building a better chronic care - national academy of ... · building a better chronic care system i...

Building a Better

System

MAKING MEDICARE RESTRUCTURING WORK

F i n a l R e p o r t o f t h e S t u d y P a n e l o n M e d i c a r e

a n d C h r o n i c C a r e i n t h e 2 1 s t C e n t u r y

J a n u a r y 2 0 0 3

Chronic CareN

ASI

Medicare

inthe

21stC

entury:Building

aB

etterC

hronicC

areSystem

■January

2003

Medicare in the 21st Century:

The National Academy of Social Insurance (NASI) is a nonprofit, nonpartisan organi-zation made up of the nation’s leading experts on social insurance. Its mission is topromote understanding and informed policymaking on social insurance and related

programs through research, public education, training, and the open exchange of ideas.Social insurance encompasses broad-based systems for insuring workers and their familiesagainst economic insecurity caused by loss of income from work and the cost of health care.NASI’s scope covers social insurance such as Social Security, Medicare, workers’ compensa-tion, unemployment insurance, and related public assistance and private employee benefits.

The Academy convenes steering committees and study panels that are charged with conduct-ing research, issuing findings and, in some cases, reaching recommendations based on theiranalyses. Members of these groups are selected for their recognized expertise and with dueconsideration for the balance of disciplines and perspectives appropriate to the project.

The views expressed in this report do not represent an official position of the NationalAcademy of Social Insurance, which does not take positions on policy issues, or its funders.The report, in accordance with procedures of the Academy, has been reviewed by a commit-tee of the Board for completeness, accuracy, clarity, and objectivity.

The Academy wishes to thank the Robert Wood Johnson Foundation for its generous support of this project.

© 2003 National Academy of Social InsuranceISBN# 1-884902-38-3

Suggested Citation:

Eichner, June and Blumenthal, David, eds., Medicare in the 21st Century: Building a BetterChronic Care System (Washington, DC: National Academy of Social Insurance, January2003).

J a n u a r y 2 0 0 3

MAKING MEDICARE RESTRUCTURING WORK

F i n a l R e p o r t o f t h e S t u d y P a n e l o n M e d i c a r e

a n d C h r o n i c C a r e i n t h e 2 1 s t C e n t u r y

Building a Better

SystemChronic Care

Medicare in the 21st Century:

Gerard AndersonJohns Hopkins School of Public Health Baltimore, MD

Patricia ArchboldOregon Health Sciences UniversityPortland, OR

Richard BringewattNational Chronic Care ConsortiumBloomington, MN

Sophia ChangVeterns Health AdministrationPalo Alto, CA

Peter FoxPDF Inc.Chevy Chase, MD

Leslie FriedAmerican Bar AssociationWashington, DC

Glenn HackbarthConsultantBend, OR

Lisa IezzoniHarvard Medical SchoolBoston, MA

Richard KronickUniversity of California, San DiegoSchool of MedicineLa Jolla, CA

Carol LevineUnited Hospital Fund of New YorkNew York, NY

Neil PoweJohns Hopkins School of Public HealthBaltimore, MD

Edward WagnerW.A. MacColl Institute for Health CareInnovationSeattle, WA

T. Franklin WilliamsUniversity of Rochester School of Medicine and DentistryRochester, NY

National Academy of Social Insurance Study Panel on Medicare and Chronic Care

in the 21st CenturyDavid Blumenthal, Chair

Massachusetts General Hospital/Partners HealthCareBoston, MA

The views expressed in this report are of those of the Study Panel Members and do not necessarily reflect those of the organizations with which they are affiliated.

June EichnerStudy Director and Senior Research Associate

Kathleen M. KingDirector of Health Security Policy

Virginia RenoVice President for Research

Reginald D. Williams, IIHealth Security Policy Research Assistant

Contractors

Project Staff

Robert BerensonAcdemyHealth

Washington, DC

Robert Kane and Rosalie Kane

University of Minnesota School of Public Health

Minneapolis, MN

Marty Lynch, Carroll Estes, and Mauro Hernandez

University of California, San FranciscoInstitute for Health and Aging

San Francisco, CA

Bruce VladeckMount Sinai School of Medicine

New York, NY

e

Acknowledgements

The National Academy of Social Insurance and its study panel on Medicare and Chronic Care inthe 21st Century gratefully acknowledge the assistance of a number of individuals in completingthis report. Many staff members of the Centers for Medicare & Medicaid Services providedvaluable information for this report. We are also thankful to Barbara Cooper, Institute forMedicare Practice; Jane Horvath, Partnership for Solutions; Robert Reischauer, Chair, NASIMedicare Steering Committee; and David Colby, The Robert Wood Johnson Foundation. Anyerrors remain those of the authors.

Contents

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .i

Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

A. Panel’s Charge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

B. Definition of Chronic Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

C. Prevalence of Chronic Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

D. Characteristics of Beneficiaries with Chronic Conditions . . . . . . . . . . . . . . . .8

E. Financial Implications of Chronic Conditions . . . . . . . . . . . . . . . . . . . . . . . .8

F. Original Statute and Intent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

G. Characteristics of “Good” Chronic Care . . . . . . . . . . . . . . . . . . . . . . . . . . .11

H. Guiding Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

Chapter 2: Needs and Preferences of Beneficiaries with Chronic Conditions are Beyond What Medicare Currently Provides . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

A. Medical Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

B. Prescription Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

C. Function and Quality of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

D. Self-Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

E. Family Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

F. Supplemental Coverage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22

Chapter 3: The Medicare Program Faces—As Well As Poses for Providers—Considerable Barriers to Chronic Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27

A. Medicare’s Similarity to the General Health Care System . . . . . . . . . . . . . .27

B. Legal and Administrative Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27

C. Original Medicare’s Fee-for-Service Reimbursement System . . . . . . . . . . . .29

D. Medicare+Choice Opportunities to Improve Chronic Care . . . . . . . . . . . . .31

E. Graduate Medical Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .32

F. Improved Care Systems and Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . .33

G. Quality Initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35

H. Research and Demonstrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .37

Chapter 4: Past Initiatives to Improve Care to People with Chronic Conditions Provide Valuable Experience . . . . . . . . . . . . . . . . . . . . . . . . . .39

A. Integrated Financing and Delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39

B. Care Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41

C. System and Payment Redesign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .42

D. Lessons from Past Initiatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45

Chapter 5: Conclusions and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . .47

A. Avenues to Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47

B. Long-Term Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48

C. Short- to Mid-Term Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . .49

D. Priority and Low-Cost Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .57

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .59

iB u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m

This report is about how Medicare couldimprove care for beneficiaries with chronicconditions. During the mid-1960s, acutecare—not chronic care—was the major focusof medicine. When Medicare was institutedin 1965, it was modeled after the healthinsurance system of that time. Medicare wasto function primarily as a claims payer; itsbenefit package and reimbursement systemswere not designed for chronic conditions;preventive services were excluded; and reim-bursement was paid only for in-person visitsand procedures to individual providers. Sincethen, good chronic care and comprehensivecoverage have become crucial to Medicarebeneficiaries. Though some improvementshave been made to Medicare, major changesin the provision and financing of chronic carefor Medicare beneficiaries are needed.Medicare has the potential to refocus itsMedicare program—as well as the nation’shealth care system—and should take a lead-ing role in improving chronic care.

This report is the final product of theMedicare and Chronic Care in the 21stCentury study panel, a panel convened bythe National Academy of Social Insurance aspart of its Making Medicare RestructuringWork project. The panel was charged withdetermining the health care and relatedneeds of Medicare beneficiaries with chronicconditions, how well Medicare meets theirneeds, features of the current Medicare pro-gram that support or impede good chroniccare, and the experience of other chronic caremodels. The panel was also expected to set anew vision for Medicare to improve care andfinancing for beneficiaries with chronic con-ditions, and then propose recommendationsto move toward that vision.

The report is divided into five sections:

■ overview of Medicare and chronic con-ditions, including prevalence of chronicconditions, financial implications ofchronic conditions, Medicare’s originalintent, characteristics of “good” chroniccare, and the panel’s guiding principles

■ needs and preferences of beneficiarieswith chronic conditions

■ barriers to chronic care facing theMedicare program and its providers

■ past initiatives to improve care to peoplewith chronic conditions

■ long-term vision and short- to mid-range recommendations

The study panel focused on originalMedicare, Medicare’s traditional fee-for-service program. It chose this focus because35 million of Medicare’s 40 million benefi-ciaries are covered under this system. Thestudy panel also recommended changes tothe Medicare+Choice (M+C) system, aschanges to M+C may be easier to facilitate.

OVERVIEW OF CHRONICCONDITIONS AMONGBENEFICIARIES

Though there are many ways to define theterm “chronic condition,” the panel chose todefine it as an illness, functional limitation, orcognitive impairment that lasts (or is expect-ed to last) at least one year; limits what a per-son can do; and requires ongoing care.Chronic conditions are prevalent amongMedicare beneficiaries, as most (87 percent)have one or more chronic condition and 65percent have multiple chronic conditions. Inaddition, one-third of beneficiaries have one

Executive Summary

or more chronic condition defined as serious.Though poor Medicare beneficiaries are themost likely to have a chronic condition, allbeneficiaries are at-risk, either through hered-ity, environmental factors, diet, age, orchance.

The cost of managing chronic conditions issubstantial. A disproportionate amount ofMedicare dollars is spent on beneficiarieswith chronic conditions. Beneficiaries withfive or more chronic conditions account for20 percent of the Medicare population but66 percent of Medicare spending. Out-of-pocket spending increases with the numberof chronic conditions: for beneficiaries withthree or more chronic conditions and nosupplemental coverage, 1996 mean annualout-of-pocket expenditures were $1,492(compared to $455 for those with no chron-ic conditions). Beneficiaries’ high out-of-pocket expenditures suggest that Medicaredoes not provide the financial protection thatit was originally designed to ensure. In addi-tion, though expenditures for chronic careare high, the Centers for Medicare &Medicaid Services (CMS) and its beneficiariesare not getting the best value possible for thedollars spent.

NEEDS AND PREFERENCES OFBENEFICIARIES WITH CHRONICCONDITIONS ARE BEYOND WHATMEDICARE CURRENTLY PROVIDES

The quality and scope of care for beneficia-ries with chronic conditions are lacking.Though age and disability-specific care are amajor priority for this population, mostproviders lack training in geriatrics and theassessment and management of functionalstatus and cognition. Many beneficiaries withcommon chronic conditions do not thereceive care recommended by clinical guide-

lines. Systems of care do not facilitate coordi-nation of care among beneficiaries’ multipleproviders, nor do they facilitate more accessi-ble and efficient care, such as care providedby teams of providers, or by phone andemail. Support for self-management and fam-ily care participation may also be negligible.

Medicare does not pay for a substantial shareof beneficiaries’ health care spending, whichdisproportionately affects those with chronicconditions. Beneficiaries must pay out-of-pocket for Part B premiums, deductibles, andcoinsurance. Medicare also does not have alimit on beneficiary copayments for coveredservices. It does not cover prescription drugs,a major form of chronic care treatment, andprovides few benefits to prevent chronic con-ditions or delay their progression. In addi-tion, Medicare does not support manyfunctional and quality of life needs. Sensoryloss, for example, is not considered byMedicare to be a medical concern, and eye-glasses and hearing aid benefits are excludedfrom coverage by statute. Rehabilitative ser-vices are often not covered when the goal isto maintain or slow the deterioration of func-tion. Also, durable medical equipment(DME) and home health care policies maylimit beneficiaries’ ability to function in soci-ety, as DME coverage requires that theequipment be used primarily in the home,while home health coverage requires that thebeneficiary be “homebound.”

THE MEDICARE PROGRAM FACES—AS WELL AS POSES FORPROVIDERS—CONSIDERABLEBARRIERS TO CHRONIC CARE

Medicare does not adequately supportproviders in their treatment and managementof chronic conditions. Its fee-for-servicereimbursement system does not pay for many

ii N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e

of the services and tools important for thecare of beneficiaries with chronic conditions,nor does it offer providers the flexibility toutilize new and efficient methods of opera-tion. Though these limitations are character-istic of the general U.S. health care system,Medicare’s barriers to improved chronic caremay be more pronounced because Medicarebeneficiaries are over twice as likely as thenon-Medicare population to have a chronicillness, and are three times as likely to have afunctional limitation. Also, under the 1965statute, CMS has limited authority over itsproviders, as it is not permitted to “exerciseany control over the practice of medicine orthe manner in which medical services areprovided.” These and other statutes impedethe provision of chronic care services.

Original Medicare’s fee-for-service reim-bursement policies do not support qualitychronic and geriatric care. Reimbursement isnot adjusted for the additional complexityand time it takes to care for chronic condi-tions. Payment to individual providers fordiscrete services (i.e., office visits and proce-dures) discourages a team approach to careand other means of care that may be moreconducive to comprehensive and more effi-cient care. It also provides little incentive tokeep beneficiaries well. Though a number oftechniques have been developed to helpproviders manage care, most have not beenincorporated into providers’ care systems andare not reimbursable by Medicare. Capitatedpayments to health plans would appear tobypass such constraints. However, the experi-ence of M+C found that payment by capita-tion did not assure increases in the quality ofchronic care. It appears that regardless oforganizational and financial arrangements,improving our present systems of care is diffi-cult and will require comprehensive change.

Congress and CMS have implemented anumber of quality improvement initiatives.Unlike for M+C, most of CMS’ quality ini-tiatives for original Medicare do not rely onregulatory requirements. Also, its initiativesdo not focus on care at the physician level,the source of most chronic care, as it is con-strained by the political and statistical diffi-culties of monitoring individual physicians.However, the National Committee forQuality Assurance (NCQA) has begun workto report on ways of measuring the quality ofcare provided by physician practices, begin-ning with large practices. NCQA and otherlarge accreditation organizations have also setstandards for accreditation, certification, andperformance measurement of chronic diseasemanagement. As the quality of such informa-tion improves, CMS could incorporate suchmeasures into original Medicare. This couldlay the basis for paying more to providerswho deliver high standards of quality of care.

One of the primary ways CMS tests newideas is through research and demonstrationprojects. However, CMS’ ability to innovateis limited by the Office of Management andBudget’s (OMB) requirement that demon-stration projects be budget neutral. Not onlydoes OMB require that demonstration pro-jects not increase Medicare expenditures overprojected spending in the absence of thedemonstration, but in the case of demonstra-tions enrolling dual eligibles, budget neutrali-ty is calculated separately for each program sothat savings in one cannot be used to offsetincreased spending in the other. The recentchronic care demonstrations are severely con-strained by the requirement that they bebudget neutral because CMS requires thatthe demonstrations provide drugs and ser-vices not covered under original Medicare.Thus, the evaluation of these demonstrations

iiiB u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m

N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e

will be based largely on the providers’ abilityto manage Medicare expenditures of partici-pating beneficiaries—at a cost that may notbe realistic—while de-emphasizing improve-ments to quality of care. How chronic carecould best be managed under more realisticconditions—allowing modest cost increasesthat might be shared by beneficiaries, forexample—will be left untested.

PAST INITIATIVES TO IMPROVE CARETO PEOPLE WITH CHRONICCONDITIONS PROVIDE VALUABLEEXPERIENCE

A number of initiatives have been imple-mented to improve care for people withchronic conditions. CMS’ Program for All-Inclusive Care for the Elderly (PACE) andthe Social HMOs (S/HMOs) have attemptedto integrate the financing and delivery ofmedical care and community-based care sys-tems for the frail elderly. Other effortsinclude Medicare case management demon-strations for high-cost beneficiaries, and itsend-stage renal disease (ESRD) program,which redesigned the payment system forESRD. Health plans have also implementedprograms to improve chronic care. KaiserPermanente’s Northern California region’sheart failure program, for example, hasworked to improve the care system forpatients with congestive heart failure.Another approach that health plans, providergroups, and CMS participate in is theChronic Care Breakthrough Series BestPractice Collaborative, which utilizes theChronic Care Model for its redesign ofhealth care organizations’ care systems.

