onthe weekend because I’ll have more time.” Idon’t know.

THERAPIST: If you go back to your leaving thesession last time, when you got the phonecall from work, was there anything else thatderailed the plan?

MARK: No, that was really it. I was still prettyoptimistic about getting the book. It was justthat I didn’t have as much time as I thought Iwould, and I had to get to work.

THERAPIST: OK, that is good to know. So yourplan was to get the book on the weekend,and you just bought it today. On the week-end, did you think about getting it, or did itjust come up again today?

MARK: I did, but I felt so bad after the party Ijust couldn’t get myself to do it.

THERAPIST: It sounds like you were reallydown. Let’s talk about the party in a bit. I’mcurious how it was that you got yourself toget the book today. Are you feeling better, oris it something else?

MARK: I’m not feeling quite as bad, and giventhat I was already out, it was easier to go getthe book. Plus I knew that we were going tomeet and you were probably going to ask meabout it.

THERAPIST: That is so great to know! So onething we know is that I have got to keep fol-lowing up about these things, because ithelps you do them.

MARK: (laughing a little) True, not that I wasenjoying imagining being called to task on it,but it did help, I guess.

THERAPIST: And getting yourself out of thehouse to buy a book this weekend was a lotharder then getting yourself out this morn-ing, since you were already leaving to go to


work. Being out already made it easier to ac-complish your task.

MARK: Right. That sort of thing seems to hap-pen a lot lately.

THERAPIST: So one solution would be to notwait for the weekend when you’ve got a spe-cific task to do, because that seems to be aharder time for you to accomplish things.The other thing would be for us to set up asystem of phone check-ins when you are feel-ing particularly down, since it seems to helpto know that we will be following up onthese tasks when we meet. The other issue,though, is to figure out what brought youdown so much this weekend, and what to doabout that.

MARK: I think that is the biggest thing.

THERAPIST: Shall we talk a bit about the partyand the weekend in order to figure out whatis going to help most? How does that sound?And we’ll make sure to come back to the TVand reading plan.

MARK: OK. I would like not to feel as lousy as Iwas feeling.

THERAPIST: Why don’t we take a look at yourActivity Record? (Reviews the record.) Itlooks like your mood ratings were moderateon Thursday and Friday after we met. Then,Friday, at the party and for the rest of theweekend, they were high, 7’s, 8’s, and 9’s,too.

MARK: I’m not sure this is for me, honestly. Ithink I gave it a fair shot, calling Mary, goingto the party. I didn’t want to do either, but Idid. And I felt worse afterwards.

When clients report that they are increasingactivation and their mood is not improving, itis important to assess a number of possibleexplanations. First, therapists may considerwhether the activation assignments were tooambitious and did not incorporate successfulgrading. In such cases, it is important for thera-pists to acknowledge responsibility for this andrecommend assignment based on smaller com-ponents of the task. Second, therapists want toconsider whether the functional analysis wasaccurate. Is it possible that they are activatingthe client in a domain that is unlikely to yieldimprovements in mood? Third, therapists wantto consider whether ruminative thinking is in-terfering with activation. In such cases, clients“physically” engage in the activation assign-

ments, whereas “mentally” they remain disen-gaged from their context and are less likely tohave an opportunity to contact whatever rein-forcement is available. Fourth, it is possiblethat although activation may not immediatelyimprove mood, it may still be “on the righttrack,” because clients are taking active stepstoward solving problems and addressing im-portant life goals.

In Mark’s case, the therapist decided initiallyto pursue the possibility that rumination wasinterfering with activation, based on Mark’scomments in earlier sessions about frequentlyruminating about Diana and their divorce.

THERAPIST: I guess one thing we could exploretogether is what was on your mind duringthe party. When you were standing by thepool talking to other people, or even swim-ming in the water, what was on your mind?

MARK: You know, when I was diving into thewater, I do remember that those were thepleasurable moments of the party. The soundof the water splashing, the coolness, the si-lence under the water, that was all great.That is what I used to love about swimmingtoo. But the other part—I think I was men-tally checking out. I was with a lot of peopleI really like. Mary is great, and her wholefamily was back visiting from the EastCoast. I haven’t seen them in years and I re-ally enjoy all of them. They are great people.But, it didn’t really matter. I just wasn’tthere.

THERAPIST: Were you thinking about Diana oryourself in relation to her?

MARK: Yeah, that was mainly it.

THERAPIST: Were the other folks there convers-ing with you?

MARK: Yes, and I was talking with them. Imean, I could hear the words coming out ofmy mouth, but I was just not there.

THERAPIST: So you have one rating on this re-cord for the party, a 7. But, if we were tobreak these different pieces apart—theswimming, when you were fully engagedwith the activity, and the talking, when yourmind was elsewhere—what would you rateeach?

MARK: The swimming . . . it was good. I guessthat would be a 3, if 0 is feeling good; I meanit didn’t take it all away. But the talking . . .that was terrible, a 9.


The therapist has successfully identified theproblem that was interfering with the potentialbenefits of activation. She continues to assessthe nature and scope of the problem.

THERAPIST: I’m trying to figure out if when youare actively engaged in an activity, whichkind of requires some attention to it, arethose the times that are more enjoyable?

MARK: Yeah, that’s true.

THERAPIST: Is this a problem that also interfereswith your mood and accomplishment oftasks at work?

MARK: Yes, exactly. I go into my office and it’slike where does the time go? Hours go byand I haven’t done a damn thing. I’m justwandering over and over things that hap-pened with Diana, what I said, what I couldhave said. It’s awful.

THERAPIST: OK, so we know this is an impor-tant problem to address. It’s interfering withyour enjoyment of times that have the poten-tial to improve your mood, and it’s interfer-ing with managing your job well. Can wespend a little more time on what happened atthe party?

MARK: OK.

THERAPIST: How would you normally be whentalking with Mary’s family, if you were notthinking about all these things? What wouldI see differently in those times than what Imight have observed on Friday?

MARK: I’d be talking to everyone. I wouldn’t befeeling so bad.

THERAPIST: Yes, that is exactly true. What I’mreally curious about is, when you are notfeeling so bad, what would you be doing dif-ferently? Would you be asking them morequestions? Making more eye contact? Re-sponding differently?

MARK: Yeah, all of those things. I’d be more ac-tive in the conversation.

THERAPIST: So you would be more engaged.

MARK: Yes, more engaged. Less of that heavyfeeling; you know, that “this really sucks”feeling.

In the preceding portion of the session, thetherapist has begun to define behaviorally whatMark does in interpersonal interactions whenhe is not depressed. Carefully specifying thesebehaviors is an important step in developing

some possible plans for targeted change in howMark approaches similar situations.

THERAPIST: Do you think that if you couldpractice talking, when you weren’t feelingdown, more like you normally would withthese people that you might feel better?

MARK: I don’t know.

THERAPIST: I think the key is to notice what youdo in response to the ruminating and to seewhether that is helping or not helping yourmood, and then for us to begin to explorewhat you may need to do differently. Itseems that at the party, what you were doingwhen your mood was better, was to be moreengaged.

MARK: It’s true. But, when I’m like this, I don’thave much to say.

THERAPIST: Yes, when you are depressed, youare more quiet and withdrawn.

MARK: Yes, because it’s painful. I see Mary’sparents and I think, “They’ve been marriedfor 30 years. I could have had that with Di-ana.” Then, I start thinking that she is withsomeone else. It just goes downhill fromthere.

THERAPIST: You are absolutely right. There is alot of pain there. And what’s happened isthat in response to that pain, you have nar-rowed activity in your life. So you not onlyfeel the pain of being reminded of that loss,but also there is not a lot else going on inyour life. And even when you are doingthings, you are not as engaged, because youare feeling so much pain. I think we need toget you back to doing the things you did be-fore you had the breakup, and before thetwo of you got together. We need to get youback to your baseline, and once we do that,we can figure out how to get you feeling evenbetter than that.

MARK: It sounds good.

THERAPIST: I know you are thinking this is likepie in the sky, but we can figure out how todo this. The key is to figure out some con-crete and manageable steps to help you en-gage more when you are doing some of theseactivities, like going to the party. You areright. It’s worlds harder to do when you arenot feeling well, but these behaviors arepartly why you enjoyed those occasionsmore in the past. We know that you enjoyed


Mary and her family in the past, and weknow that you got a lot of pleasure fromswimming when your mind was fully presentwith the activity. So the trick is not only tocall your friends, like you did so wonderfullywith Mary, but also to go to the gathering,and to get yourself really to interact insteadof just being there at the party. For timeswhen you find yourself withdrawing intoyour thoughts, we need to develop specificstrategies to help you do less of that. Canyou think of anything that would help youdo that?

MARK: I don’t know. I just don’t seem to havemuch to say these days.

THERAPIST: I know it’s hard. There are a varietyof things you could try, such as asking morequestions, and then closely attending to theresponse. Or you could focus on somethingmore specific, such as voice or facial ex-pression, to keep your mind from wander-ing. Sometimes it works just to notice thatyou’ve drifted and to take a deep breath torefocus on your goal in that moment.

MARK: I suppose I could try it. My mind justseems to keep wandering.

THERAPIST: I know. So your job here would beto practice being more vigilant as to whenthat happens, because it will happen. Themore you notice you’re drifting, the moreyou can practice refocusing yourself back onyour friend. Does your mind wander in here?

MARK: I guess a little.

THERAPIST: Why don’t we try it in here? Let’spick something to focus on, and then youcan practice here.

MARK: OK. What do I do?

THERAPIST: I’m going to time us for the next 5minutes and, as we talk, I want you to prac-tice fully engaging in our discussion. Yourmind is going to wander, particularly if weare talking about something that remindsyou of Diana, I would guess. So let’s picksomething you can focus on to bring yourattention back to our conversation. Howabout the sound of my voice, like changes intone, how I articulate words, the pace of myspeech?

MARK: I can try.

THERAPIST: Great. So, let’s talk about some op-tions for social connections that you couldmake this weekend.

The therapist and Mark continued this dis-cussion for the next few minutes, at whichtime, the therapist interrupted their conversa-tion to ask Mark for feedback about his experi-ence.

THERAPIST: What did you notice?

MARK: I don’t know, maybe you are talkingkind of softly.

THERAPIST: How engaged were you with ourdiscussion? Why don’t you give me a rating,with 0 being not engaged at all and 10 beingtotally engaged?

MARK: I guess maybe 7. It wasn’t that hardhere, because I was really focused. I guess Idid start to think about Diana a little whenwe were talking about my calling Mary. I didremind myself to pay attention to your voice,and I guess you just sounded so interested. Itmade it harder for me to wander off in mythoughts when you seemed to be paying somuch attention to what we were talkingabout.

THERAPIST: That was my impression, too, thatyour engagement was generally high, andthat you did appear to refocus your attentiona couple times. That is terrific!

MARK: Yeah, but it was a little strange. I mean,usually people aren’t that focused when theyare just talking about usual stuff.

THERAPIST: That is very true. I might have beenpaying closer attention to what you were do-ing and saying than other folks are in typicalsocial interactions. And this may feel prettyartificial now in general. My guess, though,is that once you get more engaged in socialinteractions, it won’t be necessary to concen-trate so hard. It will just come automaticallyagain.

MARK: That makes sense.

In this way, the therapist generates a strategyto block avoidance (rumination) by substitut-ing a new behavior in the form of attending todirect and immediate experience. Although, inthis case, Mark experimented with directing hisattention to interpersonal stimuli, clients mayalso be directed to experiment with attention toother aspects of sensory stimuli, such as sights,smells, and so forth. The in-session behavioralrehearsal is very important in that it allows theclient to practice and receive direct feedbackfrom the therapist, both of which increase the


likelihood of success outside of the session. Thetherapist then returns to the specific task ofreviewing and developing behavioral assign-ments for the next session.

THERAPIST: Let’s go back to your not gettingout all weekend. Do you think going out tobuy the book was too hard? Is there some-thing easier you could’ve done to help youget a little more engaged this weekend?

MARK: I’m not sure. How hard is it to go outand buy a book?

THERAPIST: Very hard, when you’re reallydown. Let’s think about smaller steps. If youcan do a smaller step and get a little rein-forcement for it, then it becomes easier tomove toward your goal.

The therapist and client continue along theselines with graded task assignment. Given thatMark has previously enjoyed socializing, heand the therapist came up with a plan that onthe weekend, he would start by returning somephone calls from friends and inviting Mary forlunch. During lunch, he would focus specifi-cally on attending to their conversation. Thetherapist also raised the possibility of swim-ming as an exercise activity. Mark reportedthat he thought his plate was full with the as-signments they had already developed, andthey decided to table further discussion ofswimming. The therapist then uses the finalmoments of the session to review the home-work, to instill hope in Mark, to validate thedifficulty of change, and to reinforce the basictreatment model.

Session 5

At the outset of the session, Mark reports im-provement in his mood and the therapist in-cludes this as an item on the agenda. Their dis-cussion allows the therapist to emphasize animportant point about maintaining new behav-iors in consistent and regular routines. In thissession, the therapist continues to emphasizethe pattern of social connections and to assessfactors that increase Mark’s vulnerability to ex-acerbated mood when alone.

THERAPIST: Let’s understand in more detailhow it is that you are feeling better?

MARK: I think the reading plan is helping. I fin-ished the book.

THERAPIST: Great! So you probably need an-other book.

MARK: (laughing) I guess that’s true. You don’tthink just the one cured the problem?

THERAPIST: (laughing) Oh, how I wish thatwere the case! Seriously, though, Mark, Ithink that is such an important question.There is a real temptation when you startfeeling a little better to back off from someof the very things that are helping. It makessense, because making these changes re-quires so much effort, I know. But maintain-ing the routines are so important.

MARK: It’s true. I actually think I’ve been doingpretty well with that this week. I’ve beenreaching out more to other people.

THERAPIST: This is fantastic.

MARK: And Mary called me again. So, I guess Ididn’t do what we talked about in terms ofcalling her, but I did ask her about lunchwhen she called. I didn’t really want to, be-cause I was feeling down when she called. Ihad just gotten a letter from the lawyerabout some new money stuff with Diana.But, I did ask Mary, and I took the girls, too.I think they enjoyed it a lot. I did really focuson asking them all a lot of questions duringlunch. I think that helped, too.

THERAPIST: Mark, you have definitely hadmore social contact in the last few days! Youare doing a huge part of this treatment,which is acting in accordance with the goalsand plans that we are setting here, as op-posed to being directed by how you feel inthe moment.

MARK: I tried.

THERAPIST: You did it! You talked about hav-ing lunch with your coworker. Did you dothat?

MARK: I did do that.

THERAPIST: You did a lot! That’s great. OK, Imay be pushing our luck here, but what doyou think about adding swimming to ouragenda?

MARK: I knew you were going to ask aboutthat again.

THERAPIST: (laughing) You know me too well.What’s your thought about it?

MARK: It’s probably a good idea. There isactually a swim lesson that the girls like todo on the weekends, and I could take them


and do laps at the same time in the otherpool.

THERAPIST: Fantastic! Do you have them withyou this weekend? Could we schedule thatfor the weekend?

MARK: Yeah, I think that would help.

THERAPIST: Mark, do you think that reconnect-ing with people and some of these activities,like reading, are connected with your im-proved mood?

MARK: Yes, that definitely had a lot to do withit. I am still not sure that we are getting tothe real problem with all of this, but you areright that it does help.

THERAPIST: So we should talk about that, too.Before we move to that, is there anythingelse that you think is contributing to yourpositive mood, or is it mostly having moresocial contact, which you find reinforcing?

MARK: It’s the social contact and trying to dis-tract myself with the reading.

THERAPIST: That is so great! Good reminder,too. Let’s talk some about another book andhow to keep up that schedule.

At this point, the therapist and Mark focuson developing a specific plan for selecting andpurchasing a new book to continue the readingroutine. Next, the therapist returns to Mark’scomments about whether the interventions areaddressing what is most important.

THERAPIST: What you mentioned before aboutthe real problem . . . I’m curious what youmeant.

MARK: I guess I’m still thinking about Diana alot. I think that there is a part of me that hasto let go, yet just isn’t letting go. I am think-ing, just asking myself, “Is there still achance for us? What did I do to screw it allup so badly?” And then I start thinking, “Isthis all I have now—having lunch with peo-ple, reading by myself at night?” You know,the kind of stuff we’ve been focusing on . . . Idon’t know. Is it really going to fix anything?

THERAPIST: Mark, I know it feels like this stuffisn’t really getting at the real problem interms of your thinking about Diana, and Iagree that is really important to talk about.At the same time, I don’t want us to losesight of the fact that this other stuff makes ahuge difference. It’s important for you to re-

connect with ways to buoy up your moodbefore you start to tackle some of the pastproblems and those that still come up withDiana. Also, I think we will find that thereare some similar patterns, so maybe the waysyou have tended to pull away from otherpeople since you’ve been down might havesome connections with what happened withDiana.

MARK: That’s true. I guess they are not totallyseparate.

THERAPIST: Are you saying that it’s time now tostart focusing our time more directly onthose topics?

MARK: I think so. Maybe I’m more aware of itbecause I’m feeling a little better. I guess I’masking more often, “Is this all there is now?”It just seems like a damn lonely life to beleading, if this is it.

The therapist and Mark end the session byreviewing the assignments. In addition, theyagree on a plan to return to Mark’s importantquestions in the next session.

Sessions 6–9

In this next series of sessions, the therapist andMark return to Mark’s question from Session5. In repeated sessions, he reports improve-ments in mood related to making progress onprojects at home, exercising, and becomingmore socially connected in casual and friend-ship circles. These areas of progress are re-flected consistently on his Activity Recordforms, which now specifically target the areasof social engagement, reading, and swimming(see Figure 8.3). (This version of the ActivityRecord can be considered when the activationtargets are clear and well developed, and thedetailed information gained via hour-by-hourmonitoring is not as necessary. It can also beused for clients who have difficulty with themore detailed Activity Record.)

Even with clear areas of improved activationand mood, Mark also reports that his mood isvulnerable to his tendency to ruminate fre-quently about his ex-wife. The therapist andMark begin to explore the potential function ofrumination about his ex-wife. As they did withrespect to both TV watching and ruminationduring social interactions, they develop someinitial hypotheses about the consequences ofMark’s ruminating about his ex-wife.


THERAPIST: Is it possible that ruminating mightbe a form of avoidance itself? It’s like yourmind gets stuck in a broken record format.You keep replaying what you did wrong,what you could have done, and one of the ef-fects is that you are actually avoiding thepainful emotions about the loss of the rela-tionship, and maybe also avoiding exploringnew relationships?

MARK: It feels like I can’t stand the loss of it.That’s what I can’t accept—that it is lost. Ikeep thinking maybe there is a way to recap-ture it, even though I know there simply isnot. We can’t even communicate about thekids’ health care without a lawyer.

THERAPIST: So, in a way, ruminating may be away to avoid dealing with grief and sadness.I wonder if part of this comes from what youlearned about how to cope with major lossafter your dad left. It seems like no onetalked about that and you got pretty caughtup in thinking about how you might havebeen responsible. I wonder if it’s hard toknow what to do emotionally right now.

MARK: It’s certainly true about what happenedwhen I was a kid.

THERAPIST: So one possibility we could experi-ment with is taking time specifically to expe-rience the sadness and loss.

MARK: I don’t know. Thinking about her andwhat I’ve lost seems overwhelming. I justwant to be done with it and move on.

THERAPIST: I know. Exactly! The problem isthat ruminating seems to have the effect ofkeeping you from moving on. Instead ofmoving onto other relationships or pursuits

in your life, your mind keeps replaying whathappened and didn’t happen with Diana.

MARK: I just don’t know if I’m ready for otherrelationships.

THERAPIST: So, if you weren’t ruminating asmuch, do you think you might experiencemore fear?

MARK: When I think about getting into an-other relationship. . . . You know, I thinkthat there is actually a person at work who isinterested in dating, but that’s been part ofthe reason that I’ve kind of held back fromdoing things with her. She’s asked me tolunch a couple of times. I just don’t want tobe back in the same place again 2 years fromnow. I can’t take this whole thing again, andI don’t want to subject my kids to it either.

THERAPIST: So, it may be possible that ruminat-ing has the effect of keeping at bay not onlyfeelings of loss about Diana but also fearsabout future loss.

The therapist also emphasize the importanceof continuing with activation plans developedin earlier sessions to maintain adaptive routinesand improve mood. In particular, they high-light the need for consistent attention to socialcontact, exercise, and reading. In addition, thetherapist and Mark begin to discuss his returnto the writers’ group in more detail, beginningto break down that larger task in manageablepieces. Work on these targets forms the major-ity of the middle of the course of treatment. AsMark begins to address feelings of loss more di-rectly and continues his work on social connec-tions, exercise, and limiting TV watching, healso begins to express interest in dating again.


Task Monday Tuesday Wednesday Thursday Friday Saturday Sunday

Reading � � � � �

Reaching outto other people

� � � � �

Swimming � �

Mood 5 5 5 6 8 3 3

FIGURE 8.3. Sample completed Activity Record (daily).

Sessions 10–15

In this section of treatment, the therapist andMark begin to address directly the prospect ofhis developing new intimate relationships in hislife, specifically with a woman at work towhom he is attracted. They explore what isnecessary to approach rather than avoidMark’s fear of starting a new relationship, andthe therapist hypothesizes in particular thatMark’s ruminative style may have functionedto avoid learning from patterns in past rela-tionships. The therapist used the TRAP/TRACacronym as a simple way to help Mark recog-nize the conditions under which he was likelyto avoid (the TRAP), and to then engage inmore adaptive coping behavior to get back on“TRAC.” For example, Mark reported that hewould see the woman at work (trigger) and be-gin to feel nervous (response) and either nottalk with her or restrict his conversation to per-functory work issues (the avoidance pattern).His alternative coping under the same condi-tions involved asking her if she would like tohave coffee. Sessions then focus heavily on ex-amining in detail what he might learn from hisformer marriage that would be instructive infuture relationships. The following dialogueprovides an example of the types of foci thatthese sessions target.

MARK: One of the things that happened a lotwith Diana is I never felt like I was reallypresent with her or the girls. It was like theywere in this little world together and I wasalways on the outside somehow. I oftenthought that I should put myself more in thecenter, like say more of what I thought, but Ijust didn’t. I never did.

THERAPIST: Did that cause conflict with her?

MARK: Yes, absolutely. It was one of things thatshe said when she ended things. Being on theoutside is a big thing for me.

THERAPIST: What does being on the outside in-volve specifically? How would I know if youwere doing that?

MARK: It’s just not being willing to speak upabout things. She always said it was like Iwasn’t really in or out on anything, just kindof on the fence the whole time.

THERAPIST: Can you think of a specific examplewhen that was an issue?

MARK: Well, my mother and brothers never re-

ally liked Diana very much, but I didn’t domuch to stand up for her with them. I justkind of let things unfold. . . .

THERAPIST: So, that might have been a TRAPwith her? Was it a trigger that you thoughtshe wanted something from you in terms ofyour commitment?

MARK: Yes, it was, because I ended up feelingreally overwhelmed by that.

THERAPIST: And the avoidance pattern waswithdrawing.

MARK: I did. I just backed off, and she had tohandle the whole scene with my family.

THERAPIST: So with your coworker, if you wereto take a stand with her now, what wouldthat look like? What would alternative cop-ing be?

MARK: I have no idea.

THERAPIST: Do you think there is a similar trig-ger?

MARK: Maybe, because I think she is wonder-ing what’s up with me? Like am I interestedor not?

THERAPIST: Have you been clear with her aboutbeing interested in dating her?

MARK: Not really. We talk often at work, but Ican’t say that I’ve really said much about it.

THERAPIST: Would you like to ask her out?

MARK: Yes, I guess I would.

THERAPIST: Why don’t we think of some spe-cific things you could say as alternatives towithdrawing and practice with some ofthem?

In these sessions, the therapist and Mark de-fine, in very specific and concrete terms, thetypes of behaviors associated with decreasedsatisfaction and quality in his former marriage.For instance, the therapist’s following questionto Mark is a central question asked repeatedlyover the course of BA: “What does being on theoutside involve specifically? What does it looklike? How would I know if you were doingthat?” The therapist emphasizes identifyingclear, specific, and observable behaviors whenanalyzing behaviors and defining goals. Then,the therapist and Mark work to identify spe-cific strategies that he can use to practice alter-native behaviors in pursuing a future relation-ship. They continue to use the TRAP/TRAC


framework to examine situations that arise andMark’s response, and to guide him toward amore engaged approach to intimate interper-sonal relationships. As Mark begins dating,they have ample opportunity to revise and re-fine strategies through activation assignmentsthat target being direct and present in intimateinteractions.

Sessions 16–19

By Session 16, the therapist and Mark agreethat the bulk of the work of understanding andproblem-solving Mark’s depression in terms ofhis unique life context and avoidant responsepatterns that maintain his depression has beencompleted. Mark has been successfully acti-vated both in terms of his secondary problembehaviors (e.g., increased reading, exercise, so-cial contacts, and projects around the house;decreased TV watching). He has also been acti-vated toward solving his primary problem(avoidance of intimacy) via initiating a new re-lationship.

Thus, the final sessions of treatment focusedon reviewing and consolidating primarythemes and methods used in therapy. Spe-cifically, the therapist and Mark identified theimportance of continuing to practice his newskills of blocking rumination by attending toimmediate goals and to his direct and immedi-ate experience, and being more direct and ex-pressive with his new partner. In addition, thetherapist carefully reviewed with Mark theways he had learned to use the fundamentals ofbehavioral activation himself. Together they re-viewed ways that Mark would know when hewas starting to feel depressed or to engage inavoidance response patterns. They also re-viewed specific steps he could take to beginself-monitoring his mood and activities, and toproblem-solve alternative coping behaviors.They also specifically identified a number of al-ternative behaviors that were uniquely helpfulin breaking the vicious cycle of depression,avoidance, and withdrawal. Mark reportedthat he felt well equipped with these tools andthe opportunities he had had to practice themin therapy. He also reported feeling encouragedabout the positive changes he had alreadymade in his life. He ended treatment expressingoptimism about his future and warmly thankedthe therapist for all of their work together.Over time, Mark continued to maintain thegains he made in treatment. He established a

new relationship with a woman, and they be-came engaged over the course of the followingyear. He continued to practice many of theskills he had learned in therapy in the contextof this new relationship, with his children, andwith his coworkers and friends.

Case Summary

The course of treatment with Mark provides anexample of many of the core principles andstrategies of BA. The treatment followed fromcareful and ongoing functional analysis of keyproblems that Mark presented, which in turnallowed the therapist to develop the organizingcase conceptualization. This work was com-pleted in collaboration with Mark during ses-sions and was also a focus of the ongoing clin-ical consultation team meetings, of whichMark’s therapist was a key member. Duringtreatment, the therapist used a range of specificstrategies, including goal setting, self-monitoring, graded task assignment, problemsolving, behavioral rehearsal, and attention toexperience. She also addressed a number of im-portant treatment targets frequently observedin BA, including interpersonal avoidance, ru-mination, and routine disruption. Overall, thetherapist worked as a coach throughout ther-apy, helping Mark to problem-solve specificsteps to overcome patterns of avoidance and toengage in activities. She also taught Mark tounderstand the pattern of antecedents to de-pressed mood and how his responses contrib-uted to either maintaining or improving hismood. She skillfully balanced acknowledgingthe difficulty of change when depressed withemphasizing the importance of action, evenwhen mood is low. She maintained a matter-of-fact, nonjudgmental, problem-solving ap-proach to difficulties that arose during thecourse of Mark’s therapy, and returned regu-larly and persistently to the selected targets ofchange.

CONCLUSION

This chapter provides the conceptual basicsand the how-to specifics that are required touse BA with depressed clients. Evolving fromthe seminal foundation established by the workof Ferster, Lewinsohn, and Beck, BA highlightsthe power of direct and sustained attention tobehavior change. BA aims to help clients be-


come active and engaged in their lives in waysthat reduce current depression and help to pre-vent future episodes. BA therapists help de-pressed clients to increase activities that bringgreater reward and to solve important prob-lems. Clients are assisted in approaching im-portant life goals and engaging directly and im-mediately with problematic aspects of theirlives. Both outcome research and other con-verging lines of empirical inquiry suggest thatBA holds promise as an efficacious treatmentfor depression. Future research will examine ingreater detail the process of change in BA andthe ease with which BA can be transported toapplied community settings.

REFERENCES

Addis, M. E., & Carpenter, K. M. (2000). The treatmentrationale in cognitive behavioral therapy: Psychologi-cal mechanisms and clinical guidelines. Cognitiveand Behavioral Practice, 7(2), 147–156.

Addis, M. E., & Martell, C. R. (2004). Overcoming de-pression one step at a time: The new behavioral acti-vation approach to getting your life back. Oakland,CA: New Harbinger.


Armento, M. E. A., & Hopko, D. R. (2007). The Envi-ronmental Reward Observation Scale (EROS): De-velopment, validity, and reliability. Behavior Ther-apy, 38, 107–119.

Beck, A.T., Rush, A. J., Shaw, B. F., & Emery, G. (1979).Cognitive therapy of depression. New York: GuilfordPress.

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Man-ual for the BDI-II. San Antonio, TX: PsychologicalCorporation.

Bennett-Levy, J., Butler, G., Fennell, M., Hackman, A.,Mueller, M., & Westbrook, D. (2004). Oxford guideto behavioural experiments in cognitive therapy. Ox-ford, UK: Oxford University Press.

Blumenthal, J. A., Babyak, M. A., Moore, K. A.,Craighead, W. E., Herman, S., Khatri, P., et al.(1999). Effects of exercise training on older patientswith major depression. Archives of Internal Medi-cine, 159, 2349–2356.

Borkovec, T. D., Newman, M. G., Pincus, A. L., &Lytle, R. (2002). A component analysis ofcognitive-behavioral therapy for generalized anxietydisorder and the role of interpersonal problems.Journal of Consulting and Clinical Psychology, 70,288–298.

Dimidjian, S., Hollon, S. D., Dobson, K. S., Schmaling,K. B., Kohlenberg, R. J., Addis, M., et al. (2006).Randomized trial of behavioral activation, cognitivetherapy, and antidepressant medication in the acute

treatment of adults with major depression. Journal ofConsulting and Clinical Psychology, 74, 658–670.

Dobson, K. S., Hollon, S. D., Dimidjian, S., Schmaling,K. B., Kohlenberg, R., Gallop, R., et al. (2007).Behavioral activation therapy, cognitive therapy andpharmacotherapy for depression: Relapse preven-tion. Manuscript under review.

Ferster, C. B. (1973). A functional analysis of depres-sion. American Psychologist, 28, 857–870.

Ferster, C. B. (1981). A functional analysis of behaviortherapy. In L. P. Rehm (Ed.), Behavior therapy for de-pression: Present status and future directions(pp. 181–196). New York: Academic Press.

First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J.B. W. (1997). User’s guide for the Structured ClinicalInterview for DSM-IV Axis I Disorders. Washington,DC: American Psychiatric Press.

First, M. B., Spitzer, R. L., Gibbons, M., Williams, J. B.W., & Benjamin, L. (1996). User’s guide for theStructured Clinical Interview for DSM-IV Axis IIPersonality Disorders (SCID-II). New York:Biometrics Research Department, New York StatePsychiatric Institute.

Foa, E. B., Rothbaum, B. O., & Furr, J. M. (2003). Aug-menting exposure therapy with other CBT proce-dures. Psychiatric Annals, 33, 47–53.

Gloaguen, V., Cottraux, J., Cucherat, M., & Blackburn,I. M. (1998). A meta-analysis of the effects of cogni-tive therapy in depressed patients. Journal of Affec-tive Disorders, 49, 59–72.

Gollwitzer, P. M. (1999). Implementation intentions:Strong effects of simple plans. American Psycholo-gist, 54, 493–503.

Gortner, E. T., Gollan, J. K., Dobson, K. S., & Jacobson,N. S. (1998). Cognitive-behavioral treatment for de-pression: Relapse prevention. Journal of Consultingand Clinical Psychology, 66, 377–384.

Hamilton, M. A. (1960). A rating scale for depression.Journal of Neurology, Neurosurgery, and Psychiatry,23, 56–61.

Hayes, A. M., Castonguay, L. G., & Goldfried, M. R.(1996). Effectiveness of targeting the vulnerabilityfactors of depression in cognitive therapy. Journal ofConsulting and Clinical Psychology, 64(3), 623–627.

Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (1999).Acceptance and commitment therapy: An experien-tial approach to behavior change. New York:Guilford Press.

Hopko, D. R., Lejuez, C. W., LePage, J. P., Hopko, S.D., & McNeil, D. W. (2003). A brief behavioral acti-vation treatment for depression: A randomized pilottrial within an inpatient psychiatric hospital. Behav-ior Modification, 27, 458–469.

Jacobson, N. S., Dobson, K. S., Truax, P. A., Addis, M.E., Koerner, K., Gollan, J. K., et al. (1996). A compo-nent analysis of cognitive-behavioral treatment fordepression. Journal of Consulting and Clinical Psy-chology, 64, 295–304.

Jacobson, N. S., Martell, C. R., & Dimidjian, S. (2001).Behavioral activation treatment for depression: Re-


turning to contextual roots. Clinical Psychology: Sci-ence and Practice, 8, 255–270.

Kanter, J. W., Mulick, P. S., Busch, A. M., Berlin, K. S.,& Martell, C. R. (2007). The Behavioral Activationfor Depression Scale (BADS): Psychometric proper-ties and factor structure. Journal of Psychopathologyand Behavioral Assessment, 29, 191–202.

Lewinsohn, P. M. (1974). A behavioral approach to de-pression. In R. M. Friedman & M. M. Katz (Eds.),The psychology of depression: Contemporary theoryand research (pp. 157–185). New York: Wiley.

Lewinsohn, P. M., Antonuccio, D. O., Steinmetz-Breckenridge, J., & Teri, L. (1984). The coping withdepression course: A psychoeducational interventionfor unipolar depression. Eugene, OR: Castalia.

Lewinsohn, P. M., Biglan, A., & Zeiss, A. S. (1976).Behavioral treatment of depression. In P. O.Davidson (Ed.), The behavioral management of anxi-ety, depression and pain (pp. 91–146). New York:Brunner/Mazel.

Lewinsohn, P. M., & Graf, M. (1973). Pleasant activi-ties and depression. Journal of Consulting and Clini-cal Psychology, 41, 261–268.

Lewinsohn, P. M., Sullivan, J. M., & Grosscup, S. J.(1980). Changing reinforcing events: An approach tothe treatment of depression. Psychotherapy: Theory,Research, Practice, and Training, 17, 322–334.

Linehan, M. M. (1993). Cognitive-behavioral treatmentof borderline personality disorder. New York:Guilford Press.

Locke, E. A., & Latham, G. P. (2002). Building a practi-cally useful theory of goal setting and task motiva-tion: A 35-year odyssey. American Psychologist, 57,705–717.

Martell, C. R., Addis, M. E., & Jacobson, N. S. (2001).Depression in context: Strategies for guided action.New York: Norton.

Nolen-Hoeksema, S. (2000). The role of rumination indepressive disorders and mixed anxiety/depressivesymptoms. Journal of Abnormal Psychology, 109,504–511.

Scogin, F., Jamison, C., & Gochneaur, K. (1989). Com-parative efficacy of cognitive and behavioral biblio-therapy for mildly and moderately depressed olderadults. Journal of Consulting and Clinical Psychol-ogy, 57, 403–407.

Segal, Z. V., Williams, J. M. G., & Teasdale, J. D.(2002). Mindfulness-based cognitive therapy for de-pression. New York: Guilford Press.

Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-Item Short-Form Health Survey (SF-36), I: Concep-tual framework and item selection. Medical Care, 30,473–483.

Watson, D. L., & Tharp, R. G. (2002). Self directedbehavior. Belmont, CA: Wadsworth/ThomsonLearning.

Weissman, M. M., & Bothwell, S. (1976). Assessment ofsocial adjustment by patient self-report. Archives ofGeneral Psychiatry, 33(9), 1111–1115.

Zeiss, A. M., Lewinsohn, P. M., & Munoz, R. F.(1979). Nonspecific improvement effects in de-pression using interpersonal skills training, pleasantactivity schedules, or cognitive training. Journalof Consulting and Clinical Psychology, 47, 427–439.

Zettle, R. D., & Rains, J. C. (1989). Group cognitiveand contextual therapies in treatment of depression.Journal of Clinical Psychology, 45(3), 436–445.


C H A P T E R 9

Dialectical Behavior Therapyfor Borderline Personality Disorder

MARSHA M. LINEHANELIZABETH T. DEXTER-MAZZA

This chapter presents one of the more remarkable developments in all of psychotherapy.Few therapists are willing to undertake the overwhelmingly difficult and wrenching task oftreating individuals with “borderline” characteristics, yet these people are among theneediest encountered in any therapeutic setting. They also impose an enormous burdenon the health care system. Over the past two decades, Linehan and her colleagues havedeveloped a psychological treatment for individuals with borderline personality disorder(BPD). Importantly, data indicate that this treatment is effective when compared to alterna-tive interventions. If results from the initial trials continue to hold up in future clinical trials,then this treatment will constitute one of the most substantial contributions to the arma-mentarium of the psychotherapist in recent times. What is even more interesting is thatthis approach blends emotion regulation, interpersonal systems, and cognitive-behavioralapproaches into a coherent whole. To this mix Linehan adds her personal experience withEastern philosophies and religions. Among the more intriguing strategies incorporatedinto this approach are “entering the paradox” and “extending,” borrowed from aikido, a Jap-anese form of self-defense. Yet the authors remain true to the empirical foundations oftheir approach. The fascinating case study presented in this chapter illustrates Linehan’stherapeutic expertise and strategic timing in a way that will be invaluable to all therapistswho deal with personality disorders. The surprising and tragic outcome illustrates theenormous burden of clinical responsibility inherent in any treatment setting, as well as thepractical issues that arise when treatment ultimately fails.—D. H. B.

Clinicians generally agree that clients with adiagnosis of borderline personality disorder(BPD) are challenging and difficult to treat. Asa result, BPD has become a stigmatized disor-der resulting in negative attitudes, trepidation,and concern with regard to providing treat-ment (Aviram, Brodsky, & Stanley, 2006;Lequesne & Hersh, 2004; Paris, 2005). Per-haps of greatest concern is the generally highincidence of suicidal behavior among this pop-ulation. Approximately 75% of clients whomeet criteria for BPD have a history of suicide

attempts, with an average of 3.4 attempts perindividual (Soloff, Lis, Kelly, Cornelius, &Ulrich, 1994). Suicide threats and crises are fre-quent, even among those who never engagein any suicidal or nonsuicidal self-injuriousbehavior (NSSI). Although much of this behav-ior is without lethal consequence, follow-upstudies of individuals with BPD have found sui-cide rates of about 7–8%, and the percentagewho eventually commit suicide is estimated at10% (for a review, see Linehan, Rizvi, Shaw-Welch, & Page, 2000). Among all individuals

365

who have committed suicide, from 7 to 38%meet criteria for BPD when personality disor-ders are assessed via a psychological autopsy,with the higher incidence occurring primarilyamong young adults with the disorder (Brent etal., 1994; Isometsa et al., 1994, 1997; Lesage etal., 1994; Rich & Runeson, 1992). Individualswith BPD also have difficulties with anger andanger expression. Not infrequently, intense an-ger is directed at their therapists. The frequentcoexistence of BPD with both Axis I conditions(e.g., mood or anxiety disorders) and other per-sonality disorders clearly complicates treat-ment further.

