What Works in Corrections: Strategies to Reduce Recidivism

Faye S. Taxman, Ph.D.George Mason University

ftaxman@gmu.edu

0123456789

10

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

Year

In M

illi

on

s

Probation Jail Prison Parole Juveniles Other Adult

Insatiable Appetite: The Ever Expanding Population

8+M Adults, 650K Juveniles

*Bureau of Justice Statistics, 2005 adjusted with estimates from Taxman, et al, 2007.

5,613,739 adults need TX(4.5M males, 1.1M females)

7.6% receive TX Each day!

1:28 adults

Substance Use Disorders (Percentages)

70

38

2.69

4440

4.712

0

10

20

30

40

50

60

70

80

Lifetime 30 Days

Adult Justice Adult General Juvenile Justice Juvenile General

CJ Populations have 4 times the SA Disorders as the General Population

STDs (Percentages)

8.9

3.95.2

0.5 0.1 0.01

16.3

5.9

1.32.8

0.6 0.0102468

101214161820

Chlamydia Gonorrhea Syphillis

Adult Justice Adult General Juvenile Justice Juvenile General

• AIDS/HIV: 2% CJ, <1% general• TB : 25% CJ, <.01% general• HEP C: 30% CJ, 1.6% general

Co-Occurring Disorders and Mental Health Disorders (Percentages)

80

64

8.4

26.2

51

75

6.3

25

0

10

20

30

40

50

60

70

80

90

Co-Occurring Disorders MH Disorders

Adult Justice Adult General Juvenile Justice Juvenile General

Type of SA Services Offered• Few Offenders Can Access Services on Any

Given Day

• Majority are Drug-Alcohol Education

– Adult Prison—74%, 8.8% ADP

– Adult Jail—61%, 4.5% ADP

– Adult CC—53.1%, 15.5% ADP

– Juv Res—88%, 30%ADP

– Juv CC—80.2%, 8.2ADP• Low Intensity OP (<4Hrs/Week)

• CBT based therapies are reported in 1 in 5 community based programs for offenders

Taxman, Perdoni, & Harrison, 2007

0%

10%

20%

30%

40%

50%

Drug AlcoholEducation

SA GroupCounseling: up

to 4 hrs/week

SA GroupCounseling: 5-

25 hrs/week

TC-Segregated TC-Non-segregated

Prison ADP Jail ADP Community ADP

% of Adult ADP in Substance Abuse Treatment Services

% o

f A

DP

in

SA

Tx

S

ervi

ces

Estimates of Needs (Belenko & Peugh, 2005): Dependent:

31.5% Males, 52.3%Females

Taxman, Perdoni, Harrison, 2007

What Should We Do?

• Evidence-Based Practices

• Policies to Support Adoption of EBPs

• Increase Programming

What Has Been Tried: CJ Interventions?Results from Meta-Analysis

• Intensive Supervision

• Boot Camps

• Case Management

• Treatment Accountability for Safer Communities (TASC)

• DTAP (Diversion to TX, 12 Month Residential)

• Tx w/ Sanctions (e.g. Break the Cycle, Seamless System, etc.)

• Drug Treatment Courts

• In-Prison Tx (TC) with Aftercare

Taxman, 1999; MacKenzie, 2006; Aos, et al, 2006

What Has Been Tried: Clinical?Results from Meta-Analysis

• Education (Psycho-Social)

• Non-Directive Counseling

• Directive Counseling

• Motivational Interviewing

• Moral Reasoning

• 12 Step with Curriculum

• Cognitive Processing

• Cognitive Behavioral (Social Skills, Behavioral Management, etc.)