These initiatives offer lessons that can beincorporated into mainstream Medicare.Most of these initiatives found that chroniccare requires specialized training of and the

coordination of providers. They also suggestthat financial incentives that align with pro-gram goals may be helpful. In addition,information systems are important to chroniccare initiatives, as organizations must havethe ability to track patients, diagnoses, andutilization. Experience also shows that sus-tained improvement requires comprehensivesystem change, and that it may not be possi-ble to vastly improve systems of care on abudget-neutral basis.

RECOMMENDATIONS

The study panel’s recommendations includeits long-term vision for Medicare and sixshort- to mid-term recommendations. Itsrecommendations address changes across therange of policy sources, including Medicarestatute; regulations; national coverage deci-sions; contractor manuals, memoranda, orother guidance; and policy interpretations byMedicare contractors, including local medicalreview policies.

Long-Term Vision

In the panel’s long-term vision, Medicarewould provide beneficiaries with access toneeded services and financial protection fromcosts that pose barriers to chronic care. Thiswould involve adding coverage for servicesnot presently included in Medicare’s benefitpackage, including function and quality oflife-related services. Changes to the benefitpackage would be designed to meet theneeds of beneficiaries. Medicare would alsoset reasonable limits for beneficiaries’ healthrelated out-of-pocket expenditures.

The panel’s vision entails a dramatic shift toinclude a chronic care focus in Medicare.Providers’ practices would be based on evi-dence-based guidelines. Concern for function

iv

and quality of life would be integrated intothe care system. There would be a seamlesscontinuum across acute, chronic, long-term,and end-of-life care. All providers would usecomputerized information systems, whichwould support the sharing of electronic med-ical records among providers, medicationorder checks, and patient-specific protocols.

As the largest health care purchaser in thecountry, Medicare would actively work toimprove the quality of chronic care. It wouldmeet and surpass the quality standards set bythe broader health care system. Quality ofcare would be measured and reported to thepublic. Medicare would make additional pay-ments to providers who offer high qualitycare. Measures of quality of care would besensitive to the unique conditions, issues, anddiversity of concerns of beneficiaries withchronic conditions.

Reimbursement methods would cease to bean obstacle to chronic care, and wouldinstead support quality chronic care delivery.Such methods would align incentives, adjustfor risk factors, and offer providers the flexi-bility they need to provide good chroniccare. Variations on prepayment and salaries tobetter support chronic care would be consid-ered. Most providers would be affiliated witha provider network organization, a healthplan, or integrated delivery system that offersthem organizational support for chronic care.

Short- to mid-range recommendations

The following are the panel’s short- to mid-term recommendations, some of which couldbe implemented immediately; others whichmay take five to ten years, though work onall should begin immediately.

Recommendation 1:Provide beneficiaries with financial protec-tion from chronic conditions.

■ Limit cost sharing requirements byadding an annual cap on out-of-pocketexpenditures for covered services.

■ Cover services necessary for beneficia-ries’ chronic care needs (as addressed inRecommendation 2).

Recommendation 2:Support the continuum of care beyondthose services presently covered byMedicare.

■ Address gaps in Medicare’s benefitstructure. Two significant gaps are pre-scription drugs and preventive healthservices.

■ Strive to include services related tofunction and health-related quality oflife.

– Relax the requirement that to be covered for home care, beneficiariesmust be homebound.

– Cover durable medical equipmentwith the specific intent of maintain-ing or restoring function.

– Provide for assistive devices that com-pensate for sensory or neurologicaldeficits.

– Support rehabilitation as a tool toimprove, maintain, or slow thedecline of function.

■ Involve families of beneficiaries. Providefamilies information and educationabout Medicare policies and choices ofhealth plans and providers. Add anexplicit patient-family education benefit.Adequately compensate providers forfamily consultation through modifica-tion of Evaluation & Managementcodes.

vB u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m

Recommendation 3:Promote new models of care.

■ Foster delivery system change.

– Encourage improved practice organi-zation and care delivery.

– Support geriatric assessment andmanagement.

– Integrate services for those dually eli-gible for Medicare and Medicaid.

■ Increase providers’ knowledge of chron-ic and geriatric care.

– Use Graduate Medical Educationfunding to support chronic caretraining.

– Support geriatric training for allphysicians and train more academicgeriatricians.

■ Payment should support new models ofcare.

– Risk-adjust Evaluation andManagement (E&M) codes.

– Improve models for risk-adjustingprepaid arrangements.

– Test alternative payment modelswithin original Medicare.

Recommendation 4:Strengthen CMS’ role as a purchaser of care.

■ Measure and report on the quality ofchronic care.

■ Designate Medicare Partnerships forQuality Services demonstration (former-ly called the Centers of Excellence) forselect chronic conditions.

Recommendation 5:Support enhanced information systems.

■ Foster implementation of electronicinformation systems.

■ Promote the collection and standardiza-tion of health and functional assessmentdata.

Recommendation 6:Implement and support funding for researchand demonstration projects.

■ Sponsor a wide variety of chronic careresearch and demonstration projectsand readily incorporate successful ele-ments into the Medicare program.

■ Focus projects on multiple chronic conditions.

■ Redefine budget neutrality for the purpose of approving proposed demonstrations.

■ Increase CMS’ budget for research anddemonstrations to improve chroniccare.

Some of these recommendations will takelonger to enact than others; some will costthe Medicare program more than others.The panel hopes that policymakers will movequickly to put as many of these recommenda-tions in place as possible.

Along with a prescription drug benefit, therecommendations the panel believes wouldhave the most substantial impact if enactedare:

■ limiting cost-sharing requirements byadding an annual limit for out-of-pocketexpenditures;

■ supporting new models of care by risk-adjusting Evaluation and Management(E&M) codes;

■ implementing information systems thattrack beneficiaries across multipleproviders and care settings.

vi N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e

The three low-cost recommendations thatthe panel believes would significantlyimprove the quality of chronic care are:

■ using Graduate Medical Education(GME) funding to support chronic caretraining;

■ testing alternative payment models;

■ measuring and reporting on the qualityof chronic care.

Medicare has for too long short-changedbeneficiaries with chronic conditions. It hasthe opportunity to improve the value of careprovided to its beneficiaries and must takethe lead in improving chronic care.

viiB u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m

1B u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m

This report of the Medicare and ChronicCare in the 21st Century study panel ana-lyzes how well Medicare meets the needs ofbeneficiaries with chronic conditions andprovides recommendations for Medicare’simprovement of chronic care. This studypanel is part of the Making MedicareRestructuring Work project of the NationalAcademy of Social Insurance (NASI). It isthe seventh of the NASI study panels onMedicare, four of which completed theirwork before this panel began.

Although the previous NASI study panelstargeted other Medicare issues, all grappledwith Medicare’s inadequacies in caring forbeneficiaries with chronic conditions. In2000, the NASI Medicare Steering Commit-tee summarized the findings of the first fourstudy panels (Bernstein and Reischauer,2000). It concluded:

■ Medicare reform needs to be addressedin the wider context of how health careis organized, paid for, and used inAmerica.

■ The current Medicare benefit package isinadequate.

■ Market-based competition raises diffi-cult issues with respect to paymentequity and the distribution of risk inMedicare markets.

■ Regardless of other program reforms,structural changes would be necessaryto give the agency that manages Medi-

care the capacity to better manage thehealth care financed through its fee-for-service program.1

The Steering Committee recommended thata study panel be formed to address issues ofaccess to appropriate care for Medicare bene-ficiaries with complex, chronic, and long-term health care conditions and disabilities. Italso recommended additional study panels toexamine issues of the operation of Medicarein a market-based system and the governanceand management of the program.

Since the Steering Committee’s report wasissued, the quality of the U.S. health care sys-tem and its lack of management for chronicconditions has received considerable atten-tion. The Institute of Medicine’s 2001report, Crossing the Quality Chasm: A NewHealth System for the 21st Century, docu-ments the low quality of the U.S. health care system and recommends focusing onchronic illnesses as the starting point forimprovement. Similarly, the Robert WoodJohnson Foundation and others have done a substantial amount of work examining the prevalence and management of chronic conditions—all of which have pointed to the magnitude of change needed to addresschronic care. Congress and the public arebecoming increasingly aware of the U.S.health care system’s deficiencies in handlingchronic conditions and are putting pressureon the Medicare program to improve care tobeneficiaries.

Chapter 1:Introduction

1 The Centers for Medicare & Medicaid Services (CMS, formerly the Health Care Financing Administration(HCFA)).

2 N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e

A. PANEL’S CHARGE

The study panel on Medicare and ChronicCare in the 21st Century was asked to exam-ine problems and possible solutions for delivering appropriate care to the growingnumber of Medicare beneficiaries withchronic care needs. In particular, the panelsought to:

■ determine the health care and relatedneeds of Medicare beneficiaries withchronic conditions, how well Medicaremeets their needs, and features of thecurrent Medicare program that supportor impede good chronic care;

■ assess the experience of other chroniccare models, including Medicaredemonstrations, community-based care,and group, staff, and network modelhealth plans;

■ set a new vision for Medicare toimprove care for beneficiaries withchronic conditions and make recom-mendations to move toward that vision.

The study panel determined that the full spec-trum of changes necessary for optimal chroniccare will take time and resources to be imple-mented. Thus, it produced two sets of recom-mendations: one that is visionary and long-term; and a second that can be implementedwithin the next ten years and lead towardachieving that vision. It proposes changes tothe level of beneficiaries’ financial risk, thehealth care delivery system, the role of pre-vention, coverage of non-medical services,Medicare’s relationships with providers, andresearch and development to improve chroniccare. Though the study panel recognizes the

role that long-term care plays in chronic care,it concentrated on health care needs and noton the residential and supportive services thatare central elements of long-term care.

The study panel focused on originalMedicare, Medicare’s traditional fee-for-service program. It chose this as its primaryfocus because 35 million of Medicare’s 40million beneficiaries are covered under thissystem. The study panel’s recommendationsalso pertain to the Medicare+Choice (M+C)system. Change may be easier to facilitate inM+C for three reasons: CMS has moreauthority to place requirements on healthplans than on original Medicare providers,2

health plans generally have greater institu-tional capability than fee-for-service-basedsolo or group practice physicians to coordi-nate care, and capitation offers flexibility tobetter organize and deliver care.

B. DEFINITION OF CHRONICCONDITION

The study panel chose to use the term“chronic condition” to distinguish it fromthe more commonly used term “chronic illness” because chronic illness typicallyexcludes sub-clinical conditions that do notqualify as an illness (e.g., lipid abnormalities)and functional and cognitive impairmentsthat are not associated with illness (e.g.,spinal cord injury). Though there are numer-ous ways to define “chronic condition,” thepanel chose the definition used by Partner-ship for Solutions, which defines it as a con-dition which lasts (or is expected to last) ayear or longer, limits what a person can do,and requires ongoing care.

2 The term “provider” is used throughout this report to mean any professional or institution who is reimbursedby Medicare, including physicians, rehabilitation therapists, hospitals, skilled nursing facilities, and home healthagencies.

C. PREVALENCE OF CHRONICCONDITIONS

Alternative definitions of the term “chroniccondition” produce a range of estimates ofthe percentage of beneficiaries with chronicconditions. A relatively broad definition of“chronic condition” is expected to produce a higher prevalence estimate, while a morestringent definition is expected to produce a lower estimate. This section presents estimates of the percentage of beneficiarieswith chronic conditions.3 It uses a broad definition to reflect those beneficiaries with any chronic condition (serious and not-as-serious), and a more stringent definition for those with serious chronic conditions.

Any Chronic Condition

Most beneficiaries (87 percent) are eligiblefor Medicare by being age 65 or older; 13percent of beneficiaries are under age 65 andare disabled according to Social Security dis-ability insurance guidelines. In addition, ben-eficiaries with end-stage renal disease (ESRD)account for 0.8 percent of Medicare benefi-ciaries.4 Approximately 17 percent of all

Medicare beneficiaries are dually eligible forboth Medicare and Medicaid. Dual eligiblesinclude those living in nursing homes, as wellthose living in the community with limitedresources.

According to the Medicare Current Bene-ficiary Survey (MCBS), in 1999, 88 percentof Medicare beneficiaries over age 65 and 82percent of disabled beneficiaries under age 65had at least one of the following ten chronicconditions: stroke, diabetes, emphysema,heart disease, hypertension, arthritis, osteo-porosis, broken hip, Parkinson’s disease, andurinary incontinence.5 The percentage withmore than one of these chronic conditions isalso substantial: 66 percent of aged beneficia-ries and 62 percent of disabled beneficiarieshave more than one chronic condition.6

Almost all (97 percent) of those with ESRDhave at least one other chronic condition onthis list (see Chart 1 see page 4). Further-more, another dataset found that 20 percentof the aged and 14 percent of the disabledbeneficiaries have five or more chronic condi-tions (Partnership for Solutions, 2001).


3 Estimates throughout this section are derived from different data because no single source was available thatprovided complete information.Therefore, data from these sources are not directly comparable. Data from theMedicare Current Beneficiary Survey (MCBS) are self-reported and include questions on beneficiaries’ clinicalconditions, as well as functional abilities.Another data source used by Partnership for Solutions, the StandardAnalytic File (SAF), is a dataset of a five percent sample of Medicare claims data. Estimates may also differbecause of methodological differences. Prevalence estimates from the Medicare Chart Book are based on a listof twelve common conditions, while those from Partnership for Solutions are based on ICD-9 codes andinclude many more conditions in its definition.

4 In 1998, well over half of new ESRD patients were over age 65 at the time of renal failure.

5 The term “disabled” is used throughout this report to include persons with disabilities who are under age 65and who are covered by Medicare by meeting SSA’s definition of disability or who have ESRD. They must alsohave paid into the Social Security system for a minimum number of work quarters.

6 The term “aged” is used throughout this report to include persons who qualify for Medicare by being age 65or older.


As Table 1 shows (see page 5), these tencommon chronic conditions are similar forboth aged and disabled Medicare beneficia-ries. Arthritis, hypertension, and pulmonarydisease are the most common chronic condi-tions among beneficiaries. The risk of havingone chronic condition may increase the riskof having another. For example, beneficiarieswith diabetes are at increased risk of havinghypertension. Those with ESRD are also like-ly to have hypertension, pulmonary disease,and diabetes.

Analysis of the Standard Statistical File (SAF)shows that almost all beneficiaries who livelong enough will eventually have at least onechronic condition. While 74 percent of the65–69-year-olds have at least one chroniccondition, 86 percent of those 85 years andolder have at least one chronic condition.Similarly, 14 percent of the 65–69-year-oldshave five or more chronic conditions, while28 percent of 85-year-olds and older havefive or more (Partnership for Solutions,2001).

Chart 1

Percentage of Beneficiaries with Specified Chronic Conditions, by beneficiary category, 1999

Total BeneficiariesESRDDisabled < 65(no ESRD)

Aged(no ESRD)

88%

66%

82%

62%

97%

81%87%

65%

1 or more chronic conditions 2 or more chronic conditions

Source: 1999 Medicare Current Beneficiary Survey (Kaiser Family Foundation, 2001). Notes: Aged, disabled <65, and ESRD categories exclude those living in a facility.

The count for chronic conditions includes stroke, diabetes, emphysema, heart disease, hyperten-sion, arthritis, osteoporosis, broken hip, Parkinson’s disease, and urinary incontinence. ESRD includes aged and disabled with ESRD.Total beneficiaries include those in facilities (1,900,670 beneficiaries).