The criteria for BPD, as defined within thetext revision of the fourth edition of the Diag-nostic and Statistical Manual of Mental Disor-ders (DSM-IV-TR; American Psychiatric As-sociation, 2000) and the Revised DiagnosticInterview for Borderlines (DIB-R; Zanarini,Gunderson, Frankenburg, & Chauncey, 1989),the most commonly used research assessmentinstrument, reflect a pervasive pattern of insta-bility and dysregulation across all domains offunctioning. Other assessment measures usedto diagnose BPD include the InternationalPersonality Disorders Examination (IPDE;Loranger, 1995) and the Diagnostic Interviewfor DSM-IV Personality Disorders (DIPD-IV;Zanarini, Frankenburg, Sickel, & Yong, 1996).The Borderline Symptom List (BSL; Bohus etal., 2001) and the McLean Screening Instru-ment for Borderline Personality Disorder (MSI-BPD; Zanarini et al., 2003) are both screeningmeasures for BPD.

Linehan (1993a) has reorganized and sum-marized the diagnostic criteria of BPD into fivedomains. First, individuals with BPD generallyexperience emotional dysregulation and insta-bility. Emotional responses are reactive, andthe individuals generally have difficulties withepisodic depression, anxiety, and irritability, aswell as problems with anger and anger expres-sion. Second, individuals with BPD have pat-terns of behavioral dysregulation, as evidencedby extreme and problematic impulsive behav-ior. As noted earlier, an important characteris-tic of these individuals is their tendency to di-rect apparently destructive behaviors towardthemselves. Attempts to injure, mutilate, or killthemselves, as well as actual suicides, occur fre-quently in this population. Third, individualswith BPD sometimes experience cognitive dys-regulation. Brief, nonpsychotic forms ofthought and sensory dysregulation, such as de-

personalization, dissociation, and delusions(including delusions about the self), are attimes brought on by stressful situations andusually cease when the stress is ameliorated.Fourth, dysregulation of the sense of self is alsocommon. Individuals with BPD frequently re-port that they have no sense of a self at all, feelempty, and do not “know” who they are.Finally, these individuals often experience in-terpersonal dysregulation. Their relationshipsmay be chaotic, intense, and marked with diffi-culties. Even though their relationships are sodifficult, individuals with BPD often find it ex-tremely hard to relinquish relationships. In-stead, they may engage in intense and franticefforts to prevent significant individuals fromleaving them. The polythetic format of theDSM-IV-TR definition allows for considerableheterogeneity in diagnosis (indeed, the require-ment that five of nine criteria be met for the di-agnosis yields 256 ways in which the BPD diag-nosis may be met), and clinical experience withthese clients confirms that this diagnostic cate-gory comprises a heterogeneous group.

This chapter focuses primarily on describingdialectical behavior therapy (DBT), a compara-tively new approach to treatment of BPD(Linehan, 1993a, 1993b). It has the distinctionof being one of the first psychosocial treat-ments demonstrated to be effective in a ran-domized clinical trial (Linehan, Armstrong,Suarez, Allmon, & Heard, 1991). Before de-scribing DBT, we first review other treatmentsfor BPD and provide information on their theo-retical rationales and supporting data (whensuch data are available). This is followed by amore in-depth description of DBT—its philo-sophical roots, underlying theory, and treat-ment protocols.

OVERVIEW OF OTHERTREATMENT APPROACHES

Various approaches have been applied to thetreatment of BPD. Although it is not our pur-pose to present a scholarly review of all themany treatments for BPD, we believe it helpfulto review briefly the status of other, currenttreatments before presenting DBT in detail.

Psychodynamic

Psychodynamic approaches currently receivingthe greatest attention include those of Kernberg


(1975, 1984; Clarkin et al., 2001; Kernberg,Selzer, Koenigsberg, Carr, & Appelbaum,1989), Adler and Buie (1979; Adler, 1981,1985, 1993; Buie & Adler, 1982), and Batemanand Fonagy (2004). Among these, Kernberg’s(1975, 1984) theoretical contributions areclearly prominent. His object relations model iscomprehensive as to theory and technique, andhas had considerable influence on the psy-choanalytic literature. His expressive psycho-therapy for clients with “borderline personalityorganization” (BPO) or BPD, transference-focused therapy (TFT), emphasizes three pri-mary factors: interpretation, maintenance oftechnical neutrality, and transference analysis.The focus of the therapy is on exposure andresolution of intrapsychic conflict. Treatmentgoals include increased impulse control andanxiety tolerance, ability to modulate affect,and development of stable interpersonal rela-tionships. TFT also uses a target hierarchyapproach to the first year of treatment. The tar-gets are (1) containment of suicidal and self-destructive behaviors, (2) therapy-destroyingbehaviors, and (3) identification and recapitu-lation of dominant object relational patterns,as experienced in the transference relationship(Clarkin et al., 2001). Kernberg has also distin-guished a supportive psychotherapy for moreseverely disturbed clients with BPO or BPD.Like expressive psychotherapy, supportive psy-chotherapy also places great emphasis on theimportance of the interpersonal relationship intherapy (transference); however, interpreta-tions are less likely to be made early in treat-ment, and only the negative responses to thetherapist and to therapy (negative transference)are explored. Both expressive and supportivepsychotherapy are expected to last severalyears, with primary foci on suicidal behaviorsand therapy-interfering behaviors. The datasupporting the use of TFT are not extensive.Clarkin and colleagues have published resultsfrom a preliminary study of TFT. Additionally,one completed randomized clinical trial hascompared TFT to schema-focused therapy(SFT; Giesen-Bloo et al., 2006). The results ofthis study are described in the section oncognitive-behavioral treatments.

The preliminary study of TFT assessed pre-and posttreatment changes over the course of a1-year treatment for adult women with BPD (N= 23). Of the 23 clients who were considered in-tent-to-treat (ITT), 17 completed the treatment.Both the ITT sample and the completer sample

were analyzed. There were no significant reduc-tions in number of suicide attempts, number ofNSSI behaviors (referred to as “parasuicide” inthe article), medical risk of either type of self-injury, or physical condition after either type ofself-injury in the ITT sample. However, signifi-cant decreases in medical risk and physical con-dition after NSSI behaviors occurred in the com-pleter sample. Furthermore, the number ofhospitalizations over the course of the treatmentyear compared to the year prior to treatment re-duced significantly for both groups. Given thelack of a control group in this study and the smallsample size, these findings should be reviewedwith caution (Clarkin et al., 2001).

Mentalization therapy, developed by Bate-man and Fonagy (2004), is an intensive therapygrounded in attachment theory (i.e., BPD isviewed as an attachment disorder), with a fo-cus on relationship patterns and nonconsciousfactors inhibiting change. “Mentalization” re-fers to one’s perception or interpretation of theactions of others and oneself as intentional.The treatment is based on the theory that indi-viduals with BPD have an inadequate capacityfor mentalization. Treatment, therefore, is fo-cused on bringing the client’s mental experi-ences to conscious awareness, facilitating amore complete, integrated sense of mentalagency. The goal is to increase the client’s ca-pacity for recognizing the existence of thethoughts and feelings he or she is experiencing.

A randomized trial of mentalization therapyoffered in a partial hospitalization setting pro-vides additional supporting data for psychoan-alytic treatment of BPD. This study by Bate-man and Fonagy (1999) consisted of randomassignment of clients to either standard psychi-atric care constrained only by the requirementthat individual psychotherapy was not allowed(control condition) or to partial hospitaliza-tion, a treatment program with the followinggoals of therapy: (1) psychoanalytically in-formed engagement of clients in treatment; (2)reduction of psychopathology, including de-pression and anxiety; (3) reduction of suicidalbehavior; (4) improvement in social compe-tence; and (5) reduction in lengthy hospitaliza-tions. The experimental treatment group re-ceived once-weekly individual psychotherapyprovided by psychiatric nurses, once-weeklypsychodrama-based expressive therapy, thrice-weekly group therapy, a weekly communitymeeting, a monthly meeting with a case admin-istrator, and a monthly medications review. At

Borderline Personality Disorder 367

the end of the 18-month treatment, the groupreceiving mentalization therapy showed signifi-cant reductions in suicidal behavior (suicide at-tempts and self-mutilation), inpatient hospital-ization stays, measures of psychopathology(including depression and anxiety), and socialfunctioning relative to the control group. Thesegains were maintained and increased during an18-month follow-up period consisting of twice-weekly group therapy (Bateman & Fonagy,2001). The researchers note that their programcontained three characteristics that they hy-pothesize to be related to treatment effective-ness: a consistent theoretical rationale for treat-ment, a relationship focus, and consistenttreatment over time.

Psychopharmacological

Reviews of the literature regarding drug treat-ments for BPD highlight a dilemma for the pre-scribing pharmacotherapist: BPD involves dys-regulation in too many domains for a singledrug to serve as a panacea (Dimeff, McDavid,& Linehan, 1999; Lieb, Zanarini, Linehan, &Bohus, 2004; Nose, Cipriani, Biancosino,Grassi, & Barbui, 2006). In general, results in-dicate that several agents may be useful for im-proving global functioning, cognitive percep-tual symptoms (e.g., suspiciousness, ideas ofreference, transitory hallucinations), emotiondysregulation, or impulsive–behavioral dys-control (for reviews, see Lieb et al., 2004; Noseet al., 2006). Nose and colleagues (2006) con-ducted a meta-analysis of 22 randomized,placebo-controlled clinical trials, published be-tween 1986 and 2006, examining the effects ofpharmacotherapy for individuals with BPD.Organization of results was based on five pri-mary outcome measures: affective instabilityand anger, impulsivity and aggression, interper-sonal relationships, suicidality, and globalfunctioning. First, no medication had a morepositive effect than placebo on suicidality.Overall, across the studies, fluoxetine, an anti-depressant, and topiramate and lamotrigine,mood stabilizers, showed more positive effectsthan placebo for affective instability and anger.Additionally, valproate, an anticonvulsant andmood stabilizer, has effectively treated behav-ioral dysregulation in clients with BPD, includ-ing those with aggressive and impulsive be-havior (Stein, Simeon, Frenkel, Islam, &Hollander, 1995). As a class, antipsychoticswere more effective than placebo for impulsivi-

ty, interpersonal relationships, and global func-tioning, and specifically, olanzapine was betterthan placebo relative to global functioning(Nose et al., 2006) and has been shown todecrease impulsive aggression and chronic dys-phoria more effectively compared to fluoxetine(Zanarini, Frankenburg, & Parachini, 2004).In summary, although some drug treatmentsmay be effective, caution is in order when con-sidering pharmacotherapy for this particularclient population. Clients with BPD are notori-ously noncompliant with treatment regimens,may abuse the prescribed drugs or overdose,and may experience unintended effects of thedrugs. With these caveats in mind, carefullymonitored pharmacotherapy may be a usefuland important adjunct to psychotherapy in thetreatment of BPD.

Cognitive-Behavioral

Treatment of BPD has received increasing at-tention from cognitive theorists. The cognitiveapproach views the problems of the client withBPD as residing within both the content andthe process of the individual’s thoughts. Beck’sapproach to treating BPD (Beck & Freeman,1990) is representative of cognitive psychother-apy generally, with the focus of treatment onrestructuring thoughts and on developing a col-laborative relationship through which moreadaptive ways of viewing the world are devel-oped. More specifically it focuses on decreasingnegative and polarized beliefs that result in un-stable affect and destructive behaviors (Brown,Newman, Charlesworth, Crits-Christoph, &Beck, 2004). In an open clinical trial of cogni-tive therapy for clients with BPD, Brown andcolleagues found decreases in clients meetingBPD criteria, depression, hopelessness, and sui-cide ideation at the end of the 12-month treat-ment and 6-month follow-up posttreatment.

The cognitive-behavioral therapies ofYoung, Klosko, and Weishaar (Kellogg &Young, 2006; Young, 2000; Young, Klosko, &Weishaar, 2003; Pretzer, 1990); Blum and col-leagues (Blum, Pfohl, St. John, Monahan, &Black, 2002) and Schmidt and Davidson (ascited in Weinberg, Gunderson, Hennen, & Cut-ter, 2006) attempt to address some of the diffi-culties experienced in applying traditional cog-nitive approaches to the treatment of BPD.Pretzer’s (1990) approach emphasizes modify-ing standard cognitive therapy to address diffi-culties often encountered in treating clients


with BPD, such as establishing a collaborativerelationship between therapist and client,maintaining a directed treatment, and improv-ing homework compliance. Blum and col-leagues (2002) developed a twice-weekly out-patient group treatment that uses apsychoeducational approach to teaching skillsto clients with BPD and to their support sys-tems (e.g., family, friends, other care provid-ers). The treatment focuses on destigmatizationof BPD, emotional control, and behavioralcontrol. At present, outcome data are limitedfor Pretzer’s approach. A pilot study has shownpotential for the group treatment developed byBlum and colleagues and a randomized controltrial is currently being conducted (Van Wel etal., 2006).

Young’s schema-focused therapy (SFT)(Young et al., 2003) postulates that stablepatterns of thinking (“early maladaptiveschemas”) can develop during childhood andresult in maladaptive behavior that reinforcesthe schemas. SFT includes a variety of interven-tions aimed at challenging and changing theseearly schemas through the identification of aset of dysfunctional schema modes that controlthe individual’s thoughts, emotions, and behav-iors (i.e., detached protector, punitive par-ent, abandoned/abused child, angry/impulsivechild). Giesen-Bloo and colleagues (2006) com-pleted the first randomized clinical trial of TFTand SFT. Transference focused therapy wascompared to schema focused therapy in a studywhere 88 participants received three years oftwice per week individual sessions of eitherschema focused therapy or transference basedtherapy. Study results indicated an overall de-crease in BPD symptoms for both treatments,however, participants who received SFT hadsignificantly greater improvements overall anda lower attrition rate. Suicide and NSSI behav-iors were not assessed as an outcome measurein this study.

Weinberg and colleagues (2006) completed arandomized, controlled trial of Schmidt andDavidson’s manual-assisted cognitive treat-ment (MACT) and treatment as usual (TAU).MACT is a brief cognitive-behavioral treat-ment that incorporates strategies from DBT,cognitive therapy, and bibliotherapy. The treat-ment targets NSSI behaviors occurring in indi-viduals with BPD. MACT was provided as anadjunctive treatment to TAU for study partici-pants (N = 30). Participants were 30 women di-agnosed with BPD, with a history of NSSI be-

haviors and at least one in the last month; how-ever, suicide was considered one of theexclusionary criteria for study participation.Participants were randomly assigned to MACTplus TAU or to TAU-alone conditions. Uponcompletion of the 6-week treatment and at the6-month follow-up, individuals who receivedMACT had significantly fewer and less severeNSSI behaviors than those in the TAU-alonecondition. The authors state that these resultsshould be interpreted with caution due to smallsample size and the use of self-report measuresonly in assessment of NSSI behaviors.

DIALECTICAL BEHAVIOR THERAPY

DBT evolved from standard cognitive-behavioral therapy as a treatment for BPD,particularly for recurrently suicidal, severelydysfunctional individuals. The theoretical ori-entation to treatment is a blend of three theo-retical positions: behavioral science, dialecticalphilosophy, and Zen practice. Behavioral sci-ence, the principles of behavior change, iscountered by acceptance of the client (withtechniques drawn both from Zen and fromWestern contemplative practice); these polesare balanced within the dialectical framework.Although dialectics was first adopted as a de-scription of this emphasis on balance, dialecticssoon took on the status of guiding principlesthat have advanced the therapy in directionsnot originally anticipated. DBT is based withina consistent behaviorist theoretical position.However, the actual procedures and strategiesoverlap considerably with those of various al-ternative therapy orientations, including psy-chodynamic, client-centered, strategic, andcognitive therapies.

Efficacy

Although several treatments (Bateman &Fonagy, 1999, 2001; Giesen-Bloo et al., 2006;Marziali & Munroe-Blum, 1994) have shownefficacy in the treatment of individuals withBPD, DBT has the most empirical support atpresent and is generally considered thefrontline treatment for the disorder. DBT hasbeen evaluated in six randomized controlledtrials (RCTs) conducted across three indepen-dent research teams (Koons et al., 2001;Linehan et al., 1991, 1999, 2002, 2006;Linehan, Heard, & Armstrong, 1993; Linehan,


Tutek, Heard, & Armstrong, 1994; Verheul etal., 2003). Two of the RCTs specifically re-cruited clients with suicidal behaviors (Linehanet al., 1991, 1993, 1994, 1999). The results ingeneral have shown DBT to be an effectiveevidenced-based treatment for the disorder. Infour of the six studies participants treated withDBT demonstrated significantly greater reduc-tions in suicide attempts, intentional self-injury,and suicidal ideation (Koons et al., 2001;Linehan et al., 1991, 1999, 2002; Verheul etal., 2003). Treatment superiority was main-tained when DBT was compared to only thosecontrol subjects who received stable individualpsychotherapy during the treatment year, andeven after researchers controlled for number ofhours of psychotherapy and of telephone con-tacts. (Linehan & Heard, 1993; Linehan et al.,1999). Two studies with participants with sub-stance dependence and BPD found DBT to bemore effective than control treatments in re-ducing substance use, and increasing globaland social adjustment (Linehan et al., 1999,2002). In the original study of recurrently sui-cidal patients with BPD, participants treatedwith DBT were significantly less likely thanTAU participants to attempt suicide or NSSIbehaviors during the treatment year, had lessmedically severe NSSI behaviors, were lesslikely to drop out of treatment, had fewer inpa-tient psychiatric days per participant, and im-proved more on scores of both global and so-cial adjustment. More specifically, Linehan andcolleagues’ (2006) study showed that partici-pants treated with DBT were half as likely toengage in suicidal behaviors compared to par-ticipants in the treatment by community ex-perts (TBCE) condition, further indicating thatDBT is an effective treatment for reducing sui-cidal behavior. This study suggests that the effi-cacy of DBT is due to specific treatment fac-tors, and not general factors or the expertise ofthe treating psychotherapists. In two studies todate, DBT has been shown to be effective in re-ducing substance use disorders (Linehan et al.,1999, 2002).

In addition to these studies of DBT for indi-viduals with BPD, three studies have examinedits effectiveness with other disorders. First,Lynch, Morse, Mendelson, and Robins (2003)found that a DBT skills group plus anti-depressant medication showed greater reduc-tions in depressive symptoms in older (over 60-years-old) depressed individuals compared toTAU plus antidepressant medication group at

the 6-month follow-up time point in a 28-weektreatment program. DBT has also been adaptedfor use with individuals without BPD who arediagnosed with binge-eating disorder (BED).Telch, Agras, and Linehan (2001) comparedwomen diagnosed with BED receiving anadapted 20-week group DBT treatment towaiting-list controls. Participants in the DBTcondition had significantly fewer days of binge-eating episodes compared to those in the wait-list control condition and were more likely toabstain from bingeing at follow-up. Addi-tionally, those in the wait-list control groupwho were offered the DBT treatment after thestudy had similar results. DBT continues to beexamined in a variety of settings and for a vari-ety of different diagnoses. However, it is impor-tant to highlight that when DBT is adapted foruse with different populations, it may not be aseffective, given that many adaptations have notbeen rigorously tested. We recommend that un-til it is tested in an RCT, adaptations shouldnot be made to DBT. If using DBT with differ-ent populations, then the most importantchange should be in the examples used whenteaching, not in the content itself.

Philosophical Basis: Dialectics

The term “dialectics” as applied to behaviortherapy refers both to a fundamental natureof reality and to a method of persuasive dia-logue and relationship. (See Wells [1972,cited in Kegan, 1982] for documentation of ashift toward dialectical approaches across allthe sciences during the last 150 years; morerecently, Peng & Nisbett [1999] discuss bothWestern and Eastern dialectical thought.) Asa worldview or philosophical position, dialec-tics guide the clinician in developing theoreti-cal hypotheses relevant to the client’s prob-lems and to the treatment. Alternatively, asdialogue and relationship, dialectics refers tothe treatment approach or strategies used bythe therapist to effect change. Thus, centralto DBT are a number of therapeutic dialecti-cal strategies. These are described later in thischapter.

Dialectics as a Worldview

DBT is based on a dialectical worldview thatemphasizes wholeness, interrelatedness, andprocess (change) as fundamental characteristicsof reality. The first characteristic, the Principle


of Interrelatedness and Wholeness, provides aperspective of viewing the system as a wholeand how individuals relate to the system, ratherthan seeing individuals as if they exist in isola-tion. Similar to contextual and systems theo-ries, a dialectical view argues that analysis ofparts of any system is of limited value unlessthe analysis clearly relates the part to thewhole. The second characteristic is the Princi-ple of Polarity. Although dialectics focuses onthe whole, it also emphasizes the complexity ofany whole. Thus dialectics asserts that reality isnonreducible; that is, within each single thingor system, no matter how small, there is polar-ity. For example, physicists are unable to re-duce even the smallest of molecules to onething. Where there is matter there is antimatter;even every atom is made up of both protonsand electrons: A polar opposite is always pres-ent. The opposing forces are referred to as thethesis and antithesis, present in all existence.Dialectics suggests that the thesis and antithesismove toward a synthesis, and inherent in thesynthesis will be a new set of opposing forces.It is from these opposing forces that the thirdcharacteristic is developed. This characteristicof the dialectical perspective refers to the Prin-ciple of Continuous Change. Change is pro-duced through the constant synthesis of thethesis and the antithesis, and because new op-posing forces are present within the synthesis,change is ongoing. These dialectical principlesare inherent in every aspect of DBT and allowfor continuous movement throughout the ther-apy process. A very important dialectical ideais that all propositions contain within themtheir own oppositions. Or, as Goldberg (1980,pp. 295–296) put it, “I assume that truth isparadoxical, that each article of wisdom con-tains within it its own contradictions, thattruths stand side by side. Contradictory truthsdo not necessarily cancel each other out ordominate each other, but stand side by side,inviting participation and experimentation.”One way that the client and therapist addressthis in therapy is by repeatedly asking eachother or oneself the question: “What is beingleft out?” This simple question can assist infinding a synthesis and letting go of an absolutetruth, a nondialectical stance.

Dialectics as Persuasion

From the point of view of dialogue and rela-tionship, dialectics refers to change by persua-

sion and by making use of the oppositions in-herent in the therapeutic relationship ratherthan by formal impersonal logic. Through thetherapeutic opposition of contradictory posi-tions, both client and therapist can arrive atnew meanings within old meanings, movingcloser to the essence of the subject under con-sideration. The spirit of a dialectical point ofview is never to accept a proposition as a finaltruth or an undisputable fact. Thus, the ques-tion addressed by both client and therapist is“What is being left out of our understanding?”Dialectics as persuasion is represented in thespecific dialectical strategies described later inthis chapter. As readers will see, when we dis-cuss the consultation strategies, dialectical dia-logue is also very important in therapist consul-tation meetings. Perhaps more than any otherfactor, attention to dialectics can reduce thechances of what psychodynamic therapistshave labeled “staff splitting,” that is, the fre-quent phenomenon of therapists’ disagreeingor arguing (sometimes vehemently) about howto treat and interact with an individual clientwho has BPD. This “splitting” among staffmembers is often due to one or more factionswithin the staff deciding that they (and some-times they alone) know the truth about a par-ticular client or clinical problem.

Dialectical Case Conceptualization

Dialectical assumptions influence case concep-tualization in DBT in a number of ways. First,dialectics suggests that a psychological disorderis best conceptualized as a systemic dysfunctioncharacterized by (1) defining the disorder withrespect to normal functioning, (2) assumingcontinuity between health and the disorder,and (3) assuming that the disorder results frommultiple rather than single causes (Hollands-worth, 1990). Similarly, Linehan’s biosocialtheory of BPD, presented below, assumes thatBPD represents a breakdown in normal func-tioning, and that this disorder is best conceptu-alized as a systemic dysfunction of the emotionregulation system. The theory proposes thatthe pathogenesis of BPD results from numerousfactors: Some are genetic–biological predispo-sitions that create individual differences in sus-ceptibility to emotion dysregulation, known asemotion vulnerability; others result from theindividual’s interaction with the environment,referred to as the invalidating environment. As-suming a systemic view compels the theorist to


integrate work from a variety of fields and dis-ciplines.

A second dialectical assumption that under-lies Linehan’s biosocial theory of BPD is thatthe relationship between the individual and theenvironment is a process of reciprocal influ-ence, and that the outcome at any given mo-ment is due to the transaction between theperson and the environment. Within sociallearning theory, this is the principle of “recipro-cal determinism.” Besides focusing on recipro-cal influence, a transactional view also high-lights the constant state of flux and change ofthe individual–environment system. Therefore,BPD can occur in multiple environments andfamilies, including chaotic, perfect, and evenordinary families. Millon (1987) made muchthe same point in discussing the etiology ofBPD and the futility of locating the “cause” ofthe disorder in any single event or time period.

Both transactional and interactive models,such as the diathesis–stress model of psycho-pathology, call attention to the role of dysfunc-tional environments in bringing about disorderin the vulnerable individual. A transactionalmodel, however, highlights a number of pointsthat are easy to overlook in an interactivediathesis–stress model. For example, a person(Person A) may act in a manner stressful to anindividual (Person B) only because of the stressPerson B is putting on Person A. Take the childwho, due to an accident, requires most of theparents’ free time just to meet survival needs.Or consider the client who, due to the need forconstant suicide precautions, uses up much ofthe inpatient nursing resources. Both of theseenvironments are stretched in their ability torespond well to further stress. Both may invali-date or temporarily blame the victim if any fur-ther demand on the system is made. Althoughthe system (e.g., the family or the therapeuticmilieu) may have been predisposed to responddysfunctionally in any case, such responsesmay have been avoided in the absence of expo-sure to the stress of that particular individ-ual. A transactional, or dialectical, account ofpsychopathology may allow greater compas-sion, because it is incompatible with the assign-ment of blame, by highlighting the reality ofthe situation rather than judgments about theindividuals. This is particularly relevant with alabel as stigmatized among mental health pro-fessionals as “borderline” (for examples of themisuse of the diagnosis, see Reiser & Levenson,1984).

A final assumption in our discussion regardsthe definition of behavior and the implicationsof defining behavior broadly. Linehan’s theory,and behaviorists in general, take “behavior” tomean anything an organism does involvingaction and responding to stimulation(Merriam-Webster’s New Universal Un-abridged Dictionary, 1983, p. 100). Conven-tionally, behaviorists categorize behavior asmotor, cognitive/verbal, and physiological, allof which may be either public or private. Thereare several points to make here. First, dividingbehavior into these three categories is arbitraryand is done for conceptual clarity rather thanin response to evidence that these responsemodes actually are functionally separate sys-tems. This point is especially relevant to under-standing emotion regulation, given that basicresearch on emotions demonstrates that theseresponse systems are sometimes overlapping,somewhat independent, but definitely notwholly independent, thus remaining consistentwith the dialectical worldview. A related pointhere is that in contrast to biological and cogni-tive theories of BPD, biosocial theory suggeststhat there is no a priori reason for favoring ex-planations emphasizing one mode of behavioras intrinsically more important or compellingthan others. Rather, from a biosocial perspec-tive, the crucial questions are under what con-ditions a given behavior–behavior relationshipor response system–response system relation-ship holds, and under what conditions these re-lationships enter causal pathways for the etiol-ogy and maintenance of BPD.

BIOSOCIAL THEORY

Emotion Dysregulation

Linehan’s biosocial theory suggests that BPD isprimarily a dysfunction of the emotion regula-tion system. Behavioral patterns in BPD arefunctionally related to or are unavoidable con-sequences of this fundamental dysregulationacross several, perhaps all, emotions, includingboth positive and negative emotions. FromLinehan’s point of view, this dysfunction of theemotion regulation system is the core pathol-ogy; thus, it is neither simply symptomatic nordefinitional. Emotion dysregulation is a prod-uct of the combination of emotional vulnera-bility and difficulties in modulating emotionalreactions. Emotional vulnerability is conceptu-alized as high sensitivity to emotional stimuli,


intense emotional responses, and a slow returnto emotional baseline. Deficits in emotionmodulation may be due to difficulties in (1) in-hibiting mood-dependent behaviors; (2) orga-nizing behavior in the service of goals, inde-pendently of current mood; (3) increasing ordecreasing physiological arousal as needed; (4)distracting attention from emotionally evoca-tive stimuli; and/or (5) experiencing emotionwithout either immediately withdrawing orproducing an extreme secondary negative emo-tion (see Gottman & Katz, 1990, for a furtherdiscussion).

Conceptually, the deficit in the emotion reg-ulation system leads to not only immense emo-tional suffering but also multiple behavioralproblems in individuals with BPD. When clini-cian’s ratings of characteristics associated withpsychopathology are examined, tendencies to-ward being chronically anxious and unhappy,depressed, or despondent are the most highlydescriptive of the BPD (Bradley, Zittel, &Westen, 2005). Dysfunction leads the individ-ual to attempt to escape aversive emotions, of-ten leading to further suffering. For example, afemale client may be experiencing intense angerafter a fight with her partner, and in an effort toescape the anger, she engages in cutting behav-iors. She begins to feel relief from her anger fora short period of time. However, once her an-ger begins to subside, shame in response to thecutting behavior begins to increase and the cy-cle of emotion escape behavior continues. Al-though the mechanisms of the initial dysregula-tion remain unclear, it is likely that biologicalfactors play a primary role. Siever and Davis(1991) hypothesized that deficits in emotionregulation for clients with BPD are related toboth instability and hyperresponsiveness ofcatecholamine function. The etiology of thisdysregulation may range from genetic influ-ences to prenatal factors to traumatic child-hood events affecting development of the brainand nervous system. Furthermore, adoptionstudies of monozygotic (MZ) twins (Davison& Neale, 1994) suggest a genetic vulnerability.However, researchers do not claim that geneticor biological factors accounted for all pathol-ogy. If pathology were solely determined by ge-netics, then 100% of the MZ twins would havebeen presumed to share the same pathology.Because this does not occur, we can explain thedifferences through the transactions betweenbiology, as described earlier, and the environ-ment.

Invalidating Environments

Most individuals with an initial temperamentalvulnerability to emotion dysregulation do notdevelop BPD. Thus, the theory suggests furtherthat particular developmental environmentsare necessary. The crucial developmental cir-cumstance in Linehan’s theory is the transac-tion between emotion vulnerability and thepresence of the “invalidating environment”(Linehan, 1987a, 1987b, 1989, 1993a), whichis defined by its tendency to negate, punish,and/or respond erratically and inappropriatelyto private experiences, independent of the va-lidity of the actual behavior. Private experi-ences, and especially emotional experiencesand interpretations of events, are not taken asvalid responses to events by others; are pun-ished, trivialized, dismissed, or disregarded;and/or are attributed to socially unacceptablecharacteristics, such as overreactivity, inabilityto see things realistically, lack of motivation,motivation to harm or manipulate, lack of dis-cipline, or failure to adopt a positive (or, con-versely, discriminating) attitude. The invali-dating environment can be any part of anindividual’s social environment, including im-mediate or extended family, school, work, orcommunity. Within each of these environmentsare even more specific idiosyncrasies that mayimpact the environment, such as birth order,years between siblings, teachers and peers, and/or coworkers. It is important to note that be-cause two children grew up in the same homedoes not mean that they were raised in identicalenvironments. Furthermore, individuals are of-ten not aware of their invalidating behaviorsand are not acting with a malicious intent.

There are three primary characteristics of theinvalidating environment. First, the environ-ment indiscriminately rejects communicationof private experiences and self-generated be-haviors. For example a person may be told,“You are so angry, but you won’t admit it” or“You can’t be hungry, you just ate.” Second,the invalidating environment may punish emo-tional displays and intermittently reinforceemotional escalation. For example, a womanbreaks up with her partner and is feeling de-pressed. Her friends and family begin tellingher to “Get over it,” “He wasn’t worth it,”“Don’t feel sad.” Over the course of the nextweek, she becomes more depressed and is be-ginning to withdraw from daily activities.Again, her environment responds in an invali-


dating manner. Finally after another 3 days ofhigh emotional arousal she makes a suicide at-tempt. At that moment the environment jumpsin and provides support by taking care of her.Unfortunately, this type of pattern often resultsin inadvertent reinforcement of extreme dys-functional behavior. Finally, the invalidatingenvironment may oversimplify the ease ofproblem solving and meeting goals for an indi-vidual.

The high incidence of childhood sexualabuse reported among individuals with BPD(Bryer, Nelson, Miller, & Krol, 1987; Herman,1986; Herman, Perry, & van der Kolk, 1989;Wagner, Linehan, & Wasson, 1989) suggeststhat sexual abuse may be a prototypic invali-dating experience for children. The relation-ship of early sexual abuse to BPD, however, isquite controversial and is open to many inter-pretations. On the one hand, Silk, Lee, Hill,and Lohr (1995) reported that the number ofcriterion BPD behaviors met was correlatedwith severity of childhood sexual abuse in agroup of clients with BPD. On the other hand,a review by Fossati, Madeddu, and Maffei(1999) suggested that sexual abuse is not a ma-jor risk factor for BPD.

The overall results of this transactional pat-tern between the emotionally vulnerable indi-vidual and the invalidating environment are theemotional dysregulation and behavioral pat-terns exhibited by the borderline adult. Such anindividual has never learned how to label andregulate emotional arousal, how to tolerateemotional distress, or when to trust his or herown emotional responses as reflections of validinterpretations of events resulting in self-invalidation (Linehan, 1993a). In more optimalenvironments, public validation of one’s pri-vate, internal experiences results in the devel-opment of a stable identity. In the family of aperson with BPD, however, private experiencesmay be responded to erratically and with insen-sitivity. Thus, the individual learns to mistrusthis or her internal states, and instead scans theenvironment for cues about how to act, think,or feel. This general reliance on others resultsin the individual’s failure to develop a coherentsense of self. Emotional dysfunction also inter-feres with the development and maintenance ofstable interpersonal relationships, which de-pend on both a stable sense of self and a capac-ity to self-regulate emotions. The invalidatingenvironment’s tendency to trivialize or ignorethe expression of negative emotion also shapes

an expressive style later seen in the adult withBPD—a style that vacillates from inhibitionand suppression of emotional experience to ex-treme behavioral displays. Behaviors such asoverdosing, cutting, and burning have impor-tant affect-regulating properties and are addi-tionally quite effective in eliciting helping be-haviors from an environment that otherwiseignores efforts to ameliorate intense emotionalpain. From this perspective, the dysfunctionalbehaviors characteristic of BPD may be viewedas maladaptive solutions to overwhelming, in-tensely painful negative affect.

DIALECTICAL DILEMMAS

Linehan (1993a) describes “dialectical dilem-mas” as behavioral patterns of the client thatoften interfere with therapy. These behavioralpatterns, also referred to as “secondary tar-gets” in treatment (compared to other targetsthat we describe later) represent six behaviorsthat are dichotomized into a set of three dimen-sions of behavior defined by their oppositepoles (see Figure 9.1). At one end of each di-mension is the behavior that theoretically ismost directly influenced biologically via defi-cits in emotion regulation. At the other end isbehavior that has been socially reinforced inthe invalidating environment. These secondarytargets are characteristics of individuals withBPD that often interfere with change, thus in-terfering with therapy.


FIGURE 9.1. Dialectical dilemmas in DBT.

Emotion Vulnerability/Self-Invalidation

One dialectical dilemma is represented by bio-logically influenced emotional vulnerability onthe one hand (e.g., the sense of being out ofcontrol or falling into the abyss) and by sociallyinfluenced self-invalidation on the other (e.g.,hate and contempt directed toward the self,dismissal of one’s accomplishments). Along thisdimension of behavior, clients with BPD oftenvacillate between acute awareness of their ownintense, unbearable, and uncontrollable emo-tional suffering on the one hand, and dismissal,judgment, and invalidation of their own suffer-ing and helplessness on the other.

Emotion vulnerability here refers to the cli-ent’s acute experience and communication ofemotional vulnerability and excruciating emo-tional pain. “Vulnerability” here means theacute experience of vulnerability rather thanthe sensitivity to emotional cues that definesthe term when discussing the emotion dysregu-lation difficulties of the BPD person. In the casehere, vulnerability is experienced; in the lattercase, the vulnerability may not be experienced.Three reactions to emotional vulnerability arecommon in BPD: (1) freezing or dissociating inthe face of intense emotion; (2) rage, often di-rected at society in general or at individualswho are experienced as invalidating; and (3) in-tense despair. Suicide here can function to com-municate to others the depth of one’s suffering(“I’ll show you”) and/or as an escape from anunendurable life.

On the other side of this polarity is self-invalidation. What is invalidated, in essence, isone’s own emotional experiencing and dysregu-lated responses. The most typical pattern hereis a reaction to emotional pain with intenseself-blame and self-hate. These individualsidentify themselves as perpetrators, which re-sults in intense levels of shame and contempttoward the self (“There is nothing wrong withme, I’m just a bad person”). Mood-dependentperfectionism is also common. Here, the indi-vidual belittles, ignores, or discounts the diffi-culty of his or her own life or may overestimatethe ease of solving current problems. Unfortu-nately, this may initiate the start of a cycle thatmay eventually end in death. Extreme perfec-tionism often ultimately leads to failure, espe-cially in individuals who overestimate theirabilities; the failure then results in self-hatred,which cues suicidal behaviors in these individu-als. Finally, self-invalidation can also be ex-

pressed through willful suppression, meaningthat the individual actively denies the experi-ence of all emotion. Often clients who comeinto our offices simply state, “I don’t do emo-tions.” As with emotion vulnerability, self-invalidation needs to be attended to activelyand directly with individuals with BPD, due tothe lethal consequences of these behaviors.

Active Passivity/Apparent Competence

A second dimension of behavior is a tendencytoward active passivity versus the socially me-diated behavior of apparent competence. Ei-ther pole of this dimension can lead to anger,guilt, or shame on the part of the client, and atendency for the therapist to either under- oroverestimate the client’s capabilities.

“Active passivity” may be defined as passiv-ity in solving one’s own problems, while ac-tively engaging others to solve one’s problems.It can also be described as passivity that ap-pears to be an active process of shutting downin the face of seeing problems coming in the fu-ture. In a sense, individuals with BPD do notappear to have the ability to regulate them-selves internally, particularly when the reg-ulation required is non-mood-dependent be-haviors. Individuals with BPD appear to be“relational selves” rather than “autonomousselves”; that is, they are more highly regulatedby their environment than by internal dia-logues, choices and decisions. Their best formof self-regulation is to regulate their environ-ment, such that it then provides the regulationthey need. The problem here is that managingone’s environment and getting the support oneneeds requires a good deal of emotional consis-tency and regulation, characteristics that ordi-narily are difficult for individuals with BPD.Lorna Benjamin has described this charac-teristic as “My misery is your command”(1996, p. 192).

On the opposite side of the polarity is appar-ent competence. “Apparent competence” refersto the tendency of other individuals to overesti-mate the capabilities of the individual withBPD. Thus, this characteristic is defined by thebehavior of the observer rather than the behav-ior of the individual with BPD. This failure toaccurately perceive their difficulties and “dis-ability” has serious effects on individuals withBPD. Not only do they not get the help theyneed, but also their emotional pain and diffi-culties may easily be invalidated, leading to a


further sense of being misunderstood. A num-ber of behavioral patterns can precipitate thisoverestimation of the competence of the indi-vidual with BPD. Often, a significant discrep-ancy between the individual’s verbal and non-verbal presentations results in the individualwith BPD believing that he or she has suffi-ciently communicated his or her level of dis-tress, when in fact the observer interprets theindividual as effectively managing a difficultsituation. An example would be a womanspeaking nonchalantly and without emotionabout urges toward suicide after a fight withher husband. Individuals with BPD also fre-quently have difficulty generalizing behaviorsacross situations, especially in relationships.For example, the person may be able to copewell in the presence of one person, such as atherapist, but be unable to cope when he or sheis alone or with someone other than the thera-pist. The therapist, understandably, may thenfail to predict the dysregulation that occurs asthe client walks away from the therapy session.Additionally, there may be a difficulty in gener-alizing coping behaviors across differentmoods. In one mood, a problem is solvable; inanother, it is not. This may not be so difficultfor the observer to figure out if mood changesare readily apparent to the observer, but oftenthey are not. Thus, accurate estimates of theperson’s competence actually require the ob-server, such as the therapist, constantly to an-ticipate mood changes that might occur to beable to predict what a client might or might notdo. It is this characteristic, more than anyother, that leads so often to a client walking outof a session, with the therapist believing all iswell, only to end up in the emergency depart-ment with a suicide attempt 2 hours later. Attimes, client failures are nothing more than fail-ures of the therapist (and often the client, also)to predict future behavior accurately.