• Therapeutic Communities

• Contingency Management/Token Economies

NIDA, 2006

What Does NOT Work (non-Behavioral)*

• Incarceration—70% return; fairly constant

• Fear and other emotional appeals

• Threatening

• Bibliotherapy

• Talking cures

• Self-Help programs

• Vague unstructured programs

• Fostering self-regard (self-esteem)

• “Punishing smarter”Latessa, 2003

Reduction of Crime and Cost

Aos, et al 2006

What Have We Learned?• DOES NOT WORKDOES NOT WORK Boot camps, intensive supervision & control-interventions do not change

behavior

• WORKSWORKS Clinical component or rehabilitation

• MIXED RESULTSMIXED RESULTS

Positive Results IF: Target High Risk Offenders

Longer duration of TX (>90 days)

Treatment is CBT or TC and multidimensional

Address Compliance issues with immediate responses

Rewards to shape behaviors

Negative Results IF: Unfair Procedures or Processes

Less than 90 days

No clinical component

All sanctions

Low risk offenders

Target offense, not behaviors

Procedural Justice

Treat like all others, Fairness

Responsivity

Diagnosis, Address Behavioral Drivers

Behavioral Management

Shape Behavior, Reinforcers,

Contingency Management

Social Learning Models

#1: Procedural Justice: Fairness & Legitimacy

National Research Council, 2005

Reduced rearrests for DV offenders when arrestees given clear instructions about the reason for rearrest (Paternoster, Brame, Bachman, Sherman, 1996)

Police misconduct in high disadvantaged areas increases violence (Kane, 2005)

Police clear instructions increase compliance in communities (Tyler, et al., 2000, 2003, 2004)

Relevance to Corrections: legal cyncism, distrust, and milieu influence outcomes from treatment programs

Impact: Perceived Fairness on Outcomes

Procedural Justice by

PO

Procedural Justice by TX

Arrest/VOP

Drug Use at Follow-

up

-.19*

-.23*

-.31*

-.44*

Taxman & Thanner, 2004

When Offenders Believe they have a VOICE, reductions in negative outcomes occur!

APA Task Force on Empirically Supported Therapy Relationships*

• Rapport/Relationship with Counselor Increase Outcomes: 40% outcomes

• Therapeutic alliance: works with client, not against

• Goal consensus and collaboration: agree on goals for client

• Empathy: understands client

*Norcross, 2002

#2: Risk, Need, & Responsivity

• Valid Instruments to Identify Risk Factors and Criminogenic Needs

• Provide Treatment for Offenders that address Criminogenic Needs

• Match Offenders to Treatment Programs Designed to Affect Criminogenic Needs

• Basic Principle: High Risk Offenders should be placed

Failure to Match Offenders to Appropriate Services Affects Outcomes

44%41% 43%

26%

0%

10%

20%

30%

40%

50%

60%

% W

ith N

ew A

rres

t

Sanction - No Tx Sanction + Tx

Criminal Offender Addict Offender

Mean

31%

Taxman, Reedy, and Ormond (2003)

Matching Offenders to Appropriate Services

• Use Risk Tools that are Actuarial in Basis

• Use Needs Tools that Focus on Dynamic criminogenic factors (e.g. peers, antisocial personality, criminal thinking, etc.) that are subject to change

• Screen/assess on key issues of criminogenic needs and dependency issues

– Offense is not a Proxy for Risk

Criminogenic Needs:

1)Anti-Social Values

3) Low Self-Control

4) Dysfunctional Family Ties

5) Substance Abuse

6) Criminal Personality

2) Criminal Peers

Eric Shepardson & Lina Bello, Bureau of Governmental Research 2001, www.bgr.umd.edu.

Focus On “Big Six”Focus On “Big Six”

Impact of Programs Based on Number of Target Criminogenic Needs Addressed

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

<3 Needs 4+ Needs

Reduction in Recidivism

Increase in Recidivism

Adapted from Latessa, 2003. Original Source is Gendreau, P., French, S.A., and A.Taylor (2002). What Works (What Doesn’t Work) Revised 2002. Invited Submission to the International Community Corrections Association Monograph Series Project

#3 Behavioral Management Approaches

What is a reinforcer? Anything that will be of value to the offender, and that will motivate production of good behavior. Goods, Services, $

• Shapes Offender Behavior – Must be salient; valuable to the recipient

– Must be swift and certain

– Must be withheld when desired behavior does not occur

• Withdrawal of aversive conditions

– Foregoing a urine testing

– Decreasing frequency of meetings with PO

Retention in Treatment with Positive Reinforcers

020406080

100120

2 4 6 8

weeks

% R

etai

ned

Standard Cont Mgt

Petry et al., 2000

% Positive for Any Illicit Drug

0

10

20

30

40

50

%

Intake Week 4 Week 8

Standard Cont. Mgt.