In addition to physical impairments, cogni-tive and mental impairments are prevalentamong Medicare beneficiaries. As Chart 2shows (see page 6), 18 percent of aged bene-ficiaries and 52 percent of disabled beneficia-ries have a cognitive or mental impairment.Because these numbers are self-reported, theactual percentage of beneficiaries with cogni-tive or mental impairment may be even high-er. An analysis of the SAF found that in1999, 8.4 percent of aged beneficiaries haveAlzheimer’s disease or other dementia(Partnership for Solutions, personal corre-spondence). This percentage may also be anunderestimate because providers may not usedementia codes when filing Medicare claims.

Many beneficiaries have functional impair-ments. Almost half of disabled beneficiarieshave one or more functional limitations, andamong aged beneficiaries, over one-quarterhave one or more functional limitations (seeChart 3 on page 6). Assessment of functionis typically based on the ability to performactivities of daily living (ADLs) and the abili-ty to perform instrumental activities of dailyliving (IADLs). ADLs include basic tasksnecessary for independent living, such asbathing, dressing, using the toilet, feedingoneself, transferring in and out of bed, andmaintaining one’s continence. IADLs addressslightly more complex tasks that involvemore cognitive ability, such as using the tele-


Table 1

Percentage of Beneficiaries with Specified Chronic Conditions, by beneficiary category and chronic condition, 1999

Source: 1999 Medicare Current Beneficiary Survey (Kasier Family Foundation, 2001). Notes: Aged, disabled <65, and ESRD categories exclude those living in a facility.

ESRD includes aged and disabled with ESRD.Total beneficiaries includes those in facilities (1,900,670 beneficiaries).

Aged Disabled <65 Total(no ESRD) (no ESRD) ESRD beneficiaries

Arthritis 57% 52% 38% 57%

Hypertension 55% 46% 93% 55%

Pulmonary disease 38% 32% 56% 37%

Diabetes 17% 20% 51% 17%

Cancer (other than skin) 17% 12% 13% 17%

Skin cancer 18% 6% 12% 16%

Osteoporosis/ broken hip 16% 13% 17% 16%

Emphysema 14% 23% 15% 15%

Stroke 10% 13% 15% 10%

Alzheimer’s disease 2% 1% 1% 2%

Parkinson’s disease 1% 1% 1% 1%


Chart 2

Beneficiaries with Any Type of Cognitive or Mental Impairment, by beneficiary category, 1998

Total BeneficiariesESRDDisabled < 65(no ESRD)

Aged(no ESRD)

18%

52%

26%22%

Source: Urban Institute, unpublished data, 2002.

Chart 3

Beneficiaries with One or More Limitations in Activities of Daily Living (ADL), by beneficiary category, 1998

3+ ADLs2 ADLs1 ADLIADLs only

1.3

31%

12%

4%6%

2%

Aged

Disabled

4%

8% 8%

Source: Urban Institute, unpublished data, 2002.

phone, housekeeping, cooking meals, shop-ping, taking medications, and paying bills.

The risk of having a functional limitationrises as the number of chronic conditionsincrease. In 1996, 15 percent of beneficiarieswith one chronic condition reported having afunctional limitation, as did 33 percent ofthose four chronic conditions, and 43 per-cent of those with seven or more chronicconditions (Partnership for Solutions, 2001).

Serious chronic conditions

The previous section shows that 87 percentof Medicare beneficiaries—most beneficia-ries—have one or more chronic condition.For some of these beneficiaries, their chronic

condition does not restrict their lives; forothers, their condition severely affects theirability to function, their health status, andtheir health care utilization and spending.Though all beneficiaries with chronic condi-tions should benefit from better managementof their condition, the needs of those withserious chronic conditions are more urgentthan those with relatively less serious condi-tions.

As with the definition of “chronic condi-tion,” there is no standard definition of “seri-ous chronic condition.” In their 2001publication, Moon and Storeygard identifiedbeneficiaries with severe chronic conditionsas those meeting their definition of having


Chart 4

Beneficiaries with Serious Chronic Conditions as a percentage of beneficiary population, 1997

Both12.7%

Cognitive Chronic Condition

10.3%

Physical Chronic Condition

9.3%

Neither67.7%

Source: One-Third at Risk: The Special Circumstances of Medicare Beneficiaries with Health Problems(Moon and Storeygard, 2001).

physical or cognitive problems, or both.7

Using a more restrictive definition than thisreport’s definition of any chronic condition,they estimate that in 1997, 33 percent ofMedicare beneficiaries suffered from a seriousphysical problem, cognitive problem, or both.

D. CHARACTERISTICS OFBENEFICIARIES WITH CHRONICCONDITIONS

Medicare beneficiaries with chronic condi-tions are a diverse population:

■ They comprise all income brackets,although poor beneficiaries are morelikely to have physical and cognitiveimpairments. Almost 12 percent ofthose with annual family incomes of lessthan $15,000 report both cognitive andphysical impairments, while 5 percent ofthose with incomes over $50,000 reportthese conditions (Moon and Storey-gard, 2001).

■ The effects of their conditions may bemedical, cognitive, or functional. A per-son with well controlled diabetes, forexample, may need medication but nothave cognitive or functional deficits. A person with early Alzheimer’s maysuffer from dementia but be otherwisemedically healthy and physically functional.

■ The health effects of their conditionsrange from minor to severe. Of benefi-ciaries over age 65: 25 percent thosewith lipid disorders, 33 percent of thosewith hypertension, 45 percent of those

with heart disease, and 50 percent ofthose with diabetes claim to be in fair orpoor health (Anderson, 2001). Thehealth status of those within a diseasecategory may also vary across the spec-trum. For example, the physical andfunctional effects of heart disease rangefrom minor to debilitating.

The risk of having a chronic conditionapplies to all beneficiaries. Few individualscan predict from an early age what theirhealth status will be as they grow older. Itcould include cancer, stroke, diabetes,Alzheimer’s disease, or spinal cord injury.The way these risks play out for individuals,families, and communities may be shaped bythe epidemiology of disease and perhaps bygenetics, and is linked to socio-economic fac-tors, risk behaviors, and sometimes luck(Moss, 1998). Thus, all beneficiaries dependon access to chronic care services and qualitysystems of care.

E. FINANCIAL IMPLICATIONS OFCHRONIC CONDITIONS

The need for managing chronic conditionshas increased over the past two decades, duein part to an increase in life expectancy. Lifeexpectancy at birth increased from 70.2 yearsin 1965 to 76.5 years in 1997. In addition,the baby boom generation’s entrance intoMedicare will place increased pressure onMedicare and the health care system. Thenumber of beneficiaries is projected to con-tinue to grow from 40 million in 2001 to 77million by 2030. The number of beneficiariesover age 85—those with the greatest chronic


7 A beneficiary is classified as having a physical condition if they report three or more diagnoses, includingrheumatoid arthritis, diabetes, Parkinson’s disease and emphysema; if they have lived in a nursing home for anypart of the year; have difficulty performing three or more activities of daily living (ADLs); or report being in“poor” health.A beneficiary is classified as having cognitive difficulty if they report problems using the telephoneor paying bills, or have ever been told they have Alzheimer’s disease or certain other mental conditions.

care needs—is projected to grow from 4.3million to 8.5 million over this same period(Kaiser Family Foundation, 2001). Theunder age 65 disabled population has growneven faster than the aged population: enroll-ment rose from 2.2 million in 1975 to 5.6million in 2000. By 2017, Medicare isexpected to cover 8.8 million disabled per-sons (MedPAC, 2002a).

Expectations for the treatment of chronicconditions have also grown. While little wasknown about treatment of chronic conditionsin 1965, advances in prevention, treatment,and management of many chronic conditionshave fostered an attitude to “do something”(Vladeck, 2002). Pharmaceuticals, surgicaltreatments, and technological proceduresused to address acute conditions now serveto treat or palliate some chronic conditions.Major joint replacement, for example, canprovide relief and renewed function toseverely arthritic joints.

Emerging technologies and interventions,though they may improve the quality of careprovided, are often costly. The high cost ofcare for Medicare beneficiaries with chronicconditions has become a major concern:

■ A disproportionate amount of Medicareexpenditures is spent on beneficiarieswith chronic conditions. For example,beneficiaries with five or more chronicconditions comprise 20 percent of theMedicare population but 66 percent ofprogram spending (Berenson andHorvath, 2002).

■ More chronic conditions equates tohigher Medicare expenditures per bene-ficiary. As beneficiaries’ number ofchronic conditions increases, average

Medicare expenditures increase. MeanMedicare annual expenditures per bene-ficiary with two chronic conditions is$7,64; those with seven or more condi-tions have mean Medicare annualexpenditures of $22,056 (Partnershipfor Solutions, 2001).

■ Care for beneficiaries who simultane-ously have a chronic illness, disability,and a functional limitation is expensive.Direct medical costs for those with oneor more chronic illness averages $3,482;for those with a chronic illness and adisability, costs rise to $6,193; for thosewith a chronic illness, disability and afunctional limitation, costs rise furtherto $11,477 (Anderson, 2001).

As the number of beneficiaries with chronicconditions increases, and as expenditures fortreatment of chronic conditions rise, there isconcern that Medicare expenditures willdeplete the Hospital Insurance (Part A) TrustFund and increase Supplementary MedicalInsurance (Part B) expenditures paid fromfederal general funds and beneficiaries’ pre-miums. Total Medicare spending increasedfrom $35 billion in 1980 to $241 billion in2001. Part A expenditures are projected torise by 72 percent between 2001 and 2011;Part B expenditures are expected to increaseby 92 percent over this same period (Boardof Trustees, HI and SMI Trust Funds,2002).8

Despite the attention to Medicare expendi-tures, the panel believes that quality of carefor beneficiaries with chronic conditionsshould be policymakers’ and the Medicareprogram’s primary concern. In addition, thevalue of Medicare’s expenditures should be


8 Intermediate assumptions.

considered: beneficiaries should receive thehighest quality of care for the dollars spent.The panel also believes that long-term pro-gram costs cannot be controlled withoutaddressing the quality of chronic care.Management of chronic conditions—or lackof management—will greatly influenceMedicare spending.

F. ORIGINAL STATUTE AND INTENT

Medicare was created to ensure that theelderly would have health benefits compara-ble to those of the working age population,and at a reasonable cost to them and to soci-ety. At the time of its passage in 1965, thecosts of health care were unaffordable tomany elderly people. Persons aged 65 or

older faced health care costs that averagedthree times more than for younger persons,while at the same time they had only half asmuch income. Hospital costs were rising dra-matically. Availability of health insurance forthe elderly was a major problem, as only halfof the elderly had health insurance and thesepolicies typically covered only one-quarter oftheir hospital expenses (Blumenthal, et al.,1988).

Medicare was designed as a social insuranceprogram. Under Part A, workers and theiremployers pay into the Medicare program inreturn for health insurance when workers andtheir spouses become elderly. The drafters ofthe original Medicare legislation emphasizedcoverage of hospital costs (Part A) becausesuch costs accounted for the bulk of healthcare expenditures; outpatient and otherhealth care–related costs at this time were asmaller share of costs than they are today.Part B (principally outpatient services) doesnot technically follow the social insurancemodel. It is funded through federal generalfunds and beneficiary premiums. Both PartsA and B provide equal coverage to all benefi-ciaries, regardless of income or assets. In1972 Medicare eligibility was expanded toinclude persons under age 65 with long-term disabilities and those with ESRD who hadpaid into the Social Security system for aminimum number of years.

Medicare’s design was consistent with com-mercial indemnity insurance of the 1960sand incorporated the insurance principles ofthis period:

■ The payer functioned as a passive claimspayer.

■ The benefit package and reimbursementsystems focused on acute care and werenot designed for chronic conditions.

■ Preventive services were excluded.

■ Reimbursement was limited to in-person visits with providers (the pre-dominant way that physicians andpatients interacted).

Since 1965, millions of elderly and disabledpersons have benefited from Medicare.Beneficiaries’ support for Medicare has beenoverwhelming, even among those with gen-


“No longer will older Americans bedenied the healing miracle of modernmedicine. No longer will illness crushand destroy the savings they have socarefully put away over a lifetime so

that they might enjoy dignity in their lateryears. No longer will young families seetheir own incomes and their own hopes

eaten away simply because they are carrying out their deep moral

obligations.”

President Lyndon B. Johnson at the signing of theMedicare legislation in July 1965

erally negative views of the federal govern-ment (Bernstein and Stevens, 1999). In1998, Robert Ball, Social SecurityCommissioner from 1962 to 1973 declared:

“Medicare has done well what itwas designed to do. Because of theprogram, hundreds of millions ofolder people and their childrenhave been better off. Not only hasthe cost of medical bills been madebearable, but lives have been savedand the quality of life of the elderlyhas been greatly improved….But itis a lifesaver itself in need of sav-ing. Once a leader in providinghealth care, the program has fallenbehind.”

Referring to deficiencies in coverage, particu-larly when compared to health insurance forthe employed population, the study panelagrees that Medicare has fallen behind andno longer provides adequate financial protec-tion to its beneficiaries. Medicare’s acute carefocus, with limitations in coverage and sup-port for a fragmented system of care, keepsMedicare from achieving its full potential.Thus, policymakers must grapple with howto ensure that beneficiaries’ chronic careneeds are well served, while at the same timecontrol the expenditures of beneficiaries andoverall Medicare spending.

G. CHARACTERISTICS OF “GOOD”CHRONIC CARE

What is “good” chronic care? For what typeof care should Medicare strive? In a paper

prepared for the study panel, former HCFAAdministrator, Bruce Vladeck, describedseven characteristics of good chronic care(Vladeck, 2002):9

■ It is continuous. The same health pro-fessional, or coherent group of profes-sionals, manages the patient’s care overthe protracted time periods made neces-sary by the characteristics of chronicconditions.

■ It is multidisciplinary. The effects and complications of serious chronicconditions—both medical and non-medical—are sufficiently diverse that no single professional discipline can beexpected to adequately address them all.In good chronic care, physicians, nurses,social workers, therapists, nutritionists,and others work together in a highlycommunicative and mutually supportiveway.

■ It is accessible. Patients should not haveto devote an inordinate amount of timeand energy to obtain the services theyneed.

■ It is coordinated and seamless. Thedifferent professionals and organizationsinvolved in the patient’s care worktogether in an efficient and harmoniousmanner, in a way that is largely invisibleto the patient.

■ It encourages “activation” of patientsto be involved in their own care. Thebetter educated patients are about theirproblems and issues of self-care, thebetter off they are likely to be.


9 Based on the authority of Edward H.Wagner, et al., “Organizing Care for Patients with Chronic Illness,” TheMilbank Quarterly 74(4), (1996), and Christine K. Cassel, Richard W. Besdine, and Lydia C. Siegel,“RestructuringMedicare for the Next Century: What Will Beneficiaries Need?” Health Affairs 18(1): 118–131, January–February, 1999.

■ It supports patients’ families andother caregivers. Chronic illness hap-pens to families, not just individualpatients. Most care of the chronically ill,including chronically ill Medicare bene-ficiaries, is provided by non-professionalrelatives and friends. Good systems ofchronic care reinforce such “informal”caregiving, rather than frustrate it.

The study panel agrees with Vladeck’s conclusion that Medicare, as it is currentlyconstituted, does not support these characteristics.

H. GUIDING PRINCIPLES

The study panel adopted the following threeprinciples for selecting policy changes to pro-mote better chronic care:

■ Focus on beneficiaries’ needs andpreferences. Medicare should recognizeand aim to meet the needs and prefer-ences of beneficiaries. The panelacknowledges, however, that givenresource constraints, Medicare cannotbe expected to fulfill all of beneficiaries’needs and preferences.

■ Payment should support recommend-ed models of care delivery. Whilechanges to reimbursement policies areoften recommended as solutions toMedicare’s chronic care problems, thepanel believes that it is important to firstrecommend appropriate care models,and then structure reimbursement poli-cies around such models (rather thanvice-versa).