Unrelenting Crisis/Inhibited Grieving

The third dimension of behavior is the ten-dency of the client with BPD to experience lifeas a series of unrelenting crises as opposed tothe behavior of “inhibited grieving” (i.e., an in-ability to experience emotions associated withsignificant trauma or loss). The client experi-ences each of these extremes in a way that facil-itates movement to the other extreme; forexample, attempting to inhibit emotional expe-

riences related to current crises may result inproblem behaviors that add to existing crises.As with all of these dialectical dilemmas, thesolution is for therapist and client to work to-ward a more balanced position that representsa synthesis of the opposing poles.

Individuals with BPD who experience unre-lenting crises have lives that are often charac-terized as chaotic and in crisis. “Crisis” is de-fined as the occurrence of problems that areextreme, with significant pressure to resolvethem quickly. The consequence of the unrelent-ing crisis is that the individual with BPD, aswell as the person’s environmental resources,such as family, friends, coworkers, and even thetherapist, slowly wear down. There are threetypical scenarios that result in a pattern of un-relenting crisis. First, individuals with extremeimpulsivity and emotion dysregulation engagein behaviors that result in crisis situations. Poorjudgment is a key element to assess when ana-lyzing the impulsive behaviors of individualswith BPD. Second, situations that do not startout as crises can quickly become critical due tothe lack of resources available to many individ-uals with BPD. This may be due to socioeco-nomic status, or to lack of family or peer sup-port. Finally, unrelenting crises can be duesimply to fate or bad luck at a given moment, aphenomenon that is out of the person’s control.For example, an unexpected disaster in a cli-ent’s apartment due to the neighbors runningthe water in their sink for an extended periodof time, might occur. The floors in the client’sapartment are damaged by the water and he orshe does not have the financial resources to payfor renter’s insurance or to replace the carpet inthe apartment. His or her apartment is now un-inhabitable but he or she does not have anyplace else to stay. This problem is out of theperson’s control, but it is still that person’s re-sponsibility to solve.

At the other extreme, and often precipitatedby a crisis, is the phenomenon of “inhibitedgrieving.” In this context, “grief” refers to theprocess of grieving, including experiencing mul-tiple painful emotions associated with loss, par-ticularly traumatic loss, not just the one emotionof deep sadness or grief. Individuals with BPDmay not be able to experience or process the griefrelated to the loss of the life they had expectedfor themselves, and ordinarily do not believethey will recover from the grief if they actuallytry to experience or to cope with it on their own.


As one client said to us, “I don’t do sadness.” An-other said, “I feel sad, I die.” Individuals withBPD may not recognize their own emotionalavoidance and shutdown. Thus, it is crucial thatthe therapist attend to emotional avoidance,particularly of sadness and grief, and assist cli-ents through the grief process. Areas that mustbe confronted, grieved, and finally accepted in-clude an insurmountably painful childhood, abiological makeup that makes life harder ratherthan easier, inability to “fit in” in many environ-ments, absence of loving people in the currentenvironment, or loss of hope for a particular fu-ture for which one had ardently hoped. Whatmust be confronted by the therapist is that egre-gious losses can be real and clients might beright: They really cannot get out of the abyss ifthey fall into it. Regardless of the situation to begrieved, avoidance of these situations may leadto increased shame. The shame is a result of be-lieving that one is unloved, being alone, or fear-ing that one will not be able to cope in the face ofemotional situations. Many of our clients be-lieve that if they begin to address any of these ar-eas, they will not be able to function in their lives,and often this is true. They do not have the skillsor resources to assist them with the process of ex-periencing emotions. We often tell clients thatmanaging grief or processing emotions requiresgoing to the cemetery to pay tribute to what islost, but building a house at the cemetery and liv-ing there is not a good idea. It is a place to visit,experience the sadness of the loss, and thenleave. The use of this metaphor has helped manyof our clients to experience emotion without fall-ing into the abyss.

STAGES OF THERAPYAND TREATMENT GOALS

In theory, treatment of all clients with BPD canbe organized and determined based on theirlevels of disorder, and is conceptualized as oc-curring in stages. “Level of disorder” is definedby the current severity, pervasiveness, complex-ity, disability, and imminent threat presented bythe client. Clients can enter into five stages oftreatment based on their current level of disor-der. First, a pretreatment stage prepares the cli-ent for therapy and elicits a commitment towork toward the various treatment goals. Ori-entation to specific goals and treatment strate-gies, and commitment to work toward goals

addressed during this stage, are likely to be im-portant throughout all stages of treatment. InStage 1 of therapy, the primary focus is on sta-bilizing the client and achieving behavioralcontrol. Out-of-control behaviors constitutethose that are disordered due to the severity ofthe disorder (e.g., as seen in an actively psy-chotic client) or due to severity combined withcomplexity of multiple diagnoses (e.g., as seenin a suicidal client who has BPD with comorbidpanic disorder and depression). Generally, thecriteria for putting a client in Stage 1 are basedon level of current functioning, together withthe inability of the client to work on any othergoals before behavior and functioning comeunder better control. As Mintz (1968) sug-gested in discussing treatment of the suicidalclient, all forms of psychotherapy are ineffec-tive with a dead client. In the subsequent stages(2–4), the treatment goals are to replace “quietdesperation” with nontraumatic emotional ex-periencing (Stage 2); to achieve “ordinary”happiness and unhappiness, and to reduce on-going disorders and problems in living (Stage3); and to resolve a sense of incompleteness andto achieve freedom (Stage 4). In summary, theorientation of the treatment is first to get actionunder control, then to help the client to feelbetter, to resolve problems in living and resid-ual disorder, and to find freedom (and, forsome, a sense of transcendence). All research todate has focused on the severely or multiplydisordered clients who enter treatment at Stage1. Understanding a client’s severity of disorderand level of treatment through accurate andthorough assessment can assist a therapist intwo ways. First, it aids in treatment planningand conceptualization with the client, and inidentifying the appropriate level of careneeded. Second, it can assist a therapist in de-termining whether to accept the client into carebased on the level of severity. For example, if atherapist has multiple Stage 1 clients, he or shemay not want to take on one more Stage 1 cli-ent, until treatment is either completed with theothers or far enough along that the therapistdoes not have multiple clients in crisis at onetime. Furthermore, some therapists may alsouse level of severity to determine that a client’scondition is not severe enough for the type oftreatment the therapist provides; for example,DBT may be too intensive a treatment forsomeone with a single diagnosis of major de-pressive disorder.


Pretreatment: Orienting and Commitment

Specific tasks of orientation are twofold. First,client and therapist must arrive at a mutuallyinformed decision to work together. Typically,the first one to four sessions are presented tothe client as opportunities for client and thera-pist to explore this possibility. Diagnostic inter-viewing, history taking, and formal behavioralanalyses of high-priority, targeted behaviorscan be woven into initial therapy sessions or beconducted separately. Second, client and thera-pist must negotiate a common set of expectan-cies to guide the initial steps of therapy. Agree-ments outlining specifically what the client andtherapist can expect from each other are dis-cussed and agreed to. When necessary, the ther-apist attempts to modify the client’s dysfunc-tional beliefs regarding the process of therapy.Issues addressed include the rate and magni-tude of change that can reasonably be ex-pected, the goals of treatment and generaltreatment procedures, and various myths theclient may have about the process of therapy ingeneral. The dialectical/biosocial view of BPDis also presented. Orientation covers severaladditional points. First, DBT is presented as asupportive therapy requiring a strong collabo-rative relationship between client and thera-pist. DBT is not a suicide prevention program,but a life enhancement program in which clientand therapist function as a team to create a lifeworth living. Second, DBT is described as acognitive-behavioral therapy with a primaryemphasis on analyzing problematic behaviorsand replacing them with skillful behaviors, andon changing ineffective beliefs and rigid think-ing patterns. Third, the client is told that DBTis a skills-oriented therapy, with special empha-sis on behavioral skills training. The commit-ment and orienting strategies, balanced by vali-dation strategies described later, are the mostimportant strategies during this phase of treat-ment. The therapist places a strong effort intogetting the client to commit to not engaging insuicidal or NSSI behaviors for some specifiedperiod of time before allowing the client toleave the session; it can be for 1 year, 6 months,until the next session, or until tomorrow.

Stage 1: Attaining Basic Capacities

The primary focus of the first stage of ther-apy is attaining behavioral control in order tobuild a life pattern that is reasonably func-

tional and stable. Furthermore, DBT does notpromote itself as a suicide prevention pro-gram; instead, it focuses on life improvement.Therefore, the primary treatment goal in DBTand in Stage 1 specifically is to assist clientsin building a life worth living. DBT attainsthis goal by focusing the treatment on specificbehavioral targets agreed upon by both thera-pist and client. Specific targets in order of im-portance are to reduce life-threatening behav-iors (e.g., suicide attempts, increase in suicideideation, NSSI behaviors, homicidal threatsand behaviors), therapy-interfering behaviors(e.g., late to session, missing sessions, not fol-lowing treatment plan, hostile attacks on thetherapist), and quality-of-life–interfering be-haviors (e.g., substance abuse, eating disorder,homelessness, serious Axis I disorders), andto increase behavioral skills. These targets areapproached hierarchically and recursively ashigher-priority behaviors reappear in eachsession. However, this does not mean thatthese behaviors must be addressed in this spe-cific order during a session; it means thatbased on the hierarchy, all relevant behaviormust be addressed at some point within thesession. For example, if a client is 10 minuteslate to session (therapy-interfering behavior)and has cut within the last week (life-threatening behavior), the therapist maychoose to address the therapy-interferingbehavior first, then move on to address thelife-threatening behaviors.

With severely dysfunctional and suicidal cli-ents, significant progress on first stage targetsmay take up to 1 year or more. In addition tothese therapy targets, the goal of increasing di-alectical behaviors is universal to all modes oftreatment. Dialectical thinking encourages cli-ents to see reality as complex and multifaceted,to hold contradictory thoughts simultaneouslyand learn to integrate them, and to be comfort-able with inconsistency and contradictions. Forindividuals with BPD, who are extreme and di-chotomous in their thinking and behavior, thisis a formidable task indeed. A dialectical em-phasis applies equally to a client’s patterns ofbehavior, because the client is encouraged tointegrate and balance emotional and overtbehavioral responses. In particular, dialecticaltensions arise in the areas of skills enhancementversus self-acceptance, problem solving ver-sus problem acceptance, and affect regulationversus affect tolerance. Behavioral extremes,whether emotional, cognitive, or overt re-


sponses, are constantly confronted while morebalanced responses are taught.

Life-Threatening Behaviors

Keeping a client alive must, of course, be thefirst priority in any psychotherapy. Thus, re-ducing suicide crisis behaviors (any behaviorsthat place the client at high and imminent riskfor suicide or threaten to do so, including credi-ble suicide threats, planning, preparations, ob-taining lethal means, and high suicide intent) isthe highest priority in DBT. The target and itspriority are made explicit in DBT during orien-tation and throughout treatment, simply be-cause suicidal behavior and the risk of suicideare of paramount concern for clients with BPD.Similarly, any acute, intentional NSSI behav-iors share the top priority. The priority here isdue both to the risk of suicidal and NSSIbehavior as the single best predictor of subse-quent suicide. Similarly, DBT also targets sui-cide ideation and client expectations about thevalue and long-term consequences of suicidalbehavior, although these behaviors may notnecessarily be targeted directly.

Therapy-Interfering Behaviors

Keeping clients and therapists working to-gether collaboratively is the second explicitlytargeted priority in DBT. The chronic nature ofmost problems among clients with BPD, in-cluding their high tendency to end therapy pre-maturely, and the likelihood of therapist burn-out and iatrogenic behaviors when treatingBPD require such explicit attention. Both clientand therapist behaviors that threaten the rela-tionship or therapeutic progress are addresseddirectly, immediately, consistently, andconstantly—and most importantly, beforerather than after either the therapist or the cli-ent no longer wants to continue. Interfering be-haviors of the client, including those that actu-ally interfere with receiving the therapy (e.g.,lateness to sessions, missed sessions, lack oftransportation to sessions, dissociating in ses-sions) or with other clients benefiting fromtherapy (in group or milieu settings; e.g., sellingdrugs to other clients in the program), andthose that burn out or cross the personal limitsof the therapist (e.g., repeated crisis calls atthree in the morning, repeated verbal attackson the therapist) are treated within therapy ses-sions. Behaviors of the therapist include any

that are iatrogenic (e.g., inadvertently reinforc-ing dysfunctional behaviors), as well as anythat cause the client unnecessary distress ormake progress difficult (e.g., therapist arrivinglate to sessions, missing sessions, not returningphone calls within a reasonable time frame).These behaviors are dealt with in therapy ses-sions, if brought up by either the client or thetherapist, and are also discussed during theconsultation/supervision meeting.

Quality-of-Life–Interfering Behaviors

The third target of Stage 1 addresses all otherbehaviors that interfere with the client havinga reasonable quality of life. Typical behaviorsin this category include serious substanceabuse, severe major depressive episodes, se-vere eating disorders, high-risk and out-of-control sexual behaviors, extreme financialdifficulties (uncontrollable spending or gam-bling, inability to handle finances), criminalbehaviors that are likely to lead to incarcera-tion, employment- or school-related dysfunc-tional behaviors (a pattern of quitting jobs orschool prematurely, getting fired or failing inschool, not engaging in any productive activi-ties), housing-related dysfunctional behaviors(living with abusive people, not finding stablehousing), mental health–related patterns (go-ing in and out of hospitals, failure to take orabuse of necessary medications), and health-related problems (failure to treat serious med-ical disorders). The goal here is for the clientto achieve a stable lifestyle that meets reason-able standards for safety and adequate func-tioning.

Behavioral Skills

The fourth target of Stage 1 is for the client toachieve a reasonable capacity for acquiring andapplying skillful behaviors in the areas of dis-tress tolerance, emotion regulation, interper-sonal effectiveness, self-management, and thecapacity to respond with awareness withoutbeing judgmental (“mindfulness” skills). In ouroutpatient program, the primary responsibilityfor skills training lies with the weekly DBTskills group. The individual therapist monitorsthe acquisition and use of skills over time, andaids the client in applying skills to specificproblem situations in his or her own life. Addi-tionally, it is the role of the individual therapist,not the skills group leader, to provide skills


coaching to the client as needed when problemsarise.

Stage 2: Posttraumatic Stress Reduction

Stage 1 of DBT takes a direct approach to man-aging dysfunctional behavioral and regulatingemotional patterns. Although the connectionbetween current behavior and previous trau-matic events (including those from childhood)may be explored and noted, the focus of thetreatment is distinctly on analyzing the rela-tionship among current thoughts, feelings, andbehaviors, and on accepting and changing cur-rent patterns. The aim of Stage 2 DBT is to re-duce “quiet desperation,” which can be definedas extreme emotional pain in the presence ofcontrol of action (Linehan et al., 1999). A widerange of emotional experiencing difficulties(e.g., avoidance of emotions and emotion-related cues) are targeted in this stage, with thegoal of increasing the capacity for normativeemotional experiencing (i.e., the ability to ex-perience a full range of emotions without eithersevere emotional escalation or behavioraldyscontrol). Because many individuals withBPD have histories of severe and chronic trau-matic experiences, these problems frequentlytake the form of posttraumatic stress disorder(PTSD) and related behaviors and are treatedthrough exposure therapy (formal and infor-mal). Stage 2 addresses four goals: remember-ing and accepting the facts of earlier traumaticevents; reducing stigmatization and self-blamecommonly associated with some types of trau-ma; reducing the oscillating denial and intru-sive response syndromes common among indi-viduals who have suffered severe trauma; andresolving dialectical tensions regarding place-ment of blame for the trauma.

Stage 3: Resolving Problems in Livingand Increasing Respect for Self

In the third stage, DBT targets the client’s unac-ceptable unhappiness and problems in living.At this stage, the client with BPD has eitherdone the work necessary to resolve problems inthe prior two stages or was never severely dis-ordered enough to need it. Although problemsat this stage may still be serious, the individualis functional in major domains of living. Thegoal here is for the client to achieve a level ofordinary happiness and unhappiness, as well as

independent self-respect. To this end, the clientis helped to value, believe in, trust, and validatehim- or herself. The targets here are theabilities to evaluate one’s own behaviornondefensively, to trust one’s own responses,and to hold on to self-evaluations, independentof the opinions of others. Ultimately, the thera-pist must pull back and persistently reinforcethe client’s independent attempts at self-validation, self-care, and problem solving. Al-though the goal is not for clients to become in-dependent of all people, it is important thatthey achieve sufficient self-reliance to relate toand depend on others without self-invalidating.

Stage 4: Attaining the Capacityfor Freedom and Sustained Contentment

The final stage of treatment in DBT targets theresolution of a sense of incompleteness and thedevelopment of a capacity for sustained con-tentment. The focus on freedom encompassesthe goal of freedom from the need to have one’swishes fulfilled, or one’s current life or behav-ioral and emotional responses changed. Herethe goals are expanded awareness, spiritual ful-fillment, and the movement into experiencingflow. For individuals at Stage 4, long-terminsight-oriented psychotherapy, spiritual direc-tion or practices, or other organized experien-tial treatments and/or life experiences may beof most benefit.

STRUCTURING TREATMENT:FUNCTIONS AND MODES

Functions of Treatment

Treatment in DBT is structured around the fiveessential functions it serves. Treatment func-tions to (1) enhance behavioral capabilities byexpanding the individual’s repertoire of skillfulbehavioral patterns; (2) improve the client’smotivation to change by reducing reinforce-ment for dysfunctional behaviors and high-probability responses (cognitions, emotions,actions) that interfere with effective behaviors;(3) ensure that new behaviors generalize fromthe therapeutic to the natural environment; (4)enhance the motivation and capabilities of thetherapist so that effective treatment is ren-dered; and (5) structure the environment sothat effective behaviors, rather than dysfunc-tional behaviors, are reinforced.


Modes of Treatment:Who Does What and When

Responsibility for performing functions andmeeting target goals of treatment in DBT isspread across the various modes of treatment,with focus and attention varying according tothe mode of therapy. The individual therapist(who is always the primary therapist in DBT)attends to one order of targets and is also, withthe client, responsible for organizing the treat-ment so that all goals are met. In skills training,a different set of goals is targeted; during phonecalls, yet another hierarchy of targets takes pre-cedence. In the consultation/supervision mode,therapists’ behaviors are the targets. Therapistsengaging in more than one mode of therapy(e.g., individual, group, and telephone coach-ing) must stay cognizant of the functions andorder of targets specific to each mode, andswitch smoothly from one hierarchy to anotheras the modes of treatment change.

Individual Therapy

DBT assumes that effective treatment must at-tend both to client capabilities and behavioralskills deficits, and to motivational and behav-ioral performance issues that interfere with useof skillful responses (function 2). Althoughthere are many ways to effect these principles,in DBT the individual therapist is responsiblefor the assessment and problem solving of skilldeficits and motivational problems, and for or-ganizing other modes to address problems ineach area.

Individual outpatient therapy sessions arescheduled on a once-a-week basis for 50–90minutes, although twice-weekly sessions maybe held as needed during crisis periods or at thebeginning of therapy. The priorities of specifictargets within individual therapy are the sameas the overall priorities of DBT discussedearlier. Therapeutic focus within individualtherapy sessions is determined by the highest-priority treatment target relevant at the mo-ment. This ordering does not change over thecourse of therapy; however, the relevance of atarget does change. Relevance is determinedby either the client’s most recent, day-to-daybehavior (since the last session) or by currentbehavior during the therapy session. If satisfac-tory progress on one target goal has beenachieved or the behavior has never been a

problem, or if the behavior is currently not evi-dent, then the therapist shifts attention toanother treatment target according to the hier-archy. The consequence of this priority alloca-tion is that when high-risk suicidal behaviorsor intentional self-injury, therapy-interferingbehaviors, or serious quality-of-life–interferingbehaviors are occurring, at least part of the ses-sion agenda must be devoted to each of thesetopics. If these behaviors are not occurring atthe moment, then the topics to be discussedduring Stages 1, 3, and 4 are set by the client.The therapeutic focus (within any topic areadiscussed) depends on the stage of treatment,the skills targeted for improvement, and anysecondary targets. During Stage 1, for example,any problem or topic area can be conceptual-ized in terms of interpersonal issues and skillsneeded, opportunities for emotion regulation,and/or a necessity for distress tolerance. DuringStage 3, regardless of the topic, the therapist fo-cuses on helping the client decrease problems inliving and achieve independent self-respect,self-validation, and self-acceptance both withinthe session and within everyday life. (These are,of course, targets all through the treatment, butthe therapist pulls back further during Stage 3and does less work for the client than duringthe two preceding stages.) During Stage 2, themajor focus is on reducing pervasive “quietdesperation,” as well as changing the extremeemotions and psychological meanings associ-ated with traumatizing cues.

For highly dysfunctional clients, it is likelythat early treatment will necessarily focus onthe upper part of the hierarchy. For example, ifsuicidal or NSSI behavior has occurred duringthe previous week, attention to it takes prece-dence over attention to therapy-interfering be-havior. In turn, focusing on therapy-interferingbehaviors takes precedence over working onquality-of-life–interfering behaviors. Althoughit is often possible to work on more than onetarget (including those generated by the client)in a given session, higher-priority targets al-ways take precedence, but all relevant targetsmust be addressed adequately during the ses-sion. Again, targets do not need to be ad-dressed in sequential order, they just have to beaddressed during the session. Determining therelevance of targeted behaviors is assisted bythe use of diary cards. These cards are filled outby the client during at least the first two stagesof therapy and are brought to weekly sessions.


Failure to complete or to bring in a card is con-sidered a therapy-interfering behavior andshould be openly addressed as such. Diarycards record daily instances of suicidal andNSSI behavior, urges to self-harm or to engagein suicide behaviors (on a 0- to 5-point scale),“misery,” use of substances (licit and illicit),and use of behavioral skills. Other targeted be-haviors (bulimic episodes, daily productive ac-tivities, flashbacks, etc.) may also be recordedon the blank area of the card. The therapist do-ing DBT must develop the pattern of routinelyreviewing the card at the beginning of each ses-sion. The card acts as a road map for each ses-sion; therefore, a session cannot begin until adiary card has been completed. If the card indi-cates that a life-threatening behavior has oc-curred, it is noted and discussed. If high suicideor self-harm urges are recorded, or there is asignificant increase (e.g., an increase of 3 pointsor higher on the 0- to 5-point scale for urges)over the course of the week, they are assessedto determine whether the client is at risk forsuicide. If a pattern of substance abuse or de-pendence appears, it is treated as a quality-of-life–interfering behavior.

Work on targeted behaviors involves a coor-dinated array of treatment strategies, describedlater in this chapter. Essentially, each session isa balance between structured, as well as un-structured, problem solving (including simpleinterpretive activities by the therapist) and un-structured validation. The amount of the thera-pist’s time allocated to each—problem solvingand validating—depends on (1) the urgency ofthe behaviors needing change or problems tobe solved, and (2) the urgency of the client’sneeds for validation, understanding, and accep-tance without any intimation of change beingneeded. However, there should be an overallbalance in the session between change (prob-lem solving) and acceptance (validation) strate-gies. Unbalanced attention to either side mayresult in a nondialectical session, in addition toimpeding client progress.

Skills Training

The necessity of crisis intervention and atten-tion to other primary targets makes skills ac-quisition within individual psychotherapy verydifficult. Thus, a separate component of treat-ment directly targets the acquisition of behav-ioral skills (function 1). In DBT this usuallytakes the form of separate, weekly, 2- to 2½-

hour group skills training sessions that clientsmust attend, ordinarily for a minimum of 6months and preferably for a year. Skills train-ing can also be done individually, although it isoften more difficult to stay focused on teachingnew skills in individual than in group therapy.After a client has gone through all skills mod-ules twice (i.e., for 1 year), remaining in skillstraining is a matter of personal preference andneed. Some DBT programs have developedgraduate groups for individuals who have ac-quired the skills but still need weekly consulta-tion in applying the skills effectively to every-day difficulties. It is important to note thatthere is no research to date on the effectivenessof graduate groups. In adolescent programs,family members are usually invited. Some pro-grams include a separate friends and familiesskills training group as well.

Each group typically has a leader and acoleader. Whereas the primary role of theleader is to teach the skills, the coleader focuseson managing group process by keeping mem-bers both focused and attending to the materialbeing taught, as well as processing the informa-tion (e.g., ensuring everyone is on the correctpage, noticing when the leader’s invalidationhas led to a member shutting down, wakingsomeone up, sitting next to a member who iscrying during group). We have found that it isdifficult to keep the group focused and theleader on schedule for teaching the skills if theleader attempts to manage both roles on his orher own. Oftentimes, the coleader role is themore difficult position to learn.

Skills training in DBT follows a psychoedu-cational format. In contrast to individual ther-apy, in which the agenda is determined primar-ily by the problem to be solved, the skillstraining agenda is set by the skill to be taught.As mentioned earlier, skills training also has ahierarchy of treatment targets that are used tokeep the group focused: (1) therapy-destroyingbehaviors (e.g., using drugs on premises, whichcould lead to the clinic being shut down; prop-erty damage; threatening imminent suicide orhomicidal behavior to a fellow group memberor therapist); (2) increasing skills acquisitionand strengthening; and (3) decreasing therapy-interfering behaviors (e.g., refusing to talk in agroup setting, restless pacing in the middle ofsessions, attacking the therapist and/or thetherapy). However, therapy-interfering behav-iors are not given the attention in skills trainingthat they are given in the individual psycho-


therapy mode. If such behaviors were a pri-mary focus, there would never be time forteaching behavioral skills. Generally, therapy-interfering behaviors are put on an extinctionschedule, while a client is “dragged” throughskills training and simultaneously soothed. InDBT, all skills training clients are required to bein concurrent individual psychotherapy.Throughout group or individual skills training,each client is urged to address other problem-atic behaviors with his or her primary thera-pist; if a serious risk of suicide develops, theskills training therapist refers the problem tothe primary therapist.

Although all of the strategies described be-low are used in both individual psychotherapyand skills training, the mix is decidedly dif-ferent. Skills acquisition, strengthening, andgeneralization strategies are the predominantchange strategies in skills training. In addition,skills training is highly structured, much moreso than the individual psychotherapy compo-nent. Half of each skills training session is de-voted to reviewing homework practice of theskills currently being taught, and the other halfis devoted to presenting and practicing newskills. Except when interpersonal process issuesseriously threaten progress, the agenda andtopics for discussion in skills training are usu-ally set by the group leader.

Four skills modules are taught on a rotatingbasis over the course of 6 months. In standardDBT, mindfulness skills are taught in 2 consec-utive weeks at the beginning of each of the sub-sequent modules. New members are able tojoin a group during either the 2 weeks of mind-fulness or the first 2 weeks of the subsequentmodule. If a new member is not ready to joinafter this point, he or she must wait until thestart of the next mindfulness module.

Mindfulness skills are viewed as central inDBT; thus, they are labeled the “core” skills.These skills represent a behavioral translationof meditation (including Zen and contempla-tive prayer) practice and include observing,describing, spontaneous participating, beingnonjudgmental, focusing awareness, and focus-ing on effectiveness. Unlike standard behaviorand cognitive therapies, which ordinarily focuson changing distressing emotions and events, amajor emphasis of DBT is on learning to man-age pain skillfully. Mindfulness skills reflect theability to experience and to observe one’sthoughts, emotions, and behaviors withoutevaluation, and without attempting to change

or control them. Distress tolerance skills com-prise two types of skills. First, crisis survivalskills are used to regulate behavior in order tomanage painful situations without makingthem worse (e.g., without engaging in life-threatening behavior) until the problem can besolved. Second, accepting reality skills are usedto tolerate the pain of problems that cannot besolved in either the short-term future or thatmay have occurred in the past and, therefore,cannot be changed ever. Emotion regulationskills target the reduction of emotional distressthrough exposure to the primary emotion in anonjudgmental atmosphere. Emotion regula-tion skills include affect identification and la-beling, mindfulness to the current emotions(i.e., experiencing nonjudgmentally), identify-ing obstacles to changing emotions, increasingpositive emotional events, and behavioral ex-pressiveness opposite to the emotion. Interper-sonal effectiveness skills teach effective meth-ods for deciding on objectives within conflictsituations (either asking for something or say-ing “no” to a request) and teach strategies thatmaximize the chances of obtaining those objec-tives without harming the relationship or sacri-ficing self-respect. Self-management skills aretaught in conjunction with the other behavioralskills; however, there is not a specific moduleallocated to these skills, because behavioralprinciples are inherent in all of DBT. Self-management skills include knowledge of thefundamental principles of learning and behav-ior change, and the ability to set realistic goals,to conduct one’s own behavioral analysis, andto implement contingency management plans.

Telephone Consultation

Telephone calls between sessions (or otherextratherapeutic contact when DBT is con-ducted in other settings, e.g., inpatient units)are an integral part of DBT. Telephone consul-tation calls also follow a target hierarchy: (1) toprovide emergency crisis intervention and si-multaneously break the link between suicidalbehaviors and therapist attention; (2) to pro-vide coaching in skills and promote skills gen-eralization (function 3); and (3) to provide acontext for repairing the therapeutic relation-ship, without requiring the client to wait untilthe next session. With respect to calls for skillscoaching, the focus of a phone call varies de-pending on the complexity and severity of theproblem to be solved and the amount of time


the therapist is willing to spend on the phone. Itis important to note that these calls are notconsidered therapy sessions and should not beused as such. Therapists must keep the functionof the call in mind so that they do not beginconducting sessions over the phone; this behav-ior could easily lead to therapist burnout withthe client. With easy or already clear situations,in which it is reasonably easy to determinewhat the client can or should do in the situa-tion, the focus is on helping the client usebehavioral skills (rather than dysfunctional be-haviors) to address the problem. Alternatively,with complex problems, or with problems toosevere for the client to resolve soon, the focus ison ameliorating and tolerating distress, and in-hibiting dysfunctional problem-solving behav-iors until the next therapy session. In the lattercase, resolving the problem that set off the cri-sis is not the target of telephone coaching calls.

With the exception of taking necessary stepsto protect the client’s life when he or she hasthreatened suicide, all calls for help are handledas much alike as possible. This is done to breakthe contingency between suicidal and NSSI be-haviors, and increased phone contact. To dothis, the therapist can do one of two things: re-fuse to accept any calls (including suicide crisiscalls), or insist that the client who calls duringsuicidal crises also call during other crises andproblem situations. As Linehan (1993b) notes,experts on suicidal behaviors uniformly saythat therapist availability is necessary with sui-cidal clients. Thus, DBT chooses the lattercourse and encourages (and at times insists) oncalls during nonsuicidal crisis periods. In DBT,calling the therapist too infrequently, as wellas too frequently, is considered therapy-interfering behavior. Through orientation tocoaching calls during pretreatment the clientlearns what to expect during the calls. For ex-ample, a therapist may communicate to the cli-ent in session what the therapist will ask duringthe call, “What’s the problem? What skills haveyou used? Where is your skills book? Go get it,and let’s figure out what other skills you canuse to get through this situation.” It is impor-tant to highlight that clients and therapists caneasily fall into the trap of considering the act ofcalling for phone consultation a skill. Althoughasking for help may be a current target of treat-ment, it is not considered a skill to be usedwhen the client is in distress. Therapists wantto reinforce the client for effectively reachingout; however, they do not want to reinforce the

client who does not try using actual skills tomanage the problem at hand prior to callingthe therapist.

Additionally, the therapist is balancing thechange-focused strategies with validationthroughout the call. It is important that thetherapist be aware of the contingency manage-ment principles that may be occurring duringthe phone calls to avoid inadvertently reinforc-ing crisis behaviors and to increase therapist–client contact between sessions.

A skills trainer uses phone calls for only onereason: to keep a client in the therapy (includ-ing, of course, when necessary, keeping the cli-ent alive). All other problems are handled bythe primary therapist, and suicidal crises areturned over to the primary therapist as soon aspossible. We have learned that this can be oneof the most difficult distinctions for group lead-ers to uphold. Clients may call group leadersfor a variety of reasons, and it is the role of thegroup leader consistently to refer the clientback to the individual therapist. For example, aclient may call a group leader to ask for assis-tance with the homework assigned the previousweek. Although this may seem appropriate forthe skills trainer to address, it should be re-ferred back to the individual therapist. Atmost, the group leader may repeat what the as-signment was but should not provide anycoaching in how to complete the assignment.

The final priority for phone calls to individ-ual therapists is relationship repair. Clientswith BPD often experience delayed emotionalreactions to interactions that have occurredduring therapy sessions. From a DBT perspec-tive, it is not reasonable to require clients towait up to a whole week before dealing withthese emotions, and it is appropriate for a cli-ent to call for a brief “heart-to-heart” talk. Inthese situations, the role of the therapist is tosoothe and to reassure. In-depth analysesshould wait until the next session.

Consultation Team

DBT assumes that effective treatment of BPDmust pay as much attention to the therapist’sbehavior and experience in therapy as it does tothe client’s. Treating clients with BPD is enor-mously stressful, and staying within the DBTtherapeutic frame can be tremendously difficult(function 4). Thus, an integral part of the ther-apy is the treatment of the therapist. Everytherapist is required to be on a consultation


team either with one other person or with agroup. DBT consultation meetings are heldweekly and are attended by therapists currentlyproviding DBT to clients. At times, the clinicalsetting may require that the team be part of anadministrative meeting due to time and spacerestraints. When this occurs, it is important toset a specific agenda and time limitations oneach part of the meeting (administration, DBT)to ensure that therapist consultation issues areaddressed. The roles of consultation are to holdthe therapist within the therapeutic frame andto address problems that arise in the course oftreatment delivery. Thus, the fundamental tar-get is increasing adherence to DBT principlesfor each member of the consultation group.The DBT consultation team is viewed as an in-tegral component of DBT; that is, it is consid-ered peer group therapy for the therapists, inwhich each member is simultaneously a thera-pist to other members and a client. The focus ison applying DBT strategies to increase DBT-adherent behaviors and decrease non-DBT be-haviors.

There are three primary functions of consul-tation to the therapist in DBT. First, a consulta-tion team helps to keep each individual thera-pist in the therapeutic relationship. The rolehere is to cheerlead and to support the thera-pist. Second, the supervisor or consultationteam balances the therapist in his or her inter-actions with the client. In providing balance,consultants may move close to the therapist,helping him/her maintain a strong position. Orconsultants may move back from the therapist,requiring the therapist to move closer to the cli-ent to maintain balance. Third, within pro-grammatic applications of DBT, the team pro-vides the context for the treatment.

JOINING THE CONSULTATION TEAM

Each team comprises therapists who are cur-rently treating a DBT client or are available totake on a DBT client. Prior to joining the team, itis important that the therapist be completelyaware of his or her commitment. As with clientsduring the pretreatment phase of DBT, thera-pists must make a commitment to the team (seeTable 9.1). A commitment session between thenew member and either the team leader, a teammember, or, in some cases, the entire team can beextraordinarily helpful here. The team memberconducting the commitment session will use thesame strategies and techniques used in a first


TABLE 9.1. DBT Consultation TeamCommitment Session

1. To keep the agreements of the team, especiallyremaining compassionate, mindful, anddialectical.

2. To be available to see a client in whatever roleone has joined the team for (e.g., individualtherapist, group skills trainer, clinicalsupervisor, pharmacotherapist).

3. To function as a therapist in the group (to thegroup) and not just be a silent observer or aperson that only speaks about his or her ownproblems.

4. To treat team meetings in the same way onetreats any other group therapy session (i.e.,attending the weekly meetings [not doublescheduling other events or clients], on time,until the end, with pagers, PDAs, and phonesout of sight and off or, if necessarily on, onsilent).

5. To come to team meetings adequatelyprepared.

6. To be willing to give clinical advice to peoplewho have more experience (especially whenit’s hard to imagine yourself as being able tooffer anything useful).

7. To have the humility to admit your mistakes/difficulties and the willingness to have thegroup help you solve them.

8. To be nonjudgmental and compassionate ofyour fellow clinicians and clients. To ring thebell of nonjudgmentalness to remind yourselfto not be judgmental or unmindful, but not toring it as a proxy for criticizing someone. Thebell is a reminder, not a censor.

9. To properly assess the problem before givingsolutions (do unto others as you wish theywould more often do unto you).

10. To call out “Elephant in the room” whenothers are ignoring or not seeing the elephant.

11. To be willing to go through a chain analysiseven though you were only 31 seconds lateand you would have been there on time if itwere not for that traffic light that alwaystakes all day to change.

12. To participate in team by sharing the roles ofLeader, Observer, Note Taker or other taskscritical to team functioning.

13. If you feel that the consult team is not beinguseful or don’t like the way it is being run,then say something about it rather thansilently stewing in frustration.

14. To repair with the team in some way whenteam meetings are missed, because the team isonly as strong as the weakest link. Therefore,the absence of any team member is felt.

15. To carry on even when feeling burnt out,frustrated, tired, overworked,underappreciated, hopeless, ineffective (easiercommitted to than done, of course).

session with a DBT client (e.g., devil’s advocate,pros and cons, troubleshooting). In addition tothe commitment items listed in Table 9.1, twofundamental commitments must be agreed to byeach member of the team. First, as mentionedpreviously, the primary function of the team isto increase therapists’ motivation and capabilityin providing DBT. Therefore, each memberagrees to work actively toward increasing theteam member’s effectiveness and adherencewhen applying DBT principles and strategies toclients and to other team members. Second, theconsultation team is a community of therapiststreating a community of clients. Thus, eachteam member agrees to be responsible for treat-ment and outcomes of all clients treated by theteam. For example, members of the team areagreeing that if a client being treated by anymember of the team commits suicide, then allmembers will say “Yes” when asked if they haveever had a client commit suicide.

CONSULTATION MEETING FORMAT

There are multiple ways to run a DBT teammeeting. The following is the way we conductour meetings at the University of Washington(although it is important to note that even thisformat could change as needs of memberschange). Each of our DBT teams has an identi-fied team leader. This person is typically themost experienced DBT therapist on the team,and his or her role is to articulate the DBT prin-ciples when necessary for overseeing the fidel-ity of the treatment provided. Additionally, ateam may have an observer who rings a bellwhenever team members make judgmentalcomments (in content or tone) about them-selves, each other, or a client; stay polarizedwithout seeking synthesis; fall out of mindful-ness by doing two things at once; or jump in tosolve a problem before assessing the problem.The point of these observations is not to layblame, but to focus the team’s awareness on thebehavior and move past it.