Petry et al., 2000

Re-Arrest Rates From Maryland PCS

0 10 20 30 40 50

Rearrest Rates*

Non-PCS PCS

•38% Reduction in Odds of Rearrest Rates

*p<.01

Taxman, 2008

Requests for VOP Warrants (Maryland PCS)

0 5 10 15 20 25 30 35

Warrant Rates*

Non-PCS PCS

• 40% Reduction in Odds of VOP Warrants

*p<.05

Taxman, 2008

Challenges: Limited Knowledge and Use of Evidence-Based Practices

Use of EBPS

0 2 4 6 8

MeanNo. ofEBP

Adult Prison Adult JailAdult CC Juv ResJuv CC Ad Drug Court

Use of techniques to engage and retain clients in treatment

Addressing co-occurring disorders

Treatment duration of 90 days or longer

Assessment of treatment outcomes

Family involvement in treatment

Availability of qualified treatment staff

Comprehensive Services

Developmentally appropriate treatment

Use of therapeutic community/CBT

Standardized risk assessment

Standardized substance abuse assessment

Continuing care or aftercare

Use of graduated sanctions and incentives

Use of drug testing in treatment

Systems integration

Friedmann, Taxman, & Henderson, 2007: Young, Dembo, & Henderson, 2007

Factors* Associated with the Use of EBPs Factors* Associated with the Use of EBPs in Adult Corrections Systemsin Adult Corrections Systems

All factors listed were statistically significant in multivariate analyses. Factors not impacting use of EBPs: Physical Plant, Staffing, Leadership

Community based programs Administrators: - Background human

service - Knowledge about EBP - Belief in rehabilitation Performance driven culture Emphasis on training Emphasis on internal support

Friedmann, Taxman, & Henderson, 2007; Grella,et al, 2007

State of Practice• Drug Abusing Offenders are Unlikely to Receive

Adequate Treatment Services—too few offenders to have an impact on behavior/outcomes

• Risk-need-responsivity model is still “under construction”, but more in place in prison-based TCs

• System needs strategies to make gains in implementation– Few knowledge barriers, lack of tools– Adoption is Affected by policies – Adoption is Affected by failure to connect to

sentencing

• Continue to develop practices to provide for a continuum of care of similar treatment orientations and philosophies

Unintended Consequences: Our Current Tx & CJS Practices

• With the majority of offenders in need of servicesand the minority receiving servicesno one can not “feel” the impact of treatment

• The continual failure to provide access contributes to an offender’s disbelief and defiance

• Strides in SA treatment do not carry over into CJS with the same, inappropriate processes

• Motivational Engagement practices need to incorporated in CJS actions at all points

Steps to Move Ahead• Adopt Risk and Need Instruments at Sentencing to Define the

Sentence

– Triage: High Risk Offenders Should Be First Priority for Programming

– Programs need to be CBT, focused on continuum of care

• Advance the use of Programming to ensure that 50% of the offenders are involved in educational, vocational, and treatment programming

• Ensure that programming is evidence-based

• Have Correctional Officers/Supervision Staff be part of the plan by using motivational strategies (change the tone of corrections)

Tools of the Trade: A Guide to Incorporating Science into Practice

http://www.nicic.org/Library/020095

WWW.CJDATS.ORG

--as noted by criminologist Joan Petersilia

““The empirical evidence The empirical evidence regarding intermediate regarding intermediate sanctions is decisive. Without a sanctions is decisive. Without a rehabilitative component, rehabilitative component, reductions in recidivism are reductions in recidivism are elusive,”elusive,”