■ Aim for efficient care. Noting thatCMS and its beneficiaries are not get-ting the best value possible for the dol-lars spent, the panel seeks policies thatminimize waste, expense, and unneces-sary effort.

The panel also strove to base its recommen-dations on evidence. Evidence, however, isfrequently limited, particularly in many areasof chronic care. The panel believes it is notalways feasible to wait for definitive evidencewhen addressing critical and pressing policyissues. As the research base increases, policiesshould be modified according to new evidence.


B u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m

Beneficiaries have needs and preferences thatcall for improved medical care, as well as anexpansion of Medicare’s scope. In addition tocontrolling medical illness, they would liketheir health care to promote function, qualityof life, self-management, and family involve-ment. For most beneficiaries with chronicconditions, attention to these factors is close-ly related to medical outcomes.

A. MEDICAL CARE

Age and disability-specific care is a priorityfor Medicare beneficiaries. Many aged benefi-ciaries have difficulty finding a physicianknowledgeable about their aged-relatedproblems. Though both geriatricians and pri-mary care physicians could provide such care,many primary care physicians have littleknowledge of geriatric principles and do notaddress functional and cognitive issues. Manypersons with disabilities have difficulty find-ing providers who are sensitive to theirneeds, and medical facilities and equipmentare not always accessible. Beneficiaries mayalso struggle to find participating specialists,particularly for mental health services.

Many beneficiaries do not receive appropriatecare for their chronic conditions. One studyfound that, in 1999, 12 percent of beneficia-ries with five chronic conditions were hospi-talized with a condition that might have beenavoidable with appropriate ambulatory care,and the percentage rose to almost 30 percentfor those with nine chronic conditions.(Wolff, et al., 2002). Another found that 30

percent of beneficiaries, many of whom hadchronic conditions, were not getting the follow-up care they needed (Foote andHogan, 2001). Low quality of care forchronic conditions, however, is not unique to Medicare: recent surveys show that fewerthan half of all U.S. patients with hyperten-sion, depression, diabetes, or asthma arereceiving appropriate treatment (Wagner, etal., 2001). One study found that 17 percentof U.S. adults with chronic conditions reportreceiving conflicting information fromproviders (Partnership for Solutions, 2001).

Beneficiaries’ adherence to providers’ orderscan be difficult because many have a hardtime understanding and remembering theirproviders’ orders. Written instructions andreminders for appointments and medicationrefills are rare. Many also have a hard timearranging transportation to providers’ offices.Though many of their questions could behandled over the phone or by email,providers’ staff usually insist they make anappointment for an office visit (in partbecause providers are not reimbursed forphone or email consultations). Those whohave difficulty coordinating their care andfollowing orders may benefit from havinghelp managing their care (as in the case studyon page 14).

Those with chronic conditions find that theirproviders frequently do not work together tocoordinate their care. The need to coordinatecare becomes more important for beneficia-ries with multiple conditions because the

Chapter 2:Needs and Preferences of Beneficiaries with Chronic Conditions are Beyond What MedicareCurrently Provides

13

N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e14

more chronic conditions a beneficiary has,the more physicians they see on average (seeChart 5 on page 15): those with one chroniccondition visited 4.0 different physiciansannually, those with three chronic conditionsvisited 6.5 different physicians, those withfive chronic conditions visited 13.8 differentphysicians.10 Many of these beneficiaries donot have a primary care provider and insteadreceive care from a variety of specialists. Eventhose with a primary care provider find thattheir provider and specialists do not ade-quately communicate with each other.Consequently, providers treat beneficiariesand prescribe medications without knowingwhat other medications the beneficiary hasbeen prescribed. Such treatment practiceslead to medical errors.

Medicare covers few benefits to preventchronic conditions or to delay their progres-sion. The principles of disease prevention andhealth promotion for older persons weredeveloped after Medicare was implementedand have steadily gained acceptance. Despite

concepts of “healthy and successful aging,”the notion that health promotion and pre-vention for the elderly “is not worth it”lingers (Omenn, 1990). Nonetheless, agrowing literature now indicates that chronicdisease and functional disability can be mea-surably reduced or postponed through clini-cal preventive services and lifestyle changes(DHHS, 2003).

While the original Medicare statute excludedpreventive health services from the benefitpackage, over the years Congress has addedspecific services. Some of these services are toprevent the onset of chronic conditions (e.g.,certain immunizations); others are screeningtools (e.g., pap smear, pelvic exam, mam-mography); others reduce the burden of dis-ease for those who already have chronicconditions. For example, Congress hasauthorized a number of preventive healthbenefits exclusively for a small subset of ben-eficiaries with chronic conditions, includingdiabetes self-management training and sup-plies; glaucoma screening for those at high-

Case StudyA 78-year-old woman with multiple chronic illnesses had been able to successfully manage her carewith the help of her primary care physician. Over the last month, her health and her functional andcognitive abilities have markedly declined. Her judgment and memory are now seriously impairedand her ability to care for herself has been compromised. In light of these new symptoms, she wasreferred for an outpatient assessment; a home health evaluation; and to a dietician, social worker, andtwo medical specialists. While she had previously been able to navigate the health care system effec-tively, the additional burden of coordinating numerous contacts with multiple health care providerswas impossible for her to manage. Also, though she tried to follow the recommendations of thesenumerous health care providers, the multiple—and sometimes contradictory—recommendations weretoo complex. She was hospitalized soon after.

Family Care Case Study

10 As distinct from the definition used by SSA, Partnership for Solutions defines disability as any one of the follow-ing characteristics: 1) the use of assistive technology, 2) difficulty walking, climbing stairs, grasping objects, reachingoverhead, lifting, bending or stooping, or standing for long periods of time, 3) any limitation in work, 4) social/recreational limitations, 5) cognitive limitations, 6) vision problems, 7) deafness or difficulty hearing.


risk for glaucoma (diabetes or a family histo-ry of glaucoma); and medical nutrition thera-py for those with diabetes, chronic renaldisease, or kidney transplants. Nonetheless,other preventive health services, includingcholesterol management, and hearing impair-ment screenings, remain uncovered.

B. PRESCRIPTION DRUGS

Beneficiaries need and want coverage for pre-scription drugs. An August 2002 poll foundthat 80 percent of those age 45 and abovebelieve that a prescription drug benefitshould be added to Medicare that year(Love, 2002). Prescription drug coverage ismore urgent for beneficiaries with chronicconditions because ongoing drug treatmentis crucial to the treatment of almost allchronic conditions. Also, the greater number

of chronic conditions a beneficiary has, thegreater are his/her drug needs and expendi-tures. As Chart 6 (see page 16) shows, bene-ficiaries with two chronic conditions filled anaverage of 18 prescriptions annually; thosewith five or more chronic conditions aver-aged 49 prescriptions annually.

Lack of drug coverage under Medicare cre-ates financial strain for many beneficiaries andtheir families, particularly for those withchronic conditions. A 2001 survey of agedbeneficiaries in eight states found that amongthose with three or more chronic conditionsand without drug coverage, approximatelyone-third reported not filling prescriptionsbecause of cost and 36 percent skipped dosesto make their medication last longer. Thepercentage of those with drug coverage andthree or more chronic conditions who also

Chart 5

Average Annual Number of Physician Services, by number of chronic conditions, 1999

5+43210

1.3 2.04.0

7.85.2

11.3

6.5

14.9

Number of Chronic Conditions

Unique Physicians

Physician Visits

8.1

19.5

13.8

37.1

Source: 1999 SAF. Partnership for Solutions, Oct. 19, 2001.Note: Data for Aged 65+ beneficiaries


reported not filling prescriptions and skip-ping dosages due to cost is also substantial(17 percent and 19 percent, respectively).The survey also found that among those withcongestive heart failure, diabetes, or hyper-tension, over 25 percent did not fill at leastone prescription in the past year because ofcost (compared to 12–14 percent of thosewith coverage), more than one-tenth did notfill three or more prescriptions (compared to4–6 percent of those with coverage), andapproximately one-third skipped dosages tomake prescriptions last longer (as did 14–17percent of those with coverage) (Safran, etal., 2002).

Most beneficiaries (73 percent) have prescrip-tion drug coverage from sources other thanMedicare (Kaiser Family Foundation, 2001).(See page 22 for discussion of supplementalinsurance.) Nevertheless, many of those withprescription drug coverage find that it is notcomprehensive.

■ Only three of the ten standard Medigapbenefit packages (Plans H, I, J) coverprescription drugs.11 In part becausepremiums for these policies are expen-sive, in 2000, only nine percent of thosewith Medigap had one of these policies(Super, 2002). Plans H and I pay 50percent of drug charges up to $1,250with a $250 deductible; Plan J pays 50

16

Chart 6

Medicare Beneficiaries’ Average Annual Number of Prescriptions (including refills), by number of chronic conditions

5+43210

3.7

10.4

17.9

24.1

Number of Chronic Conditions

33.3

49.2

Num

ber

of P

resc

ript

ions

Source: MEPS, 1996 (Berenson and Horvath, 2002).

11 To facilitate comparison shopping, OBRA 1990 required that all Medigap policies sold after 1992 conform toone of ten uniform benefit packages.The ten Medigap options (labeled A, B, C,…J) cover a core set of servicesand the benefits generally increase in comprehensiveness from A through J.


percent of charges up to $3,000 peryear, also with a $250 deductible. Forthose with high prescription costs, thesemaximum limits will not cover theircosts. Also, these plans are often notoffered to disabled beneficiaries.

■ M+C as a source of drug coverage iseroding. The percentage of M+Cenrollees with prescription drug cover-age fell from 84 percent in 1999 to 68percent in 2003. Plans are simultane-ously decreasing the amount of covereddrug spending while increasing benefi-ciaries’ copayments. Most plans (86 per-cent) limited drug benefits in 2000 andan increasing number of plans set annu-al benefit limits at $500 or less (21 per-cent of plans in 1999 and 32 percent ofplans in 2000). Only 18 percent ofplans in 2000 offered drug coverageabove a $2,000 level (CRS, 2002).

■ Nearly all (99%) beneficiaries withemployer-sponsored health insurancehave prescription drug coverage. How-ever, firms are shifting prescription drugcosts onto retirees. Among firms with200 or more workers offering retireehealth benefits, 32 percent of thesefirms increased prescription drug costsharing requirements for retirees from1999 to 2001 (Kaiser Family Found-ation, et al., 2002).

■ All beneficiaries with full Medicaid cov-erage receive prescription drug benefits.Medicaid plays a critical role in provid-ing drug coverage to low-income bene-ficiaries but does not close the coveragegap entirely. Though Medicaid drugcoverage is relatively generous, not allof those who are eligible are enrolled.Also, increased drug costs and strainedstate budgets have prompted states to

control Medicaid prescription drug uti-lization and spending. Many states arenow placing limits on the number ofconcurrent prescriptions, the amount ofdrug supplied at one time, or the num-ber of refills permitted. States may alsocharge minimal copayments of $.50 to$3.00, which may be burdensome forMedicaid beneficiaries.

■ At least 34 states have established orauthorized some type of program toprovide pharmaceutical coverage orassistance, primarily to low-incomeelderly or persons with disabilities whodo not qualify for Medicaid. Most pro-grams subsidize a portion of the costs,but others use discounts or bulk pur-chasing approaches (NCSL, 2002). Aswith Medicaid coverage, not all of thosewho are eligible are enrolled, and mostof these state programs do not providecomplete coverage.

■ CMS issued the final regulation for theMedicare-Endorsed Prescription DrugInitiative on September 4, 2002. Underthis initiative, CMS will endorse cardsponsors that secure rebates or dis-counts from drug manufactures onbrand name and/or generic drugs.Beneficiaries can enroll in one Medi-care-endorsed card program at a time.According to administration officials,this initiative is expected to yield savingsthat will be shared with enrollees, eitherdirectly or indirectly through pharmacydiscounts or pharmacy services.

Though the above sources of prescriptiondrug coverage may increase accessibility tobeneficiaries, most do not protect fully andadequately against out-of-pocket costs. Also,the future of most of these coverage sourcesis not secure.

N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e18

C. FUNCTION AND QUALITY OF LIFE

Many beneficiaries with chronic conditionshave function and quality of life needs thatare not covered by Medicare. For many ben-eficiaries, function and quality of life consid-erations are as important to them as theirmedical needs.

The case study below describes the experi-ence of a relatively healthy woman who facedproblems walking. The medical system didnot address her functional issues before shebroke her leg. Though it did ensure that herleg healed properly, it did not appropriatelyaddress her long-term functional concernsand issues.

Sensory aids

Medicare considers sensory loss a functionaland quality of life issue—and not a medicalconcern. Yet, such loss may prompt a declinein medical status. Eyeglasses and hearing aidbenefits, for example, are specifically exclud-ed by statute (eyeglasses are generally cov-

ered only after cataract surgery). However,eye care, including glasses, can slow visualloss and reduce the incidence of disablingfalls. Falls are often the cause of a broken hip,requiring hip repair or replacement surgery,rehabilitation, and a need for long-term per-sonal care. Similarly, hearing loss is a majorcause of isolation and a risk factor for depres-sion (Cassel, et al., 1999).

Rehabilitative services

Unlike expectations for acute conditions,functional (and clinical) improvement is notexpected for many chronic conditions.Instead, the goal may be to maintain a levelof function, or in other instances, slow thedeterioration of function. For many benefi-ciaries with chronic conditions, rehabilitativeservices are a means of achieving this goal.Coverage for outpatient occupational thera-py, physical therapy, and speech therapy,however, is sometimes denied by theMedicare carriers unless improvement orrestoration of function is expected.

Case StudyA 68-year-old woman had a number of falls over the past few years, the most serious of which result-ed in a broken tooth. One day she twisted her foot as she stepped on the bottom stair of her homeand broke five bones in her leg. After five days in the hospital recovering from surgery, she was dis-charged to her home, with arrangements made for daily visits by a physical therapy aide. No oneaddressed why she fell and what she could do to keep from falling again.

A year after her surgery, she felt that her balance was more precarious than before her fall. She hadbecome fearful of walking and began limiting her outside activities. She made an appointment withher orthopedist to relay her concerns. He said that the leg had healed fine and no further treatmentwas warranted. After telling her internist of her concerns, he referred her to a neurologist to rule out abrain tumor. The neurologist said that mild neuropathy could possibly be the cause of her falls. Heprescribed medication for the neuropathy but offered no other suggestions.

Her daughter insisted that there must be something that could be done to prevent further falls. She hadheard the term “balance therapy” and told her mother to discuss it with her orthopedist. During hernext office visit, the orthopedist claimed he had never heard of it. He also said he doubted the neurol-ogist’s neuropathy diagnosis.

Mother of chronic care panel staff person


According to the Medicare Part A Manualfor intermediaries (for beneficiaries receivingpost-hospital coverage at a skilled nursingfacility or by a home health agency),Medicare covers skilled therapy to preventdeterioration or maintain capabilities.However, for rehabilitation services coveredunder Part B, the Medicare Part B Manualfor carriers states that coverage is discontin-ued when a beneficiary reaches his/her fullrehabilitation potential. It recognizes, howev-er, that a therapeutic maintenance programto delay or minimize muscular and functionaldeterioration in patients suffering from achronic condition may be considered reason-able and necessary. Yet Medicare will onlyreimburse for a minimal number of therapeu-tic sessions to instruct the beneficiary or fam-ily members in carrying out the maintenanceprogram. It does not cover assistance inimplementing the plan.

Durable medical equipment

Durable medical equipment (DME) is anoth-er instance where Medicare policies fall shortof helping beneficiaries to maximize functionand quality of life. DME coverage policy

requires that the equipment be used primari-ly in the home. This policy is an obstacle,particularly for younger, disabled beneficiarieswho would like to try to work outside thehome, as well as those would prefer to lessentheir dependence on others by doing theirown shopping and other errands. For exam-ple, beneficiaries who have trouble walkingmay need the use of a motorized scooteroutside of the home but have difficulty get-ting it approved because of the “primarily inthe home” requirement.