A team may begin with a mindfulness prac-tice. There are several functions of mindfulnesson a team. First, it helps members transitioninto the team by participating fully and focus-ing on only one thing in the moment, using aDBT mindset. Second, it can provide an oppor-tunity for team members to enhance their skillsin leading and providing feedback about thepractice with other team members. Consulta-

tion team agreements (see Table 9.2) have beendeveloped to facilitate a DBT frame and help tocreate a supportive environment for managingclient–therapist and therapist–therapist diffi-culties. Therefore, a team may elect to read oneor all of the team agreements during the teammeeting. Most importantly, an agenda is set bythe team following the DBT hierarchy of tar-gets, with a specific focus on the needs of thetherapist rather than the problems of the cli-ents. Our agenda at the University of Washing-ton uses the following format; however, thefollowing items can be prioritized differentlybased on the needs of an individual team: (1)the therapists’ need for consultation around cli-ents’ suicidal crises or other life-threatening be-haviors; (2) therapy-interfering behaviors (in-cluding client absences and dropouts, as wellas therapist therapy-interfering behaviors); (3)therapist team-interfering behaviors and burn-out; (4) severe or escalating deterioration inquality-of-life behaviors; (5) reportage of goodnews and therapists’ effective behaviors; (6) asummary of the work of the previous skillsgroup and graduate group by group leaders;and (7) discussion of administrative issues (re-quests to miss team or be out of town, new cli-ent contacts; changes in skills trainers or grouptime, format of consultation group, etc.). Thisagenda spans the 1-hour consultation meeting.Although the agenda may look impossiblylong, therapists ordinarily manage the time bybeing explicit about their need for help andconsultation from the team.

Ancillary Care

When problems in the client’s environment in-terfere with the client’s functioning or progress,the therapist moves to the case managementstrategies. Although not new, case managementstrategies direct the application of core strate-gies (discussed later) to case management prob-lems. There are three case management strat-egies: the consultant-to-the-client strategy,environmental intervention, and the consul-tation/supervision team meeting (describedabove). Because DBT is grounded in dialecticsand avoids becoming rigid, a therapist inter-venes in the client’s environment only undervery specific conditions: (1) The client is unableto act on her own behalf and outcome is ex-tremely important; (2) the environment willonly speak with someone who is in high power


(e.g., the therapist instead of the client); (3)when the client’s or others lives are in imminentdanger; (4) when it is the humane thing to doand will cause no harm; and (5) when the clientis a minor.

CONSULTATION TO THE CLIENT STRATEGY

The consultation-to-the-client strategy was de-veloped with three objectives in mind. First, cli-ents must learn how to manage their own livesand care for themselves by interacting effec-tively with other individuals in the environ-ment, including health care professionals. Theconsultation-to-the-client strategy emphasizesclients’ capacities and targets their ability totake care of themselves. Second, this strategywas designed to decrease instances of “split-ting” between DBT therapists and other indi-viduals interacting with clients. Splitting occurswhen different individuals in a client’s networkhold differing opinions on how to treat the cli-ent. A fundamental tenet of this strategy is thattherapists do not tell others, including otherhealth care professionals, how to treat the cli-ent. The therapist may suggest, but may not de-mand. What this means in practice is that thetherapist is not attached to others treating a cli-ent in a specific way. By remaining in the roleof a consultant to the client, the therapiststays out of such arguments. Finally, theconsultation-to-the-client strategy promotes re-spect for clients by imparting the message thatthey are credible and capable of performing in-terventions on their own behalf.

As mentioned previously, it is the responsi-bility of the individual DBT therapist to coordi-nate and organize care with ancillary treatmentproviders (function 5; e.g., case managers,pharmacotherapists). The consultation-to-the-client strategy balances the consultation-to-the-therapist strategy described earlier, primarilyby providing direct consultation to the client inhow to interact with other providers, ratherthan consulting with the environment on howto interact with the client. Except for specialcircumstances listed earlier, DBT therapists donot discuss clients with ancillary providers, orother individuals in the client’s environment,without the client present. The therapist workswith the client to problem-solve difficulties heor she has with his or her network, leaving theclient to act as the intermediary between thetherapist and other professionals.


TABLE 9.2. DBT ConsultationTeam Agreements

1. Dialectical agreement: We agree to accept adialectical philosophy: There is no absolutetruth. When caught between two conflictingopinions, we agree to look for the truth inboth positions and to search for a synthesis byasking questions such as “What is being leftout?”

2. Consultation to the client agreement: We agreethat the primary goal of this group is toimprove our own skills as DBT therapists, andnot serve as a go-between for clients to eachother. We agree to not treat clients or eachother as fragile. We agree to treat other groupmembers with the belief that others can speakon their own behalf.

3. Consistency agreement: Because change is anatural life occurrence, we agree to acceptdiversity and change as they naturally comeabout. This means that we do not have toagree with each others’ positions about how torespond to specific clients, nor do we have totailor our own behavior to be consistent witheveryone else’s.

4. Observing limits agreement: We agree toobserve our own limits. As therapists andgroup members, we agree to not judge orcriticize other members for having differentlimits from our own (e.g., too broad, toonarrow, “just right”).

5. Phenomenological empathy agreement: Allthings being equal, we agree to search fornonpejorative or phenomenologically empathicinterpretations of our clients’, our own, andother members’ behavior. We agree to assumethat we and our clients are trying our best andwant to improve. We agree to strive to see theworld through our clients’ eyes and throughone another’s eyes. We agree to practice anonjudgmental stance with our clients and withone another.

6. Fallibility agreement: We agree ahead of timethat we are each fallible and make mistakes.We agree that we have probably either donewhatever problematic things we’re beingaccused of, or some part of it, so that we canlet go of assuming a defensive stance to proveour virtue or competence. Because we arefallible, it is agreed that we will inevitablyviolate all of these agreements, and when this isdone, we will rely on each other to point outthe polarity and move to a synthesis.

ENVIRONMENTAL INTERVENTION

As outlined earlier, the bias in DBT is towardteaching the client how to interact effectivelywith his or her environment. The consultation-to-the-client strategy is thus the dominant casemanagement strategy and is used wheneverpossible. There are times, however, when inter-vention by the therapist is needed. In general,the environmental intervention strategy isused over the consultation-to-the-client strat-egy when substantial harm may befall the clientif the therapist does not intervene. The generalrule for environmental intervention is thatwhen clients lack abilities that they need tolearn or that are impossible to obtain, or arenot reasonable or necessary, the therapist mayintervene.

Client Variables

DBT was developed to treat the multidiag-nostic, difficult-to-treat individuals. Therefore,there are a number of requisite client character-istics for Stage 1 DBT. Of these, voluntary par-ticipation and a commitment to a specifiedtime period (e.g., 16 weeks, 6 months to 1 year)are critical. The effective application of DBTrequires a strong interpersonal relationship be-tween therapist and client. The therapist mustfirst work to become a major reinforcer in thelife of the client, then use the relationship topromote change in the client. Continuing therelationship can only be used as a positive con-tingency when a client wants to be in treat-ment; thus, contingency management is seri-ously compromised with involuntary clients.Court-ordered treatment is acceptable, if cli-ents agree to remain in therapy even if the or-der is rescinded. A client characteristic neces-sary for group therapy is the ability to controlovertly aggressive behavior toward others.DBT was developed and evaluated with per-haps the most severely disturbed portion of thepopulation with BPD; all clients accepted intotreatment had histories of multiple suicidal andNSSI behaviors. However, the treatment hasbeen designed flexibly and is likely to be effec-tive with less severely disturbed individuals.

Therapist Variables

In comparison to other aspects of therapy, thetherapist characteristics that facilitate DBThave received comparatively little attention.

However, evidence supports the assumptionthat effective therapy for clients with BPD re-quires the proficient balancing of acceptanceand change strategies (Shearin & Linehan,1992). This research also found that therapists’nonpejorative perceptions of clients were asso-ciated with less suicidal behavior.

Linehan (1993b) describes requisite therapistcharacteristics in terms of three bipolar dimen-sions that must be balanced in the conduct oftherapy. The first dimension represents the bal-ance of an orientation of acceptance with an ori-entation of change. The therapist must be able toinhibit judgmental attitudes (often under verytrying circumstances) and to practice acceptanceof the client, of him- or herself, and of the thera-peutic relationship and process exactly as theseare in the current moment. Nevertheless, thetherapist remains cognizant that the therapeuticrelationship has originated in the necessity ofchange, and he or she assumes responsibility fordirecting the therapeutic influence. Second, thetherapist must balance unwavering centerednesswith compassionate flexibility. “Unwaveringcenteredness” is the quality of believing in one-self, the therapy, and the client. “Compassionateflexibility” is the ability to take in relevant infor-mation about the client and to modify one’s posi-tion accordingly by letting go of a previouslyheld position. In balancing these two dimen-sions, the therapist must be able to observe his orher own limits without becoming overly rigid,especially in the face of attempts by the client tocontrol the therapist’s behaviors. Finally, theDBT therapist must be able to balance a high de-gree of nurturing with benevolent demanding.“Nurturing” refers to teaching, coaching, assist-ing, and strengthening the client, whereas “be-nevolent demanding” requires the therapist torecognize existing capabilities, to reinforceadaptive behavior, and to refuse to “do” for theclient when the client can “do” for him- or her-self. Above all, the ability to demand requires aconcomitant willingness to believe in the client’sability to change; the effective DBT therapistmust see his or her client as empowered.

TREATMENT STRATEGIES

“Treatment strategies” in DBT refer to the roleand focus of the therapist, as well as to a coor-dinated set of procedures that function toachieve specific treatment goals. AlthoughDBT strategies usually consist of a number of


steps, use of a strategy does not necessarily re-quire the application of every step. It is consid-erably more important that the therapist applythe intent of the strategy than that he or sheshould inflexibly lead the client through a se-ries of prescribed maneuvers.

DBT employs five sets of treatment strategiesto achieve the previously described behavioraltargets: (1) dialectical strategies, (2) core strate-gies, (3) stylistic strategies, (4) case manage-ment strategies (discussed earlier), and (5) inte-grated strategies. DBT strategies are illustratedin Figure 9.2. Within an individual session andwith a given client, certain strategies may beused more than others, and all strategies maynot be necessary or appropriate. An abbrevi-ated discussion of the first three types of DBTtreatment strategies follows. For greater detail,the reader is referred to the treatment manual(Linehan, 1993a).

Dialectical Strategies

Dialectical strategies permeate the entire ther-apy, and their use provides the rationale foradding the term “dialectical” to the title of thetherapy. There are three types of dialecticalstrategies: those having to do with how thetherapist structures interactions; those pertain-ing to how the therapist defines and teaches

skillful behaviors; and certain specific strate-gies used during the conduct of treatment.

Dialectics of the Relationship:Balancing Treatment Strategies

“Dialectical strategies” in the most generalsense of the term have to do with how the ther-apist balances the dialectical tensions withinthe therapy relationship. As noted earlier, thefundamental dialectic within any psychothera-py, including that with a client who has BPD, isthat between acceptance of what is and effortsto change what is. A dialectical therapeutic po-sition is one of constant attention to combiningacceptance and change, flexibility and stability,nurturing and challenging, and a focus on ca-pabilities and a focus on limitations and defi-cits. The goals are to bring out the opposites,both in therapy and in the client’s life, and toprovide conditions for syntheses. The pre-sumption is that change may be facilitated byemphasizing acceptance, and acceptance byemphasizing change. The emphasis upon oppo-sites sometimes takes place over time (i.e., overthe whole of an interaction), rather than simul-taneously or in each part of an interaction. Al-though many, if not all, psychotherapies, in-cluding cognitive and behavioral treatments,attend to these issues of balance, placing the


FIGURE 9.2. Treatment strategies in DBT. From Linehan (1993b). Copyright 1993 by The GuilfordPress. Reprinted by permission.

concept of balance at the center of the treat-ment ensures that the therapist remains atten-tive to its importance.

Three primary characteristics are needed tomaintain a dialectical stance in the therapeuticrelationship: movement, speed, and flow.Movement refers to acting with certainty,strength, and total commitment on the part ofthe therapist. If the therapist only moves half-heartedly, the client will only move halfheart-edly. Speed is of the essence and entails keepingthe therapy moving, so that it does not becomerigid or stuck. Finally, flow refers to beingmindful to the moment-to-moment unfoldingof a session and responding smoothly, and withapparent effortlessness.

Teaching Dialectical Behavior Patterns

Dialectical thinking is emphasized throughoutthe entire treatment. Not only does the thera-pist maintain a dialectical stance in his or hertreatment of the client but he or she also fo-cuses on teaching and modeling dialecticalthinking to the client. The therapist helps theclient move from an “either–or” position to a“both–and” position, without invalidating thefirst idea or its polarity when asserting thesecond. Behavioral extremes and rigidity—whether cognitive, emotional, or overtlybehavioral —are signals that synthesis has notbeen achieved; thus, they can be considerednondialectical. Instead, a “middle path” simi-lar to that advocated in Buddhism is advocatedand modeled. The important thing in followingthe path to Enlightenment is to avoid beingcaught and entangled in any extreme and al-ways follow the Middle Way (Kyokai, 1966).This emphasis on balance is similar to the ap-proach advocated in relapse prevention modelsproposed by Marlatt and his colleagues (e.g.,Marlatt & Gordon, 1985) for treating addic-tive behaviors.

Specific Dialectical Strategies

There are eight specific dialectical treatmentstrategies: (1) entering and using paradox, (2)using metaphor, (3) playing the devil’s advo-cate, (4) extending, (5) activating the client’s“wise mind,” (6) making lemonade out of lem-ons (turning negatives into positives), (7) al-lowing natural change (and inconsistencieseven within the therapeutic milieu), and (8) as-

sessing dialectically by always asking the ques-tion “What is being left out here?” Due tospace limitations, a selection of these strategiesis included in the following sections. For acomplete review, the interested reader is re-ferred to the DBT treatment manual (Linehan,1993a).

ENTERING THE PARADOX

Entering the paradox is a powerful techniquebecause it contains the element of surprise. Thetherapist presents the paradox without ex-plaining it and highlights the paradoxical con-tradictions within the behavior, the therapeuticprocess, and reality in general. The essence ofthe strategy is the therapist’s refusal to step inwith rational explanation; the client’s attemptsat logic are met with silence, a question, or astory designed to shed a small amount of lighton the puzzle to be solved. The client is pushedto achieve understanding, to move toward syn-thesis of the polarities, and to resolve the di-lemma him- or herself. Linehan (1993b) hashighlighted a number of typical paradoxes andtheir corresponding dialectical tensions en-countered over the course of therapy. Clientsare free to choose their own behavior but can-not stay in therapy if they do not work atchanging their behavior. They are taught toachieve greater independence by becomingmore skilled at asking for help from others. Cli-ents have a right to kill themselves, but if theyever convince the therapist that suicide is immi-nent, they may be locked up. Clients are not re-sponsible for being the way they are, but theyare responsible for what they become. In high-lighting these paradoxical realities, both clientand therapist struggle with confronting and let-ting go of rigid patterns of thought, emotion,and behavior, so that more spontaneous andflexible patterns may emerge.

USING METAPHOR: PARABLE, MYTH, ANALOGY,AND STORYTELLING

The use of metaphor, stories, parables, andmyth is extremely important in DBT and pro-vides an alternative means of teaching dialecti-cal thinking. Stories are usually more interest-ing, are easier to remember, and encourage thesearch for other meanings of events under scru-tiny. Additionally, metaphors allow clients todistance themselves from the problem being


discussed and can therefore be less threatening.In general, the idea of metaphor is to takesomething the client does understand and use itas an analogy for something the client does notunderstand. Used creatively, metaphors aid un-derstanding, suggest solutions to problems,and reframe the problems of both clients andthe therapeutic process. Furthermore, meta-phors and stories can be developed by boththerapist and client taking turns throughout asession or over the course of treatment. Whenthe therapist and client relate to a metaphor, itcan be a powerful tool to use throughout thetreatment, reminding the client what he or sheis working on. For example, changing behaviorby learning new skills can be compared tobuilding a new hiking trail in the woods. Atfirst, the current trail is defined and easy tonavigate; however, it always leads to a deadend (old dysfunctional behavior). To build anew trail (skillful behaviors), the hiker must re-peatedly go through a new, undefined area un-til it becomes worn in. This takes time and thehiker moves slowly and deliberately clearingaway the brush. Additionally, while the newpath is developing, the old path is slowly be-coming grown over. This story can be returnedto throughout treatment, each time the clientbegins to struggle between trying new skillsand returning to old dysfunctional behavior.

PLAYING DEVIL’S ADVOCATE

The devil’s advocate technique is quite simi-lar to the argumentative approach used inrational–emotive and cognitive restructuringtherapies as a method of addressing a client’sdysfunctional beliefs or problematic rules.With this strategy, the therapist presents apropositional statement that is an extreme ver-sion of one of the client’s own dysfunctional be-liefs, then plays the role of devil’s advocate tocounter the client’s attempts to disprove the ex-treme statement or rule. For example, a clientmay state, “Because I’m overweight, I’d bebetter off dead.” The therapist argues in favorof the dysfunctional belief, perhaps by suggest-ing that because this is true for the client, itmust be true for others as well; hence, all over-weight people would be better off dead. Thetherapist may continue along these lines: “Andsince the definition of what constitutes beingoverweight varies so much among individuals,there must be an awful lot of people who

would be considered overweight by someone.That must mean they’d all be better off dead!”Or “Gosh, I’m about 5 pounds overweight. Iguess that means I’d be better off dead, too.”Any reservations the client proposes can becountered by further exaggeration, until theself-defeating nature of the belief becomes ap-parent. The devil’s advocate technique is oftenused in the first several sessions to elicit astrong commitment from the client and in com-mitment sessions with new therapists joiningthe DBT team. The therapist argues to the cli-ent that since the therapy will be painful anddifficult, it is not clear how making such a com-mitment (and therefore being accepted intotreatment) could possibly be a good idea. Thisusually has the effect of moving the client totake the opposite position in favor of therapeu-tic change. To employ this technique success-fully, it is important that the therapist’s argu-ment seem reasonable enough to invitecounterargument by the client, and that the de-livery be made with a straight face, in a naivebut offbeat manner.

EXTENDING

The term “extending” has been borrowed fromaikido, a Japanese form of self-defense. In thatcontext, extending occurs when the student ofaikido waits for a challenger’s movements toreach their natural completion, then extends amovement’s endpoint slightly further thanwhat would naturally occur, leaving the chal-lenger vulnerable and off balance. In DBT, ex-tending occurs when the therapist takes the se-verity or gravity of what the client iscommunicating more seriously than the clientintends. This strategy is the emotional equiva-lent of the devil’s advocate strategy. It is partic-ularly effective when the client is threateningdire consequences of an event or problem to in-duce change in the environment. Take the inter-action with the following client, who threatenssuicide if an extra appointment time for thenext day is not scheduled. The following inter-change between therapist and client occurs af-ter attempts to find a mutually acceptable timehave failed.

CLIENT: I’ve got to see you tomorrow, or I’msure I will end up killing myself. I just can’tkeep it together by myself any longer.

THERAPIST: Hmm, I didn’t realize you were so


upset! We’ve got to do something immedi-ately if you are so distressed that you mightkill yourself. What about hospitalization?Maybe that is needed.

CLIENT: I’m not going to the hospital! Whywon’t you just give me an appointment?

THERAPIST: How can we discuss such a mun-dane topic as session scheduling when yourlife is in danger? How are you planning tokill yourself?

CLIENT: You know how. Why can’t you cancelsomeone or move an appointment around?You could put an appointment with one ofyour students off until another time. I can’tstand it any more!

THERAPIST: I’m really concerned about you. Doyou think I should call an aid car?

The aspect of the communication that thetherapist takes seriously (suicide as a possibleconsequence of not getting an appointment) isnot the aspect (needing an extra appointmentthe next day) that the client wants taken seri-ously. The therapist takes the consequencesseriously and extends the seriousness even fur-ther. The client wants the problem taken seri-ously, and indeed is extending the seriousnessof the problem.

MAKING LEMONADE OUT OF LEMONS

Making lemonade out of lemons is similar tothe notion in psychodynamic therapy of utiliz-ing a client’s resistances; therapeutic problemsare seen as opportunities for the therapist tohelp the client. The strategy involves takingsomething that is apparently problematic andturning it into an asset. Problems become op-portunities to practice skills; suffering allowsothers to express empathy; weaknesses becomeone’s strengths. To be effective, this strategy re-quires a strong therapeutic relationship be-tween therapist and client; the client must be-lieve that the therapist has a deep compassionfor his or her suffering. The danger in using thisstrategy is that it is easily confused with the in-validating refrain repeatedly heard by clientswith BPD. The therapist should avoid the ten-dency to oversimplify a client’s problems, andrefrain from implying that the lemons in the cli-ent’s life are really lemonade. While recogniz-ing that the cloud is indeed black, the therapistassists the client in finding the positive charac-teristics of a situation—thus, the silver lining.

Core Strategies

Validation

Validation and problem-solving strategies, to-gether with dialectical strategies, make up thecore of DBT and form the heart of the treat-ment. Validation strategies are the most obvi-ous acceptance strategies, whereas problem-solving strategies are the most obvious changestrategies. Both validation and problem-solvingstrategies are used in every interaction with theclient, although the relative frequency of eachdepends on the particular client, the current sit-uation, and the vulnerabilities of that client.However, throughout an entire session, thereshould be an overall balance between the ac-ceptance and change strategies. Many treat-ment impasses are due to an imbalance of onetype of strategy over the other. We discuss vali-dation strategies in this section and problem-solving strategies in the next.

Clients with BPD present themselves clini-cally as individuals in extreme emotional pain.They plead, and at times demand, that theirtherapists do something to change this state ofaffairs. It is very tempting to focus the energyof therapy on changing the client by modifyingirrational thoughts, assumptions, or schemas;critiquing interpersonal behaviors or motivescontributing to interpersonal problems; givingmedication to change abnormal biology; reduc-ing emotional overreactivity and intensity; andso on. In many respects, this focus recapitulatesthe invalidating environment by confirming theclient’s worst fears: The client is the problemand indeed cannot trust his or her own reac-tions to events. Mistrust and invalidation ofhow one responds to events, however, are ex-tremely aversive and can elicit intense fear, an-ger, and shame, or a combination of all three.Thus, the entire focus of change-based therapycan be aversive, because the focus by necessitycontributes to and elicits self-invalidation.However, an entire focus of acceptance-basedtherapy can also be invalidating when it ap-pears to the client that the therapist does nottake his or her problems seriously. Therefore,once again, a dialectical stance focuses on abalance between the two poles.

Validation (according to the Oxford EnglishDictionary; Simpson & Weiner, 1989) means“the action of validating or making valid . . . astrengthening, reinforcement, confirming; an es-tablishing or ratifying.” It also encompasses ac-tivities such as corroborating, substantiating,


verifying, and authenticating. The act of validat-ing is “to support or corroborate on a sound orauthoritative basis . . . to attest to the truth or va-lidity of something” (Merriam-Webster, Inc.,2006). To communicate that a response is validis to say that it is “well-grounded or justifiable:being at once relevant and meaningful . . . logi-cally correct . . . appropriate to the end in view[or effective] . . . having such force as to compelserious attention and [usually] acceptance”(Webster’s Dictionary, 1991). Being “valid im-plies being supported by objective truth or gen-erally accepted authority” (Webster’s Dictio-nary, 1991); “being well-founded on fact, orestablished on sound principles, and thoroughlyapplicable to the case or circumstances,” and“soundness and strength,” “value or worth,”and “efficacy” (Simpson & Weiner, 1989).These are precisely the meanings associated withthe term when used in the context of psychother-apy in DBT:

The essence of validation is this: The therapistcommunicates to the client that her [sic] responsesmake sense and are understandable within her[sic] current life context or situation. The thera-pist actively accepts the client and communicatesthis acceptance to the client. The therapist takesthe client’s responses seriously and does not dis-count or trivialize them. Validation strategies re-quire the therapist to search for, recognize, and re-flect to the client the validity inherent in her [sic]response to events. With unruly children, parentshave to catch them while they’re good in order toreinforce their behavior; similarly, the therapisthas to uncover the validity within the client’s re-sponse, sometimes amplify it, and then reinforceit. (Linehan, 1993b, pp. 222–223, original em-phasis)

Two things are important to note here. First,validation means the acknowledgment of thatwhich is valid. It does not mean “making”valid. Nor does it mean validating that which isinvalid. The therapist observes, experiences,and affirms, but he or she does not create valid-ity. Second, “valid” and “scientific” are notsynonyms. Science may be one way to deter-mine what is valid, logical, sound in principle,and/or generally accepted as authority or nor-mative knowledge. However, an authentic ex-perience or apprehension of private events (atleast, when similar to the same experiences ofothers or when in accord with other, more ob-servable events) is also a basis for claiming va-lidity. Validation can be considered at any one

of six levels. Each level is correspondingly morecomplete than the previous one, and each leveldepends on one or more of the previous levels.They are definitional of DBT and are requiredin every interaction with the client. These levelsare described most fully in Linehan (1997), andthe following definitions are taken from herdiscussion.

LISTENING AND OBSERVING (V1)

Level 1 validation requires listening to and ob-serving what the client is saying, feeling, anddoing, as well as a corresponding active effortto understand what is being said and observed.The essence of this step is that the therapist isstaying awake and interested in the client, pay-ing attention to what the client says and does inthe current moment. The therapist notices thenuances of response in the interaction. Valida-tion at Level 1 communicates that the client perse, as well as the client’s presence, words, andresponses in the session have “such force as tocompel serious attention and [usually] accep-tance” (see earlier definitions of validation;pp. 360–361)

ACCURATE REFLECTION (V2)

The second level of validation is the accuratereflection back to the client of his or her ownfeelings, thoughts, assumptions, and behaviors.The therapist conveys an understanding of theclient by hearing what the client has said andseeing what the client does, and how he or sheresponds. Validation at Level 2 sanctions, em-powers, or authenticates that the individual iswho he or she actually is (p. 362).

ARTICULATING THE UNVERBALIZED (V3)

In Level 3 of validation, the therapist commu-nicates understanding of aspects of the client’sexperience and response to events that have notbeen communicated directly by the client. Thetherapist “mind-reads” the reason for the cli-ent’s behavior and figures out how the clientfeels and what he or she is wishing for, think-ing, or doing just by knowing what has hap-pened to the client. The therapist can make thelink between precipitating event and behaviorwithout being given any information about thebehavior itself. Emotions and meanings the cli-ent has not expressed are articulated by thetherapist (p. 364).


VALIDATING IN TERMS OF PAST LEARNINGOR BIOLOGICAL DYSFUNCTION (V4)

At Level 4, behavior is validated in terms of itscauses. Validation here is based on the notionthat all behavior is caused by events occurringin time; thus, in principle, it is understandable.The therapist justifies the client’s behavior byshowing that it is caused by past events. Eventhough information may not be available to de-termine all the relevant causes, the client’s feel-ings, thoughts and actions make perfect sensein the context of the client’s current experience,physiology, and life to date. At a minimum,what “is” can always be justified in terms ofsufficient causes; that is, what is “should be,”in that whatever was necessary for it to occurhad to have happened (p. 367).

VALIDATION IN TERMS OF PRESENT CONTEXTOR NORMATIVE FUNCTIONING (V5)

At Level 5, the therapist communicates thatbehavior is justifiable, reasonable, well-grounded, meaningful, and/or efficacious interms of current events, normative biologicalfunctioning, and/or the client’s ultimate lifegoals. The therapist looks for and reflects thewisdom or validity of the client’s response andcommunicates that the response is understand-able. The therapist finds the relevant facts inthe current environment that support the cli-ent’s behavior. The therapist is not blinded bythe dysfunctionality of some of the client’s re-sponse patterns to those aspects of a responsepattern that may be either reasonable or ap-propriate to the context. Thus, the therapistsearches the client’s responses for their inherentaccuracy or appropriateness, or reasonableness(as well as commenting on the inherentdysfunctionality of much of the response, ifnecessary) (pp. 370–371).

RADICAL GENUINENESS (V6)

In Level 6, the task is to recognize the person ashe or she is, seeing and responding to thestrengths and capacities of the client, whilekeeping a firm empathic understanding of his orher actual difficulties and incapacities. The ther-apist believes in the client and his or her capacityto change and move toward ultimate life goalsjust as the therapist may believe in a friend orfamily member. The client is responded to as aperson of equal status, due equal respect. Vali-

dation at the highest level is the validation of theindividual as “is.” The therapist sees more thanthe role, more than a “client” or “disorder.”Level 6 validation is the opposite of treating theclient in a condescending manner or as overlyfragile. It is responding to the individual as ca-pable of effective and reasonable behaviorrather than assuming that he or she is an invalid.Whereas Levels 1–5 represent sequential stepsin validation of a kind, Level 6 representschange in both level and kind (p. 377).

Cheerleading strategies constitute anotherform of validation and are the principal strate-gies for combating the active passivity and ten-dencies to hopelessness in clients with BPD. Incheerleading, therapists communicate the beliefthat clients are doing their best and validate cli-ents’ ability to eventually overcome their diffi-culties (a type of validation that, if not handledcarefully, can simultaneously invalidate clients’perceptions of their helplessness). In addition,therapists express a belief in the therapy rela-tionship, offer reassurance, and highlight anyevidence of improvement. Within DBT, cheer-leading is used in every therapeutic interaction.Although active cheerleading by therapistsshould be reduced as clients learn to trust andto validate themselves, cheerleading strategiesalways remain an essential ingredient of astrong therapeutic alliance.

Finally, functional validation, another formof validation that is used regularly in DBT, is aform of nonverbal or behavioral validationthat at times may be more effective than verbalvalidation. For example, a therapist drops a50-pound block on the client’s foot. It would beconsidered invalidating for the therapist simplyto respond verbally, saying, “Wow, I can seethat really hurts! You must be in a lot of pain.”Functional validation would entail the thera-pist removing the block from the client’s foot.

Problem Solving

We have previously discussed how therapieswith a primary focus on client change are typi-cally experienced as invalidating by clients withBPD. However, therapies that focus exclusivelyon validation can prove equally problematic.Exhortations to accept one’s current situationoffer little solace to an individual who experi-ences life as painfully unendurable. WithinDBT, problem-solving strategies are the corechange strategies, designed to foster an activeproblem-solving style. For clients with BPD,


however, the application of these strategies isfraught with difficulties. The therapist mustkeep in mind that with clients with BPD theprocess will be more difficult than with manyother client populations. In work with clientswho have BPD, the need for sympathetic un-derstanding and interventions aimed at en-hancing current positive mood can be ex-tremely important. The validation strategiesjust described, as well as the irreverent commu-nication strategy described later, can be tre-mendously useful here. Within DBT, problemsolving is a two-stage process that concentratesfirst on understanding and accepting a selectedproblem, then generating alternative solutions.The first stage involves (1) behavioral analysis;(2) insight into recurrent behavioral contextpatterns; and (3) giving the client didactic in-formation about principles of behaviors,norms, and so on. The second stage specificallytargets change through (4) analysis of possiblesolutions to problems; (5) orienting the clientto therapeutic procedures likely to bring aboutdesired changes; and (6) strategies designed toelicit and strengthen commitment to these pro-cedures. The following sections specifically ad-dress behavioral analysis, solution analysis,and problem-solving procedures.

Behavioral Analysis

Behavioral analysis is one of the most impor-tant strategies in DBT. It is also the most diffi-cult. The purpose of a behavioral analysis isfirst to select a problem, then to determine em-pirically what is causing it, what is preventingits resolution, and what aids are available forsolving it. Behavioral analysis addresses fourprimary questions:

1. Are ineffective behaviors being reinforced,are effective behaviors followed by aversiveoutcomes, or are rewarding outcomes de-layed?

2. Does the client have the requisite behavioralskills to regulate his or her emotions, re-spond skillfully to conflict, and manage hisor her own behavior?

3. Are there patterns of avoidance, or are ef-fective behaviors inhibited by unwarrantedfears or guilt?

4. Is the client unaware of the contingenciesoperating in his or her environment, or areeffective behaviors inhibited by faulty be-liefs or assumptions?

Answers to these questions guide the thera-pist in the selection of appropriate treatmentprocedures, such as contingency management,behavioral skills training, exposure, or cogni-tive modification. Thus, the value of an analy-sis lies in helping the therapist assess and un-derstand a problem fully enough to guideeffective therapeutic response. The first step inconducting a behavioral analysis is to help theclient identify the problem to be analyzed anddescribe it in behavioral terms. Identifying theproblem can be the most difficult task for thetherapist, and if not done accurately and specif-ically, can lead the therapist and client down apath of solving only a related problem, withoutgetting to the true heart of the problem behav-ior at hand. Problem definition usually evolvesfrom a discussion of the previous week’s events,often in the context of reviewing diary cards.The assumption of facts not in evidence is per-haps the most common mistake at this point.Defining the problem is followed by a chainanalysis—an exhaustive, blow-by-blow de-scription of the chain of events leading up toand following the behavior. In a chain analysis,the therapist constructs a general road map ofhow the client arrives at dysfunctional re-sponses, including where the road actuallystarts (highlights vulnerability factors andprompting events), and notes possible alterna-tive adaptive pathways or junctions along theway. Additional goals are to identify eventsthat automatically elicit maladaptive behavior,behavioral deficits that are instrumental inmaintaining problematic responses, and envi-ronmental and behavioral events that may beinterfering with more appropriate behaviors.The overall goal is to determine the function ofthe behavior, or, from another perspective, theproblem the behavior was instrumental in solv-ing.

Chain analysis always begins with a specificenvironmental event. Pinpointing such an eventmay be difficult, because clients are frequentlyunable to identify anything in the environmentthat set off the problematic response. Never-theless, it is important to obtain a descriptionof the events co-occurring with the onset of theproblem. The therapist then attempts to iden-tify both environmental and behavioral eventsfor each subsequent link in the chain. Here thetherapist must play the part of a very keen ob-server, thinking in terms of very small chunksof behavior, and repeatedly identifying whatthe client was thinking, feeling, and doing, and


what was occurring in the environment frommoment to moment. The therapist asks the cli-ent, “What happened next?” or “How did youget from there to there?” Although, from theclient’s point of view, such links may be self-evident, the therapist must be careful not tomake assumptions. For example, a client whohad attempted suicide once stated that she de-cided to kill herself because her life was toopainful for her to live any longer. From the cli-ent’s point of view, this was an adequate expla-nation for her suicide attempt. For the thera-pist, however, taking one’s life because life istoo painful was only one solution. One coulddecide life is too painful, then decide to changeone’s life. Or one could believe that deathmight be even more painful and decide to toler-ate life despite its pain. In this instance, carefulquestioning revealed that the client actually as-sumed she would be happier dead than alive.Challenging this assumption, then, became akey to ending her persistent suicide attempts. Itis equally important to pinpoint exactly whatconsequences are maintaining the problematicresponse. Similarly, the therapist should alsosearch for consequences that serve to weakenthe problem behavior. As with antecedentevents, the therapist probes for both environ-mental and behavioral consequences, obtainingdetailed descriptions of the client’s emotions,somatic sensations, actions, thoughts, and as-sumptions. A rudimentary knowledge of therules of learning and principles of reinforce-ment is crucial.

The final step in behavioral analysis is toconstruct and test hypotheses about events thatare relevant to generating and maintaining theproblem behavior. The biosocial theory of BPDsuggests several factors of primary importance.For example, DBT focuses most closely on in-tense or aversive emotional states; the amelio-ration of negative affect is always suspected asbeing among the primary motivational vari-ables for dysfunctional behavior in BPD. Thetheory also suggests that typical behavioralpatterns, such as deficits in dialectical thinkingor behavioral skills, are likely to be instrumen-tal in producing and maintaining problematicresponses.

Solution Analysis

Once the problem has been identified and ana-lyzed, problem solving proceeds with an activeattempt at finding and identifying alternative

solutions. DBT posits that there are five re-sponses to any one problem: (1) Solve theproblem; (2) change the emotional reaction tothe problem; (3) tolerate the problem; or (4)stay miserable. An alert client suggested an-other response, which we have added: (5)Make things worse. These five options are pre-sented to the client at pretreatment andthroughout sessions prior to problem solvingto ensure that therapist and client are workingtoward the same goal at any given point.

At times, solutions are discussed throughoutthe behavioral analysis, and pointing to thesealternative solutions may be all that is required,rather than waiting until the behavioral analy-sis is completed. The therapist may ask, “Whatdo you think you could have done differentlyhere?” Throughout this process, the therapist isactively modeling effective problem solvingand solution generation, with a heavier empha-sis on modeling and guiding the client early onin treatment. At other times, a more completesolution analysis is necessary. Here the task isto “brainstorm” or generate as many alterna-tive solutions as possible. Solutions should thenbe evaluated in terms of the various outcomesexpected. The final step in solution analysis isto choose a solution that will somehow be ef-fective. Throughout the evaluation, the thera-pist guides the client in choosing a particularbehavioral solution. Here, it is preferable thatthe therapist pay particular attention to long-term over short-term gain, and that chosen so-lutions render maximum benefit to the clientrather than benefit to others.

Problem-Solving Procedures

DBT employs four problem-solving procedurestaken directly from the cognitive and behavior-al treatment literature. These four—skills train-ing, contingency procedures, exposure, andcognitive modification—are viewed as primaryvehicles of change throughout DBT, since theyinfluence the direction that client changes takefrom session to session. Although they are dis-cussed as distinct procedures by Linehan(1993b), it is not clear that they can in fact bedifferentiated in every case in clinical practice.The same therapeutic sequence may be effec-tive because it teaches the client new skills(skills training), provides a consequence thatinfluences the probability of preceding clientbehaviors occurring again (contingency proce-dures), provides nonreinforced exposure to


cues associated previously but not currentlywith threat (exposure procedures), or changesthe client’s dysfunctional assumptions or sche-matic processing of events (cognitive modifica-tion). In contrast to many cognitive and behav-ioral treatment programs in the literature, theseprocedures (with some exceptions noted be-low) are employed in an unstructured manner,interwoven throughout all therapeutic dia-logue. Thus, the therapist must be well awareof the principles governing the effectiveness ofeach procedure in order to use each in immedi-ate response to events unfolding in a particularsession. The exceptions are in skills training,where skills training procedures predominate,and Stage 2, where exposure procedures pre-dominate.

Skills Training

An emphasis on skills building is pervasivethroughout DBT. In both individual and grouptherapy, the therapist insists at every opportu-nity that the client actively engage in the acqui-sition and practice of behavioral skills. Theterm “skills” is used synonymously with “abil-ity” and includes, in its broadest sense, cogni-tive, emotional, and overt behavioral skills, aswell as their integration, which is necessary foreffective performance. Skills training is calledfor when a solution requires skills not currentlyin the individual’s behavioral repertoire, orwhen the individual has the component behav-iors but cannot integrate and use them effec-tively. Skills training in DBT incorporates threetypes of procedures: (1) skills acquisition (mod-eling, instructing, advising); (2) skills strength-ening (encouraging in vivo and within-sessionpractice, role playing, feedback); and (3) skillsgeneralization (phone calls to work on apply-ing skills; taping therapy sessions to listen tobetween sessions; homework assignments).

Contingency Procedures

Every response within an interpersonal interac-tion is potentially a reinforcement, a punish-ment, or a withholding or removal of rein-forcement. Contingency management requirestherapists to organize their behavior strategi-cally so that client behaviors that representprogress are reinforced, while unskillful ormaladaptive behaviors are extinguished or leadto aversive consequences. Natural conse-quences should be used over arbitrary conse-

quences whenever possible. An important con-tingency for most clients with BPD is the thera-pist’s interpersonal behavior with such clients.The ability of the therapist to influence the cli-ent’s behavior is directly tied to the strength ofthe relationship between the two. Thus, contin-gency procedures based on the relationship areless useful in the very early stages of treatment(except, possibly, when the therapist is the“only game in town”).

A first requirement for effective contingencymanagement is that the therapist orient the cli-ent to the principles of contingency manage-ment and explain how learning takes place.The therapist must attend to the client’s behav-iors and use the principles of shaping to rein-force those behaviors that represent progresstoward DBT targets. Equally important is thatthe therapist takes care not to reinforce behav-iors targeted for extinction. In theory, this mayseem obvious, but in practice, it can be quitedifficult. The problematic behaviors of clientswith BPD are often quite effective in obtainingreinforcing outcomes or in stopping painfulevents. Indeed, the very behaviors targeted forextinction have been intermittently reinforcedby mental health professionals, family mem-bers, and friends. Contingency management attimes requires the use of aversive consequences,similar to “setting limits” in other treatmentmodalities. Three guidelines are importantwhen using aversive consequences. First, pun-ishment should “fit the crime,” and a clientshould have some way of terminating its appli-cation. For example, in DBT, a detailed behav-ioral analysis follows a suicidal or NSSI act;such an analysis is an aversive procedure formost clients. Once it has been completed, how-ever, a client’s ability to pursue other topics isrestored. Second, it is crucial that therapists usepunishment with great care, in low doses, andvery briefly, and that a positive interpersonalatmosphere be restored following any clientimprovement. Third, punishment should bejust strong enough to work. Although the ulti-mate punishment is termination of therapy, apreferable fallback strategy is putting clients on“vacations from therapy.” This approach isconsidered when all other contingencies havefailed, or when a situation is so serious that atherapist’s therapeutic or personal limits havebeen crossed. When utilizing this strategy, thetherapist clearly identifies what behaviors mustbe changed and clarifies that once the condi-tions have been met, the client can return. The


therapist maintains intermittent contact byphone or letter, and provides a referral orbackup while the client is on vacation. (In col-loquial terms, the therapist kicks the client out,then pines for his or her return.) Observing lim-its constitutes a special case of contingencymanagement involving the application ofproblem-solving strategies to client behaviorsthat threaten or cross a therapist’s personal lim-its. Such behaviors interfere with the therapist’sability or willingness to conduct the therapy,thus constituting a special type of therapy-interfering behavior. Therapists must take re-sponsibility for monitoring their own personallimits and clearly communicate to their clientswhich behaviors are tolerable and which arenot. Therapists who do not do this eventuallyburn out, terminate therapy, or otherwise harmtheir clients. DBT favors natural over arbitrarylimits. Thus, limits vary among therapists, andwith the same therapist over time and circum-stance. Limits should also be presented as forthe good of the therapist, not for the good ofthe client. The effect of this is that although cli-ents may argue about what is in their own bestinterests, they do not have ultimate say overwhat is good for their therapists.

Cognitive Modification

The fundamental message given to clients inDBT is that cognitive distortions are just aslikely to be caused by emotional arousal as tobe the cause of the arousal in the first place.The overall message is that, for the mostpart, the source of a client’s distress is the ex-tremely stressful events of his or her liferather than a distortion of events that are ac-tually benign. Although direct cognitive re-structuring procedures, such as those advo-cated by Beck and colleagues (Beck, Brown,Berchick, Stewart, & Steer, 1990; Beck, Rush,Shaw, & Emery, 1979) and by Ellis (1962,1973), are used and taught as part of emo-tion regulation, they do not hold a dominantplace in DBT. In contrast, contingency clarifi-cation strategies are used relentlessly, high-lighting contingent relationships operating inthe here and now. Emphasis is placed onhighlighting immediate and long-term effectsof clients’ behavior (both on themselves andon others), clarifying the effects of certain sit-uations on clients’ own responses, and exam-ining future contingencies that clients arelikely to encounter. An example here is ori-

enting a client to DBT as a whole and totreatment procedures as they are imple-mented.

Exposure

All of the change procedures in DBT can bereconceptualized as exposure strategies. Manyof the principles of exposure as applied to DBThave been developed by researchers in expo-sure techniques (see Foa & Kozak, 1986; Foa,Steketee, & Grayson, 1985). These strategieswork by reconditioning dysfunctional associa-tions that develop between stimuli (e.g., anaversive stimulus, hospitalization, may becomeassociated with a positive stimulus, nurturingin the hospital; a client may later work to behospitalized) or between a response and a stim-ulus (e.g., an adaptive response, healthy ex-pression of emotions, is met with an aversiveconsequent stimulus, rejection by a loved one;a client may then try to suppress emotions). Asnoted earlier, the DBT therapist conducts achain analysis of the eliciting cue, the problembehavior (including emotions), and the conse-quences of the behavior. Working within abehavior therapy framework, the therapist op-erates according to three guidelines for expo-sure in DBT. First, exposure to the cue thatprecedes the problem behavior must be non-reinforced (e.g., if a client is fearful that dis-cussing suicidal behavior will lead to his or herbeing rejected, the therapist must not reinforcethe client’s shame by ostracizing him/her). Sec-ond, dysfunctional responses are blocked in theorder of the primary and secondary targets oftreatment (e.g., suicidal or NSSI behavior re-lated to shame is blocked by getting the client’scooperation in throwing away hoarded medi-cations). Third, actions opposite to the dys-functional behavior are reinforced (e.g., thetherapist reinforces the client for talking aboutpainful, shame-related suicidal behavior).

Therapeutic exposure procedures are usedinformally throughout the whole of therapyand formally during Stage 2, in which the clientis systematically exposed to cues of previoustraumatic events. Exposure procedures of theDBT therapist involve first orienting the clientto the techniques and to the fact that exposureto cues is often experienced as painful or fright-ening. Thus, the therapist does not remove thecue to emotional arousal, and at the same timehe or she blocks both the action tendencies (in-cluding escape responses) and the expressive


tendencies associated with the problem emo-tion. In addition, the DBT therapist works toassist the client in achieving enhanced controlover aversive events. A crucial step of exposureprocedures is that the client be taught how tocontrol the event. It is critical that the clienthave some means of titrating or ending expo-sure when emotions become unendurable. Thetherapist and client should collaborate in devel-oping positive, adaptive ways for the client toend exposure voluntarily, preferably after somereduction in the problem emotion has oc-curred.

Stylistic Strategies

DBT balances two quite different styles of com-munication that refer to how the therapist exe-cutes other treatment strategies. The first, re-ciprocal communication, is similar to thecommunication style advocated in client-centered therapy. The second, irreverent com-munication, is quite similar to the style advo-cated by Whitaker (1975) in his writings onstrategic therapy. Reciprocal communicationstrategies are designed to reduce a perceivedpower differential by making the therapistmore vulnerable to the client. In addition, theyserve as a model for appropriate but equal in-teractions within an important interpersonalrelationship. Irreverent communication is usu-ally riskier than reciprocity. However, it can fa-cilitate problem solving or produce a break-through after long periods when progress hasseemed thwarted. To be used effectively, irrev-erent communication must balance reciprocalcommunication, and the two must be woveninto a single stylistic fabric. Without such bal-ancing, neither strategy represents DBT.

Reciprocal Communication

Responsiveness, self-disclosure, warm engage-ment, and genuineness are the basic guidelinesof reciprocal communication. Responsivenessrequires attending to the client in a mindful (at-tentive) manner and taking the client’s agendaand wishes seriously. However, this does notmean that the therapist gives priority to the cli-ent’s agenda over the treatment hierarchy. It re-fers to the therapist validating the importanceof the client’s agenda openly. It is a friendly, af-fectionate style reflecting warmth and engage-ment in the therapeutic interaction. Both self-involving and personal self-disclosure, used in

the interests of the client, are encouraged to in-crease problem solving or to reinforcetherapeutic activities. Self-involving self-disclosure is the therapist’s immediate, personalreactions to the client and his or her behavior.This strategy is used frequently throughoutDBT. For example, a therapist whose clientcomplained about his coolness said, “Whenyou demand warmth from me, it pushes meaway and makes it harder to be warm.” Simi-larly, when a client repeatedly failed to fill outdiary cards but nevertheless pleaded with hertherapist to help her, the therapist responded,“You keep asking me for help, but you won’tdo the things I believe are necessary to helpyou. I feel frustrated because I want to helpyou, but I feel that you won’t let me.” Suchstatements serve both to validate and to chal-lenge. They constitute both an instance of con-tingency management, because therapist state-ments about the client are typically experiencedas either reinforcing or punishing, and an in-stance of contingency clarification, because theclient’s attention is directed to the conse-quences of his or her interpersonal behavior.Self-disclosure of professional or personal in-formation is used to validate and model copingand normative responses. The key point here isthat a therapist should only use personal exam-ples in which he or she has successfully mas-tered the problem at hand. This may seem likean obvious point, but it is very easy to fall intothis pit by trying actively to validate the client’sdilemma. For example, when working with aclient whose goal is to wake up early eachmorning to exercise but who is having diffi-culty getting out of bed, the therapist may at-tempt to validate the behavior as normative bystating, “Yeah, I struggle with getting up everymorning, too, even though I tell myself everynight that I am going to exercise in the morn-ing.” However, this self-disclosure is only beuseful to the client if the therapist continues bystating what skillful behavior he or she uses toget up each morning and exercise successfully.

Irreverent Communication

Irreverent communication is used to push theclient “off balance,” get the client’s attention,present an alternative viewpoint, or shift affec-tive response. It is a highly useful strategy whenthe client is immovable, or when therapist andclient are “stuck.” It has an “offbeat” flavorand uses logic to weave a web the client cannot


escape. Although it is responsive to the client,irreverent communication is almost never theresponse the client expects. For irreverence tobe effective it must be both genuine (vs. sarcas-tic or judgmental) and come from a place ofcompassion and warmth toward the client.Otherwise, the client may become even morerigid. When using irreverence the therapisthighlights some unintended aspect of the cli-ent’s communication or “reframes” it in an un-orthodox manner. For example, if the clientsays, “I am going to kill myself,” the therapistmight say, “I thought you agreed not to dropout of therapy.” Irreverent communication hasa matter-of-fact, almost deadpan style that is insharp contrast to the warm responsiveness ofreciprocal communication. Humor, a certainnaivete, and guilelessness are also characteristicof the style. A confrontational tone is also ir-reverent, communicating “bullshit” to re-sponses other than the targeted adaptive re-sponse. For example, the therapist might say,“Are you out of your mind?” or “You weren’tfor a minute actually believing I would thinkthat was a good idea, were you?” The irrever-ent therapist also calls the client’s bluff. For theclient who says, “I’m quitting therapy,” thetherapist might respond, “Would you like a re-ferral?” The trick here is to time the bluff care-fully, with the simultaneous provision of asafety net; it is important to leave the client away out.

CASE STUDY

Background

At the initial meeting, “Cindy,” a 30-year-old,white, married woman with no children, wasliving in a middle-class suburban area with herhusband. She had a college education and hadsuccessfully completed almost 2 years of medi-cal school. Cindy was referred to one of us (M.M. L.) by her psychiatrist of 1½ years, whowas no longer willing to provide more thanpharmacotherapy following a recent hospital-ization for a near-lethal suicide attempt. In the2 years prior to referral, Cindy had been hospi-talized at least 10 times (once for 6 months) forpsychiatric treatment of suicidal ideation; hadengaged in numerous instances of both NSSIbehavior and suicide attempts, including atleast 10 instances of drinking Clorox bleach,multiple deep cuts, and burns; and had hadthree medically severe or nearly lethal suicide

attempts, including cutting an artery in herneck. At the time of referral, Cindy met DSM-III-R (American Psychiatric Association, 1987)as well as Gunderson’s (1984) criteria for BPD.She was also taking a variety of psychotropicdrugs. Until age 27, Cindy was able to functionwell in work and school settings, and her mar-riage was reasonably satisfactory to both part-ners, although her husband complained aboutCindy’s excessive anger. When Cindy was in thesecond year of medical school, a classmate sheknew only slightly committed suicide. Cindystated that when she heard about the suicide,she immediately decided to kill herself also, buthad very little insight into what about the situa-tion actually elicited her inclination to kill her-self. Within weeks she left medical school andbecame severely depressed and actively sui-cidal. Although Cindy self-presented as a per-son with few psychological problems beforethe classmate’s suicide, further questioning re-vealed a history of severe anorexia nervosa,bulimia nervosa, and alcohol and prescriptionmedication abuse, originating at the age of 14years. Indeed, she had met her husband at anAlcoholics Anonymous (AA) meeting while at-tending college. Nevertheless, until the stu-dent’s suicide in medical school, Cindy hadbeen successful at maintaining an overall ap-pearance of relative competence.

Treatment

At the initial meeting, Cindy was accompaniedby her husband, who stated that he and Cindy’sfamily considered his wife too lethally suicidalto be out of a hospital setting. Consequently, heand Cindy’s family were seriously contemplat-ing the viability of finding long-term outpatientcare. However, Cindy stated a strong prefer-ence for inpatient treatment, although no ther-apist in the local area other than M. M. L. ap-peared willing to take her into outpatienttreatment. The therapist agreed to acceptCindy into therapy, contingent on the client’sstated commitment to work toward behavioralchange and to stay in treatment for at least 1year. (It was later pointed out repeatedly thatthis also meant the client had agreed not tocommit suicide.) Thus, the therapist began thecrucial first step of establishing a strong thera-peutic alliance by agreeing to accept the clientdespite the fact that no one else was willing todo so. She pointed out, however, that accep-tance into therapy did not come without a cost.


In this manner, the therapist communicated ac-ceptance of the client exactly as she was in thecurrent moment, while concomitantly makingclear that Cindy’s commitment toward changewas the foundation of the therapeutic alliance.At the fourth therapy session, Cindy reportedthat she felt she could no longer keep herselfalive. When reminded of her previous commit-ment to stay alive for 1 year of therapy, Cindyreplied that things had changed and she couldnot help herself. Subsequent to this session, al-most every individual session for the next 6months revolved around the topic of whether(and how) to stay alive versus committing sui-cide. Cindy began coming to sessions wearingmirrored sunglasses and would slump in herchair or ask to sit on the floor. Questions fromthe therapist were often met with a minimalcomment or long silences. In response to thetherapist’s attempts to discuss prior self-injurious behavior, Cindy would become angryand withdraw (slowing down the pace of ther-apy considerably). The client also presentedwith marked dissociative reactions, whichwould often occur during therapy sessions.During these reactions, Cindy would appearunable to concentrate or hear much of whatwas being said. When queried by the therapist,Cindy would describe her experience as feeling“spacey” and distant. The client stated that shefelt she could no longer engage in many activi-ties, such as driving, working, or attendingschool. Overall, the client viewed herself as in-competent in all areas.

The use of diary cards, which Cindy filledout weekly (or at the beginning of the session,if she forgot), assisted the therapist in carefullymonitoring Cindy’s daily experiences of sui-cidal ideation, misery, and urges to harm her-self, as well as actual suicide attempts and NSSIbehaviors. Behavioral analyses that attemptedto identify the sequence of events leading up toand following Cindy’s suicidal behavior soonbecame an important focus of therapy. At everypoint the therapist presented self-injuriousbehavior as to be expected, given the strengthof the urge (but considered it ultimatelybeatable), and pointed out repeatedly that ifthe client committed suicide, therapy would beover, so they had better work really hard now,while Cindy was alive.

Over the course of several months, thebehavioral analyses began to identify a fre-quently recurring behavioral pattern that pre-ceded suicidal behaviors. For Cindy, the chain

of events would often begin with an interper-sonal encounter (almost always with herhusband), which culminated in her feelingthreatened, criticized, or unloved. These feel-ings were often followed by urges either to self-mutilate or to kill herself, depending somewhaton the covarying levels of hopelessness, anger,and sadness. Decisions to self-mutilate and/orto attempt suicide were often accompanied bythe thought, “I’ll show you.” At other times,hopelessness and a desire to end the pain per-manently seemed predominant. Both are exam-ples of emotional vulnerability. Following theconscious decision to self-mutilate or to at-tempt suicide, Cindy would then immediatelydissociate and at some later point cut or burnherself, usually while in a state of “automaticpilot.” Consequently, Cindy often had diffi-culty remembering specifics of the actual acts.At one point, Cindy burned her leg so badly(and then injected it with dirt to convince thedoctor that he should give her more attention)that reconstructive surgery was required.Behavioral analyses also revealed that dissocia-tion during sessions usually occurred followingCindy’s perception of the therapist’s disap-proval or invalidation, especially when thetherapist appeared to suggest that change waspossible. The therapist targeted in-session dis-sociation by immediately addressing it as it oc-curred.

By several months into therapy, an appar-ently long-standing pattern of suicidal behav-iors leading to inpatient admission was ap-parent. Cindy would report intense suicidalideation, express doubts that she could resistthe urge to kill herself, and request admissionto her preferred hospital; or, without warning,she would cut or burn herself severely and re-quire hospitalization for medical treatment. At-tempts to induce Cindy to stay out of the hospi-tal or to leave the hospital before she was readytypically resulted in an escalation of suicidality,followed by her pharmacotherapist’s (a psychi-atrist) insistence on her admission or the hospi-tal’s agreement to extend her stay. Observationof this behavioral pattern led the therapist tohypothesize that the hospitalization itself wasreinforcing suicidal behavior; consequently,she attempted to change the contingencies forsuicidal behaviors. Using didactic and con-tingency clarification strategies, the therapistattempted to help Cindy understand how hos-pitalization might be strengthening the verybehavior they were working to eliminate. This


issue became a focal point of disagreementwithin the therapy, with Cindy viewing thetherapist’s position as unsympathetic and lack-ing understanding of her phenomenal experi-ence. In Cindy’s opinion, the intensity of heremotional pain rendered the probability of sui-cide so high that hospitalization was necessaryto guarantee her safety. She would buttress herposition by citing frequently her difficultieswith dissociative reactions, which she reportedas extremely aversive and which, in her opin-ion, made her unable to function much of thetime. From the therapist’s perspective, the dele-terious long-term risk of suicide created by re-peated hospitalization in response to suicidalbehavior was greater than the short-term riskof suicide if hospitalization stays were reduced.These differences in opinion led to frequent dis-agreements within sessions. It gradually be-came clear that Cindy viewed any explanationsof her behavior as influenced by reinforcementas a direct attack; she implied that if hospital-ization was reinforcing her suicidal behavior,then the therapist must believe that the purposeof her suicidality was for admission into thehospital. This was obviously not the case (atleast some of the time), but all attempts to ex-plain reinforcement theory in any other termsfailed. The therapist compensated somewhatfor insisting on the possibility that she (thetherapist) was correct by doing three things.First, she repeatedly validated the client’s expe-rience of almost unendurable pain. Second, shemade certain to address the client’s dissociativebehavior repeatedly, explaining it as an auto-matic reaction to intensely painful affect (or thethreat of it). Third, she frequently addressedthe quality of the relationship between Cindyand herself to strengthen the relationship andmaintain Cindy in therapy, even though to doso was a source of even more emotional pain.By the fifth month, the therapist became con-cerned that the current treatment regimen wasgoing to have the unintended consequence ofkilling the client (via suicide). At this point, thetherapist’s limits for effective treatment werecrossed; therefore, she decided to employ theconsultation-to-the-client strategy to addressCindy’s hospitalizations. The first-choice strat-egy would have been to get Cindy to negotiatea new treatment plan with her preferred hospi-tal and admitting psychiatrist. Cindy refused togo along, however, because she disagreed withthe wisdom of changing her current unlimitedaccess to the inpatient unit. The therapist was

able to get her to agree to a consultationmeeting with all of her treatment providers,and, with some tenacity, the therapist actuallygot Cindy to make all the calls to set up themeeting (including inviting her insurance moni-tor, who was coordinating payment for treat-ment).

At the case conference, the therapist pre-sented her hypothesis that contingent hospital-ization was reinforcing Cindy’s suicidal behav-ior. She also assisted Cindy in making the casethat she (the therapist) was wrong. Using recip-rocal communication and contingency man-agement, the therapist stated that she simplycould not conduct a therapy she thought mightkill the client (and she had to go along withwhat she thought was best even if she werewrong—“to do otherwise would be unethi-cal”), and she requested that a new system ofcontingencies be agreed upon to disrupt thefunctional relationship between Cindy’s sui-cidal behavior and hospitalization. Therefore,a plan was developed wherein the client wasnot required to be suicidal to gain hospital ad-mittance. Under this new set of contingencies,Cindy could elect, at will, to enter the hospitalfor a stay of up to 3 days, at the end of whichtime she would always be discharged. If sheconvinced people that she was too suicidal fordischarge, she would be transferred to herleast-preferred hospital for safety. Suicidal andNSSI behaviors would no longer be groundsfor admission except to a medical unit, whenrequired. Although there was some disagree-ment as to the functional relationship betweensuicidal behavior and hospitalization, this sys-tem was agreed upon. Following this meeting,Cindy’s husband announced that he was nolonger able to live with or tolerate his wife’ssuicidal behavior, and that the constant threatof finding her dead had led to his decision tofile for divorce. The focus of therapy thenshifted to helping Cindy grieve over this eventand find a suitable living arrangement. Cindyalternated between fury that her husbandwould desert her in her hour of need (or “ill-ness,” as she put it) and despair that she couldever cope alone. She decided that “getting herfeelings out” was the only useful therapy. Thisled to many tearful sessions, with the therapistsimultaneously validating the pain; focusing onCindy’s experiencing the affect in the moment,without escalating or blocking it; and cheer-leading Cindy’s ability to manage without go-ing back into the hospital. Due to Cindy’s high


level of dysfunctionality, she and her therapistdecided that she would enter a residential treat-ment facility for a 3-month period. The facilityhad a coping skills orientation and providedgroup but not individual therapy. Cindy sawher therapist once a week and talked to her sev-eral times a week during this period. With somecoaching, Cindy looked for and found a room-mate to live with and returned to her ownhome at the end of 3 months (the ninth monthof therapy). Over the course of treatment, thetherapist used a number of strategies to treatCindy’s suicidal, NSSI, and therapy-interferingbehaviors. In-depth behavioral chain and solu-tion analysis helped the therapist (and some-times the client) gain insight into the factors in-fluencing current suicidal behavior. For Cindy,as for most clients, performing these analyseswas quite difficult, because the process usuallygenerated intense feelings of shame, guilt, oranger. Thus, behavioral analysis also func-tioned as an exposure strategy, encouraging theclient to observe and experience painful affect.It additionally served as a cognitive strategy inhelping to change Cindy’s expectancies con-cerning the advantages and disadvantages ofsuicidal behavior, especially as the therapist re-peatedly made statements such as “How doyou think you would feel if I got angry at youand then threatened suicide if you didn’tchange?” Finally, behavioral analysis served ascontingency management, in that the client’sability to pursue topics of interest in therapysessions was made contingent on the successfulcompletion of chain and solution analysis.

Cindy presented early in therapy with ex-ceedingly strong perceptions as to her needsand desires, and with a concomitant willing-ness to engage in extremely lethal suicidalbehavior. As previously mentioned, several ofthese acts were serious attempts to end her life,whereas others functioned as attempts to gainattention and care from significant others. Thisclient also presented with an extreme sensitiv-ity to any attempts at obvious change proce-dures, which she typically interpreted as com-municating a message about her incompetenceand unworthiness. Although Cindy initiallycommitted herself to attending weekly groupskills training for the first year of therapy, herattendance at group meetings was quite erratic,and she generally tended either to miss entiresessions (but never more than three in a row) orto leave during the break. Cindy answered thetherapist’s attempts to address this issue by

stating that she could not drive at night due tonight blindness. Although considered atherapy-interfering behavior and frequently ad-dressed over the course of therapy, missingskills training was not a major focus of treat-ment, due to the continuing presence of higher-priority suicidal behavior. The therapist’sefforts to engage the client in active skills ac-quisition during individual therapy sessionswere also somewhat limited and were alwayspreceded by obtaining Cindy’s verbal commit-ment to problem solving. The stylistic strategyof irreverent communication was of value tothe therapeutic process. The therapist’s irrever-ence often served to “shake up” the client, re-sulting in a loosening of dichotomous thinkingand maladaptive cognitions. The result of thiswas Cindy’s increased willingness to explorenew and adaptive behavioral solutions. Finally,relationship strategies were heavily employedas tools to strengthen the therapeutic allianceand to keep it noncontingent on suicidal and/ordissociative behaviors. Included here werebetween-session therapist-initiated telephonecalls to see how Cindy was doing, the therapistroutinely giving out phone numbers when shewas traveling, and sending the client postcardswhen she was out of town.

By the 12th month of therapy, Cindy’s sui-cidal and self-injurious behavior, as well asurges to engage in such behavior, receded. Inaddition, her hospital stays were reducedmarkedly, with none occurring after the eighthmonth. While living at home with a roommate,Cindy was readmitted to medical school. Partof the reason for returning to school was toturn her life around, so that she could try to re-gain her husband’s love and attention, or atleast his friendship. As the therapy continuedto focus on changing the contingencies of sui-cidal behavior, reducing both emotional painand inhibition, and tolerating distress, a furtherfocus on maintaining sobriety and reasonablefood intake was added. During the first monthsof living in her home without her husband,Cindy had several alcoholic binges, and herfood intake dropped precipitously. These be-haviors became immediate targets. The thera-pist’s strong attention to these behaviors alsocommunicated to Cindy that the therapistwould take her problems seriously even if shewere not suicidal. Therapy focused as well onexpanding her social network. As with suicidalbehaviors, attention to these targets served as apathway to treating associated problems. As


crisis situations decreased in frequency, muchgreater attention was paid to analyzing familypatterns, including experiences of neglect andinvalidation, that might have led to Cindy’sproblems in later life. Cindy did not report ahistory of sexual or physical abuse. Thus, theexplicit goal of Stage 2 (which was being cau-tiously entered as an overlap to Stage 1) was tounderstand Cindy’s history and its relationshipto her current problems.

In other cases, especially when there hasbeen sexual and/or physical abuse in child-hood, movement to Stage 2 before Stage 1 tar-gets have been mastered is likely to result inretrogression to previously problematic behav-iors. For example, another client treated by thesame therapist (M. M. L.), Terry, had beenquite seriously abused physically by her motherthroughout childhood and sexually abused byher father, beginning at age 5. The sexual ad-vances were nonviolent at first but becamephysically abusive at approximately age 12.Prior to this therapy, Terry had not disclosedthe incidents of abuse to anyone.

After successful negotiation of Stage 1 tar-gets, the therapist proceeded to expose Terry totrauma-related cues by simply having her beginto disclose details of the abuse. These exposuresessions were intertwined with work on currentproblems in Terry’s life. Following one expo-sure session focused on the sexual abuse, Terryreverted to some of her previously problematicbehaviors, evidenced by withdrawal and si-lence in sessions, suicidal ideation, and medica-tion noncompliance. The appearance of suchbehavior marked the necessity of stoppingStage 2 discussions of previous sexual abuse toaddress Stage 1 targets recursively. Three ses-sions were devoted to a behavioral analysis ofTerry’s current suicidal, therapy-interfering,and quality-of-life–interfering behaviors; thesewere eventually linked both to fears about howthe therapist would view her childhood emo-tional responses to her father, and to holidayvisits with her father that precipitated conflictsover how Terry should be feeling about him inthe present. This two-steps-forward, one-step-back approach is common to therapy for cli-ents with BPD, and in particular may mark thetransition between Stage 1 and Stage 2.

As previously mentioned, Stage 3 targets theclient’s self-respect, regardless of the opinions ofothers. Betty, who was also in treatment with thesame therapist (M. M. L.), had successfully ne-gotiated Stages 1 and 2, and had become a highly

competent nurse with training and supervisoryresponsibilities. Therapy with Betty was then fo-cused on maintaining her self-esteem in the faceof very powerful significant others (e.g., her su-pervisor) who constantly invalidated her. Com-ponents of the treatment included the therapist’snoting and highlighting for Betty her tendency tomodify her self-opinion in accordance with thatof others, persistent attempts to extract fromBetty self-validation and self-soothing, and im-agery exercises wherein the client imagined andverbalized herself standing up to powerful oth-ers. Much of the therapy focus was on Betty’s in-terpersonal behavior within the therapy session,with attention to relating this behavior to her in-teractions with other important people. Thus,treatment at that point was very similar to thefunctional-analytic psychotherapy regimen de-veloped by Kohlenberg and Tsai (1991). Over-all, this third stage of therapy involved the move-ment to a more egalitarian relationship betweenthe client and the therapist, in which emphasiswas placed on the client’s standing up for herown opinions and defending her own actions.This approach required that the therapist bothreinforce the client’s assertions, and step backand refrain from validating and nurturing theclient in the manner characteristic of Stages 1and 2. In addition, therapy sessions were re-duced to every other week, and issues surround-ing eventual termination were periodically dis-cussed.

Stage 4 of DBT targets the sense of incom-pleteness that can preclude the experience ofjoy and freedom. Sally started Stage 1 treat-ment with the same therapist (M. M. L.) 15years ago. Stage 1 lasted 2 years; this was fol-lowed by a break of 1 year, after which treat-ment resumed for several years of bimonthlysessions leading to monthly sessions, and cur-rently consists of four or five sessions a year.Sally has been married for 30 years to an irreg-ularly employed husband who, though devotedand loyal, is quite invalidating of her. Althoughapparently brilliant, he is usually dismissedfrom jobs for his interpersonal insensitivity. Shehas been employed full-time at the same placefor years, working with children. The son shefelt closest to died in a plane accident 2 yearsago; her mother died last year, and her father isvery ill. Despite having a stable marriage,working in a stable and quite fulfilling job,having raised two well-adjusted sons, and stillbeing athletic, life feels meaningless to Sally. Inthe past she was very active in spiritual activi-


ties; following meditation retreats or extendedperiods of daily meditation, she would reportcontentment and some sense of joy. Since herson died, Sally has let go of most of her spiri-tual activities. Following 2 years of focusing ongrieving, she is now ready for Stage 4. Treat-ment planning focused on actively practicingand keeping track of progress in radical accep-tance (or “letting go of ego,” in Zen terminol-ogy), either alone or with group support.

TRANSCRIPTS

The following (composite) transcripts repre-sent actual examples of the process of therapyoccurring over several sessions with differentclients. These particular dialogues betweentherapist and client have been chosen to pro-vide the reader with comprehensive examplesof the application of a wide range of DBT treat-ment strategies. The session targets in the fol-lowing transcript were orienting and commit-ment. The strategies used were validation,problem solving (insight, orienting, and com-mitment), dialectical (devil’s advocate), andintegrated (relationship enhancement).

Obtaining the client’s commitment is a cru-cial first step in beginning therapy with clientswho have BPD. As illustrated in the followingtranscript, the dialectical technique of devil’sadvocate can be highly effective when used as acommitment strategy. In this first therapy ses-sion, the therapist’s ultimate goal was to obtainthe client’s commitment to therapy, as well as acommitment to eliminate suicidal behavior. Shebegan by orienting the client to the purpose ofthis initial session.

THERAPIST: So are you a little nervous aboutme?

CLIENT: Yeah, I guess I am.

THERAPIST: Well, that’s understandable. For thenext 50 minutes or so, we have this opportu-nity to get to know each other and see if wewant to work together. So what I’d like to dois talk a little bit about the program and howyou got here. So tell me, what do you wantout of therapy with me, and what are youdoing here?

CLIENT: I want to get better.

THERAPIST: Well, what’s wrong with you?

CLIENT: I’m a mess. (Laughs.)

THERAPIST: How so?

CLIENT: Umm, I don’t know. I just can’t evencope with everyday life right now. And Ican’t even . . . I’m just a mess. I don’t knowhow to deal with anything.

THERAPIST: So what does that mean exactly?

CLIENT: Umm, well, everything I try these daysjust seems overwhelming. I couldn’t keep upon my job, and now I’m on medical leave.Plus everyone’s sick of me being in the hospi-tal so much. And I think my psychiatristwants to send me away because of all myself-harming.

THERAPIST: How often do you self-harm?

CLIENT: Maybe once or twice a month. I usemy lighter or cigarettes, sometimes a razorblade.

THERAPIST: Do you have scars all over?

CLIENT: (Nods yes.)

THERAPIST: Your psychiatrist tells me you’vealso drunk Clorox. Why didn’t you mentionthat?

CLIENT: I guess it didn’t enter my mind.

THERAPIST: Do things just not enter your mindvery often?

CLIENT: I don’t really know. Maybe.

THERAPIST: So maybe with you I’m going tohave to be a very good guesser.

CLIENT: Hmm.

THERAPIST: Unfortunately, though, I’m not thegreatest guesser. So we’ll have to teach youhow to have things come to mind. So what isit exactly that you want out of therapy withme? To quit harming yourself, quit trying tokill yourself, or both?

CLIENT: Both. I’m sick of it.

THERAPIST: And is there anything else you wanthelp with?

CLIENT: Um, well, I don’t know how to handlemoney, and I don’t know how to handle rela-tionships. I don’t have friends; they don’tconnect with me very often. I’m a former al-coholic and a recovering anorexic/bulimic. Istill have a tendency toward that.

THERAPIST: Do you think maybe some of whatis going on with you is that you’ve replacedyour alcoholic and anorexic behaviors withself-harm behaviors?

CLIENT: I don’t know. I haven’t thought about


it that way. I just feel that I don’t know howto handle myself, and—you know, and Iguess work through stuff, and that is obvi-ously getting to me, because if it wasn’t, Iwouldn’t be trying to kill myself.

THERAPIST: So from your perspective, oneproblem is that you don’t know how to dothings. A lot of things.

CLIENT: Yeah, and a lot of it is, I do know how,but for some reason I don’t do it anyway.

THERAPIST: Um hmm.

CLIENT: You know, I mean I know I need tosave money, and I know that I need to bud-get myself, and I do every single month, butevery single month I get in debt. But, um,you know, it’s really hard for me. You know,it’s like sometimes I know it, or I know Ishouldn’t eat something and I do it anyway.

THERAPIST: So it sounds like part of the prob-lem is you actually know how to do things;you just don’t know how to get yourself todo the things you know how to do.

CLIENT: Exactly.

THERAPIST: Does it seem like maybe your emo-tions are in control—that you are a personwho does things when you’re in the mood?

CLIENT: Yes. Everything’s done by the mood.

THERAPIST: So you’re a moody person.

CLIENT: Yes. I won’t clean the house for 2months, and then I’ll get in the mood toclean. Then I’ll clean it immaculately andkeep it that way for 3 weeks—I mean, justimmaculate—and then when I’m in themood I go back to being a mess again.

THERAPIST: So one of the tasks for you and mewould be to figure out a way to get yourbehavior and what you do less hooked upwith how you feel?

CLIENT: Right.

The therapist used insight to highlight forthe client the observed interrelationship be-tween the client’s emotions and her behavior.She then began the process of shaping a com-mitment through the dialectical strategy ofdevil’s advocate.

THERAPIST: That, of course, is going to be hellto do, don’t you think? Why would youwant to do that? It sounds so painful.

CLIENT: Well, I want to do it, because it’s so in-

consistent. It’s worse, you know, becausewhen I’m . . . I know that, like with budget-ing money or whatever, I know I need to doit, and then when I don’t do it, it makes meeven more upset.

THERAPIST: Why would you ever want to dosomething you’re not in the mood for?

CLIENT: Because I’ve got to. Because I can’t sur-vive that way if I don’t.

THERAPIST: Sounds like a pretty easy life to me.

CLIENT: Yeah, but I can’t afford to live if I justspend my money on fun and stupid, frivo-lous things that I . . .

THERAPIST: Well, I guess maybe you shouldhave some limits and not be too off the wall,but in general, I mean, why clean the house ifyou’re not in the mood?

CLIENT: Because it pisses me off when it’s amess. And I can’t find things, like I’ve lostbills before and then I end up not payingthem. And now I’ve got collection agencieson my back. I can’t deal with all this, and Iend up self-harming and going into the hos-pital. And then I just want to end it all. But itstill doesn’t seem to matter, because if I’mnot in the mood to clean it, I won’t.

THERAPIST: So the fact that it makes horriblethings happen in your life so far hasn’t beenenough of a motivation to get you to dothings against your mood, right?

CLIENT: Well, obviously not (laughs), becauseit’s not happening.

THERAPIST: Doesn’t that tell you, though? Thisis going to be a big problem, don’t youthink? This isn’t going to be something sim-ple. It’s not like you’re going to walk in hereand I’m going to say, “OK, magic wand,”and then all of a sudden you’re going towant to do things that you’re not in themood for.

CLIENT: Yeah.

THERAPIST: Yeah, so it seems to me that ifyou’re not in the mood for things, if you’rekind of mood-dependent, that’s a very toughthing to crack. As a matter of fact, I think it’sone of the hardest problems there is to dealwith.

CLIENT: Yeah, great.

THERAPIST: I think we could deal with it, but Ithink it’s going to be hell. The real questionis whether you’re willing to go through hell


to get where you want to get or not. Now Ifigure that’s the question.

CLIENT: Well, if it’s going to make me happier,yeah.

THERAPIST: Are you sure?

CLIENT: Yeah, I’ve been going through thissince I was 11 years old. I’m sick of this shit.I mean, excuse my language, but I really am,and I’m backed up against the wall. Either Ineed to do this or I need to die. Those are mytwo choices.

THERAPIST: Well, why not die?

CLIENT: Well, if it comes down to it, I will.

THERAPIST: Um hmm, but why not now?

CLIENT: Because, this is my last hope. Becauseif I’ve got one last hope left, why not take it?

THERAPIST: So, in other words, all things beingequal, you’d rather live than die, if you canpull this off.

CLIENT: If I can pull it off, yeah.

THERAPIST: OK, that’s good; that’s going to beyour strength. We’re going to play to that.You’re going to have to remember that whenit gets tough. But now I want to tell youabout this program and how I feel about youharming yourself, and then we’ll see if youstill want to do this.

As illustrated by the foregoing segment, thetherapist’s relentless use of the devil’s advocatestrategy successfully “got a foot in the door”and achieved an initial client commitment. Thetherapist then “upped the ante” with a brief ex-planation of the program and its goals.

THERAPIST: Now the most important thing tounderstand is that we are not a suicide pre-vention program; that’s not our job. But weare a life enhancement program. The way welook at it, living a miserable life is noachievement. If we decide to work together,I’m going to help you try to improve yourlife, so that it’s so good that you don’t wantto die or hurt yourself. You should alsoknow that I look at suicidal behavior, includ-ing drinking Clorox, as problem-solvingbehavior. I think of alcoholism the same way.The only difference is that cutting, burning,unfortunately—it works. If it didn’t work,nobody would do it more than once. But itonly works in the short term, not the longterm. So to quit cutting, trying to hurt your-

self, is going to be exactly like quitting alco-hol. Do you think this is going to be hard?

CLIENT: Stopping drinking wasn’t all that hard.

THERAPIST: Well, in my experience, giving upself-harm behavior is usually very hard. Itwill require both of us working, but you willhave to work harder. And like I told youwhen we talked briefly, if you commit tothis, it’s for 1 year—individual therapy withme once a week, and group skills trainingonce a week. So the question is, are you will-ing to commit for 1 year?

CLIENT: I said I’m sick of this stuff. That’s whyI’m here.

THERAPIST: So you’ve agreed to not drop out oftherapy for a year, right?

CLIENT: Right.

THERAPIST: And do you realize that if you don’tdrop out for a year, that really does, if youthink about it, rule out suicide for a year?

CLIENT: Logically, yeah.

THERAPIST: So we need to be absolutely clearabout this, because this therapy won’t workif you knock yourself off. The most funda-mental mood-related goal we have to workon is that, no matter what your mood is, youwon’t kill yourself or try to.

CLIENT: All right.

THERAPIST: So that’s what I see as numberone priority—not our only one but numberone—that we will work on that. And gettingyou to agree—meaningfully, of course—andactually follow through on staying alive andnot harming yourself and not attempting sui-cide, no matter what your mood is. Now thequestion is whether you agree to that.

CLIENT: Yes, I agree to that.