Home health

Under Medicare statute, home health bene-fits require that a beneficiary be “confined tothe home,” a condition characterized by theword “homebound.” The basis of thisrequirement is that if beneficiaries are able toleave their home to obtain care in an outpa-tient facility, they do not require home care.Many beneficiaries who need this benefit,however, are able to leave the home occa-sionally or with substantial assistance (e.g.,with a motorized wheelchair or carried to thecar by family members).

Case StudyA 78-year-old man with mild dementia was hospitalized with pneumonia. After he was admitted to anursing home, he received physical therapy under Part A of Medicare. After his Part A benefits ran out,his son hoped that the physical therapy would continue under Part B, but the nursing home conclud-ed that his condition would not improve. The physical therapist wrote up a maintenance plan for him(reimbursed by Medicare) but no assistance was provided to help implement the plan. He was re-hospitalized that year for pneumonia and then again released to the nursing home.

After four hospitalizations, the son requested ongoing physical therapy services. He claimed hisfather’s hospitalizations were the direct result of lack of activity. The father’s physician concurred. The physical therapist, however, was hesitant to reinstate services because she did not believe that the Medicare carrier would approve coverage. She told the son that though ongoing physical therapywould prevent future hospitalizations, it could not be expected to improve his medical condition—a requirement for rehabilitation coverage under Part B.

Son of Medicare beneficiary


The statute did not initially specify what theterm “homebound” meant; however, overthe years, Congress has included a definitionof “homebound” in the statute and expand-ed it, though the definition is not clear-cut.Congress enacted two exceptions to thehomebound rule in 2000: it allowed forthose who leave home to go to church or to

attend adult day care centers.12 These excep-tions affected very few beneficiaries.

In cases where the statute is not specific,CMS provides guidance to carriers throughits manuals. In July 2002, CMS altered itshome health agency manual to be more sen-sitive to cases such as the David Jayne case(see case study on page 21). The manualexpands the list of examples used to illustrateoccasional absences from the home and notesthat the list is still not all-inclusive. It addsALS and other neurodegenerative disabilitiesto examples of conditions that may indicate abeneficiary is homebound. It suggests that“homebound” determinations be made overa period of time, not on a daily or weeklybasis, so a few special trips outside the homewould not be used to disqualify a beneficiarywho has a normal inability to leave the home.

Because there is no clear-cut policy for“homebound” in both statute and CMSmanuals, carriers have considerable discretionin defining it. Defining “homebound” bringsinto question: for how long could a benefi-ciary leave home, how often could they leave,and for what purpose could they leave andstill qualify as “homebound?” Thus, becausethe policy is unclear and because the carriersvary in its enforcement, beneficiaries arefaced with the dilemma of remaining athome vs. leaving home and risking loss oftheir home health care.

In addition to meeting the “homebound”requirement, beneficiaries with chronic con-ditions face other barriers to home healthnursing coverage. To qualify for home healthnursing services, a beneficiary must need“skilled nursing care… on an intermittent

20

The current statute describing the “home-bound” definition reads as follows:

For purposes of paragraph (2)(C)[home healthagency certification], an individual shall beconsidered to be “confined to his home” if theindividual has a condition, due to an illness orinjury, that restricts the ability of the individualto leave his or her home except with the assis-tance of another individual or the aid of a sup-portive device (such as crutches, a cane, awheelchair, or a walker), or if the individual hasa condition such that leaving his or her home ismedically contraindicated. While an individualdoes not have to be bedridden to be considered“confined to his home,” the condition of theindividual should be such that there exists anormal inability to leave the home, that leavinghome requires a considerable and taxing effortby the individual. Any absence of an individualfrom the home attributable to the need toreceive health care treatment, including regularabsences for the purpose of participating in thetherapeutic, psychosocial or medical treatmentin an adult day-care program that is licensed orcertified by a State, or accredited, to furnishadult day-care services in the State shall not dis-qualify an individual from being considered tobe “confined to his home.” Any other absenceof an individual from the home shall not so dis-qualify an individual if the absence is of infre-quent or of relatively short duration. Forpurposes of the preceding sentence, anyabsence for the purpose of attending a religiousservice shall be deemed to be an absences ofinfrequent or short duration.”

Sections 1814(a) and 1835(a) Medicare statute

12 BIPA, Section 507.

basis,” along with being homebound.(Qualifying criteria for receiving physical andspeech therapy in the home include meetingthe homebound definition but does notrequire the need for skilled nursing care orcare on an intermittent basis.) The “skilled”nursing care requirement excludes those whoneed only personal care. The “intermittent”requirement conflicts with a basic characteris-tic of chronic conditions: that the conditionis ongoing. Thus, the home health benefitdoes not meet the needs of those withchronic conditions, particularly younger dis-abled beneficiaries who could leave the homewith the support of home health nursing services.

D. SELF-MANAGEMENT

The goals of clinical self-management pro-grams are increased patient knowledge, better adherence to the regimen, timelyadaptation of the regimen to change in dis-ease status, improved disease control, and,ultimately, reduced complications. Beneficia-ries with diabetes, for example, may need tomonitor their own glucose levels, regulatetheir food intake, adjust their insulin dosage,and self-inject insulin. This takes considerabletraining, knowledge, and discipline on thepart of the beneficiary.

Comprehensive self-management, however,includes more than clinical self-management.A more complete picture of the physical, psy-chological, functional, and quality of life tasksthat beneficiaries with chronic conditionsmust deal with include:

■ engaging in activities that promotehealth and build physiological reserve,such as exercise, proper nutrition, smok-ing cessation, social activities, and sleep;

■ interacting with health care providersand systems and adhering to recom-mended treatment protocols;

■ monitoring their own physical and emo-tional status and making appropriatemanagement decisions on the basis ofsymptoms and signs;

■ managing the impact of the illness ontheir ability to function in importantroles, on emotions and self-esteem, andon relations with others. (Wagner, et al.,1996)

Despite the importance of self-managementfor chronic conditions, many beneficiariesreceive little or no self-management support.This may be due, in part, to providers’ lackof training and tools in self-management orbecause they have little time to spend on it.


Case StudyA 41-year-old man with amyotrophic lateral sclerosis (ALS, also known as Lou Gehrig’s disease) isdependent on home health care. His degenerative neuromuscular condition had caused his musclesto atrophy, leaving him unable to eat, breathe, or move on his own. He is attached to a life supportsystem and a computerized voice simulator. His life depends on a ventilator, intravenous feeding, anddaily care from a home health aide. He leads an active life by leaving his home in his motorizedscooter. He is also an advocate for changing the homebound rule. His home health agency discontin-ued its services after he attended a football game. The agency reinstated its services after members ofCongress became involved and the beneficiary agreed to pay his home health provider for any claimdenied by Medicare.

David Jayne, disability advocate

Health plans are more likely than privatepractice providers to provide support for such activities—including group classes,meetings with nurse educators, and com-puter programs to monitor and support self-management. Some health plans contract the self-management support out to diseasemanagement firms.

Though appropriate self-management isimportant for all chronic conditions, it mayrequire major involvement of caregivers inthe case of dementia and those with othersevere cognitive limitations. Such cases mayrequire training for the caregiver, more inten-sive care from a physician, or home and community-based care.

E. FAMILY PARTICIPATION

Many beneficiaries depend on their familiesto coordinate their medical care and to pro-vide assistance with activities of daily living.Family members are often the majorproviders of advice and guidance to benefi-ciaries about choice of health plans, ofproviders, and of treatment. They also dealwith the “blizzard of paper frequently associ-ated with chronic illness and high service uti-lization.” They find, however, that Medicareand its providers do not recognize the crucialrole that the family plays in beneficiaries’ careand that Medicare and its contractors makelittle effort to inform or educate them(Vladeck, 2002).

Caregivers involved in their family member’scare often have a difficult time engaging thebeneficiary’s physicians. While the beneficiary

is the “customer” in this setting, familymembers can often be extremely helpful toboth beneficiaries and physicians. When theyaccompany a beneficiary to a medicalappointment, the physician may spend insuf-ficient time answering their questions andaddressing their concerns. Caregivers whowould benefit from discussing the beneficia-ry’s diagnosis and treatment plan with thephysician without the beneficiary present maynot given the opportunity to do so.

F. SUPPLEMENTAL COVERAGE

Medicare does not cover a substantial shareof beneficiaries’ health care spending.Uncovered costs include Part B premiums,deductibles, and coinsurance, as well as forservices that Medicare does not cover.Medicare also does not have an annual orlifetime cap that limits beneficiaries’ out-of-pocket spending beyond a certain level, evenif the expenses are for covered services. Itdoes not, for example, pay for hospital staysbeyond 90 days per benefit period and skillednursing facilities beyond 100 days.13 There isalso no limit on coinsurance. Three percentof beneficiaries are projected to exceed$5,000 in total cost sharing for covered ser-vices in 2000 (MedPAC, 2002b). Approxi-mately 6,000 beneficiaries exhaust theirlifetime reserve inpatient hospital days eachyear (Vladeck, 2002).

Those with chronic conditions have higherout-of-pocket expenditures than those withno chronic condition. As shown in Table 2(see page 23), for those with no supplementalcoverage in 1996, mean annual out-of-pocket


13 Benefit period begins the day the beneficiary is admitted to a hospital or skilled nursing facility and ends whenthe beneficiary has not received hospital or skilled nursing care for 60 days in a row. Medicare may help pay foran additional 60 inpatient hospital days through a beneficiaries’ lifetime reserve of 60 days. Reserve days may beused only once in a lifetime.

expenditures for beneficiaries with three ormore chronic conditions was $1,492 com-pared to $455 for those with no chronicconditions. Those with private supplementalinsurance face similarly large out-of-pocketexpenditures ($1,394 for those with three ormore chronic conditions compared to $484for those with no chronic conditions). Thosewith both Medicare and Medicaid have sub-stantially lower out-of-pocket costs, thoughthese expenditures may be even more signifi-cant to low-income beneficiaries.

Supplemental coverage helps close Medi-care’s gaps. In 1999, 87 percent of benefici-aries with chronic conditions had somesupplemental coverage. Such coverageoptions include individually purchasedMedigap, employer-sponsored insurance,Medicaid, and M+C. Coverage options, how-ever, are not available or affordable to all.Disabled beneficiaries were less likely thanthe aged to have supplemental coverage: 28percent of the disabled lacked supplementalcoverage, while 10 percent of the aged lackedsuch coverage (Kaiser Family Foundation,

2001). (See page 15 for a discussion of pre-scription drug coverage.)

■ Employer-sponsored health insurance(most of which is retiree insurance) cov-ers 33 percent of beneficiaries. Thesebenefit packages are almost always moregenerous than Medigap polices andM+C plans and almost all have an annu-al cap. The percentage of firms with 200or more workers offering retiree healthbenefits, however, fell from 41 percentto 34 percent from 1999 to 2001. Thegenerosity of such coverage also hasdeclined and many employers are con-sidering increases in premiums or costsharing for retirees in the future (KaiserFamily Foundation, et al., 2002).

■ Individual Medigap policies are pur-chased by 24 percent of beneficiaries(Tapay and Smolka, 1999). Medigapoffers protection against the cost ofdeductibles and copayments forMedicare-covered services. Most poli-cies, however, offer no assistance indefraying the cost of services thatMedicare does not cover. Nationally, the


Table 2

Mean Annual Out-of-Pocket Spending Per Person Over Age 65, by coverage status and number of chronic conditions, 1996

Source: Hwang, W., et al., 2001. Note: Out-of pocket expenditures as reported in MEPS represent self-reported payments for coinsurance

and deductibles, as well as cash outlays for services, supplies, and other items not covered byMedicare. Medicare and other health insurance premiums were not included in the analysis.

Number of chronic conditions

Coverage status All None 1 2 3+

Medicare only $924 $455 $643 $966 $1,492

Medicare/private $910 $484 $636 $875 $1,394

Medicare/Medicaid $434 $262 $247 $447 $649

average annual 1999 Medigap premiumwas $1,311. From 1999 to 2000, pre-miums for Medigap plans without pre-scription drug coverage rose by 4 to 10percent. Premiums for the three Medi-gap plans offering prescription drugcoverage increased even more rapidly—by 17 to 34 percent (GAO, 2002).

Aged beneficiaries are granted a six-month open enrollment period begin-ning on their first day of Medicareenrollment. During this period, aninsurance company cannot denyMedigap coverage, place conditions ona policy, or charge higher premiumsbecause of pre-existing medical condi-tions. Those who do not buy aMedigap policy during the open enroll-ment period lose these protections.

Disabled beneficiaries have limitedaccess to Medigap coverage. They donot have the federal Medigap protec-tions that the aged do, including a man-dated open enrollment period. Instead,Medigap protections must be developedand implemented at the state level.Only 19 states mandate an open enroll-ment period and only two (NY andMO) limit premiums. Consequently,disabled beneficiaries account for onlyone percent of Medigap policyholders(Fox, et al., 2002).

■ M+C plans are available to about 59percent of Medicare beneficiaries in2003 (though less than 10 percent ofbeneficiaries in rural areas have access toat least one M+C plan) and about 14percent of all beneficiaries are enrolledin a M+C plan (CRS, 2002). However,plan withdrawals from the M+C pro-gram have caused some enrollees tochoose new M+C plans or return to the

original Medicare program. After therapid growth rate of Medicare managedcare enrollment in the 1990s, enroll-ment reached a peak in 1999, withalmost 17 percent of beneficiariesenrolled, and has declined since then.Over this same period, mean premiumand cost-sharing levels in M+C plansincreased, while coverage of prescriptiondrugs and other benefits not coveredunder original Medicare (e.g., hearingservices and preventive dental benefits)were reduced (Achman and Gold,2002).

■ Medicaid is available to those with lowincomes, with 17 percent of all benefi-ciaries dually eligible for both Medicareand Medicaid. Most (83 percent) ofdual eligibles receive full Medicaid ben-efits, including comprehensive prescrip-tion drug coverage and full payment ofMedicare premiums and cost sharingrequirements. The rest (17 percent)—generally those with slightly higherincomes—receive assistance with premiums and/or copayments anddeductibles through Medicare SavingsPrograms (Kaiser Family Foundation,2001). Despite efforts to increaseenrollment in these programs, only 47 percent of those who meet the eligibility requirements are enrolled.Low enrollment may be due to lack ofawareness of the programs, confusionabout eligibility requirements, the timeand effort costs of enrolling, and a“welfare stigma” associated withMedicaid (DHHS, 1999).

The more chronic conditions a person has,the more likely they are to have supplementalcoverage. While 82 percent of beneficiarieswithout chronic conditions have supplemen-


tal coverage, the percentage rises to 90 per-cent for those with five or more chronic con-ditions. This appears to be explained bybeneficiaries with chronic conditions’ higherrates of Medicaid coverage: the more chronicconditions a beneficiary has, the more likelythey are to have Medicaid (10 percent ofbeneficiaries without a chronic conditionhave Medicaid; 21 percent of those with fiveor more chronic conditions have Medicaid.)

Beneficiaries who cannot afford out-of-pocket expenditures may rely on relatives orfriends to help them pay for health care.Those who cannot rely on relatives or friendsmay delay care or forgo other necessities.Beneficiaries most at risk of delaying or goingwithout care are those in poor health, thosewho are disabled, the poor, and those with-out supplemental insurance. A recentMedPAC analysis found that 17 percent ofbeneficiaries in poor health delayed care dueto cost, as did 18 percent of disabled benefi-ciaries, 10 percent of those within the 125 to

200 percent of the federal poverty standard,and 20 percent of those without supplemen-tal insurance. It concluded that though thosewith incomes just above the poverty level aremost at risk, substantial proportions of bene-ficiaries with higher incomes, including thosewith supplemental coverage, also appear tobe at risk from high health care costs(MedPAC, 2002b).