The therapist, having successfully obtainedthe client’s commitment to work on suicidalbehavior again employed the strategy of devil’sadvocate to reinforce the strength of the com-mitment.

THERAPIST: Why would you agree to that?

CLIENT: I don’t know. (Laughs.)

THERAPIST: I mean, wouldn’t you rather be in atherapy where, if you wanted to kill yourself,you could?

CLIENT: I don’t know. I mean, I never reallythought about it that way.


THERAPIST: Hmm.

CLIENT: I don’t want to . . . I want to be able toget to the point where I could feel like I’mnot being forced into living.

THERAPIST: So are you agreeing with me be-cause you’re feeling forced into agreeing?

CLIENT: You keep asking me all these ques-tions.

THERAPIST: What do you think?

CLIENT: I don’t know what I think right now,honestly.

A necessary and important skill for the DBTtherapist is the ability to sense when a clienthas been pushed to his or her limits, as well asthe concomitant skill of being willing and ableto step back and at least temporarily refrainfrom further pressuring the client. In these in-stances, continued pressure from the therapistis likely to boomerang and have the oppositeeffect of what the therapist intends. Here thetherapist noticed the client’s confusion andsensed that further pushing was likely to resultin the client’s reducing the strength of her com-mitment. Consequently, the therapist steppedback and moved in with validation.

THERAPIST: So you’re feeling pushed up againstthe wall a little bit, by me?

CLIENT: No, not really. (Starts to cry.)

THERAPIST: What just happened just now?

CLIENT: (pause) I don’t know. I mean, I don’tthink I really want to kill myself. I think Ijust feel like I have to. I don’t think it’s reallyeven a mood thing. I just think it’s when Ifeel like there’s no other choice. I just say,“Well, you know there’s no other choice, sodo it.” You know. And so right now, I don’tsee any ray of hope. I’m going to therapy,which I guess is good. I mean, I know it’sgood, but I don’t see anything any betterthan it was the day I tried to kill myself.

THERAPIST: Well, that’s probably true. Maybe itisn’t any better. I mean, trying to kill yourselfdoesn’t usually solve problems. Although itactually did do one thing for you.

CLIENT: It got me in therapy.

THERAPIST: Yeah. So my asking you all thesequestions makes you start to cry. You looklike you must be feeling pretty bad.

CLIENT: Just overwhelmed, I guess the word is.

THERAPIST: That’s part of the reason we’re hav-ing this conversation, to try to structure ourrelationship so that it’s very clear for both ofus. And that way, at least, we’ll try to cutdown on how much you get overwhelmed bynot knowing what’s going on with me. OK?

CLIENT: Um hmm.

THERAPIST: And so I just want to be clear onwhat our number one goal is, and how hardthis is, because if you want to back out,now’s the time. Because I’m going to takeyou seriously if you say, “Yes, I want to doit.”

CLIENT: I don’t want to back out.

THERAPIST: OK. Good. Now I just want to saythat this seems like a good idea right now.You’re in kind of an energized mood today,getting started on a new program. But in 5hours, it might not seem like such a goodidea. It’s kind of like it’s easy to commit to adiet after a big meal, but it’s much harderwhen you’re hungry. But we’re going towork on how to make it keep sounding like agood idea. It’ll be hell, but I have confidence.I think we can be successful working to-gether.

Note how the therapist ended the session bypreparing the client for the difficulties she waslikely to experience in keeping her commitmentand working in therapy. Cheerleading and rela-tionship enhancement laid the foundation for astrong therapeutic alliance. The following ses-sion occurred approximately 4 months intotherapy. The session target was suicidal behav-ior. The therapist used validation, problemsolving (contingency clarification, didactic in-formation, behavioral analysis, and solutionanalysis), stylistic (irreverent communication),dialectical (metaphor, making lemonade out oflemons), and skills training (distress tolerance)strategies.

The therapist reviewed the client’s diary cardand noted a recent, intentional self-injury, inwhich the client opened up a previously self-inflicted wound following her physician’s re-fusal to provide pain medication. The therapistbegan by proceeding with a behavioral analy-sis.

THERAPIST: OK. Now you were in here lastweek telling me you were never going to hurtyourself again because this was so ridicu-


lous, you couldn’t stand it, you couldn’t hurtyourself any more. So let’s figure out howthat broke down on Sunday, so we can learnsomething from it. OK. So when did youstart having urges to hurt yourself?

CLIENT: My foot began to hurt on Wednesday. Istarted to have a lot of pain.

THERAPIST: It hadn’t hurt before that?

CLIENT: No.

THERAPIST: So the nerves were dead before thator something, huh? So you started having alot of pain. Now when did you start havingthe pain, and when did the urge to harmyourself come?

CLIENT: At the same time.

THERAPIST: They just come at the identical mo-ment?

CLIENT: Just about.

The specification of an initial prompting en-vironmental event is always the first step inconducting a behavioral chain analysis. Herethe therapist began by directly inquiring whenthe urges toward suicide and NSSI began. Notealso the therapist’s use of irreverent communi-cation early in the session.

THERAPIST: So how is it that feeling pain setsoff an urge to self-harm? Do you know howthat goes? How you get from one to theother?

CLIENT: I don’t know. Maybe it wasn’t tillThursday, but I asked my nurse. I go, “Look,I’m in a lot of pain, you know. I’m throwingup my food because the pain is so bad.” Andthe nurse tried. She called the doctor andtold him I was in a lot of pain, and asked ifhe’d give me some painkillers. But no! So Ikept asking, and the answer kept being no,and I got madder and madder and madder.So I felt like I had to show somebody that ithurt, because they didn’t believe me.

THERAPIST: So let’s figure this out. So is it thatyou’re assuming that if someone believed ithurt as bad as you said it does, they wouldactually give you the painkillers?

CLIENT: Yes.

THERAPIST: OK. That’s where the faulty think-ing is. That’s the problem. You see, it’sentirely possible that people know how badthe pain is but still aren’t giving you medica-tion.

CLIENT: I believe firmly, and I even wrote it inmy journal, that if I’d gotten pain medica-tion when I really needed it, I wouldn’t haveeven thought of self-harming.

The therapist proceeded by obtaining a de-scription of the events co-occurring with theonset of the problem. Here it became apparentthat maladaptive thinking was instrumental inthe client’s decision to self-harm. In the follow-ing segment, the therapist used the dialecticalstrategy of metaphor to highlight for the clienther cognitive error.

THERAPIST: Now let me ask you something—you’ve got to imagine this, OK? Let’s imag-ine that you and I are on a raft together outin the middle of the ocean. Our boat hassunk and we’re on the raft. And when theboat sank, your leg got cut really badly. Andtogether we’ve wrapped it up as well as wecan. But we don’t have any pain medicine.And we’re on this raft together and your legreally hurts, and you ask me for pain medi-cine, and I say no. Do you think you wouldthen have an urge to hurt yourself and makeit worse?

CLIENT: No, it would be a different situation.

THERAPIST: OK, but if I did have the pain medi-cation and I said no because we had to saveit, what do you think?

CLIENT: If that were logical to me, I’d go alongwith it and wouldn’t want to hurt myself.

THERAPIST: What if I said no because I didn’twant you to be a drug addict?

CLIENT: I’d want to hurt myself.

THERAPIST: OK. So we’ve got this clear. Thepain is not what’s setting off the desire toself-harm. It’s someone not giving you some-thing to help, when you feel they could ifthey wanted to.

CLIENT: Yes.

The therapist used contingency clarificationto point out the effects of others’ responses onthe client’s own behavior. In the following seg-ment, the therapist again employed contin-gency clarification in a continued effort tohighlight for the client the communicationfunction of NSSI.

THERAPIST: So, in other words, hurting yourselfis communication behavior, OK? So what we


have to do is figure out a way for the com-munication behavior to quit working.

CLIENT: Why?

THERAPIST: Because you’re not going to stopdoing it until it quits working. It’s like tryingto talk to someone; if there’s no one in theroom, you eventually quit trying to talk tothem. It’s like when a phone goes dead, youquit talking.

CLIENT: I tried three nights in a row in a per-fectly assertive way and just clearly stated Iwas in a lot of pain.

THERAPIST: You know, I think I’ll switch chairswith you. You’re not hearing what I’m say-ing.

CLIENT: And they kept saying, “No,” and thensome little light came on in my head.

THERAPIST: I’m considering switching chairswith you.

CLIENT: And it was like, “Here, now can youtell that it hurts a lot?”

THERAPIST: I’m thinking of switching chairswith you.

CLIENT: Why?

THERAPIST: Because if you were sitting overhere, I think you would see that no matterhow bad the pain is, hurting yourself to getpain medication is not a reasonable re-sponse. The hospital staff may not have beenreasonable either. It may be that they shouldhave given you pain medicine. But we don’thave to say they were wrong in order to saythat hurting yourself was not the appropri-ate response.

CLIENT: No, I don’t think it was the appropri-ate response.

THERAPIST: Good. So what we’ve got to do isfigure out a way to get it so that the responsedoesn’t come in, even if you don’t get painmedicine. So far, it has worked very effec-tively as communication. And the only wayto stop it is to get it to not work any more.And of course, it would be good to get otherthings to work. What you’re arguing is“Well, OK, if I’m not going to get it this way,then I should be able to get it another way.”

CLIENT: I tried this time!

THERAPIST: Yes, I know you did, I know youdid.

CLIENT: A lady down the hallway from me was

getting treatment for her diabetes, and it gotreal bad, and they gave her pain medication.

THERAPIST: Now we’re not on the same wave-length in this conversation.

CLIENT: Yes, we are. What wavelength are youon?

THERAPIST: I’m on the wavelength that it mayhave been reasonable for you to get painmedicine, and I certainly understand yourwanting it. But I’m also saying that no mat-ter what’s going on, hurting yourself is some-thing we don’t want to happen. You’re func-tioning like if I agreed with you that youshould get pain medication, I would thinkthis was OK.

CLIENT: Hmm?

THERAPIST: You’re talking about whether theyshould have given you pain medication ornot. I’m not talking about that. Even if theyshould have, we’ve got to figure out how youcould have gotten through without hurtingyourself.

As illustrated by the foregoing exchange, aclient with BPD often wants to remain focusedon the crisis at hand. This poses a formidablechallenge for the therapist, who must necessar-ily engage in a back-and-forth dance betweenvalidating the client’s pain and pushing forbehavioral change. This segment also illus-trates how validation does not necessarily im-ply agreement. Although the therapist vali-dated the client’s perception that the nurse’srefusal to provide pain medication may havebeen unreasonable, she remained steadfast inmaintaining the inappropriateness of the cli-ent’s response.

CLIENT: I tried some of those distress tolerancethings and they didn’t work.

THERAPIST: OK. Don’t worry, we’ll figure out away. I want to know everything you tried.But first I want to be sure I have the pictureclear. Did the urges start building afterWednesday and get worse over time?

CLIENT: Yeah. They started growing with thepain.

THERAPIST: With the pain. OK. But also theystarted growing with their continued refusalto give you pain medicine. So you werethinking that if you hurt yourself, theywould somehow give you pain medicine?


CLIENT: Yeah. ‘Cause if they wouldn’t listen tome, then I could show them.

THERAPIST: OK, so you were thinking, “If theywon’t listen to me, I’ll show them.” Andwhen did that idea first hit? Was that onWednesday?

CLIENT: Yeah.

THERAPIST: OK. Well, we’ve got to figure out away for you to tolerate bad things withoutharming yourself. So let’s figure out all thethings you tried, and then we have to figureout some other things, because those didn’twork. So what was the first thing you tried?

At this juncture the behavioral analysis re-mained incomplete, and it would normallyhave been premature to move to the stage ofsolution analysis. However, in the therapist’sjudgment, it was more critical at this point toreinforce the client’s attempts at distress toler-ance by responding to the client’s communica-tion that she had attempted behavioral skills.

CLIENT: I thought that if I just continued to beassertive about it that the appropriate mea-sures would be taken.

THERAPIST: OK, but that didn’t work. So whydidn’t you harm yourself right then?

CLIENT: I didn’t want to.

THERAPIST: Why didn’t you want to?

CLIENT: I didn’t want to make it worse.

THERAPIST: So you were thinking about prosand cons —that if I make it worse, I’ll feelworse?

CLIENT: Yeah.

One aspect of DBT skills training stresses theusefulness of evaluating the pros and cons oftolerating distress as a crisis survival strategy.Here the therapist employed the dialecticalstrategy of turning lemons into lemonade byhighlighting for the client how she did, in fact,use behavioral skills. Note in the following re-sponse how the therapist immediately rein-forced the client’s efforts with praise.

THERAPIST: That’s good thinking. That’s whenyou’re thinking about the advantages anddisadvantages of doing it. OK, so at thatpoint the advantages of making it worsewere outweighed by the disadvantages. OK.

So you keep up the good fight here. Nowwhat else did you try?

CLIENT: I tried talking about it with other cli-ents.

THERAPIST: And what did they have to say?

CLIENT: They said I should get pain medica-tion.

THERAPIST: Right. But did they say you shouldcut yourself or hurt yourself if you didn’t getit?

CLIENT: No. And I tried to get my mind off mypain by playing music and using mindful-ness. I tried to read and do crossword puz-zles.

THERAPIST: Um hmm. Did you ever try radicalacceptance?

CLIENT: What’s that?

THERAPIST: It’s where you sort of let go and ac-cept the fact that you’re not going to get thepain medication. And you just give yourselfup to that situation. You just accept that itain’t going to happen, that you’re going tohave to cope in some other way.

CLIENT: Which I did yesterday. I needed a littleAtivan to get me there, but I got there.

THERAPIST: Yesterday?

CLIENT: Yeah. I took a nap. When I woke up Ibasically said, “Hey, they’re not going tochange, so you’ve just got to deal with thisthe best that you can.”

THERAPIST: And did that acceptance helpsome?

CLIENT: I’m still quite angry about what Ibelieve is discrimination against borderlinepersonalities. I’m still very angry about that.

THERAPIST: OK. That’s fine. Did it help,though, to accept?

CLIENT: Um hmm.

THERAPIST: That’s good. That’s great. That’s agreat skill, a great thing to practice. Whenpush comes to shove, when you’re really atthe limit, when it’s the worst it can be, radi-cal acceptance is the skill to practice.

CLIENT: That’s AA.

During a solution analysis, it is often neces-sary that the therapist facilitate the process byhelping the client “brainstorm,” or by makingdirect suggestions for handling future crises.


Here the therapist suggested a solution that isalso taught in the DBT skills training moduleon distress tolerance. The notion of radical ac-ceptance stresses the idea that acceptance ofone’s pain is a necessary prerequisite for endingemotional suffering.

THERAPIST: OK. Now let’s go back to how yougave in to the urge. Because you really man-aged to battle all the way till then, right?OK. Usually, with you, we can assume thatsomething else happened. So let’s figure outSunday and see if there wasn’t an interper-sonal situation that day that made you feelcriticized, unloved, or unacceptable.

CLIENT: Well, on Saturday I was so pissed offand I went to an AA meeting. And it got onmy brain how alcohol would steal away mypain. I went looking all around the neighbor-hood for an open store. I was going to go getdrunk. That’s how much my pain was influ-encing me. But I couldn’t find a store thatwas open, so I went back to the hospital.

THERAPIST: So you got the idea of getting alco-hol to cure it, and you couldn’t find any, soyou went back to the hospital. You were in alot of pain, and then what happened?

CLIENT: I told the nurse, “I’ve been sober al-most 10 years and this is the first urge I’vehad to drink; that’s how bad my pain is.”And that wasn’t listened to.

THERAPIST: So you figured that should havedone it?

CLIENT: Yeah.

THERAPIST: Yeah. ‘Cause that’s a high-levelcommunication, that’s like a suicide threat.Very good, though. I want you to know,that’s better than a suicide threat, becausethat means you had reduced the severity ofyour threats.

The response above was very irreverent, inthat most clients would not expect their thera-pists to view making a threat as a sign of thera-peutic progress. The therapeutic utility of irrev-erence often lies in its “shock” value, whichmay temporarily loosen a client’s maladaptivebeliefs and assumptions, and open the client upto the possibility of other response solutions.

CLIENT: And I just told her how I was feelingabout it, and I thought that would do it. Andthe doctor still wouldn’t budge.

THERAPIST: So what did she do? Did she say shewould call?

CLIENT: She called.

THERAPIST: OK. And then what happened?

CLIENT: She came back. She was really sweet,and she just said, “I’m really sorry, but thedoctor said no.”

THERAPIST: Then did you feel anger?

CLIENT: I don’t know if I was really angry, but Iwas hurt.

THERAPIST: Oh, really? Oh, that’s pretty inter-esting. OK. So you were hurt . . .

CLIENT: Because I ended up hugging my teddybear and just crying for a while.

THERAPIST: Before or after you decided to hurtyourself?

CLIENT: Before.

THERAPIST: OK. So you didn’t decide rightaway to hurt yourself. You were thinkingabout it. But when did you decide to do it?

CLIENT: Later on Saturday.

THERAPIST: When?

CLIENT: After I got sick of crying.

THERAPIST: So you laid in bed and cried, feelinguncared about and hurt, abandoned proba-bly, and unlovable, like you weren’t worthhelping?

CLIENT: Yes.

THERAPIST: That’s a really adaptive response.That’s what I’m going to try to teach you.Except that you’ve already done it withoutmy teaching it to you. So how did you getfrom crying, feeling unloved and not caredabout, and you cry and sob—how did youget from there to deciding to hurt yourself,instead of like going to sleep?

CLIENT: Because then I got angry. And I said,“Fuck this shit, I’ll show him.”

THERAPIST: Now did you quit crying beforeyou got angry, or did getting angry make youstop crying?

CLIENT: I think getting angry made me stopcrying.

THERAPIST: So you kind of got more energized.So you must have been ruminating while youwere lying there, thinking. What were youthinking about?

CLIENT: For a long time I was just wantingsomebody to come care about me.


THERAPIST: Um hmm. Perfectly reasonable feel-ings. Makes complete sense. Now maybethere you could have done something differ-ent. What would have happened if you hadasked the nurse to come in and talk to you,hold your hand?

An overall goal of behavioral analysis is theconstruction of a general road map of how theclient arrives at dysfunctional responses, withnotation of possible alternative pathways. Herethe therapist was searching for junctures in themap where possible alternative responses wereavailable to the client.

CLIENT: They don’t have time to do that.

THERAPIST: They don’t? Do you think thatwould have helped?

CLIENT: I don’t know. She couldn’t help me.

THERAPIST: She could have made you feel caredabout. That would have been a caring thingto do.

CLIENT: Yeah, but I don’t think it would havehelped.

THERAPIST: What would have helped?

CLIENT: Getting pain medication.

THERAPIST: I thought you’d say that. You havea one-track mind. Now listen, we’ve got tofigure out something else to help you, be-cause it can’t be that nothing else can help.That can’t be the way the world works foryou. There’s got to be more than one way toget everywhere, because we all run into boul-ders on the path. Life is like walking on apath, you know, and we all run into boul-ders. It’s got to be that there are other pathsto places. And for you, it really isn’t the painin your ankle that’s the problem; it’s the feel-ing of not being cared about. And probably afeeling that has something to do with anger,or a feeling that other people don’t respectyou—a feeling of being invalidated.

CLIENT: Yes.

THERAPIST: So I think it’s not actually the painin your ankle that’s the problem. Because ifyou were out on that raft with me, youwould have been able to handle the pain if Ihadn’t had any medicine, right? So it’s reallynot the pain; it’s the sense of being invali-dated and the sense of not being cared about.That’s my guess. Do you think that’s correct?

CLIENT: Yes.

THERAPIST: See, the question is, is there anyother way for you to feel validated and caredabout, other than them giving it to you?

CLIENT: No.

THERAPIST: Now is this a definite, like “I’m notgoing to let there be any other way,” or is itmore open, like “I can’t think of anotherway, but I’m open to the possibility?”

CLIENT: I don’t think there’s another way.

THERAPIST: Does that mean you’re not evenopen to learning another way?

CLIENT: Like what?

THERAPIST: I don’t know. We have to figure itout. See, what I think is happening is thatwhen you’re in a lot of pain and you feel ei-ther not cared about or not taken seriously,invalidated, that’s what sets you up to hurtyourself, and also to want to die. The prob-lem that we have to solve is how to be in asituation that you feel is unjust without hav-ing to harm yourself to solve it. Are youopen to that?

CLIENT: Yeah.

As illustrated here, behavioral analysis is of-ten an excruciating and laborious process forclient and therapist alike. The therapist oftenfeels demoralized and is tempted to abandonthe effort, which may be likened to trying tofind a pair of footprints hidden beneath layersof fallen leaves; the footprints are there, but itmay take much raking and gathering of leavesbefore they are uncovered. With repeated anal-yses, however, the client learns that the thera-pist will not “back down.” Such persistence onthe part of the therapist eventually extinguishesa client’s refusal to attempt new and adaptiveproblem-solving behaviors. As clients increas-ingly acquire new behavioral skills, more adap-tive attempts at problem resolution eventuallybecome discernible.

In the following session (approximately 10months into therapy), the client arrived wear-ing mirrored sunglasses (again) and was angrybecause collection agencies were persistent inpressuring her for payment on delinquent ac-counts. In addition, her therapist had been outof town for a week. The session targets wereemotion regulation and interpersonal effective-ness. Dialectical (metaphor), validation (cheer-leading), problem solving (contingency clar-ification, contingency management), stylistic(reciprocal communication, irreverent commu-


nication), and integrated (relationship en-hancement) strategies were used. In this firstsegment, the therapist used cheerleading, con-tingency clarification, and the contingencymanagement strategy of shaping to get the cli-ent to remove her sunglasses and work on ex-pressing her anger.

THERAPIST: It’s not a catastrophe that the col-lector did this to you, and it’s not a catastro-phe to be mad at the collector. It’s made yourlife a lot harder, but you can handle this. Youcan cope with this. This is not more than youcan cope with. You’re a really strongwoman; you’ve got it inside you. But you’vegot to do it. You’ve got to use it. I’m willingto help you, but I can’t do it alone. You haveto work with me.

CLIENT: How?

THERAPIST: Well, by taking off your sunglasses,for starters.

The therapist began the exchange by at-tempting to normalize the issue (“It’s not a ca-tastrophe”), validating the client (“It’s madeyour life a lot harder”), and cheerleading (“Youcan handle this. You can cope. . . . You’re areally strong woman”). The therapist thenmoved to contingency clarification by pointingout that provision of the therapist’s assistancewas contingent on the client’s willingness towork. She immediately followed this by re-questing a response well within the client’sbehavioral repertoire.

CLIENT: I knew you’d say that.

THERAPIST: And I knew you knew I’d say that.

CLIENT: Sunglasses are your biggest bitch, Ithink.

THERAPIST: Well, how would you like to look atyourself talking to someone else? (longpause) They make it difficult for me. And Ifigure they make it harder for you. I thinkyou do better when you’re not wearing thosesunglasses. It’s like a step; you always dobetter when you go forward. And when youdo, you feel better. I’ve noticed that. (longpause) So that’s what you should do; youshould take off your sunglasses, and then weshould problem-solve on how to cope whenyou can’t get angry. There’s nothing freakishabout that. Something has happened in yourlife that has made it so that you’re afraid to

be angry, and we just have to deal with that,you and me. It’s just a problem to be solved.It’s not a catastrophe; it’s not the worst thinganyone ever did. It’s just a problem that youhave, and that’s what you and I do. We solveproblems; we’re a problem-solving team.(pause)

CLIENT: (Removes sunglasses.) All right.

THERAPIST: Thank you. That’s a big step, Iknow, for you.

The therapist’s use of reciprocal communica-tion informed the client of her feelings regard-ing the sunglasses. Note the matter-of-fact atti-tude taken by the therapist and her continuedattempt to normalize the issue (i.e., “There’snothing freakish about that . . . it’s not theworst thing anyone ever did”). Also note theframing of the issue as a problem to be solved,as well as the therapist’s use of the relationshipstrategy to enhance the therapeutic alliance.The therapist also made a point of validatingthe client by letting her know that she realizedthis was difficult.

THERAPIST: Now, c’mon, I want you to find itinside yourself. I know you’ve got it; I knowyou can do it. You can’t give up. You can’t letyour feet slip. Keep going. Just express di-rectly to me how you feel. That you’re angryat yourself, that you’re angry at the collec-tion agency, and that you’re damn angrywith me. (long pause)

CLIENT: (barely audible) I’m angry at you, atmyself, and the collection agency.

The therapist continued to rely on cheerlead-ing and praise as she continued the shapingprocess in an attempt to get the client to ex-press her anger directly.

THERAPIST: Good, did that kill you? (longpause) That’s great. Is that hard? (longpause) It was, wasn’t it? Now say it with alittle vigor. Can’t you say it with a little en-ergy?

CLIENT: (Shakes her head no.)

THERAPIST: Yes, you can. I know you’ve got itin you. I have a good feel for what yourstrengths are. I don’t know how I’ve got thisgood feel, but I do. And I know you can do itand you need to do it, and you need to say itwith some energy. Express how angry you


are. You don’t have to yell and scream orthrow things. Just say it aloud—“I’m an-gry!” (long pause) You can scream, ofcourse, if you want; you can say, “I’m an-gry!”

CLIENT: That’s it. That’s all I can do.

THERAPIST: Listen, you have to take the risk.You’re not going to get past this or throughthis. You have to take the risk. You are like aperson mountain climbing and we’ve cometo this crevasse and it’s very deep, but wecan’t go back because there’s an avalanche,and the only way to go forward is for you tojump over this crevasse. You’ve got to do it.Tell me how mad you are, in a way that I canunderstand how you really feel.

CLIENT: (long pause) I can’t do any of it.

THERAPIST: That is bullshit.

CLIENT: You want me to get angry at you, don’tyou?

THERAPIST: I don’t care who you get angry at. Ithink you already are angry. I just want youto express it. I’m not going to ask you to doanything more today, by the way. I figure theonly thing today you have to do is say “I’mangry,” in a voice that sounds angry, and Ifigure you’re capable of that. And I might beangry if you don’t do it. I don’t think I willbe, but I might. That’s OK. I can be angry,you can be angry, we can be angry some-times, and it isn’t going to kill either one ofus.

Cheerleading and metaphor were unsuccess-ful in moving the client to express her angermore forcefully. Consequently, the therapistswitched to irreverent communication in an at-tempt to get the client to “jump track.” Alsonote how the therapist communicated to theclient the potential negative consequences ofher continued refusal to express her anger (i.e.,“. . . I might be angry . . .”). In this manner, thetherapist used the relationship as a contingencyin order to promote change in the client.

THERAPIST: OK, so how angry are you? On ascale of 1 to 100, how angry would you sayyou are? At 100, you’re ready to kill. You’reso enraged, you’d go to war if you could.

CLIENT: (barely audible) Maybe 100.

THERAPIST: Really?

CLIENT: They know my situation.

THERAPIST: Um hmm.

CLIENT: They’re persistent.

THERAPIST: Um hmm. (pause) Who’s the safestto be angry at? Yourself, me, or the collec-tion agency?

CLIENT: Collection agency.

THERAPIST: OK, then, tell me how angry youare. You don’t have to make it sound like100. Try to make it sound like 50.

CLIENT: They really pissed me off! (said in aloud, angry voice)

THERAPIST: Well, damn right. They piss me off,too.

As illustrated by the foregoing exchange, aprimary difficulty in working with clients whohave BPD is their not uncommon tendency torefuse to engage in behavioral work. Thus, it isabsolutely necessary that the therapist main-tain persistence and not give up in the face of aclient’s “I can’t” statements. In situations likethese, the use of irreverent communication of-ten succeeds in producing a breakthrough andgaining client compliance.

POSTVENTION

After completing the writing of Cindy’s casehistory for publication in this Handbook, 14months into therapy, Cindy died of a prescrip-tion drug overdose plus alcohol. We considereddropping the case history and replacing it witha more successful case. However, in Cindy’shonor, and because we think much can belearned from both failed and successful ther-apy, we decided to leave the case in. The imme-diate precipitant for Cindy’s overdose was acall to her estranged husband, during whichshe discovered that another woman was livingwith him. As Cindy told her therapist during aphone call the next morning, her unverbalizedhope that they might someday get back to-gether, or at least be close friends, had beenshattered. She phoned again that evening intears, stating that she had just drunk half a fifthof liquor. Such drinking incidents had occurredseveral times before, and the phone call wasspent “remoralizing” Cindy, offering hope,problem-solving how she could indeed livewithout her husband, and using crisis interven-tion techniques to get her through the evening,until her appointment the following day.


Cindy’s roommate was home and agreed totalk with her, watch a TV movie together, andgo to bed (plans on which the roommate didfollow through). Cindy stated that althoughshe felt suicidal, she would stop drinking andwould not do anything self-destructive beforeher appointment. She was instructed to call thetherapist back later that evening if she wantedto talk again. The next day, when Cindy didnot arrive for her appointment, the therapistcalled her home, just as her roommate discov-ered Cindy dead, still in bed from the night be-fore. At this point, the therapist was faced witha number of tasks. The therapist called to in-form other therapists who had been treatingthe client, and she spoke with a legal consultantto review the limits of confidentiality when aclient has died. Once the family (Cindy’s par-ents and estranged husband) were alerted, thetherapist called each to offer her condolences.The next day, the therapist (who was the seniortherapist and supervisor on the treatmentteam) called a meeting of the treatment team todiscuss and process the suicide. It was espe-cially important to notify the individual thera-pists of the remaining three members of Cindy’sskills training group. Group members were no-tified of the suicide by their individual psycho-therapists. Within minutes of the beginning ofthe next group session, however, two membersbecame seriously suicidal, and one of them hadto be briefly hospitalized. (By the third weekfollowing the suicide, however, both had re-gained their forward momentum.) A thirdgroup member took this occasion to quit DBTand switch to another therapy, saying that thisproved the treatment did not work. In the daysand weeks following the suicide, the therapistattended the funeral and met with Cindy’sroommate and with her parents.

What can we learn from this suicide? First, itis important to note that even when a treat-ment protocol is followed almost to the letter, itmay not save a client. Even an effective treat-ment can fail in the end. In this case, DBTfailed. This does not mean that the progressmade was unimportant or not real. Had this“slippery spot over the abyss” been negotiatedsafely, perhaps the client would have been ableto develop, finally, a life of quality. Risk is noteliminated, however, just because an individualmakes substantial progress. In this case, thetherapist did not believe during the last phonecall that the client was at higher than ordinaryrisk for imminent suicide. In contrast to manyprevious phone calls and therapy sessions in

which the client had cried that she might not beable to hold on, during the last call the clientmade plans for the evening, agreed to stopdrinking and not to do anything suicidal orself-destructive, and seemed to the therapist(and the roommate) to be in better spirits fol-lowing the phone call. Her roommate washome and available. Thus, the therapist did nottake extraordinary measures that evening toprevent suicide. Indeed, the problem behaviorfocused on during the call was the drinking.The topic of suicide was brought up by thetherapist, in the course of conducting a risk as-sessment.

Could the therapist have known? Only (per-haps) if she had paid more attention to the pre-cipitant and less to the affect expressed at theend of the phone call. In reviewing notes aboutthe client, the therapist saw that each previousnear-lethal attempt was a result of the client’sbelieving that the relationship with her hus-band had irrevocably ended. Although the cli-ent could tolerate losing her husband, shecould not tolerate losing all hope for a reconcil-iation at some point, even many years hence.Had the therapist linked these two ideas (com-plete loss of hope and suicide attempt), shemight have been able to work out a better planwith the client for a reemergence of the crisislater in the evening. The value of both conduct-ing thorough behavioral assessments and orga-nizing them into a coherent pattern is high-lighted in this case. Second, when all is said anddone, an individual with BPD must ultimatelybe able and willing to tolerate the almost un-imaginable pain of his or her life until the ther-apy has a chance to make a permanent differ-ence. Ultimately, the therapist cannot save theclient; only the client can do that. Even if mis-takes are made, the client must nonetheless per-severe. In this case, the DBT protocol of “no le-thal drugs for lethal people” was violated, eventhough the client had a past history of near-lethal overdoses. Why was the protocol not en-forced? There were two primary reasons. First,the client came into therapy with a strong beliefthat the host of medications she was on wereessential to her survival. Any attempt on thetherapist’s part to manage her medicationswould have been met by very strong resistance.Although the drugs were dispensed in smalldoses, the only safe alternative would havebeen to have the person living with her (herhusband at first, then her roommate) manageher medications, which the client also resisted.In addition, the “no lethal drugs” protocol of


DBT is regularly criticized by some mentalhealth professionals, who believe that psycho-active medications are a treatment of choice forsuicidal individuals. In the face of professionaland client resistance to the policy in this casethe therapist relented. The second reason wasthat the lethal behavior of the client duringtherapy consisted of cutting and slashing; thus,her using drugs to commit suicide did not seemlikely, and the therapist allowed herself a falsesense of safety with respect to them. Third, agroup member’s suicide is extraordinarilystressful for clients with BPD who are in grouptherapy. Although it is easy to believe that alli-ances are not strong in a psychoeducationalbehavioral skills group, this has universally notbeen our experience. The suicide of one mem-ber is a catastrophic event and can lead to con-tagious suicide and NSSI behavior, and therapydropouts. Thus, extreme care is needed in theconduct of group meetings for some time fol-lowing a suicide. Similar care is needed withthe treatment team, where the thread of hopethat maintains therapists in the face of a daunt-ing task is also strained. It is important that thepersonal reactions of therapists, as well as a pe-riod of mourning and grieving, be shared andaccepted. Fears of legal responsibility, never farfrom the surface, must be confronted directly;legal counsel must be sought as necessary; and,in time, a careful review of the case and thetherapy must be conducted, if only to improvetreatment in the future.

ACKNOWLEDGMENTS

The writing of this chapter was supported by NationalInstitute of Mental Health Grant No. MH34486 toMarsha M. Linehan. Parts of this chapter are drawnfrom Linehan (1993b), Linehan and Koerner (1992),Koerner and Linehan (1992), and Linehan (1997). Thequotations from Linehan (1997) in the section on vali-dation are reprinted with the permission of the Ameri-can Psychological Association. Finally, this chapter is arevision of the same chapter in the previous edition ofthis book. Many contributions to this chapter weremade by previous authors Bryan M. Cochran and Con-stance A. Kehrer.

REFERENCES

Adler, G. (1981). The borderline–narcissistic personalitydisorder continuum. American Journal of Psychiatry,138, 46–50.

Adler, G. (1985). Borderline psychopathology and itstreatment. New York: Aronson.

Adler, G. (1993). The psychotherapy of core borderlinepsychopathology. American Journal of Psychothera-py, 47, 194–206.

Adler, G., & Buie, D. H. (1979). Aloneness and border-line psychopathology: The possible relevance of childdevelopment issues. International Journal of Psycho-analytic Psychotherapy, 60, 83–96.

American Psychiatric Association. (1987). Diagnosticand statistical manual of mental disorders. (3rd ed.,rev.). Washington, DC: Author.


Aviram, R. B., Brodsky, B. S., & Stanley, B. (2006). Bor-derline personality disorder, stigma, and treatmentimplications. Harvard Review of Psychiatry, 14,249–256.

Bateman, A., & Fonagy, P. (1999). Effectiveness of par-tial hospitalization in the treatment of borderline per-sonality disorder: A randomized controlled trial.American Journal of Psychiatry, 156, 1563–1569.

Bateman, A., & Fonagy, P. (2001). Treatment of border-line personality disorder with psychoanalytically ori-ented partial hospitalization: An 18-month follow-up. American Journal of Psychiatry, 158, 36–42.

Bateman, A. W., & Fonagy, P. (2004). Mentalization-based treatment of BPD. Journal of Personality Dis-orders, 18, 36–51.

Beck, A. T., Brown, G., Berchick, R. J., Stewart, B. L., &Steer, R. A. (1990). Relationship between hopeless-ness and ultimate suicide: A replication with psychi-atric outpatients. American Journal of Psychiatry,147, 190–195.

Beck, A. T., & Freeman, A. (1990). Cognitive therapy ofpersonality disorders. New York: Guilford Press.


Benjamin, L. S. (1996). Interpersonal diagnosis andtreatment of personality disorders (2nd ed.). NewYork: Guilford Press.

Blum, N., Pfohl, B., St. John, D., Monahan, P., & Black,D. W. (2002). STEPPS: A cognitive-behavioralsystems-based group treatment for outpatients withborderline personality disorder—a preliminary re-port. Comprehensive Psychiatry, 43, 301–310.

Bohus, M., Limberger, M. F., Frank, U., Sender, I.,Gratwohl, T., & Stieglitz, R. D. (2001). [Develop-ment of the borderline symptom list]. Psychothera-pies, Psychosomatik, Medizinische Psychologie, 51,201–211.

Bradley, R., Zittel, C. C., & Westen, D. (2005). The bor-derline personality diagnosis in adolescents: Genderdifferences and subtypes. Journal of Child Psychol-ogy and Psychiatry, 46, 1006–1019.

Brent, D. A., Johnson, B. A., Perper, J., Connolly, J.,Bridge, J., Bartle, S., et al. (1994). Personality disor-der, personality traits, impulsive violence, and com-pleted suicide in adolescents. Journal of the AmericanAcademy of Child and Adolescent Psychiatry, 33,1080–1086.


Brown, G. K., Newman, C. F., Charlesworth, S. E.,Crits-Christoph, P., & Beck, A. T. (2004). An openclinical trial of cognitive therapy for borderline per-sonality disorder. Journal of Personality Disorders,18, 257–271.

Bryer, J. B., Nelson, B. A., Miller, J. B., & Krol, P. A.(1987). Childhood sexual and physical abuse as fac-tors in adult psychiatric illness. American Journal ofPsychiatry, 144, 1426–1430.

Buie, D. H., & Adler, G. (1982). Definitive treatment ofthe borderline personality. International Journal ofPsychoanalytic Psychotherapy, 9, 51–87.

Clarkin, J. F., Foelsch, P. A., Levy, K. N., Hull, J. W.,Delaney, J. C., & Kernberg, O. F. (2001). The devel-opment of a psychodynamic treatment for patientswith borderline personality disorder: A preliminarystudy of behavioral change. Journal of PersonalityDisorders, 15, 487–495.

Davison, G. C., & Neale, J. M. (1994). Abnormal psy-chology (6th ed.). New York: Wiley.

Dimeff, L. A., McDavid, J., & Linehan, M. M. (1999).Pharmacotherapy for borderline personality disor-der: A review of the literature and recommendationsfor treatment. Journal of Clinical Psychology in Med-ical Settings, 6, 113–138.

Ellis, A. (1962). Reason and emotion in psychotherapy.New York: Lyle Stuart.

Ellis, A. (1973). Humanistic psychotherapy: Therational–emotive approach. New York: Julian Press.

Foa, E. B., & Kozak, M. J. (1986). Emotional process-ing of fear: Exposure to corrective information. Psy-chological Bulletin, 99, 20–35.

Foa, E. B., Steketee, G., & Grayson, J. B. (1985).Imaginal and in vivo exposure: A comparison withobsessive–compulsive checkers. Behavior Therapy,16, 292–302.

Fossati, A., Madeddu, F., & Maffei, C. (1999). Border-line personality disorder and childhood sexual abuse:A meta-analytic study. Journal of Personality Disor-ders, 13, 268–280.

Giesen-Bloo, J., Van Dyck, R., Spinhoven, P., vanTilburg, W., Dirksen, C., van Asselt, T., et al. (2006).Outpatient psychotherapy for borderline personalitydisorder: Randomized trial of schema-focused ther-apy vs transference-focused psychotherapy. Archivesof General Psychiatry, 63, 649–658.

Goldberg, C. (1980). The utilization and limitations ofparadoxical intervention in group psychotherapy. In-ternational Journal of Group Psychotherapy, 30,287–297.

Gottman, J. M., & Katz, L. F. (1990). Effects of mari-tal discord on young children’s peer interaction andhealth. Developmental Psychology, 25, 373–381.

Gunderson, J. G. (1984). Borderline personality disor-der. Washington, DC: American Psychiatric Press.

Herman, J. L. (1986). Histories of violence in an outpa-tient population: An exploratory study. AmericanJournal of Orthopsychiatry, 56, 137–141.

Herman, J. L., Perry, J. C., & van der Kolk, B. A. (1989).

Childhood trauma in borderline personality disorder.American Journal of Psychiatry, 146, 490–495.

Hollandsworth, J. G. (1990). The physiology of psycho-logical disorders. New York: Plenum Press.

Isometsa, E., Heikkinen, M., Henriksson, M.,Marttunen, M., Aro, H., & Lonnqvist, J. (1997). Dif-ferences between urban and rural suicides. ActaPsychiatrica Scandinavica, 95, 297–305.

Isometsa, E. T., Henriksson, M. M., Aro, H. M.,Heikkinen, M. E., Kuoppasalmi, K. I., & Lonnqvist,J. K. (1994). Suicide in major depression. AmericanJournal of Psychiatry, 151, 530–536.

Kegan, R. (1982). The evolving self: Problem and pro-cess in human development. Cambridge, MA: Har-vard University Press.

Kellogg, S. H., & Young, J. E. (2006). Schema therapyfor borderline personality disorder. Journal of Clini-cal Psychology, 62, 445–458.

Kernberg, O. F. (1975). Borderline conditions andpathological narcissism. New York: Aronson.

Kernberg, O. F. (1984). Severe personality disorders:Psychotherapeutic strategies. New Haven, CT: YaleUniversity Press.

Kernberg, O. F., Selzer, M. A., Koenigsberg, H. W., Carr,A. C., & Appelbaum, A. H. (1989). Psychodynamicpsychotherapy of borderline patients. New York: Ba-sic Books.

Koerner, K., & Linehan, M. M. (1992). Integrative ther-apy for borderline personality disorder: Dialecticalbehavior therapy. In J. C. Norcross & M. R.Goldfried (Eds.), Handbook of psychotherapy inte-gration (pp. 433–459). New York: Basic Books.

Kohlenberg, R. J., & Tsai, M. (1991). Functioned ana-lytic psychotherapy: Creating intense and curativetherapeutic relationships. New York: Plenum Press.

Koons, C. R., Robins, C. J., Tweed, J. L., Lynch, T. R.,Gonzalez, A. M., Morse, J. Q., et al. (2001). Efficacyof dialectical behavior therapy in women veteranswith borderline personality disorder. Behavior Ther-apy, 32, 371–390.

Kyokai, B. D. (1966). The teachings of Buddha. Japan:Bukkyo Dendo Kyokai.

Lequesne, E. R., & Hersh, R. G. (2004). Disclosure of adiagnosis of borderline personality disorder. Journalof Psychiatric Practice, 10, 170–176.

Lesage, A. D., Boyer, R., Grunberg, F., Vanier, C.,Morissette, R., Menard-Buteau, C., et al. (1994). Sui-cide and mental disorders: A case–control study ofyoung men. American Journal of Psychiatry, 151,1063–1068.

Lieb, K., Zanarini, M., Linehan, M. M., & Bohus, M.(2004). Seminar section: Borderline personality dis-order. Lancet, 364, 453–461.

Linehan, M. M. (1987a). Dialectical behavior therapy forborderline personality disorder: Theory and method.Bulletin of the Menninger Clinic, 51, 261–276.

Linehan, M. M. (1987b). Dialectical behavioral ther-apy: A cognitive behavioral approach to parasuicide.Journal of Personality Disorders, 1, 328–333.

Linehan, M. M. (1989). Cognitive and behavior therapy


for borderline personality disorder. In A. Tasman, R.E. Hales, & A. J. Frances (Eds.), Review of psychia-try, Volume 8 (pp. 84–102). Washington DC: Ameri-can Psychiatric Press.

Linehan, M. M. (1993a). Cognitive-behavioral treat-ment of borderline personality disorder. New York:Guilford Press.

Linehan, M. M. (1993b). Skills training manual fortreating borderline personality disorder. New York:Guilford Press.

Linehan, M. M. (1997). Validation and psychotherapy.In A. Bohank & L. Greenberg (Eds.), Empathy re-considered: New directions in psychotherapy (pp.353–392). Washington, DC: American PsychologicalAssociation.

Linehan, M. M., Armstrong, H. E., Suarez, A., Allmon,D., & Heard, H. L. (1991). Cognitive-behavioraltreatment of chronically parasuicidal borderline pa-tients. Archives of General Psychiatry, 48, 1060–1064.

Linehan, M. M., Comtois, K. A., Murray, A., Brown,M. Z., Gallop, R. J., Heard, H. L., et al. (2006). Two-year randomized trial + follow-up of dialecticalbehavior therapy vs. therapy by experts for suicidalbehaviors and borderline personality disorder. Ar-chives of General Psychiatry, 63, 757–766.

Linehan, M. M., Dimeff, L. A., Reynolds, S. K.,Comtois, K., Shaw-Welch, S., Heagerty, P., et al.(2002). Dialectical behavior therapy versus compre-hensive validation plus 12-step for the treatment ofopioid dependent women meeting criteria for border-line personality disorder. Drug and Alcohol Depend-ence, 67, 13–26.

Linehan, M. M., & Heard, H. L. (1993). Impact of treat-ment accessibility on clinical course of parasuicidalpatients: In reply to R. E. Hoffman [Letter to the edi-tor]. Archives of General Psychiatry, 50, 157–158.

Linehan, M. M., Heard, H. L., & Armstrong, H. E.(1993). Naturalistic follow-up of a behavioral treat-ment for chronically parasuicidal borderline patients.Archives of General Psychiatry, 50, 971–974.

Linehan, M. M., & Koerner, K. (1992). A behavioraltheory of borderline personality disorder. In J. Paris(Ed.), Borderline personality disorder: Etiology andtreatment (pp. 103–121). Washington, DC: Ameri-can Psychiatric Association.

Linehan, M. M., Rizvi, S. L., Shaw-Welch, S., & Page,B. (2000). Psychiatric aspects of suicidal behaviour:Personality disorders. In K. Hawton & K. vanHeeringen (Eds.), International handbook of suicideand attempted suicide (pp. 147–178). Sussex, UK:Wiley.

Linehan, M. M., Schmidt, H., III, Dimeff, L. A., Craft, J.C., Kanter, J., & Comtois, K. A. (1999). Dialecticalbehavior therapy for patients with borderline person-ality disorder and drug-dependence. American Jour-nal of Addiction, 8, 279–292.

Linehan, M. M., Tutek, D. A., Heard, H. L., &Armstrong, H. E. (1994). Interpersonal outcome ofcognitive behavioral treatment for chronically sui-

cidal borderline patients. American Journal of Psy-chiatry, 151, 1771–1776.

Loranger, A. W. (1995). International Personality Dis-order Examination (IPDE) manual. White Plains,NY: Cornell Medical Center.

Lynch, T. R., Morse, J. Q., Mendelson, T., & Robins, C.J. (2003). Dialectical behavior therapy for depressedolder adults: A randomized pilot study. AmericanJournal of Geriatric Psychiatry, 11, 33–45.

Marlatt, G. A., & Gordon, J. R. (1985). Relapse preven-tion: Maintenance strategies in the treatment of ad-dictive behaviors. New York: Guilford Press.

Marziali, E., & Munroe-Blum, H. (1994). Interpersonalgroup psychotherapy for borderline personality dis-order. New York: Basic Books.

Merriam-Webster’s New Universal Unabridged Dictio-nary. (1983). Cleveland, OH: Dorset & Baber.

Merriam-Webster, Inc. (2006). Merriam-Webster On-line Dictionary. Available online at www.m-w.com

Millon, T. (1987). On the genisis and prevalence of theborderline personality disorder: A social learning the-sis. Journal of Personality Disorders, 1, 354–372.

Mintz, R. S. (1968). Psychotherapy of the suicidal pa-tient. In H. L. P. Resnick (Ed.), Suicidal behaviors:Diagnoses and management (pp. 271–296). Boston:Little, Brown.

Nose, M., Cipriani, A., Biancosino, B., Grassi, L., &Barbui, C. (2006). Efficacy of pharmacotherapyagainst core traits of borderline personality disorder:Meta-analysis of randomized controlled trials. Inter-national Clinical Psychopharmacology, 21, 345–353.

Paris, J. (2005). Borderline personality disorder. Cana-dian Medical Association Journal, 172, 1579–1583.

Peng, K., & Nisbett, R. E. (1999). Culture, dialectics,and reasoning about contradiction. American Psy-chologist, 54, 741–754.

Pretzer, J. (1990). Borderline personality disorder. In A.Freeman, L. Pretzer, B. Fleming, & K. M. Simon(Eds.), Clinical applications of cognitive therapy(pp. 181–202). New York: Plenum Press.

Reiser, D. E., & Levenson, H. (1984). Abuses of the bor-derline diagnosis: A clinical problem with teachingopportunities. American Journal of Psychiatry, 141,1528–1532.

Rich, C. L., & Runeson, B. S. (1992). Similarities in di-agnostic comorbidity between suicide among youngpeople in Sweden and the United States. ActaPsychiatrica Scandinavica, 86, 335–339.

Shearin, E. N., & Linehan, M. M. (1992). Patient–therapist ratings and relationship to progress in dia-lectical behavior therapy for borderline personalitydisorder. Behavior Therapy, 23, 730–741.

Siever, L. J., & Davis, K. L. (1991). A psychobiologicalperspective on the personality disorders. AmericanJournal of Psychiatry, 148, 1647–1658.

Silk, K. R., Lee, S., Hill, E. M., & Lohr, N. E. (1995).Borderline personality disorder symptoms and sever-ity of sexual abuse. American Journal of Psychiatry,152, 1059–1064.

Simpson, J. A., & Weiner, E. S. (1989). Oxford English


Dictionary (2nd ed.) [Online]. Retrieved December17, 2000, from www.oed.com

Soloff, P. H., Lis, J. A., Kelly, T., Cornelius, J., & Ulrich,R. (1994). Risk factors for suicidal behavior in bor-derline personality disorder. American Journal ofPsychiatry, 151, 1316–1323.

Stein, D. J., Simeon, D., Frenkel, M., Islam, M. N., &Hollander, E. (1995). An open trial of valproate inborderline personality disorder. Journal of ClinicalPsychiatry, 56, 506–510.

Telch, C. F., Agras, W. S., & Linehan, M. M. (2001). Di-alectical behavior therapy for binge eating disorder:A promising new treatment. Journal of Consultingand Clinical Psychology, 69, 1061–1065.

Van Wel, B., Kockmann, I., Blum, N., Pfohl, B., Black, D.W., & Heesterman, W. (2006). STEPPS group treat-ment for borderline personality disorder in The Neth-erlands. Annals of Clinical Psychiatry, 18, 63–67.

Verheul, R., van den Bosch, L. M. C., Koeter, M. W. J.,de Ridder, M. A. J., Stijnen, T., & van den Brink, W.(2003). Dialectical behaviour therapy for womenwith borderline personality disorder: 12-month, ran-domised clinical trial in The Netherlands. BritishJournal of Psychiatry, 182, 135–140.

Wagner, A. W., Linehan, M. M., & Wasson, E. J. (1989).Parasuicide: Characteristics and relationships tochildhood sexual abuse. Unpublished work, Univer-sity of Washington, Seattle.

Weinberg, I., Gunderson, J. G., Hennen, J., & Cutter, C.J., Jr. (2006). Manual assisted cognitive treatment fordeliberate self-harm in borderline personality disor-der patients. Journal of Personality Disorders, 20,482–492.

Wells, H. K. (1972). Alienation and dialectical logic.Kansas Journal of Sociology, 3, 7–32.

Whitaker, C. A. (1975). Psychotherapy of the absurd:With a special emphasis on the psychotherapy of ag-gression. Family Process, 14, 1–16.

Young, J. E. (2000). Schema therapy for borderline per-sonality disorder: Detailed treatment model. NewYork: Cognitive Therapy Center of New York.

Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003).Schema therapy: A practitioner’s guide. New York:Guilford Press.

Zanarini, M. C., Frankenburg, F. R., & Parachini, E.A. (2004). A preliminary, randomized trial of flu-oxetine, olanzapine, and the olanzapine–fluoxetinecombination in women with borderline personalitydisorder. Journal of Clinical Psychiatry, 65, 903–907.

Zanarini, M. C., Frankenburg, F. R., Sickel, A. E., &Yong, L. (1996). The Diagnostic Interview for DSM-IV Personality Disorders. Belmont, MA: McLeanHospital, Laboratory for the Study of Adult Develop-ment.

Zanarini, M. C., Gunderson, J. G., Frankenburg, F. R.,& Chauncey, D. L. (1989). The Revised DiagnosticInterview for Borderlines: Discriminating borderlinepersonality disorder from other Axis II disorders.Journal of Personality Disorders, 3, 10–18.

Zanarini, M. C., Vujanovic, A. A., Parachini, E. A.,Boulanger, J. L., Frankenburg, F. R., & Hennen, J.(2003). A screening measure for BPD: The McLeanScreening Instrument for Borderline Personality Dis-order (MSI-BPD). Journal of Personality Disorders,17, 568–573.


C H A P T E R 1 0

Bipolar Disorder

DAVID J. MIKLOWITZ

Our goal in this book is to present creative and important psychological treatments withempirical support. This chapter on bipolar disorder by David J. Miklowitz was new to theprevious edition but existing evidence for the efficacy of this approach continues to grow.Based on years of systematic research on psychological factors contributing to the onsetand maintenance of bipolar disorder, this sophisticated family therapy approach targetsthe most important psychosocial factors linked to the disorder and associated with pooroutcome (e.g., disruptions in circadian rhythms and certain specific styles of interpersonalinteractions). This chapter, and especially the very useful case study, also illustrate an es-sential linkage between psychological and pharmacological approaches to this very se-vere form of psychopathology.—D. H. B.

Bipolar disorder is one of the oldest and mostreliably recognized psychiatric disorders. Ourthinking about this disorder has evolved overthis century, but the original descriptions(Kraepelin, 1921) of “manic–depressive insan-ity” greatly resemble our current conceptual-izations. This chapter begins with a review ofbasic information about the disorder, its diag-nosis, its longitudinal course, and drug treat-ment. This information about the illness is in-teresting in its own right, but it also providesthe rationale for using psychosocial treat-ment as an adjunct to pharmacotherapy. Themajority of the chapter describes a fo-cused, time-limited, outpatient psychosocialtreatment—family-focused treatment (FFT)—

that comprises three interrelated modules: psy-choed- ucation, communication enhancementtraining (CET), and problem-solving skillstraining (Miklowitz & Goldstein, 1997). It isdesigned for patients who have had a recent ep-isode of mania or depression.

THE DIAGNOSISOF BIPOLAR DISORDER

DSM-IV Criteria

The core characteristic of bipolar disorder isextreme affective dysregulation, or mood statesthat swing from extremely low (depression) toextremely high (mania). Patients in a manic ep-

421

isode have euphoric, elevated mood or irritablemood; behavioral activation (e.g., increasedgoal-directed activity, excessive involvement inhigh-risk activities, decreased need for sleep,increased talkativeness or pressure of speech);and altered cognitive functioning (grandiosedelusions or inflated self-worth, flight of ideasor racing thoughts, distractibility)—typicallyfor more than 1 week. For the diagnosis of amanic episode, there must be evidence that theperson’s psychosocial functioning (marital, oc-cupational, or social) is disrupted; that hospi-talization is required; or that psychotic features(e.g., grandiose delusions) are present (see thefourth edition of the Diagnostic and StatisticalManual of Mental Disorders [DSM-IV; Ameri-can Psychiatric Association, 1994a]).

A patient in a hypomanic episode showsmany of the same symptoms, but the durationis typically shorter (i.e., 4 days or more).Hypomanic symptoms also do not bring aboutsevere impairment in social or occupationalfunctioning and are not associated with theneed for hospitalization or with psychosis.However, the symptoms must reflect a realchange in a person’s ordinary behavior—onethat is observable by others. The distinction be-tween mania and hypomania, which is reallyone of degree rather than type of illness, is diffi-cult for clinicians to make reliably. Often thedegree to which the behavioral activation af-fects a patient’s functioning is underestimatedby the patient, who can see nothing but good inhis or her behavior. A theme of this chapter isthe value of including significant others (i.e.,parents, spouses/partners, and siblings) in pa-tients’ assessment and treatment.

Some patients with bipolar disorder (40%or more; Calabrese, Fatemi, Kujawa, &Woyshville, 1996) experience mixed episodes,in which the criteria for a major depressive epi-sode and a manic episode are met nearly everyday for a minimum duration of 1 week. Thesymptoms of both poles of the illness are expe-rienced simultaneously. The boundaries be-tween “pure” mania and mixed mania are notentirely clear, because depression often lurksbeneath the manic exterior and is easily evokedby situational factors (Young & Harrow,1994). Children and adolescents with the dis-order are particularly prone to mixed states(Geller et al., 2002).

DSM-IV proceeds with the diagnosis of bi-polar disorder somewhat differently than pastDSM systems have done. First, the diagnosti-

cian determines whether the patient satisfiesthe cross-sectional criteria for a manic ormixed episode. If he or she does meet these cri-teria, the diagnosis of bipolar I disorder is ap-plied. If the patient currently meets DSM-IVcriteria for a major depressive episode, he orshe is diagnosed as having bipolar disorderonly if there is a past history of manic, mixed,or hypomanic episodes; otherwise, the diagno-sis is likely to be major depressive disorder (sin-gle episode or recurrent) or another mood dis-order. If the patient presents in remission, theremust be evidence of prior manic or mixed epi-sodes. One implication of this rather compli-cated set of diagnostic rules is that a singlemanic or mixed episode, even in the absence ofdocumentable depression, is enough to warrantthe bipolar I diagnosis. The key word here is“documentable,” because patients oftenunderreport their depression histories and re-veal them only upon careful questioning.

How Has the Diagnosisof Bipolar Disorder Changed?

Every version of the DSM has brought changesin the way we think about bipolar disorder, andmodifications are likely to continue as new edi-tions are published. For example, DSM-V willprobably contain separate criteria for diagnos-ing the disorder in children and adolescents.Currently, DSM-IV uses the same criteria to di-agnose mania in adults and children, despitethe clear developmental differences in presenta-tion (Leibenluft, Charney, Towbin, Bhangoo,& Pine, 2003).

DSM-IV distinguishes between bipolar I andbipolar II disorders. In the former, patientshave fully syndromal manic or mixed episodes.In bipolar II illness, patients only have hypo-manic episodes. Unlike bipolar I disorder, amajor depressive episode must have occurredfor a diagnosis of bipolar II disorder. (In thischapter, the term “bipolar disorder” refers ei-ther to bipolar I or bipolar II disorder as de-fined in DSM-IV, unless otherwise specified).DSM-IV also includes a course descriptor,“rapid cycling,” which appears to characterizebetween 13 and 20% of patients (Calabrese etal., 1996), and more often females (Coryell,Endicott, & Keller, 1992; Schneck et al., 2004).This much misunderstood course subtype is of-ten confused with mixed episodes. A mixed ep-isode refers to the presentation of mania andmajor depression within a single episode,


whereas in rapid cycling a patient has had fouror more discrete major depressive, manic,mixed, or hypomanic episodes in a single year.The confusion in applying this course de-scriptor lies in the fact that it is difficult to tellwhen one episode has ended and another be-gins: If a patient quickly switches from maniato depression in a 48-hour period (what somerefer to as “ultrarapid cycling”), is this truly anew episode or just a different presentation ofthe same episode? Rapid cycling appears to bea transient state of the disorder and not a life-long phenomenon (Coryell et al., 1992).

Finally, DSM-IV deals with the thorny prob-lem of patients with depression who developmanic, mixed, or hypomanic episodes that arebrought on by antidepressants or other activat-ing drugs. Because of the effects of antidepres-sants on the serotonin, norepinephrine, and do-pamine systems, there is the potential for thesedrugs to induce activation, particularly in a pa-tient who is already biologically vulnerable tomood swings. If a patient has never had amanic episode but then develops one after tak-ing an antidepressant, the likely diagnosis issubstance-induced mood disorder. The diagno-sis of bipolar disorder is only then considered ifthe symptoms of mania preceded the antide-pressant (a difficult historical distinction), or ifthe mania symptoms continue for at least amonth after the antidepressant is withdrawn.Similar diagnostic considerations apply to pa-tients who abuse drugs (e.g., cocaine, amphet-amine) that are “psychotomimetics” and caninduce manic-like states.

Comorbidity and Differential Diagnosis

Bipolar disorder, in its various forms, affects3.9% of the population (Kessler, Berglund,Demler, Jin, & Walters, 2005). It is oftencomorbid with other conditions. When 1-yearprevalence rates were considered within a com-munity epidemiological sample, the highestcorrelations were observed between mania/hypomania and attention-deficit/hyperactivitydisorder, followed by oppositional defiant dis-order, agoraphobia, panic disorder, generalizedanxiety disorder, alcohol dependence, and drugabuse (Kessler, Chiu, Demler, & Walters,2005). The disorders with which bipolar disor-der is comorbid have the common underpin-ning of affective dysregulation. In theEpidemiologic Catchment Area study (Regieret al., 1990), 46% of patients with bipolar dis-

order, as defined by DSM-III (AmericanPsychiatric Association, 1980), also met DSM-III criteria for lifetime alcohol abuse (15%) ordependence (31%); 41% met the criteria forother substance abuse (13%) or dependence(28%). A total of 61% met criteria for any sub-stance abuse or dependence.

The distinction between bipolar disorderand Axis II disorders is especially difficult. No-tably, the hallmark of borderline personalitydisorder is affective instability. Akiskal (1996)has argued that what is commonly seen by cli-nicians as Axis II pathology is actuallyundertreated subsyndromal mood disorder.When studies of the overlap of bipolar disorderand Axis II disorders are done carefully, the es-timates of Axis II comorbidity are actuallyquite conservative. For example, studies byCarpenter, Clarkin, Glick, and Wilner (1995)and George, Miklowitz, Richards, Simoneau,and Taylor (2003) estimate that only about 22–28% of patients with bipolar disorder meet thediagnostic criteria for Axis II disorders whenpatients are evaluated during a period of remis-sion. Furthermore, the comorbid Axis II diag-nosis is not always borderline personality dis-order; it is often a disorder from cluster C (e.g.,avoidant or dependent personality disorder).

The boundaries between bipolar and unipo-lar illness are sometimes difficult to draw. De-pressions in bipolar and unipolar disordersmay look quite similar. There is some evidencethat bipolar depressions are more likely to becharacterized by greater psychomotor retarda-tion but less agitation, anxiety, and somatiza-tion than unipolar depressions (Beigel &Murphy, 1971; Goodwin & Jamison, 1990;Katz, Robins, Croughan, Secunda, & Swann,1982). Other evidence indicates that bipolardepressions are shorter than unipolar depres-sions and are more often characterized bymood lability, diurnal variation, feeling worsein the morning, and derealization (Goldberg &Kocsis, 1999; Mitchell et al., 1992). Even morecomplicated is the distinction between agitateddepression of the unipolar type and mixed ma-nia of the bipolar type; both are characterizedby sadness and a highly anxious, restless, acti-vated state. Goldberg and Kocsis (1999) rec-ommend that clinicians attempting to makethese distinctions place emphasis on attributessuch as goal-drivenness and undiminished en-ergy (despite lack of sleep), both of which tipthe scales toward bipolar disorder rather thanunipolar depressive illness.

Bipolar Disorder 423

The distinction between bipolar disorderand schizophrenia is also a difficult judgmentcall. When a patient who has schizophreniapresents with a psychotic episode, he or she canappear acutely activated, grandiose in thinkingand actions, and elated or depressed. In ahighly influential paper published prior to theappearance of DSM-III, Pope and Lipinski(1978) argued strongly that patients with tradi-tionally schizophrenic “Schneiderian first-ranksymptoms” (e.g., thought broadcasting, delu-sions of control) often have bipolar or evenunipolar depressions.

In part to deal with these diagnostic ambigu-ities, DSM-IV makes a distinction betweenschizoaffective disorder and major mood disor-ders with psychotic features. In schizoaffectivedisorder, delusions and hallucinations havebeen present for at least 2 weeks, even whenthere have been no prominent affective symp-toms. In major mood disorders, psychoticsymptoms occur only during periods of signifi-cant mood disturbance. Importantly, psychoticsymptoms, particularly “mood-incongruent”features (delusions or hallucinations that haveno clear content related to sadness or elation,such as the belief that thoughts are being in-serted into one’s head), are a poor prognos-tic sign in bipolar and unipolar disorders(Brockington, Hillier, Francis, Helzer, & Wain-wright, 1983; Coryell, Keller, Lavori, &Endicott, 1990; Grossman, Harrow, Fudula, &Meltzer, 1984; Grossman, Harrow, Goldberg,& Fichtner, 1991; Kendler, 1991; Miklowitz,1992; Tohen, Waternaux, & Tsuang, 1990).

DSM-IV also describes a subsyndromal orsubaffective condition: cyclothymic disorder.Patients with cyclothymic disorder alternatebetween periods of hypomanic symptoms andbrief periods of depression that fall short of thecriteria for major depressive illness. As soon asthe person develops a full manic, mixed, or de-pressive episode, the diagnosis of bipolar I or IIdisorder is substituted. Again, these distinc-tions really concern the degree and duration ofsymptoms rather than their form. In my ownexperience, clinicians are prone to “push” pa-tients with cyclothymia into the bipolar II cate-gory, especially if they feel that the patients arenot reliable in their historical reporting. Some-times it is better to observe the mood lability ofa patient over time than to attempt to distin-guish cyclothymic disorder and bipolar disor-der cross-sectionally.

The accuracy of patients’ reporting is in-creased by self-report records kept over periodsof a year or more. Examples are “life charts”that show, prospectively, the duration, severity,polarity, and functional impairment associatedwith mood disorder fluctuations (Leverich &Post, 1998).

DRUG TREATMENT ANDTHE COURSE OF BIPOLAR DISORDER

Standard Pharmacotherapy

The course of bipolar illness (its pattern of re-lapsing and remitting over time) is best consid-ered with reference to the drug treatments thathelp stabilize patients and allow most to func-tion in the community. In the prepharmaco-logical era (i.e., prior to 1960), patients werehospitalized for years at a time (Cutler & Post,1982). Nowadays, the availability of moodstabilizers such as lithium carbonate, the anti-convulsants (e.g., divalproex sodium [Depa-kote], lamotrigine [Lamictal], and otheragents), and the atypical antipsychotics (e.g.,olanzapine [Zyprexa], quetiapine [Seroquel],and risperidone [Risperdal], Ziprasidone[Geodon] or aripiprazole [Abilify]) has donemuch to ameliorate the course of bipolar illness(Goldberg, 2004). Some of these drugs not onlycontrol the acute episodes of the illness but alsohave “prophylactic value,” meaning that theyhelp prevent future episodes or minimize theduration or severity of episodes that do occur.

Most psychiatrists describe three phases ofdrug treatment: an acute phase, in which thegoal is to control the most severe symptoms ofthe manic, mixed, or depressive disorder; a sta-bilization phase, in which the goal is to help thepatient recover fully from the acute phase,which often means treating residual symptoms(e.g., mild depression) or levels of social–occupational impairment; and a maintenancephase, in which the goal is to prevent recur-rences and continue to treat residual symp-toms. The drugs recommended for bipolar dis-order vary according to the phase of treatment.During the acute and stabilization phases, anantipsychotic medication may accompany amood stabilizer. An antidepressant may be rec-ommended after a manic episode has stabilizedif a patient has ongoing, residual depressionsymptoms. These phases of treatment are alsorelevant to the psychosocial–psychotherapeutic


treatment of bipolar disorder, as discussedlater.

Symptomatic Outcome

If drug treatment is so effective, then why dowe need psychosocial treatment? The problemthat consistently arises in the drug treatment ofbipolar disorder is “breakthrough episodes.”With lithium or anticonvulsant treatment, ratesof relapse over 1 year vary between 37 (Gitlin,Swendsen, Heller, & Hammen, 1995) and 67%(Shapiro, Quitkin, & Fleiss, 1989). Gelenbergand colleagues (1989) found that the 2-year“survival rate” (i.e., the proportion of patientsnot relapsing) for bipolar disorder was 40% ontypical lithium regimens. Gitlin and colleagues(1995) found that the 5-year survival rate wasonly 27%.

In a representative longitudinal study, Keckand colleagues (1998) examined the 12-monthcourse of bipolar disorder among 134 patientswho began in an acute manic or mixed episode.The majority of the patients (N = 104) weretreated with mood stabilizers, with or withoutaccompanying antipsychotics or antidepres-sants. The investigators made a distinctionamong “syndromic recovery,” in which pa-tients no longer met the DSM-III-R (AmericanPsychiatric Association, 1987) criteria for amanic, mixed, or depressive episode for at least8 weeks; “symptomatic recovery,” a toughercriterion by which patients had to have mini-mal or no mood disorder symptoms for 8weeks; and “functional recovery,” which re-quired that patients regain their premorbid(preillness) level of employment, friendships,interests, and independent living status. Of the106 patients who completed the study, 51(48%) achieved syndromic recovery by 12-month follow-up. Only 28 (26%) achievedsymptomatic recovery, and 25 (24%) reachedfunctional recovery at follow-up. Predictors ofpoor outcome included low socioeconomic sta-tus, medication noncompliance, and longer du-ration of illness.

Even more problematic are the residualsymptoms between episodes. In a 13-yearstudy, subsyndromal symptoms, particularlydepression, were present during approximatelyhalf of the weeks of follow-up (Judd et al.,2002). Patients generally take longer to recoverfrom depressive episodes than from manic epi-sodes (Kupfer et al., 2000).

Social–Occupational Functioning

The problems patients with bipolar disorderexperience in social–occupational recovery aresubstantial, even when they are given mood-stabilizing medications. Dion, Tohen, Anthony,and Waternaux (1989) found that about one inevery three patients with mania could not workin the 6 months following a manic episode, andonly about one in five worked at their expectedlevel. Coryell and colleagues (1993) found thatsocial–occupational dysfunction—includinghigh rates of marital separations and divorces—could be observed up to 5 years after a moodepisode. Goldberg, Harrow, and Grossman(1995), in a 4½-year follow-up of patients withmajor mood disorders, found that patientswith bipolar illness had lower work function-ing scores than patients with unipolar illness.Furthermore, only 41% were rated as having a“good overall outcome” at follow-up. Patientsin this study were usually taking lithium aloneor lithium with antipsychotic medications.

A study of 253 bipolar I and II patients re-vealed that only about 33% of the patientsworked full time and only 9% worked parttime outside of the home (Suppes et al., 2001);57% of patients reported being unable to workor able to work only in protected settings.

What predicts social–occupational function-ing? In our 9-month follow-up of lithium- oranticonvulsant-treated young adults with bipo-lar disorder, family conflict, as measured bynegative parent-to-patient verbal interactionsduring a family problem-solving task, pre-dicted levels of social functioning at follow-up(Miklowitz, Goldstein, Nuechterlein, Snyder,& Mintz, 1988). Patient gender also appears tobe important: Harrow, Goldberg, Grossman,and Meltzer (1990) found that women withmania had better work functioning at follow-up than did men with mania.

Not surprisingly, patients with higher socio-economic status have more favorable func-tional outcomes than patients with lower so-cioeconomic status (Harrow et al., 1990; Kecket al., 1998). It appears that social functioningand symptomatic outcome are two correlatedbut separate domains of outcome in bipolardisorder, or “open-linked systems” (Strauss &Carpenter, 1972). Whereas many assume thatsevere symptoms predispose patients to worsefunctional outcomes, at least one study (Gitlinet al., 1995) found the reverse: Poor social


functioning, especially job functioning, pre-dicted a shorter time to relapse in a 41

3-yearfollow-up.

Medication Nonadherence

Part of the reason why patients with bipolardisorder have breakthrough episodes is drugnonadherence. In the Keck and colleagues(1998) study, only 47% of the patients werefully adherent with their medications at follow-up. Miklowitz and colleagues (1988) foundthat only about 30% of young patients withrecent-onset mania took their medications on aroutinely scheduled basis during a 9-monthfollow-up. Whereas rates vary with the meth-ods and criteria for measuring adherence, it isclearly a significant risk factor in this disorder.Jamison and colleagues (Jamison & Akiskal,1983; Jamison, Gerner, & Goodwin, 1979)identified the following factors as predictors ofnonadherence: negative feelings about havingone’s moods controlled by a medication; miss-ing high or euphoric periods; and side effects,which for lithium can include weight gain,thirst, and tremor of the hands.

Why Psychotherapy?

What is the role of psychosocial treatment in adisorder with such a heavy biological and ge-netic basis? There is little doubt that medica-tion is the first-line treatment for bipolar dis-order. The evidence that lithium and theanticonvulsants reduce relapse rates and im-prove functioning is substantial. But can we dobetter? An optimal and perhaps overly optimis-tic view of the outcome of patients with bipolardisorder would include symptom stability forextended periods, minimal disruptions in socialrole functioning after episodes, and becomingconsistent contributors to society. One role ofpsychotherapy as an adjunct to medication isto teach skills for symptom management, toaugment social and occupational role function-ing, and to keep patients adherent to their drugregimens. Implicit in this objective is that thephysiology and psychology of major psychiat-ric disorders are not fully separable. We knowthat changes in brain function (as revealed inpositron emission tomography scans) inobsessive–compulsive disorder can be detectedbefore and after treatment among respondersto behavior therapy and responders tofluoxetine (Prozac) (Baxter et al., 1992). The

time has come to think about psychotherapyand medication as working synergistically inthe major mood disorders.

The strongest argument for including psy-chotherapy in an outpatient treatment programis to help patients to cope with stress triggers.As noted in the next section, certain forms oflife events and family tensions are risk factorsin the course of the disorder. Psychotherapycan target these factors and teach patientsadaptive coping mechanisms, which can thenbe brought to bear during periods of wellnessto help stave off the likelihood of a future re-lapse.

A VULNERABILITY–STRESS MODELOF RECURRENCES

Implicit in the notion that psychotherapywould be helpful to a patient with bipolar dis-order is the notion that stress plays a role ineliciting symptoms of mood disorder. What isthe evidence for this view? What are the targetsfor psychosocial intervention? Figure 10.1summarizes a model that takes into accountthe background of biochemical (e.g., seroto-nin, dopamine, norepinephrine, or gamma-aminobutyric acid [GABA]) imbalances, ge-netic vulnerability, and the two types of stresstargeted in FFT: life events and disturbances infamily functioning (see Goodwin & Jamison,1990; Miklowitz & Frank, 1999; Miklowitz &Goldstein, 1997).

Life Events Stress

In this model, life events affect the onset of bi-polar mood symptoms through the avenue ofdisrupting daily routines and sleep–wake cycles(“social rhythms”). The social rhythm stabilityhypothesis (Ehlers, Kupfer, Frank, & Monk,1993) states that major life events can disruptdaily rhythms in mood disorders via one of twoavenues. They can act as zeitstorers, which dis-rupt established social and circadian rhythms(e.g., the production of neuroendocrines as afunction of the time of day). For example, apreviously unemployed patient who gets a jobwith constantly shifting work hours is forced toadopt a new pattern of daily routines, whichmay include changes in sleep–wake habits. Ma-jor events can also result in the loss ofzeitgebers, which maintain the stability ofrhythms. For example, a spouse or partner


helps maintain a person on predictable socialand sleep schedules. A relationship separation,in addition to being a significant emotionalevent, results in the loss of this human time-keeper.

Patients with bipolar disorder are exquisitelysensitive to even minor changes in sleep–wakehabits. A study by Malkoff-Schwartz and col-leagues (1998) found that manic episodes wereoften precipitated by life events that changedsleep–wake habits (e.g., changing time zonesdue to air travel). However, depressive episodeswere not differentially associated with rhythm-disruptive life events. Similar findings emergedfrom a second study (Malkoff-Schwartz et al.,2000).

One of the clinical implications of these find-ings is that if patients can be taught to regular-ize their social rhythms, especially in the face oflife events that normally disrupt those rhythms,then the outcome of bipolar disorder should beimproved. Thus, variability in the sleep–wakecycle is a target for treatment. This is a centraltenet of interpersonal and social rhythm ther-apy, an individual psychotherapy discussed be-low (Frank, 2005).

Family Stress

Family conflicts may also become a breedingground for increased cycling of bipolar disor-der (see Figure 10.1). One method of measur-ing family stress is to evaluate a family’s level of“expressed emotion” (EE). In this procedure, aresearcher administers the Camberwell FamilyInterview (Vaughn & Leff, 1976) to a familymember (parent, spouse/partner, or sibling) forapproximately 1 hour to assess the relative’s re-actions to the patient’s psychiatric disorder,with particular emphasis on a recent illness epi-sode. Later, a trained judge evaluates tapes ofthese interviews on three primary dimensions:“critical comments” (e.g., “When I talk to him,I get upset that he just shuts down. It’s likethere’s no one there!”); “hostility,” or personal,generalized criticism of the patient (e.g., “I likenothing about him”); and “emotional over-involvement,” or the tendency to be over-concerned, overprotective, or to use inordi-nately self-sacrificing behaviors in the patient’scare (e.g., “I don’t invite people to the housebecause Allen [son] doesn’t like it”). Familymembers who score high on one or more of


FIGURE 10.1. A vulnerability–stress “instability” model of mood episodes in bipolar disorder. FromMiklowitz and Goldstein (1997). Copyright 1997 by The Guilford Press. Reprinted by permission.

these dimensions are called “high-EE”; thosewho do not are called “low-EE.”

EE is a well-established predictor of thecourse of schizophrenia. In Butzlaff and Hoo-ley’s (1998) meta-analysis of 28 longitudinalstudies of EE in schizophrenia, 23 studies repli-cated the same core finding: Patients who re-turn after an illness episode to high-EE familiesare two to three times more likely to relapse in9-month to 1-year prospective follow-ups thanthose returning to low-EE families. Severalstudies have also documented a link betweenhigh EE in families and relapse among patientswith bipolar disorder followed either prospec-tively or retrospectively (Honig, Hofman,Rozendaal, & Dingemanns, 1997; Miklowitzet al., 1988; O’Connell, Mayo, Flatow,Cuthbertson, & O’Brien, 1991; Priebe,Wildgrube, & Muller-Oerlinghausen, 1989;Yan, Hammen, Cohen, Daley, & Henry, 2004).A 2-year study of EE and bipolar adolescentsundergoing family treatment also replicatedthis longitudinal association (Miklowitz,Biuckians, & Richards, 2006).

On first examination, one might concludethat patients with bipolar disorder are sensitiveto stress in the family milieu, and that levels ofEE elicit an underlying biological vulnerability.But the relationship is far from simple. First, itappears that the high-EE relatives of patientswith bipolar illness, unipolar illness, or schizo-phrenia are more likely than low-EE relativesto interpret the patients’ negative problem be-haviors as controllable by the patients(see, e.g., Brewin, MacCarthy, Duda, &Vaughn, 1991; Hooley, 1987; Hooley & Licht,1997; Weisman, Lopez, Karno, & Jenkins,1993; Wendel, Miklowitz, Richards, &George, 2000). Second, relatives and patientscoping with bipolar disorder are often lockedinto verbally aggressive, negative cycles of face-to-face interaction. We (Simoneau, Miklowitz,& Saleem, 1998) found that high-EE relativesof patients with bipolar disorder were morenegative than low-EE relatives during face-to-face problem-solving interactions. The relativesand patients in high-EE families were also morelikely to engage in counterproductive “attack–counterattack” cycles. Often the patientswere provocateurs in these interchanges; theywere not the “victims” of verbally aggressiveor punitive relatives (Miklowitz, Wendel, &Simoneau, 1998; Simoneau et al., 1998).