Medicare’s lack of support for persons withchronic conditions may be more of a prob-lem for lower and middle-income beneficia-ries than for the wealthy, but the care andreimbursement systems pose difficulties foralmost all beneficiaries. Both the wealthy andthe poor may find that their providers do not coordinate their care with their otherproviders, their providers devote little atten-tion to preventive measures to maximizehealth, function, and quality of life, and they receive insufficient self-managementinstruction.


In the panel’s view, Medicare has failed toadequately support providers in their treat-ment and management of chronic condi-tions. Its fee-for-service reimbursementsystem does not reimburse for many of theservices and tools important for the care ofbeneficiaries with chronic conditions, nordoes it offer providers the flexibility to utilizenew and efficient methods of operation. Inthis and other respects, Medicare reflects thebroader health care system in which it exists.

A. MEDICARE’S SIMILARITY TO THEGENERAL HEALTH CARE SYSTEM

Medicare’s benefit and reimbursement struc-ture generally mirrors that of the U.S. healthcare system. Major characteristics of the gen-eral U.S. health care system are:

■ focus on acute care and curative treat-ments;

■ limited implementation of promisingtools and techniques for managing care(e.g., evidence-based protocols, healthassessments, telephone follow-up care,self-management support);

■ provider reimbursement systems that donot support chronic care.

Such characteristics are perpetuated throughmedical education and training, making theacute care construct of the U.S. health caresystem difficult to change. However, because

Medicare is the largest health care payer inthis country and because its beneficiaries are more likely to have chronic conditions,Medicare’s efforts to improve chronic care toits beneficiaries should also improve the careprovided by the general health care system.

B. LEGAL AND ADMINISTRATIVECONSTRAINTS

Barriers to chronic care may be even morepronounced for the Medicare populationthan for the general health care system. First,the needs of the Medicare population aregreater than that of the general population.Medicare beneficiaries are over twice as likelyto have a chronic condition than the non-Medicare population. They are also approxi-mately three times more likely to have afunctional limitation (Partnership for Solu-tions, 2001). Second, under law, CMS is lim-ited in its authority over providers, as it is notpermitted to “exercise any control over thepractice of medicine or the manner in whichmedical services are provided.” While healthplans can impose requirements on theirproviders, CMS is expected to function pri-marily as a claims payer and is unable to usebasic care management tools to try toimprove the health care delivery system(Berenson, 2002).

The basic structure of Medicare coveragemakes Medicare’s policies on chronic care


Chapter 3:The Medicare Program Faces — As Well As Poses forProviders — Considerable Barriers to Chronic Care14

14 Much of this chapter is based on two papers written for the study panel:“Confronting the Barriers to ChronicCare Management in Medicare,” by Robert Berenson, and “Round Pegs and Square Holes: Medicare andChronic Care,” by Bruce Vladeck.

resistant to change. Medicare’s statutes didnot envision ongoing treatment of chronicconditions when they were written in 1965.Despite many changes in our society andmedical system—in the demographic charac-teristics of beneficiaries, medical technology,and care delivery—some statutes that impedeoptimal chronic care services remainunchanged. One such example is coveragefor inpatient services that is still tied to theconcept of “spell of illness.” This conceptlimits inpatient care to 90 days and skillednursing facility care to 100 days per spell ofillness (Section 1861(a)). It ignores the factthat many chronic conditions are ongoingand may last for the remainder of the benefi-ciary’s lifetime.

Medicare’s statutes do not support coveragefor many services vital to those with chronicconditions. Medicare payments are prohibit-ed for services that “are not reasonable andnecessary for the diagnosis or treatment ofillness or injury or to improve the function-ing of a malformed body member….”(Section 1862(a)). A strict standard of “med-ical necessity” is often the rationale for denialof coverage for services and equipment thatare important to beneficiaries with chronicconditions. This statute also prohibits pay-ment for “personal comfort items…eyeexaminations…hearing aids…custodialcare…and dental services.” Also, nursing ser-vices in the home are covered by Medicareonly when a beneficiary is “homebound,”and in need of “skilled” services on an“intermittent basis.”

As stated on page 14, preventive health bene-fits were excluded from coverage in the origi-nal Medicare statute. Over time, Congresshas added a number of preventive healthbenefits to Medicare. Congress’ decisions

about which preventive health services toinclude, however, have been ad hoc, and theresulting set of benefits does not reflect theconsensus of experts in the field of preven-tion and health promotion, namely those rec-ommendations of the U.S. PreventiveServices Task Force (USPSTF). For example,some services that received the endorsementby the USPSTF for the over-age-65 popula-tion remain uncovered (e.g., blood pressurescreening, cholesterol measurement, hearingimpairment screening), while others thatwere not recommended were added to the benefits package (e.g., prostate cancerscreening by PSA and digital rectal exam)(MedPAC, 2002b).

Coverage decisions are based on statute butrequire interpretation by CMS and its con-tractors. CMS may issue national coveragedecisions, memoranda, or specify its inten-tions in manuals for intermediaries or carri-ers. National coverage decisions grant, limit,or exclude Medicare coverage for a specificmedical service, procedure, or device and arebinding on all contractors. In general,national coverage decisions are made fordecisions with broader policy and paymentimplications and where there is a sufficientbasis on which to make informed decisions.National coverage decisions are rare—fewerthan 400 have been issued since the incep-tion of the Medicare program. Thus, mostcoverage decisions are made by local carriersand fiscal intermediaries.

Four examples highlight the complexity ofMedicare coverage policies and how interme-diaries and carriers influence benefitapprovals and denials:

■ The line between covered “treatmentservices” and uncovered services thatinvolve “personal comfort” or “custodi-


al care” is not clear-cut. For example,special mattresses or wheelchair seatsdesigned to reduce the risk of decubitusulcers can be an important preventivemeasure for older patients with chronicconditions. Medicare carriers may bewilling to reimburse for such items onlyafter beneficiaries have developed actualsymptoms (Vladeck, 2002).

■ Medicare’s requirement that beneficia-ries be “homebound,” to receive homehealth services is subject to interpreta-tion. Carriers differ in their definitionsof the statute, including the phrases,“normal inability to leave the home,that leaving the home requires a consid-erable and taxing effort,” and thatabsences must be “of infrequent or ofrelatively short duration.” Advocatesbelieve that some carriers are inappro-priately denying home health services tobeneficiaries who would clearly benefitfrom them because of a strict interpreta-tion of “homebound.”

■ Medicare’s prohibition against payingfor services that “are not reasonable and necessary” leads to denials for out-patient rehabilitation in those caseswhere a beneficiary is not expected toimprove. Beneficiaries with chronic con-ditions who receive occupational, physi-cal, or speech therapy to maintainfunction or limit a decline in functionare denied such services because the car-riers deem them to be unreasonable ormedically unnecessary based on this“improvement” measuring stick.

■ Until recently, some intermediaries andcarriers restricted or denied rehabilita-tion and mental health services topatients, in part, because the beneficia-ries’ primary diagnostic was Alzheimer’s

disease or another form of dementia.Here, “treatment of illness” was inter-preted to mean active treatment of theunderlying disease, rather than interven-tion in the functional consequences ofdisease. Such patients were denied cov-erage for rehabilitation services on theassumption that a person with dementiacannot “learn,” even though researchand practice have demonstrated the suc-cess of appropriate therapies in rehabili-tation and in the prevention offunctional decline (Fried, 2000). (InSeptember 2001, to correct this and toclarify policy, CMS issued a memo to allof its intermediaries and carriers whichstated that contractors may not installedits that result in the automatic denialof services based solely on the ICD-9-CM codes for dementia.)

Lack of clarity regarding coverage decisionsmay influence providers to withhold provi-sion of the service. Providers may tell thebeneficiary that Medicare will not cover theservice, or require the beneficiary to sign astatement that if Medicare does not agree tocover the service, the beneficiary will pay thecharges. This often occurs when the providerbelieves that the beneficiary will not or can-not pay if Medicare does not. It may alsoinfluence beneficiaries to go without carethey believe may be useful to their condition.

C. ORIGINAL MEDICARE’S FEE-FOR-SERVICE REIMBURSEMENT SYSTEM

Original Medicare’s fee-for-service reim-bursement system is not designed to pro-mote effective chronic care. Reimbursementis based on payment to individual providers,with particular emphasis on physicians. Ingeneral, it pays for services provided by non-


physicians (other than those specificallyauthorized) only when they are billed by thephysician as “incident to” the physician’s ser-vices (the provider must be a direct employeeof the physician and the service must be pro-vided under the physician’s direct personalsupervision, which usually requires the physi-cal presence of the physician). This require-ment discourages a team approach to careand delegation of major aspects of the proto-col to nurses, pharmacists, and other non-physician colleagues. In addition, fee-for-service reimbursement pays providers for dis-crete services, such as procedures and officevisits. It does not, however, pay for othermeans of communication that may be con-ducive to care, including telephone and emailencounters. From a budgetary perspective,there is concern that if original Medicarereimbursed for such services, then utilizationand expenditures, as well as providers’administrative costs, would increase dramati-cally. Payment for such services would alsomake fraud and abuse detection more com-plex because it would be difficult to trackthese encounters for billing purposes.

As an “any willing provider” program, allclinicians with an appropriate license are eligi-ble to participate in Medicare. Thus, exceptin demonstrations, CMS cannot pay more toreward high-quality providers or reducereimbursement for low quality of care. Also,if reimbursement for a service is offered, itmust be payable to all participating physiciantypes, whether or not they have modifiedtheir practice styles. In a CMS chronic careinitiative to provide reimbursement for carecoordination activities, CMS staff was sur-prised to find that podiatrists billed for thisservice more often than any other physiciantype. CMS, however, was powerless to target

this reimbursement to relevant physiciantypes.

Original Medicare’s fee-for-service systemdoes not foster coordination and high-qualitycare across various levels of providers. Forexample, a beneficiary who receives a hipreplacement often under goes surgery in thehospital, is transferred to a skilled nursingfacility for rehabilitative care, and then con-tinues to receive physical therapy services athome. Instead of paying for discrete services,CMS’ Medicare Partnership for QualityServices Demonstration (previously called theCenters of Excellence) has bundled paymentfor services to high-volume providers whowork together to provide high-quality care.The demonstration has targeted invasive pro-cedures, namely selected cardiovascularsurgery and hip and knee replacement, buthas not included care for chronic conditions.

Medicare’s reimbursement for primary caredoes not support quality chronic and geri-atric care. Reimbursement for routine prima-ry care visits is insufficient for the care ofmany with chronic conditions, as care for thispopulation usually takes a considerableamount of time, particularly when self-management and multiple conditions areaddressed. Older patients tend to requiremore time with their providers than youngerpatients, and disabled patients more timethan non-disabled patients. Reimbursementrates also offer little incentive for primarycare practitioners to coordinate specialty care.In addition, Medicare does not reimburse fortime spent with family members unless morethan half of the visit is dedicated to a coordi-nation-related activity, and it does not com-pensate for time spent with a family memberwithout the beneficiary present.



Though Medicare’s resource-based relativevalue scale (RBRVS)–based Physician FeeSchedule has somewhat alleviated the reim-bursement gap between primary and specialtycare, its coding and classification of theEvaluation & Management (E&M) servicesis a barrier to chronic care. E&M codesaccount for almost half of Medicare-paidphysician services. These codes fail to ade-quately reflect the additional complexity andtime requirements associated with care formany beneficiaries with chronic conditions.Though codes and additional payments wereadded for prolonged services with physician-to-patient contact, few physicians actually billfor them. In addition, its bundling of Medi-care payments for care management servicesinto other services disproportionately andunfairly impacts practitioners who providecare to elderly patients with multiple chronicconditions (payment for care plan oversight isavailable only to providers of home healthagency and hospice patients) (Health PolicyAlternatives, 2002).

D. MEDICARE+CHOICEOPPORTUNITITES TO IMPROVECHRONIC CARE

The M+C program created opportunities toimprove beneficiaries’ care. Though Medi-care had a risk contracting program that wasexpanded in 1982 under TEFRA, M+C’senactment in 1997 came at a time whentechniques for managing care were moreavailable.15 Such techniques include assign-ment of primary care providers, assignmentof care managers to high-cost patients, tele-phone advice lines, wellness support pro-grams (e.g., patient self-care education,preventive services), and disease manage-

ment. Health plans also updated their com-puter systems to generate patient reminderletters, physician profiles, and disease reg-istries. Many believed that prepaid, capitatedreimbursement provided incentives to con-trol costs, as well as to keep beneficiarieshealthy. It was hoped that as managed careevolved, its techniques would be incorporat-ed into original Medicare (Gluck andGinsburg, 1998, and Fox, et al., 1998).

Nevertheless, there was a lack of evidencethat M+C plans, in the aggregate, increasedor decreased quality when compared with theoriginal Medicare sector. One factor may bethat the Balanced Budget Act of 1997 con-strained payments to M+C plans during aperiod when health care costs were risingsharply: beginning in 1998, most M+C planswere limited to a two percent annual increasein capitation. The lack of impact is also likelythe result of other factors:

■ Few health plans made substantialchanges to their care systems. Thoughsome plans instituted targeted caremanagement programs, most plans con-centrated on controlling costs, mainlyby reducing inpatient utilization andnegotiating reimbursement rates.Managed care techniques to improvethe quality of chronic care were notwidely adopted.

■ Managed care techniques may be moreeffective in closed model systems (e.g.,group, staff model HMOs); instead, themajority (64 percent) of M+C enrolleesare enrolled in an independent practiceassociation (IPA) model (CRS, 2002).IPAs contract with physicians in solopractice or with associations of physi-

15 Tax Equity and Fiscal Responsibility Act of 1982.

cians. IPAs usually have little influenceover physicians, as most function as adiscounted fee-for-service third-partypayer, with its physicians operatingunder the constraints of fee-for-servicereimbursement and contracting withmultiple health plans.

■ Physician practice patterns are difficultto change.

■ Risk adjustment has been difficult toimplement. Though legislated byCongress, its implementation has beenpostponed and a phase-in plan put inplace. As a result, health plans did notmarket to beneficiaries with chronicconditions, and due to fear of adverseselection, have lacked incentive to beknown as leaders in chronic care.

■ Enrollees’ ability to switch plansprompted a lack of continuity. This dis-couraged plans from investing in ser-vices for which only long-term benefitswere expected.

Though “good” chronic care (as outlined onpage 11) should be easier to provide in well-organized, prospectively reimbursed healthcare organizations (e.g., health plans thathave defined populations, strong clinical cul-tures, comprehensive services, a preventiveorientation, data systems, and centralizedresources such as patient education), experi-ence suggests that regardless of organization-al and financial arrangements, patients withchronic conditions and their providers areconfronted with a set of formidable obstaclesto achieving quality clinical care and self-management (Wagner, et al., 1996).

E. GRADUATE MEDICAL EDUCATION

Medicare providers’ education and trainingcontributes to poor chronic care practices.Medicare provides substantial financial sup-

port for graduate medical education (GME).Its direct graduate medical education(DGME) funding pays for the direct costs oftraining of medical residents in teaching hos-pitals, while indirect medical education(IME) payments are adjustments to diagno-sis-related group (DRG) rates under theMedicare prospective payment system (PPS)to recognize the higher costs of teachinghospitals. (DGME expenditures were approx-imately $3 billion in 2001; IME expendi-tures, $5.2 billion.) At issue is whetherphysicians are being appropriately trained bythese hospitals to care for an aging popula-tion with chronic conditions.

Residency training takes place largely in hos-pitals. Because most chronic care is—andshould be—provided in outpatient facilities,medical residents complete their trainingwith insufficient experience managing chron-ic conditions. Their hospital experience alsoprovides them with few home care and nurs-ing home experiences. In a 2001 survey,physicians said that their chronic care trainingis inadequate (see Table 3 on page 33).