Clearly, a psychosocial treatment programshould consider aspects of the family’s affective

environment—such as high-EE attitudes inrelatives or the negative interchanges thatcharacterize relative–patient communication—to be targets for intervention. But does one at-tempt to change these attitudes and interactionpatterns directly, or instead make an “end run”around them? Family members coping with aspouse/partner, offspring, or sibling who hasbipolar disorder are understandably quite an-gry, and it makes little sense to tell them theyshould not be. Others feel that their overpro-tective behavior is more than warranted by thesituation.

In developing FFT, we (Miklowitz &Goldstein, 1997) concluded that at least onecomponent of dealing with these attitudes andtransaction patterns is psychoeducation, whichinvolves the provision of information to pa-tients and family members about the disorderand its manifestations. As discussed earlier, rel-atives (parents, spouses, or siblings) need to re-alize that at least some proportion of the pa-tient’s aversive behaviors (e.g., irritability,aggression, inability to work, or low productiv-ity) can be attributed to a biochemically drivenillness state. This may seem obvious to us as cli-nicians, but to family members who deal withthe patient on a day-to-day basis, it is easy toattribute aversive behaviors to personality fac-tors or laziness, or to believe that “He or she isdoing this to hurt me.” In parallel, patientsneed to become more cognizant of the way theyprovoke anger and resentment in family mem-bers.

Negative face-to-face interactions cannot beeradicated, but they can be made more produc-tive through the techniques of communicationand problem-solving skills training. Thus, fam-ilies or couples can be taught to stick with oneproblem topic rather than trying to solve manyat a time, or to use listening skills to avoidcounterproductive attack–counterattack cy-cles. Later in this chapter, these methods are ex-plained with reference to a difficult treatmentcase.

TREATMENT OUTCOME STUDIES

Controlled psychotherapy outcome studies arerelatively new to the field of bipolar disorderand certainly have not kept pace with the re-search on drug treatment. This section de-scribes several randomized controlled trials ofindividual and family/marital interventions.


More thorough reviews of the studies in thisarea are available (Miklowitz, 2006; Miklo-witz & Otto, 2006).

Individual Therapy

Two controlled trials of individual therapy de-serve special emphasis here. Each focused on aspecific risk factor in the course of bipolar dis-order: medication noncompliance (Cochran,1984) and disruptions in social rhythms (Franket al., 1999, 2005). These therapy models areto be distinguished from earlier, psychoanalyti-cally oriented models whose aim was to re-structure personality or deal with early child-hood conflicts (e.g., Cohen, Baker, Cohen,Fromm-Reichmann, & Weigert, 1954).

In Cochran’s (1984) study, 28 lithium-treated outpatients with bipolar disorder wererandomly assigned to lithium carbonate withsix sessions of individual cognitive therapy orto a lithium-only group. The cognitive therapywas oriented toward restructuring cognitionsassociated with noncompliance. At a 6-monthfollow-up, patients who received the cognitiveintervention had significantly better med-ication adherence, fewer hospitalizations, andfewer mood episodes precipitated by drugnonadherence.

Lam, Hayward, Watkins, Wright, and Sham(2005; Lam et al., 2003) examined a 6-month,12- to 18-session cognitive-behavioral therapy(CBT) model with drug treatment versus drugtreatment alone (N = 103). Patients had been inremission for at least 6 months but had had atleast three episodes in the past 5 years. At a 1-year follow-up, 44% of patients in CBT had re-lapsed compared with 75% of the patients whoreceived drug treatment alone. Twelve to 30months after treatment, CBT did not preventrelapse relative to drug treatment alone, but itdid continue to show a positive influence onmood and days spent in episodes.

A recent multicenter effectiveness trial ofCBT in the United Kingdom (N = 253) indi-cated that not all subpopulations of bipolar pa-tients are equally likely to benefit from CBT(Scott et al., 2006). The study compared 22 ses-sions of CBT plus pharmacotherapy to treat-ment as usual (TAU) plus pharmacotherapy.The patients had been in a variety of symp-tomatic states before study entry. No effectswere found for CBT on time to recurrence. Apost hoc analysis revealed that patients withless than 12 prior episodes had fewer recur-

rences in CBT than in TAU. However, patientswith 12 or more episodes were more likely tohave recurrences in CBT than in TAU. Possibly,CBT is most applicable to patients in the earlystages of their disorder or in those whosecourse is less recurrent.

Frank and colleagues (2005) investigated theefficacy of an interpersonal psychotherapy forbipolar disorder—one that included not onlythe core elements of Klerman, Weissman,Rounsaville, and Chevron’s (1984) model of in-terpersonal psychotherapy for depression butalso a component in which patients self-regulated their daily routines and sleep–wakecycles. Patients with a recent mood episodewere randomly assigned either to 45-minute in-terpersonal and social rhythm therapy (IPSRT)sessions and mood-stabilizing medications orto an active clinical management intervention,also with medications. The latter consisted of20-minute sessions with a psychotherapist whofocused on drug side effects and symptom man-agement. Session frequencies were identicalacross the two groups. Randomization wasdone first during an acute phase of treatment,with sessions held weekly, and again at the be-ginning of a preventive, maintenance phaseof treatment, with sessions held biweekly ormonthly for up to 2 years. IPSRT in the acutephase was associated with more time before re-currences in the maintenance phase than in theclinical management condition. IPSRT wasmost effective in delaying recurrences in themaintenance phase, when patients succeeded instabilizing their daily routines and sleep–wakecycles during the acute phase. Thus, consis-tency of routines may protect against a worsen-ing course of the disorder (Frank et al., 2005).

Family and Marital Therapy

There are now several studies of family inter-ventions as adjuncts to medications for patientswith bipolar disorder. The first, by a groupat the Cornell University Medical College(Clarkin et al., 1990; Glick, Clarkin, Haas,Spencer, & Chen, 1991) randomly assigned186 inpatients with psychotic or mood disor-ders to either a brief family intervention or tostandard hospital care. The family interventionaveraged nine weekly or twice-weekly sessions,and focused on helping patients and familymembers to cope with the hospital experienceand make plans for a positive posthospital ad-justment. At 6- and 18-month follow-ups, fe-


male patients who had received the family in-tervention had less severe symptoms and higherglobal functioning than those who had origi-nally received standard hospital care. Interest-ingly, among female patients from both themood disorder and psychotic disorder groups,family treatment led to improvements at 6 andat 18 months in relatives’ emotional attitudes(e.g., less rejecting feelings toward the pa-tients). In the small subsample of patients withbipolar disorder (N = 21), the effects of familyintervention were only seen among 14 femalepatients, so the findings cannot be consideredconclusive.

Another randomized controlled trial by theCornell group (Clarkin, Carpenter, Hull,Wilner, & Glick, 1998) compared a psychoedu-cational marital intervention and pharma-cotherapy with pharmacological care alone.The sample size (N = 33) was small, but pa-tients who received the marital treatment hadbetter drug adherence and better global func-tioning scores over 11 months of treatmentthan those in the comparison group. No effectwas observed for marital intervention on thesymptomatic outcome of patients.

Three randomized trials have been com-pleted on FFT, one at the University of Califor-nia, Los Angeles (UCLA—Goldstein, Rea, &Miklowitz, 1996; Rea et al., 2003;) one at theUniversity of Colorado (Miklowitz et al., 2000;Miklowitz, George, Richards, Simoneau, &Suddath, 2003), and one in the context of theSystematic Treatment Enhancement Programfor Bipolar Disorder (STEP-BD; Miklowitz etal., 2007). These studies are examined in detailbelow.

The UCLA and Colorado Studies

The UCLA and Colorado studies each exam-ined a 9-month, 21-session FFT interventionthat comprised psychoeducation, commun-ication enhancement training, and problem-solving training. Participants were patients andtheir parents or spouses. Patients were re-cruited during an index episode of bipolar dis-order and maintained on mood-stabilizingmedications, with or without antipsychotic orantidepressive agents. However, the studies dif-fered in an important respect: At Colorado, thecomparison “crisis management” group re-ceived two sessions of family education and in-dividual crisis sessions, as needed, over 9months. In the UCLA study, patients in the

comparison group received an individual casemanagement and problem-solving interventionthat was of similar intensity (21 sessions) to theFFT intervention.

Despite these design differences, the resultsthat emerged from the Colorado and UCLAstudies were quite similar. In the Coloradostudy (Miklowitz et al., 2000, 2003), FFT andmedication led to lower frequencies of relapsesand longer delays prior to relapses over a 2-year period than did crisis managementand medication. FFT was also associated withmore improvement in depression and maniasymptoms—effects that did not appear untilthe 9- and 12-month follow-ups, but continuedfor the full 24 months of follow-up. In theUCLA study (Goldstein et al., 1996; Rea et al.,2003), the effects of FFT were seen on hospital-ization rates over a 2-year follow-up. The ef-fects of FFT on time to relapse were not seen inthe first year but did appear in the second year.Notably, rates of rehospitalization in the 1- to2-year period following the 9-month treatmentwere 12% in the FFT group and 60% in the in-dividual therapy group; for relapse, the rateswere 28% and 60%, respectively. Both studiessuggested that there may be a delayed effect ofFFT: Patients and family members may need to“absorb” the treatment and incorporate the ed-ucation and skills training into their day-to-daylives before it has ameliorative effects on the ill-ness.

This latter point was clarified further bySimoneau, Miklowitz, Richards, Saleem, andGeorge (1999), who examined family interac-tion transcripts obtained in the Colorado studybefore and after FFT or crisis managementtreatment. Families (patients with their parentsor spouses) participated in interactional assess-ments that comprised 10-minute problem-solving discussions, which were transcribedand coded via the Category System for CodingPartner Interactions (Hahlweg et al., 1989).Forty-four families returned at 1 year for thesame assessment, after the FFT or crisis man-agement protocol had been completed. Inter-estingly, at the posttreatment (1 year) interac-tional assessments, patients in FFT and those incrisis management could not be distinguishedon the basis of frequency of negative interac-tional behaviors (e.g., criticisms). But therewere clear differences at posttreatment in posi-tive interactional behaviors, particularly in thenonverbal sphere. After FFT, patients and rela-tives were more likely to smile at each other,


nod when others were speaking, and lean to-ward each other when speaking. Moreover, thedegree to which patients improved in their non-verbal interactional behavior over the course ofpsychosocial treatment was correlated withtheir degree of symptom improvement over theyear of treatment.

FFT appeared to have ameliorated certaintensions within the family environments. Fu-ture studies using multiple, time-lagged assess-ments of family interaction and patients’ symp-toms would help to disentangle the directionalrelationship between improvements in familycommunication and patients’ symptomaticoutcomes.

The STEP-BD Study

STEP-BD examined the effectiveness of treat-ment with mood stabilizers in combinationwith psychosocial treatments in 15 participat-ing sites in the United States (Miklowitz et al.,2007). Bipolar patients in a depressive episode(N = 293) were randomly assigned to mood-stabilizing medications—with or withoutantidepressants—and 30 sessions of FFT, inter-personal and social rhythm therapy, cognitive-behavioral therapy, or collaborative care (CC),a three-session psychoeducational treatment.Patients assigned to any intensive psychothera-py had higher recovery rates over 1 year and re-covered an average of 110 days faster than pa-tients in CC. Patients in intensive therapy werealso more likely to remain stable in mood overthe year-long study. In FFT, 77% of the patientsrecovered by 1 year; in interpersonal therapy,65%; and in CBT, 60%. In the CC condition,52% recovered. The differences between theintensive modalities did not reach statisticalsignificance.

The findings of STEP-BD, one of the largestrandomized treatment studies for bipolar dis-order, suggests that psychotherapy is an essen-tial component of the effort to stabilize bipolarpatients in a depressive episode. The commoningredients of intensive treatments—such asteaching strategies to manage mood, identify-ing and intervening early with prodromalsymptoms, enhancing patients’ compliancewith medications, and working toward resolu-tion of key interpersonal or family problems—contributed to more rapid recoveries. FFTproved to be a particularly potent treatment inthis study, although its limitations also becameapparent: only 54% of the patients assessed for

the study had families who were accessible andwilling to participate in treatment.

Family Psychoeducationin Early-Onset Bipolar Disorder

More recent applications of family psychoedu-cation have focused on patients with juvenile-onset bipolar disorder, who most frequentlylive with or are strongly connected with theirfamilies of origin. A revision of FFT for adoles-cents has been developed (Miklowitz et al.,2004), using the same 21-session structureadapted to the developmental needs of this agegroup (e.g., the occurrence of more frequent,briefer episodes, typically with a mixed presen-tation). A 2-year open trial of 20 adolescent pa-tients revealed significant improvements overtime in manic symptoms, depressive symp-toms, and parent-rated problem behaviors(Miklowitz et al., 2006). A second open trialfound positive benefits for the combination ofFFT and CBT for school-age children with bi-polar disorder (Pavuluri et al., 2004). A two-site pilot randomized trial of FFT andpharmacotherapy versus brief psychoeducationand pharmacotherapy (N = 58) is nearing com-pletion (Miklowitz et al., 2006).

Fristad, Gavazzi, and Mackinaw-Koons(2003) have developed a multifamily psychoed-ucation group for school-age bipolar and uni-polar depressed children. Children with mooddisorders who were assigned to multifamilygroups showed greater mood stability over 6months than children on a waiting list. Im-provements extended to parents as well: Par-ents who participated in the groups reported agreater understanding of depression and bipo-lar disorder, more positively toned family inter-actions, and increased use of psychosocial andmedical services than did waiting-list parents.

Summary

The addition of psychosocial treatment—family, group, or individual—to pharmaco-therapy leads to more positive outcomes of bi-polar disorder than can be achieved withpharmacotherapy alone. In drawing conclu-sions from this small literature, we must keepin mind the different domains of outcome thathave been targeted. For some, the targeted out-comes have been relapse rates or symptom se-verity. For others, outcomes have been levels ofglobal functioning or medication compliance.


Few of these studies have examined social, oc-cupational, or school functioning, or the do-main cited as the most important to patients:quality of life.

The remainder of this chapter is devoted tothe specifics of delivering FFT. For whom is itintended? How does it proceed? How are fami-lies educated about bipolar disorder, and howdo they learn new styles of communicating orsolving problems? In reviewing these methods,the reader may wish to reflect on the varioustargets of family intervention (i.e., family atti-tudes or expectations, interpersonal conflict,medication nonadherence) and the various do-mains of outcome that are presumed to be in-fluenced by family interventions via their im-pact on these targets.

CONTEXT OF THERAPY

Treatment Objectives and Structure

FFT has six objectives, all of which concerncoping with an episode of bipolar disorder.These are summarized in Table 10.1. Some ofthese pertain to dealing with the current epi-sode; others are more focused on anticipatingepisodes in the future, and the stress triggersfor these episodes. A strong case is made for theprotective effects of medication and a stable,nonstressful family environment.

FFT is delivered in 21 outpatient sessionslasting 1 hour each. Sessions are given weeklyfor 3 months, biweekly for 3 months, andmonthly for 3 months. This structure was orig-inally proposed by Falloon, Boyd, and McGill(1984) for the behavioral treatment of families

of patients with schizophrenia. The session-by-session plan is more a guide for the clinicianthan a requirement, because some families re-quire less intensive contact at the beginning,others require more intensive contact later, andstill others simply do not need this much treat-ment. The treatment is designed to parallelstages of recovery from a mood episode. Dur-ing the stabilization phase, about seven ses-sions are devoted to psychoeducation, in whichpatients and their relatives become acquaintedwith the nature, course, and treatment of bipo-lar disorder. At this stage, patients are oftenstill symptomatic and are usually functioningsocially or occupationally at a level lower thantheir preepisode capabilities (see Keck et al.,1998; Strakowski et al., 1998). Psychoeduca-tion is an attempt to hasten clinical stabiliza-tion by reducing the family tensions that oftenaccompany the stabilization phase. This isdone through helping a patient and his or herfamily members make sense of the differentevents that have precipitated the acute episode,come to a common understanding of the causesand the treatment of the illness, develop plansfor how the family will act if there are signs of adeveloping recurrence, and modulate expecta-tions for the patient’s and family’s functioningin the recovery period.

Once the family has begun the communica-tion training module (7–10 sessions), the pa-tient is usually fully stabilized from the acuteepisode, although he or she may still have re-sidual mood symptoms. At this point, the pa-tient is usually able to tolerate exercises ori-ented toward resolving family conflict andpromoting behavior change. For example, heor she can practice listening while another fam-ily member speaks, and family members can dothe same for him or her. These exercises can bedifficult when a patient’s emotions are still dys-regulated, but the structure introduced by com-munication training can help the patient modu-late how he or she expresses emotions.

During the final phase, problem-solvingtraining (four to five sessions), the mood epi-sode is largely remitted and the patient hasmoved into the maintenance phase of drugtreatment. At this stage, and sometimes evenearlier, the patient and family are motivated toidentify and address quality-of-life issues thathave been disrupted by the illness (e.g., how amarried/cohabiting patient can find work; howparents can help a young adult offspring moveout of the home and gradually become more in-


TABLE 10.1. The Six Objectivesof Family-Focused Treatment

Assist the patient and relatives in the following:

• Integrating the experiences associated withmood episodes in bipolar disorder

• Accepting the notion of a vulnerability to futureepisodes

• Accepting dependency on mood-stabilizingmedication for symptom control

• Distinguishing between the patient’s personalityand his or her bipolar disorder

• Recognizing and learning to cope with stressfullife events that trigger recurrences of bipolardisorder

• Reestablishing functional relationships after amood episode

dependent). The last few sessions of FFT, heldmonthly, help to consolidate gains made duringthe 9-month treatment.

Setting

Our laboratory, the Colorado Family Project,was established in 1989 at the University ofColorado at Boulder’s Department of Psychol-ogy. The staff members are psychology post-doctoral fellows, psychology graduate stu-dents, and psychiatrists from the communitywith whom we have developed collaborativerelations. Since 1989, we have involved over200 adults and adolescents with bipolar disor-der in various forms of psychosocial care andresearch follow-up. We identify patients forour programs through inpatient screening orthrough referral from outpatient psychiatristswhen patients have acute episodes.

In the Colorado study of FFT, we recruited82 patients from hospital settings and 19 fromoutpatient clinics. It is worth noting that as theeffects of managed care cost containment havetaken hold, fewer and fewer of our patients arerecruited from hospital wards. The initial diag-nostic evaluation (see below) is done in the hos-pital (for inpatients) or in our outpatient re-search clinic at the Department of Psychology.A pretreatment family assessment (see below)is also done in this laboratory setting. Treat-ment and research follow-up interviews aretypically done in this same laboratory, al-though some of our earlier work (Miklowitz etal., 2003) involved family sessions in patients’homes. The UCLA FFT study was entirely out-patient clinic–based. Home-based treatmentcan enhance the generalization of skills training(Falloon et al., 1984), and may also allow clini-cians to treat patients of lower socioeconomicstatus, who ordinarily would not have the re-sources to travel to a university clinic. How-ever, meeting at families’ homes raises many lo-gistical and financial problems for therapists.Thus, we have moved almost entirely to an out-patient clinic–based model, unless home-basedtreatment would greatly enhance complianceor treatment success.

Client Variables

FFT is usually administered to adult patientswith bipolar disorder who live with (or inclose proximity to) their parents, siblings, orspouses/partners. Patients can be of any age,

including adolescents or preteen children. Al-though there is no contraindication for offeringFFT to patients with remitted bipolar disorder,in our experience they and their family mem-bers are less motivated for treatment than thosewho have recently coped with a mood episode.

Patients with bipolar disorder can present asmanic, mixed, hypomanic, depressed, or rapid-cycling. The polarity of the most recent epi-sode, however, is a moving target—it maychange before the patient is seen next. Patientswho are manic or hypomanic, particularlythose who are elated and grandiose, are oftenin denial about whether they are really ill, andmay believe that the disorder and its treatmentare simply ways for others to control them. De-pressed patients may be more motivated forpsychosocial treatment but may have cognitivedifficulty assimilating the educational contentof sessions. Patients with a mixed episode orrapid-cycling bipolar disorder are candidatesfor FFT, but their illness often follows a morepernicious course (Keller et al., 1986; McElroyet al., 1992; Post et al., 1989).

Patients with comorbid alcohol or other sub-stance use disorders pose special problems.These patients are usually resistant to psycho-social treatment and medication. They are alsodifficult to diagnose; the effects of drugs or al-cohol can mimic the cycling of a mood disor-der. Generally, patients with active substanceuse disorders are more successfully treated ifthey are “dry” before FFT commences. If not,it is usually necessary to supplement FFT withchemical dependency programs (e.g., dual-diagnosis Alcoholics Anonymous groups).

Concurrent Drug Treatment

We require that our patients be seen simulta-neously by a psychiatrist, who monitors thepatient’s medications. Typically, a regimen in-cludes a primary mood stabilizer, usually lith-ium carbonate or divalproex sodium (Depa-kote), or newer agents, such as lamotrigine(Lamictal). The choice of these mood stabiliz-ers is at least in part a function of whether thepatient presents with clear-cut episodes of ma-nia or depression, in which case lithium is oftenrecommended. If there is evidence of mixedstates or rapid cycling, the anticonvulsants areoften recommended (Goldberg, 2004). Moreand more, we are seeing patients treated withatypical antipsychotics (e.g., olanazapine[Zyprexa], risperidone [Risperdal], quetiapine


[Seroquel], or clozapine [Clozaril]), either asadjuncts to traditional mood stabilizers or assubstitutes for them. There is evidence thatthese agents are highly effective in controllingmania, and some (notably, olanzapine andquetiapine) have antidepressant properties aswell (Goldberg, 2004; Kowatch et al., 2005;Kowatch & DelBello, 2003). These agents areparticularly valuable if the patient is highly agi-tated or psychotic. Antidepressants (e.g.,paroxetine [Paxil], venlafaxine [Effexor], bu-propion [Wellbutrin]) are still recommended asadjuncts to mood stabilizers or atypicalantipsychotics if the patient’s depression doesnot remit, but they are given sparingly, becausethere is a risk of switching from depression intomanic, mixed, or rapid-cycling states (Altshuleret al., 1995). In children, there is concern abouta slightly increased risk of suicidal ideation orbehaviors (Vitiello & Swedo, 2004). Very fewof our patients (i.e., fewer than 5%) receiveelectroconvulsive therapy in addition to drugtreatment.

A core principle of FFT is that the familytherapist must have regular contact with thepatient’s psychiatrist. This contact is estab-lished early in treatment. A close affiliation be-tween the psychosocial and pharmacologicaltreatment team enhances the likelihood of thepatient’s remaining compliant with his or hermedications; it also decreases the likelihood of“splitting,” or the tendency for a patient (oreven family members) to have a “good doctor”and a “bad doctor.” For example, patients fre-quently complain about their physicians andsay to their FFT clinicians, “I wish you couldjust monitor my medications.” An FFT clini-cian who has a regular dialogue with a patient’sphysician can avoid the trap that is being set byencouraging the patient to bring up these prob-lems with the physician directly.

Some patients who refuse all medications as-sume that coming to therapy will be a substi-tute for drug treatment. These patients oftenhave had bad experiences with pharmaco-therapy and psychiatrists, and may also believethat they are not ill, or that the illness they dohave can be treated using “alternative medi-cine.” We have generally taken a hard line withthese patients and do not accept them into FFTunless they commit to standard pharmaco-therapy (usually lithium, anticonvulsants, and/or atypical antipsychotics). Patients with bipo-lar disorder who are unmedicated are highlylikely to have relapses, and it is not in their best

interests for the clinician to imply that their ill-ness can be managed with psychosocialtreatment alone.

Therapist Variables

In the Colorado Treatment Outcome Study ofadults (Miklowitz et al., 2003), therapist agevaried from 23 to 46 years and from 1 to 14years of clinical experience. The majority weregraduate students in clinical psychology or cli-nicians with recent doctorates. Few had exten-sive background in family therapy before learn-ing FFT. The 15-site Systematic TreatmentEnhancement Program for Bipolar Disorder(STEP-BD; Miklowitz et al., 2007) involvedtherapists with doctorates or social workerswho varied considerably in treatment experi-ence. In other words, there are no requirementsthat an FFT therapist have a certain degree oramount of clinical training at the outset.

There are no studies of therapist variables aspredictors of the course of family psychoeduca-tional treatments. Our own clinical sense hasbeen that there are two predictors of success inlearning FFT. The first positive predictor is theability to think of a family or couple as a sys-tem in which members are interdependent andmutually influence other members’ behaviors.Therapists who have trouble with FFT oftenhave difficulty making the transition to thissystemic way of thinking. They tend, for exam-ple, to conduct family sessions as if they wereindividual sessions, with one patient and sev-eral observers. Some of these same problemsarise in learning other forms of family therapy.

The second positive predictor is the willing-ness to think biopsychosocially—that is, to seebipolar disorder as a biologically based illnessthat requires medication treatment, even if itssymptoms are partially evoked by concurrentstressors. Thus, a therapist often must arguefor the patient’s drug adherence even whenpsychosocial issues are more interesting andseem more pressing.

We have found that the following trainingprotocol works well for learning FFT. First, cli-nicians attend group supervision sessions inwhich trained FFT therapists discuss theircases. They read the published treatment man-ual (Miklowitz & Goldstein, 1997) and an up-dated manual for adolescents (Miklowitz &George, 2004), and watch sample FFT video-tapes. Then they serve as cotherapists totrained FFT therapists. After treating two cases


with close supervision, they are usually readyto see families or couples independently, or totake on trainees themselves.

The cotherapy model has advantages sepa-rate from training. It has a long history in thefamily therapy literature (see, e.g., Napier &Whitaker, 1978). Cotherapists have a way ofkeeping their fellow therapists on track. Also, ifone member of the family appears to be feeling“ganged up on” by one clinician and otherfamily members, the other therapist can bridgethe gap by allying him- or herself with this fam-ily member. In-session dialogue between the cli-nicians can also provide effective modeling ofcommunication skills for members of a familyor couple.

PRETREATMENT ASSESSMENTS

Diagnostic Evaluation

Bipolar disorder is becoming an increasinglycommon diagnosis in inpatient and outpatientcommunity settings. Although this is a positivedevelopment given its underidentification inthe past, there is also an element of sloppinessin modern diagnostic evaluations. Nowhere isthis more obvious than in the diagnosis of chil-dren and adolescents, who are now beingcalled “bipolar” with little supporting evidence(McClellan, 2005). The inadequacy of commu-nity diagnostic evaluations derives in part frominadequate insurance reimbursement for theevaluation phases of a patients’ treatment.Some of the patients referred to us have beenmore aptly diagnosed as having cyclothymicdisorder, borderline personality disorder, oreven major depressive disorder, recurrent.Many adolescents are referred with “rage at-tacks.” Our colleagues in community practicehave often noted the same problems when pa-tients who presumably have bipolar disorderare referred to them.

Upon seeing a new patient, a clinician oftenfinds it useful to do a formal assessment usingall or part of a structured diagnostic interviewto determine the reliability of the diagnosis.Within our research protocols, we have usedthe Structured Clinical Interview for DSM-IV—Patient Version (SCID; First, Spitzer, Gib-bon, & Williams, 1995) as the diagnostic as-sessment device. The SCID is well describedelsewhere (Spitzer, Williams, Gibbon, & First,1992). When the patient is under age 18, weuse the Schedule for Affective Disorders and

Schizophrenia for School-Age Children—Present and Lifetime Version (K-SADS-PL;Chambers et al., 1985; Kaufman et al., 1997),which requires separate interviews with thechild and at least one parent, and then a con-sensus rating.

Some of the factors that can affect the reli-ability of the data obtained from the SCID orK-SADS-PL include whether the patient isacutely ill or stable; acutely ill patients are lessreliable in their symptom reports. Typically, pa-tients in a manic state minimize their symp-toms, whereas depressed patients may do thereverse. Patients with bipolar disorder alsohave trouble with retrospective reporting: “I’vehad over 1,000 episodes” and “I’ve been con-stantly manic–depressed since I was an infant”are common responses to diagnostic inter-views.

Whether one uses the structured or an open-ended clinical interview, it is often difficult todetermine whether a patient’s mood dysregu-lations and associated changes in activity are atthe subsyndromal or syndromal levels. Somepatients report brief periods of hypomania orirritability that alternate with more severe de-pressions. These brief, activated periods do notalways reach the DSM-IV duration thresholdfor hypomania (4 days or more), especiallyamong children and teens. In some cases, thepatient is “one symptom short.” Some of thesepatients are better labeled as having cyclo-thymic disorder or major depressive disorder.For children with manic symptoms who do notmeet the duration or symptom count criteria,the diagnosis of “bipolar disorder, not other-wise specified” is often given (Birmaher et al.,2006).

Hagop Akiskal (1996; Akiskal & Akiskal,1992) has encouraged clinicians to consider abroader bipolar spectrum that includes coretemperamental disturbances. In addition tocyclothymia, Akiskal describes a “bipolar III”or “pseudounipolar” subtype, in which pa-tients have recurrent major depressions, withan underlying temperament marked by hyper-thymia (exuberance, overoptimism, grandios-ity, stimulus seeking, physical intrusivenesswith others), and/or a family history of bipolardisorder. He also describes a subgroup of pa-tients with “subbipolar dysthymia,” marked bymild depressions with hypomanias evoked byantidepressant treatment. Akiskal argues thatthese “soft-spectrum” patients respond to thesame medications as traditional patients with


bipolar I disorder and have similar family his-tories.

In FFT, the broadening of the bipolar spec-trum to include these patients introduces aquandary: Does the clinician proceed with suchpatients in the same way as with patients withbipolar I or bipolar II disorder? How does theclinician educate the patient and family aboutthe factors that bring about manic or depres-sive episodes if discrete episodes cannot beidentified? If a patient has never had a truemanic episode in the absence of antidepres-sants, then should the treatment proceed underthe assumption that eventually he or she willdevelop mania spontaneously? Do we tell thefamily of a child with bipolar disorder, not oth-erwise specified, that he or she is at risk for de-veloping the full syndrome, or are we on shakyground in doing so? Do the same self-management techniques (e.g., using problemsolving to minimize family conflict) apply?

For our research, we have opted to includeonly patients with bipolar I or II disorder as de-fined by DSM-IV criteria. Our general impres-sion has been that patients who do not gothrough clear-cut cycles of mood episodes are adifferent population of patients. However, wesee value in further research on Akiskal’s con-ceptualization of the soft spectrum of bipolardisorder, and particularly research that evalu-ates whether psychosocial interventions shouldplay the same or a different—perhaps even amore intensive—role in the treatment of suchpatients.

Mood Chart

Clarity on the diagnosis, as well as the patient’sprogress in treatment, is aided by asking thepatient to keep a daily mood record. One suchinstrument is the Social Rhythm Metric(Monk, Flaherty, Frank, Hoskinson, & Kupfer,1990; Monk, Kupfer, Frank, & Ritenour,1991), which asks the patient to documentdaily mood on a –5 (Depressed) to +5 (Eu-phoric/activated) scale, along with social rou-tines that may influence these moods (e.g.,sleep–wake times, times when the patient so-cializes, the intensity of this social stimulation,the patient’s exercise habits, and other factors).

Leverich and Post (1998) have developed aself-rated “life chart” that requires the patientto keep track of daily mood variations, medica-tions, life stressors, and sleep. Data frommood/activity charts help the clinician and pa-tient to evaluate collaboratively the type of cy-cling the patient experiences and the degree towhich social stressors contribute to mood fluc-tuations. Figure 10.2 is an example of a moodchart; note the cycling of the disorder in rela-tion to specific social stressors and sleep pat-terns reported by the patient. In this example, astressor (a pet’s illness) is associated with sleepdisruption and the appearance of mixed moodsymptoms at the subsyndromal level.

Family Assessments

Psychoeducational approaches usually beginwith a thorough assessment of family attitudes


FIGURE 10.2. Example of a self-rated mood chart.

and behaviors to identify the targets of inter-vention. In our research studies, we have begunwith the Camberwell Family Interview, the in-strument for rating EE discussed earlier. Thisinterview, usually done when the patient isacutely symptomatic, focuses on the prior 3-month period, which usually includes theprodromal phases of symptom buildup. For thepurposes of clinical planning over research, theinterview yields answers to the following ques-tions: What is the current level of tension in thehousehold and in the relative/patient relation-ship? Which of the patient’s behaviors are elic-iting stimuli for family arguments or hostility?Do family members understand that the patienthas bipolar disorder, or are they likely to attrib-ute the patient’s negative behaviors to internalor controllable factors?

A problem with the EE/Camberwell FamilyInterview method is its lack of easy exportabilityto community care settings. Interviews with twoparents can total 3 hours, and the coding of in-terview tapes can add an additional 6 person-hours per family. If a clinician’s purpose is treat-ment planning rather than research, he or shemay be able to substitute a self-report measuresuch as the Perceived Criticism Scale (Hooley &Teasdale, 1989). This measure simply asks thepatient to rate, on a 1–10 scale, the degree towhich close relatives express critical commentstoward him or her, and the degree to which he orshe expresses critical comments toward rela-tives. In one prospective study, self-reportedscores on this instrument were just as strong inpredicting depressive relapse among patientswith recurrent major depression as were EEscores from the Camberwell Family Interviewmethod (Hooley & Teasdale, 1989). Evaluationof this scale as a predictor of the course of adultbipolar disorder was recently accomplished inthe multicenter STEP-BD study (N = 360;Miklowitz, Wisniewski, Miyahara, Otto, &Sachs, 2005). In that study, the degree to whichpatients reported being upset or distressed bycriticism from relatives was a strong predictor oftheir levels of depression over a 1-year prospec-tive period. Interestingly, the amount of criticismthey perceived from relatives was notprognostically significant.

In our research protocols, we typically bringthe family in for an interactional assessmentonce the patient has achieved some degree ofremission. First, each member of the family, in-cluding the patient, is interviewed individually

and asked to identify two family problem top-ics. Then each participant is asked to verbalizethe nature of this problem as if the family mem-ber to whom the problem statement is directedwere sitting across from him/her (e.g., a motherstates, “Ralph, you never tell me when some-thing is wrong, even though I know it is”). Atape recording of this problem statement isthen played for the opposing family member,who is asked to respond on tape contiguously(e.g., “Mom, I don’t talk to you because when Ido, it makes things worse”). Two simulated“conversations” are obtained from each familymember, and the clinician or experimenterchooses one from the patient and one from arelative as stimuli for family discussions.

Next, the family is brought into a meetingroom with videotaping capability. The first ofthese two problem cue–response sequences isplayed, and the family members are asked todiscuss the problem for 10 minutes; to tell eachother why they said what they did on the tape,and how they each feel about the issue; and totry to reach a resolution. The clinician then ex-its the room, and the discussion ensues. Thenhe or she reenters the room and plays the sec-ond cue–response sequence, and a second 10-minute discussion ensues. Transcripts of these10-minute problem-solving discussions canthen be coded, using the Category System forCoding Partner Interactions (discussed earlier)or a similar interactional coding system. Unfor-tunately, obtaining this information from tran-scripts is very labor-intensive and cannot usu-ally be carried out before the treatment begins.

The clinician can also rely on simple obser-vations of the family’s communication andproblem-solving behavior to inform the skillstraining modules of FFT. To quote Yogi Berra,“You can observe a lot by watching.” First,many family members or patients are unable tofocus on a single problem, and instead begin to“cross-complain” or to accuse other familymembers to counteract the accusations directedat them. Some engage in attack–counterattackcycles, which may include negative verbal ornonverbal behaviors. For any particular family,the clinician must first identify the form theseinterchanges take, which dyadic or triadic rela-tionships they involve; the content areas thattrigger the interchanges (e.g., medication-taking habits, independence, boundaries withinthe family), and whether members of the familyare able to stop these cycles before they spiral


out of control. Who criticizes whom, and howoften? How does the target person respond?Does the original problem ever get solved?

During these unsupervised family interac-tions, some family members combine negativeaffect with unclear or fragmented communica-tion, which has been called “communicationdeviance” in the schizophrenia literature(Wynne, Singer, Bartko, & Toohey, 1977). Wehave found that the relatives of patients withbipolar disorder and the relatives of patientswith schizophrenia do not differ in levels ofcommunication deviance (Miklowitz et al.,1991). Patients with bipolar disorder and theirrelatives often speak in a hurried, driven man-ner, and often jumble phrases, get words out oforder, or express fragments of ideas. The extentto which these communication styles reflect sit-uational stress or enduring patterns of relatingin the family is not clear. Nonetheless, FFT caninclude exercises that focus on communicationclarity, as well as affective balance.

PROCESS OF TREATMENT

Psychoeducation

The initiation of the psychoeducation moduleof FFT requires three conditions. First, the pa-tient must be seeing a psychiatrist and have be-gun a medication regimen. Second, he or shemust have achieved some degree of clinical sta-bility, although there is no requirement that theacute episode be fully remitted. Third, the fam-ily should have undergone an assessment of at-titudes and interactional behaviors, even if notas formalized as described for the research pro-tocols. Table 10.2 summarizes the topical do-mains that are covered in the first module ofFFT, psychoeducation. In the seven or moreweekly sessions that comprise this module, par-ticipants (patients and their close relatives) be-come acquainted with the symptoms of bipolardisorder; the way episodes develop; the roles ofgenetics, biology, and stress; pharmacologicaltreatments; and the role of stress managementstrategies.

The Initial Sessions: Providing a Rationale

As in most other forms of therapy, the clini-cians begin by explaining the rationale for theFFT program. Many participants ask whyfamily or couple sessions should accompanymedication for a patient adjusting to a recent

episode of bipolar disorder. Particularly help-ful in orienting participants is the “reentrymodel”:

“An episode of mood disorder can be quite trau-matic to all members of the family. . . . In bipolardisorder, when the person returns home and be-gins to recover, there is a ‘getting reacquainted’period in which everyone has to get to know ev-eryone else again, and when everyone tries tomake sense of what happened. This is a toughtime for any family, and part of our purpose hereis to make this ‘reacquaintance period’ less dis-turbing to all of you. We’d like during this year toget you, as a family, back to where you were be-fore became ill. We want to giveyou some tools to deal with this recovery period.”(Miklowitz & Goldstein, 1997, p. 93)

There are two purposes for this introduction.First, it communicates to the family membersthat their emotional reactions to the patient’sillness—even if quite negative—are normal andexpectable. Second, it implies that the therapywill include exploration and clarification ofparticipants’ emotional reactions to informa-tion about the disorder. This feature of thetherapy can be made even more explicit:

“If feelings come up for you when we’re discuss-ing this material, please bring them up. We’re in-terested in knowing how this material applies to


TABLE 10.2. Issues in Psychoeducation

The symptoms and course of the disorder• The signs and symptoms of bipolar disorder• The development of the most recent episode• The recent life events survey• Discussing the hospitalization experience• Variations in prognosis: The course

of the disorder

The etiology of bipolar disorder• The vulnerability–stress model• The roles of stress and life events• Genetic and biological predispositions• Risk and protective factors

Intervening within the vulnerability–stress model• Types of medication and what they do• Psychosocial treatments• How the family can help• The self-management of the disorder• The relapse drill

Note. From Miklowitz and Goldstein (1997). Copyright1997 by The Guilford Press. Reprinted with permission.

clinical handbook of psychological...

Documents