There is also a shortage of primary care andspecialty physicians trained in geriatrics.Providers who are trained in geriatric medi-cine are much better prepared to treat olderpatients than those with chronic conditions.Reasons for geriatric medicine training aretwofold: first, caring for older personsrequires a special body of knowledge and for-mal training; second, geriatric concepts offunction, family participation in care, andcare management are also core principles forchronic care. Almost all physicians (with thepossible exception of pediatricians, patholo-gists, and radiologists) should be trained ingeriatric principles. However, in actuality,most physicians have little education andtraining in the care of older people.


In the U.S., only three of 125 medicalschools have a full department of geriatrics.There are currently fewer than 600 facultymembers in the U.S. who list geriatrics astheir medical specialty (geriatrics is a subspe-cialty of internal medicine and family prac-tice). Most medical schools do not requirecourses in geriatrics. Out of 98,000 residentsin 1999, only 466 individuals chose a geri-atrics subspecialty (International LongevityCenter, 2000). Also, CMS, the primary fun-der of GME, had supported a two-year geri-atrics fellowship but has cut this support to aone-year fellowship. A one-year fellowship,however, does not provide the training need-ed for an academic position. The lack of aca-demic geriatricians is a problem becausethere are few geriatric educators to train geri-atricians and general practice physicians ingeriatric principles and techniques, and fewacademic researchers who specialize in geri-atric populations.

F. IMPROVED CARE SYSTEMS AND TECHNIQUES

A number of techniques have been devel-oped to help providers manage care. These

include protocols and guidelines, diseasemanagement, care and case management,and comprehensive geriatric assessment.Though all could enhance the quality ofchronic care, their implementation has beenhampered by lack of agreement as to how tobest provide them.

Lack of standardized guidelines is a barrier to effective chronic care. For some chronicconditions, there are no well-recognized evidence-based guidelines and few guidelinesexist for the management of two or morecomorbid chronic conditions. In cases whereevidence-based guidelines are available frommultiple sources, providers are not in agree-ment with which to use. Even when guide-lines are formally accepted by providerspecialty organizations, adherence to guide-lines has been limited and sporadic. Providerspracticing in small to mid-sized groups havebeen least likely to use evidence-based guide-lines; providers working in a health plan orgroup practice that support use of guidelinesand assist practitioners to comply with themseem to be most successful with their use(Wagner, et al., 1996).


Table 3

Percentage of Physicians Who Believe Their Chronic CareTraining to be Inadequate, 2001

Source: Partnership for Solutions, 2002.

Topic Percent

Coordination of in-home and community services for people with 66%chronic illness

Approaches to educating patients with chronic illness 66%

Management of psychological and social aspects of chronic illness care 64%

Nutrition in chronic illness 63%

Management of chronic pain 63%

Disease management (defined in this reportas the systematic effort to improve the man-agement of a condition) generally incorpo-rates the use of appropriate clinical guidelineswith treatment protocols and informationsystems specifically designed for a given disease. It may also include self-managementtraining and medication management. Thereare two forms of disease management: a con-tracted “carve-out” model and a primarycare–based model. In the carve-out model,care is provided through contracts with dis-ease-management companies. Such programsusually focus on a single condition and sepa-rate such care from primary care. Becausetheir goal is to save money for their clients,they target high cost patients for whom sav-ings can be achieved in the short-run. Incontrast, the primary care–based modelworks by improving the primary care systemto enhance outcomes. It also works to coor-dinate primary care with specialists whenappropriate. This type of disease managementbetter serves those with multiple chronicconditions and those whose illness has notyet incurred high expenditures (Boden-heimer, 1999). Evidence of quality improve-ments as a result of disease management isgrowing, though the evidence is mixed onwhether such efforts save money.

Though there is general agreement that carewithin original Medicare should be managed,there is less agreement as to how to manageit and who should manage it. As with diseasemanagement, there is no consistent definitionof case management. The term has a range ofmeanings, from strictly administrative func-tions (e.g., filling out insurance forms and

arranging transportation) to sophisticatedclinical management (e.g., medication adjust-ment). According to Wagner, growing evi-dence suggests that sophisticated clinical casemanagement can improve basic care for anumber of serious chronic conditions(Ferguson and Weinberger, 1998). There iscontinued debate, however, as to who shouldserve as beneficiaries’ case managers (optionsinclude primary care physicians, specialists,nurse case managers, social workers, careteams, and disease management companies).Evidence indicates that the general case man-ager who is not an integral part of the med-ical care team, has little if any impact onoutcomes.16 As stated earlier, the MedicareFee Schedule does not generally provide for aseparate payment for case management ser-vices, as such services are generally supposedto be incorporated into other payments.

Other programs to manage chronic care areimplemented under the name “care manage-ment.” Such programs tend to move beyondclinical care management and work toimprove function and quality of life. Forexample, the HMO Workgroup on CareManagement has designed targeted interven-tions to address common geriatric condi-tions, including physical inactivity, falls,medication-related complications, dementia,depression, undernutrition, and urinaryincontinence (HMO Workgroup on CareManagement, 2002).17 The health plans andprovider groups who have implementedthese interventions have the organizationalcapacity to offer such programs to theirmembers. Capitation offers them the flexibili-ty to provide services that are most appropri-


16 See page 42,“Medicare Case Management Demonstrations for High-Cost Medicare Beneficiaries” and Fergusenand Weinberger, 1998.

17 The HMO Workgroup on Care Management is a group representing staff, group, and IPA-model health plansand large group practices that are capitated by health plans for a significant portion of revenues.


ate to their members’ needs and incentives tokeep patients healthy, as well-person care canbe less expensive to provide than acute,urgent and inpatient care.

Geriatric assessment and management is par-ticularly relevant to the aged and those withchronic conditions. It aids in the diagnosis offunctional ability, physical health, cognitiveand mental health, and the socioenvironmen-tal situation of beneficiaries. Tools such ascomprehensive geriatric assessment (CGA)may assist providers in their diagnosis andthen allow interventions to slow physical,functional, and mental decline. There is nosingle accepted standard for CGA, however,as protocols vary widely in purpose, compre-hensiveness, staffing, organization, and struc-tural and functional components. Hetero-geneous results are to be expected and havebeen found among the various trials (Stuck,et al., 1993, and Cohen, et al., 2002). Thegeriatric community is now working to moreclosely link the assessment with managementand treatment. Though it has lobbied longand hard for coverage for CGA underMedicare, payment has not been approved.

Fragmentation of the Medicare and Medicaidsystems is a barrier to care. For those who aredually eligible for both Medicare andMedicaid, these distinct systems make it diffi-cult to integrate long-term care with the restof medical services. Each has different eligi-bility criteria, coverage standards, andprovider reimbursement systems. ThoughCMS oversees both Medicare and Medicaid,the federal government runs Medicare, whilethe states run Medicaid. Within CMS there isinsufficient integration of its Medicare andMedicaid divisions and no division that ischarged with administration of dual eligibles.A NASI case study of dual eligibles with

HIV/AIDS found this caused considerableconfusion and discontinuity of care on thepart of beneficiaries and providers (Eichner,2001). Such bureaucratic barriers reflect thepolitics of developing and implementing newservice models for dual eligibles. As dual eli-gibles comprise 17 percent of all Medicarebeneficiaries but 28 percent of Medicarespending, lack of integration of these twosystems is inefficient and does not promotehigh quality chronic care.

G. QUALITY INITIATIVES

Though CMS is technically prohibited bystatute to “exercise any control over the prac-tice of medicine or the manner in whichmedical services are provided,” CMS hasimplemented initiatives to improve the quali-ty of care of both M+C and original Medi-care. As stated earlier in this report, thequality of care received by beneficiaries isheavily influenced by the general (non-Medicare) health system. However, Medi-care’s potential influence in the quality sectorshould not be underestimated. MedPAC’s2002 quality improvement standard reportstated: “As the largest single health care pur-chaser, Medicare can effectively influence ifnot set the quality improvement agenda. Aswith coverage policy, private purchasers oftenfollow Medicare’s lead with respect to qualityrequirements.” Health care experts inter-viewed by MedPAC believe that the Medi-care program needs to use this leadershipposition to further stimulate quality improve-ment efforts (MedPAC, 2002a).

In the Balanced Budget Act of 1997,Congress imposed a set of rigorous qualityimprovement standards in its requirementsfor the M+C program. CMS’ quality regula-tion of M+C is now considered by many tobe as rigorous as the most well-developed


private accreditation standards (i.e., JCAHO,NCQA, URAC) (MedPAC, 2002a). Plansmust demonstrate their efforts on multiplesets of measures: two quality improvementprojects, the Medicare version of Health Plan Employer Data and Information Set(HEDIS), and the Medicare ConsumerAssessment of Health Plan Study (CAHPS)survey. These results are released to the pub-lic to help it evaluate and compare healthplans.

CMS relies less heavily on regulatory require-ments to stimulate quality improvement inoriginal Medicare than it does in the M+Cprogram. In order to ensure that a sufficientnumber of providers participate in Medicare,CMS’ quality efforts emphasize a broad net-work and voluntary efforts, though it doesmaintain some regulatory requirements. Itsbasic original Medicare quality initiativesinclude Conditions of Participation for PartA providers; support for Quality Improve-ment Organizations (QIOs, formerly calledPeer Review Organizations (PROs)); andrequirements for providers to conduct peri-odic patient appraisals through designatedassessment instruments. Also, CMS’ qualityof care reporting requirements for nursingfacilities are extensive and it has begun releas-ing individual nursing facility quality mea-surement data to the public. CMS is applyingperformance measures to dialysis facilities andis working to develop them for home healthagencies (MedPAC, 2002a).

Despite CMS’ efforts to foster quality mea-surement systems, measures are inconsistentacross care settings. Though beneficiariesreceive treatment across the spectrum ofhealth care settings, each setting has its ownCMS-mandated measures and informationsystems. For example, nursing homes utilize

the Minimum Data set (MDS); home healthagencies, the Outcome and AssessmentInformation Set (OASIS); and inpatient reha-bilitation facilities, the Inpatient Rehabilita-tion Facility-Patient Assessment Instrument(IRF-PAI). Thus, it is difficult to track bene-ficiaries’ health, function, and quality of lifeacross settings of care. BIPA has requiredCMS to design standard instruments suchthat their common elements are readily com-parable and statistically compatible.

While HEDIS data are reported on individ-ual M+C plans, quality of care statistics arenot reported at the individual physician levelor for physician groups—the primary sourceof care for beneficiaries with chronic condi-tions. The reporting of such statistics is con-troversial because of concern that physicianswith sicker and more severely chronically illpatients will not rate as well as those withhealthier patients. Another constraint is thatthe number of Medicare patients seen by anindividual provider is small (as it is for manygroup practices), making the quality of carereporting less statistically meaningful than atthe health plan level. It is also costly andtime-consuming for individual physicians orsmall practices to collect and analyze data.Nevertheless, physicians can directly influencethe quality of chronic care, more so thanmost M+C plans and other original Medicareinstitutions (e.g., hospitals, skilled nursingfacilities, home health agencies). The NationalCommittee for Quality Assurance (NCQA) ispresently working to implement a HEDISversion for large group practices, as a largenumber of patients may be appropriate foraggregate reports.

Another quality initiative is the developmentand promotion of accreditation, certification,and performance measurement of chronic

disease management. NCQA, the AmericanAccreditation HealthCare Commission(URAC), and the Joint Commission onAccreditation of Healthcare Organizations(JCAHO) are all implementing such pro-grams. Health plans, vendors, and providersof disease management will have the oppor-tunity to apply for accreditation from one ofthese three organizations. Program standardswill address the periodic update of evidence-based guidelines, measurement and qualityimprovement activities, and performancemeasurement and improvement.

H. RESEARCH ANDDEMONSTRATIONS

Though it is not written in statute, CMS’innovation is limited by the Office ofManagement and Budget’s (OMB) require-ment that demonstration projects be budgetneutral. Not only does OMB require thatdemonstration projects not increase Medicareexpenditures to over projected spending inthe absence of the demonstration, but in thecase of demonstrations enrolling dual eligi-bles, budget neutrality is calculated separatelyfor each program so that savings in one can-not be used to offset increased spending inthe other.

Though the panel commends Congress onauthorizing these three chronic care demon-strations described below, the demonstrationsare severely constrained by the requirementthat they be budget neutral.

■ Coordinated Care Demonstration.This project tests models of coordinatedcare for beneficiaries with chronic con-ditions who represent high costs to theoriginal Medicare program. CMS willpay a monthly all-inclusive rate for theproposed coordinated care services,

which might include coordination withcommunity-based services, transporta-tion, medications, noncovered homevisits, and equipment. Statutory Medi-care services will be reimbursed as usual.The demonstration seeks to improvethe quality of care provided to specificbeneficiaries with a chronic illness andmanage Medicare expenditures so thatthey do not exceed what they wouldhave been in the absence of the demon-stration. In 2001, fifteen sites wereselected for participation.

■ Disease Management Demonstration.This project targets severely chronicallyill beneficiaries with congestive heartfailure, diabetes, and coronary heart dis-ease. It requires that prescription drugsbe included, with their costs expectedto be covered by more efficient provi-sion and utilization of Medicare-coveredservices and the prevention of avoid-able, costly medical complications.

■ Physician Group Practice Demon-stration. This project assesses whetherrewards to physicians in group practicecan improve service delivery and qualityfor Medicare beneficiaries, and ultimate-ly prove cost-effective. CMS willencourage physician groups to coordi-nate their care to chronically ill benefi-ciaries, give incentives to groups thatprovide efficient patient services, andpromote active use of utilization andclinical data to improve efficiency andpatient outcomes. Participating physi-cian groups will continue to be paidunder the existing Medicare fee sched-ules and will receive bonus paymentsfrom Medicare savings achieved by thegroup. Groups will receive 70 percentof the bonus pool solely due to financial


performance, and the remaining 30 per-cent due to quality performance.

The disease management demonstrationrequires that selected participants provideprescription drugs to those chronically illbeneficiaries enrolled in the demonstration.Similarly, the coordinated care demonstrationrequires participating organizations to pro-vide services not covered under originalMedicare. Thus, the evaluation of these

demonstrations will be based largely on theparticipating organizations’ ability to manageMedicare costs—at a cost that may not berealistic, while de-emphasizing improvementsto quality of care. How chronic care couldbest be managed under a scenario whereMedicare costs could increase—a more realis-tic scenario—or in the instance where theywould be shared by the beneficiaries, will beleft untested.


CMS, health plans, and original Medicarepractitioners have implemented initiatives toimprove care for people with chronic condi-tions. These initiatives offer lessons to guidefuture chronic care programs. This chapterincludes lessons learned from six programs,some of which are considered a success,though all have drawbacks and have facednumerous obstacles.

A. INTEGRATED FINANCING ANDDELIVERY

Because Medicare does not cover all of theservices a beneficiary may need, beneficiariesmust piece together services and programsfrom multiple sources to meet their needs.This results in a patchwork of uncoordinatedservices. Programs that attempt to integratethe financing and delivery of two or moreprograms find that each program has its ownpolicies and procedures, making it difficult todesign a single coordinated program thatmeets beneficiaries’ needs.

Program coordination problems are mostapparent for frail elderly dual eligibles. In aneffort to address this, CMS’ Medicaid andMedicare divisions and states’ Medicaid agen-cies worked together with advocates of thefrail elderly to create a financing mechanismto support coordinated programs. Two CMS demonstrations, Program for All-Inclusive Care for the Elderly (PACE) andSocial Health Maintenance Organization (S/HMO), were implemented to test modelsof integrated financing and delivery pro-grams. Under the Balanced Budget Act of

1997, PACE became a permanent Medicarebenefit. According to CMS and Congress,the S/HMO models have not been as successful.

Program for All-Inclusive Care for theElderly (PACE)

PACE has almost twenty years’ experience infully integrating medical, home and commu-nity based care for the frail elderly who arenursing home eligible. The program providesa comprehensive range of preventive, prima-ry, acute, and long-term care. It also includesnon-medical services, including meals andadult day care. Participants must be certifiedas nursing home eligible to enroll. Theyreceive intensive care coordination providedby a multidisciplinary team, which includes aPACE primary care physician, social worker,rehabilitation therapists, as well as nursing,activity, and transportation staff. The PACE-run adult day health centers are co-locatedwith a primary care clinic and enrolleesreceive the majority of their medical andcommunity-based services here.

PACE sites receive a capitated payment fromboth Medicare and Medicaid and then poolthese funds to provide services. PACE pro-grams generally receive payments 2.39 timesgreater than the average M+C payment toaccount for the frailty of PACE enrollees.Medicaid payments are reached throughagreement with the state. PACE programstypically pay salaries to primary care medicaland social service staff, with most sites con-tracting with outside entities for hospital andmedical specialty services. Capitation incen-


Chapter 4:Past Initiatives to Improve Care to People withChronic Conditions Provide Valuable Experience

tives support coordination of care and substi-tution of primary and home and communitycare for high-cost acute and nursing homecare. These savings allow the PACE site topay for social services, transportation, andmental health care—benefits beyond thosetypically provided by Medicare and Medicaid.

PACE is widely considered a model of a suc-cessfully integrated medical and chronic careprogram. Most PACE sites have managed avery complex package of care within theirbudgets. Abt Associates’ evaluation of theimpact of PACE on participant outcomesreported positive outcomes, including muchlower rates of nursing home utilization andin-patient hospitalization, higher utilizationof ambulatory services, better health statusand quality of life reports, and a lower mor-tality rate than comparison group members.Also, the benefits of PACE appeared to bemagnified for those participants with highlevels of physical impairment (DHHS, 1998).

As well regarded as PACE is, growth hasbeen slower than its supporters expected:there are only 25 PACE sites with 7,696enrollees (2000 average daily census).Widespread replicability may be limitedbecause of high start-up costs (for reservesand when the PACE site must buy or build abuilding), the relatively small frail eligibledual eligible population (Medicare-only par-ticipants are permitted but they must pay thefull Medicaid capitation themselves), andbecause some eligible participants are unwill-ing to give up their private providers for thePACE site’s providers. It is also unclearwhich factor (or factors) makes PACE suc-cessful (is it, for example, the provision ofnon-medical services, its dedicated and tal-ented staff, or its care team approach?)

Policymakers continue to seek other modelsthat may evolve from PACE.

Social Health MaintenanceOrganization I and II (S/HMO I and II)

As with PACE, S/HMOs aim to integrateacute, chronic, long-term care, and social ser-vices. Care is provided through capitatedM+C plans. Participating health plans areexpected to control high-cost utilization andauthorize additional basic services to substi-tute for more costly care. Available servicesare standard Medicare benefits plus somehome and community-based services, includ-ing personal assistance, transportation,durable medical equipment, and home modification. Outpatient drugs are a coveredbenefit. S/HMOs also use care coordinatorsto emphasize community-based services and to coordinate institutional and noninsti-tutional care. There are over 105,000 Medi-care beneficiaries enrolled in the S/HMOdemonstration.

S/HMO I has been a Medicare demonstra-tion project since 1985. There are three par-ticipating health plans: Kaiser PermanenteSenior Advantage II; SCAN Health Plan; andElderplan. Eligibility requirements are thesame as M+C, though beneficiaries under age65 and those who need long-term nursinghome care are excluded. Under S/HMO I,participating health plans receive an adjustedM+C capitation. These payments are 15 to20 percent higher than the amounts theywould receive under the standard M+C capi-tation. Approximately five percent of mem-bers are targeted for special chronic careservices.

CMS’ [then HCFA] first evaluation ofS/HMO I found that although S/HMO Isuccessfully offered long-term care services, it

40 N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e40 N a t i o n a l A c a d e m y o f S o c i a l I n s u r a n c e

did not develop a well-coordinated system ofcare with acute, chronic, and long-term med-ical benefits. The principal problem was thatS/HMO projects did not establish successfulworking relationships between physicians andcase managers. Evaluators found such rela-tionships necessary for the development of awell-coordinated system of care.

The S/HMO II demonstration, whichbecame operational in 1996, was designed toimprove the services, financing methods, andbenefit design of S/HMO I. However, onlyone health plan, Health Plan of Nevada, amixed staff- and IPA-model health plan, par-ticipates in S/HMO II. (Some claim this isbecause of a requirement to develop manage-ment information systems to facilitate patientassessments and promote information flow(Berenson, 2002); others claim that it isbecause of the risk adjustment methodemployed.) The risk-adjusted capitation isfive percent higher than what would bereceived under the standard M+C capitation.S/HMO II also attempts to encourage formal involvement of geriatricians in thecare planning process. Unlike S/HMO I,S/HMO II continues enrollment for thosewho need custodial-level nursing home care,though the plan is not at risk for such care.The single S/HMO II site enrolls over35,000 members, approximately 20 percentof whom receive some level of care coordina-tion or monitoring.

A provision in the Balanced Budget Act of1997 required the Secretary of HHS to sub-mit a plan for the integration of the S/HMOdemonstrations as an option under the M+Cprogram. In response, a comprehensiveanalysis of the two S/HMO models was con-ducted and a report was submitted toCongress. The report recommends phasing

out the S/HMO payment methodologiesbeginning in 2004 and converting the sitesto standard M+C plans in 2007. HHS basedits recommendation on the following: 1) there is no consistent evidence that theS/HMOs improve beneficiary outcomes, 2) the S/HMOs are paid more than M+Cplans despite comparable case mix, and 3) the innovative S/HMO II design has beenimplemented in just one site, indicating littleinterest in that delivery model (Personal cor-respondence with CMS staff). Nevertheless, a2002 case study analysis commissioned byCMS concluded that care coordination, geri-atric services, communications, and supportinfrastructure development of S/HMO II hasbeen extensive. Because of the difficultiesintegrating medical and social care, the tech-nical assistance contractors believe that ittook several years before key benefits couldbe adequately developed and linkages created(DHHS, 2002).

B. CARE COORDINATION

The care coordination programs describedhere, Kaiser Permanente’s heart failure pro-gram and the CMS’ case managementdemonstration, are attempts to better man-age the care of high-cost beneficiaries, mostof whom have chronic conditions. The KaiserPermanente model shows how a large group-model plan can organize resources to targetchronic conditions. The CMS case manage-ment demonstration illustrates the difficultiesof coordinating care through the use of staffwho are independent of the beneficiaries’physician.

Kaiser Permanente, Northern CaliforniaHeart Failure Program

Kaiser Permanente’s Northern Californiaregion is organized as a group-model HMO



with all Kaiser patients cared for by staffphysicians. In the early 1990s, it initiated anumber of chronic care programs to enhancequality and reduce costs. Its heart failure pro-gram is designed to improve medical man-agement, quality of life, and functional healthstatus for its members with heart failure.Kaiser’s primary care division is responsiblefor this initiative.

The program features the use of interdiscipli-nary care teams. Providers are guided by evi-dence-based guidelines, scripts, and assess-ment tools to assess and manage clinical,behavioral and social issues and functionalimpairments. Protocols are used to helpmanage medications. The program usesreminder systems, patient care monitoring,and motivational interviewing to encourageinvolvement by patients.

Two years of data, 1998–2000, show thatquality indicators are improving and patientsatisfaction is relatively high and is continu-ing to increase. Kaiser’s major constraint,however, is the cost of maintaining the pro-gram. Funding for 2000 was $3 million,including the salaries of 37 dedicated full-time employees. Though Kaiser hoped thatimproved outcomes and reduced hospitaliza-tions would pay for the cost of the program,this has not yet been realized.

Medicare Case ManagementDemonstrations for High-Cost Medicare Beneficiaries

From 1993 to 1995, three demonstrationprojects were implemented to identify groupsof original Medicare beneficiaries at risk ofneeding high-cost care and to design the spe-cific features of a case management interven-tion to reduce these costs. The projects were

conducted by: 1) a government businessholding company of a large insurer, 2) a PeerReview Organization, and 3) a tertiary careteaching hospital. All three projects had basiccase management activities in common(assessment, service coordination, condition-specific self-care education, and emotionalsupport to clients and informal caregivers),though they differed in their level of in-person client contact, the degree to whichcase management activities were structured,use of nurses and social workers, and empha-sis on education and service coordination.CMS (then HCFA) reimbursed the projectsites for related expenses but did not providereimbursement for case management servicesor financial incentives for reduced utilization.

Under contract to CMS, Mathematica PolicyResearch’s evaluation of the projects foundthat none of them improved self-care orhealth, or reduced Medicare spending.According to project evaluators, the primaryreasons for lack of impacts were: 1) physi-cians were not involved in the interventions,2) projects did not have sufficiently focusedinterventions and goals, 3) projects lackedstaff with sufficient case management experi-ence and specific clinical knowledge to gener-ate the desired reductions in hospital use,and 4) the projects had no financial incentiveto reduce Medicare spending. The demon-stration’s evaluators state, however, that casemanagement might be cost-effective if itincluded greater involvement of physicians, ismore well-defined and goal-oriented, andincorporates financial incentives to generateMedicare savings (Schore, et al., 1999).

C. SYSTEM AND PAYMENT REDESIGN

The first program described here, the Break-through Series Best Practice Collaborative

Approach, involves a redesign of health careorganizations’ systems. The second programis Medicare’s ESRD program, which isunique in its reimbursement structure.Original Medicare provides a capitated pay-ment for the renal physicians’ services and abundled payment to the dialysis facilities foreach dialysis visit. Under its M+C ESRDdemonstration, a capitated payment is madeto M+C plans for provision of comprehensivehealth care services for ESRD beneficiaries.

Chronic Care Breakthrough Series BestPractice Collaborative Approach

The Chronic Care Breakthrough Series,developed by the Institute for HealthcareImprovement (IHI), is being used toimprove chronic illness care in a variety oforganizations and across multiple chronic ill-nesses. Participants include managed careorganizations, academic health center prac-tice organizations, hospitals, and Health andHuman Services’ Health Resources andServices Administration (HRSA)-supportedcommunity health systems. The participatingorganizations assign a team to attend “collab-orative learning sessions.” During these ses-sions they use the Chronic Care Model todesign and test system changes to improvethe care of a single chronic condition (e.g.,diabetes or asthma). The Chronic CareModel is a guide to improving the manage-ment of chronic conditions within primarycare. It aims to facilitate comprehensive sys-tem change through:

■ delivery system design (coordinatedcare, care-teams, and telephone andemail consultations);

■ decision support (evidence-based guide-lines, assessment tools, medication order

entry systems, computerized diagnosis,and management assistance);

■ clinical information systems (recall andreminders, disease registries, andprovider profiles);

■ self-management support (individual orgroup training and follow-up and edu-cational materials);

■ community linkages (coordination ofmedical and social services);

■ organizational support (CMS to sup-port providers and/or insurers’ caredelivery and MCO or group practice to support individual providers’ caredelivery).

The model does not offer a quick and easyfix; it is a multidimensional solution to acomplex problem. None of the organizationshave achieved full implementation of theChronic Care Model, but all have madeimportant strides toward that goal.

The Breakthrough Series has involved 104health care organizations in year-long qualityimprovement programs. Participating organi-zations include a variety of health plan typesthat pay their providers on a fee-for-servicebasis. CMS is supporting this model throughits Quality Improvement Organizations(QIOs) as part of its QI InterventionStrategies Collaborative.

Initial evidence demonstrates that theChronic Care Model can improve chroniccare and in some cases reduce health carecosts. Using diabetes care programs as anexample, a review of research evidence ofprograms that implemented the ChronicCare Model through multifaceted interven-tions found that most (32 of 39 studies)improved process and outcome measures for


diabetes. Another review of the researchfound that whether or not the program real-ized cost savings depends on the specificchronic condition targeted (improvements inthe care of congestive heart failure, asthma,and diabetes were shown to have the capacityto reduce expenditures). Whether or not thecost savings translate to a business advantagedepends on the type of organization institut-ing chronic care improvement and the modeof reimbursement. Though the cost savingsin congestive heart failure, asthma, and dia-betes care result from lower hospital andemergency room utilization, a hospital orhealth care system reimbursed by capitationcan save money for the organization. On the other hand, a hospital or health care sys-tem that is reimbursed per diem or fee-for-service may lose revenue (Bodenheimer, etal., 2002).

End-Stage Renal Disease

In 1972, Congress enacted legislation to pro-vide eligibility to individuals with ESRD.This is the only group of individuals to beeligible for Medicare based on a clinical con-dition. The high cost of care for ESRD bene-ficiaries has prompted innovations in bothoriginal Medicare and M+C. Savings to origi-nal Medicare are limited, however, by benefi-ciaries’ high rates of comorbid conditions(see Table 1 on page 5). Capitation thatincludes the full range of beneficiaries’ care(including comorbidities), on the other hand,appears to have the opportunity to improvequality of care.

ESRD Under Original Medicare

All ESRD beneficiaries require either dialysisor kidney transplant to sustain life—care thatis very expensive. As a means of containingcosts and better managing care, Medicarepays renal physicians a monthly capitation for

the direction of renal care (an amount abouttwice the payment for physician supervisionof one inpatient dialysis session). It also paysdialysis facilities a predetermined payment foreach dialysis treatment they furnish (up to 3 per week). This prospective payment isintended to cover all operating and capitalcosts for dialysis, including services, tests,drugs (erythropoeitin is covered by Medicarebut is not included in the prospective pay-ment), and supplies routinely required fordialysis treatment.

The capitated amount is paid to renal physi-cians to manage dialysis care accounts foronly a small part (approximately 20%) of allphysician care received by ESRD patients.There were cost efficiencies in the early yearsof its implementation, but in recent years,most physician charges for dialysis patientsare for services other than renal care. Thus,ESRD capitation payments have not beensuccessful at controlling a patient’s totalexpenditures. In part, this is due to thechanging demographics of the ESRD popula-tion. Since 1978, the average age of Medi-care’s ESRD population has increased and a much higher percentage of persons withdiabetes as the cause of renal failure areenrolled. Such ESRD patients have highermorbidity and mortality rates and higheraverage per capita expenditures ($54,094 forthose with diabetes; $40,452 for those with-out) (Eggers, 2000).

Poor outcomes in the early 1990s promptedseveral quality improvement efforts. Suchefforts included the development of perfor-mance measures by CMS, development ofclinical practice guidelines by the renal com-munity, and provision of facility-specificinformation to patients by CMS (posted onits Dialysis Facility Compare website).



Medicare+Choice ESRD Demonstration

In 1984, as part of TEFRA, Medicare beganproviding a capitated payment to HMOs thatenrolled aged and disabled Medicare benefi-ciaries with ESRD. Though beneficiaries withESRD were prohibited from enrolling inHMOs, they were allowed to remain in theirHMO if they developed ESRD after enroll-ment. Medicare paid 95 percent of thestatewide average per capita original Medi-care ESRD costs, unadjusted for patient char-acteristics or treatment modality. Capitationprovided opportunities for savings over fee-for-service expenditures by reducing inpatientutilization (nearly 40 percent of the costs fororiginal Medicare ESRD patients were forinpatient care.) In 1993, Congress passedlegislation requiring a S/HMO demonstra-tion for ESRD beneficiaries. This demonstra-tion tested whether: it is feasible to haveyear-round open enrollment of ESRD benefi-ciaries in an HMO; integrated acute andchronic care services and case managementfor ESRD beneficiaries improves health out-comes; capitation rates reflecting ESRDpatients’ treatment increases the probabilityof kidney transplant; and whether additionalcoverage benefits are cost-effective. Partici-pating M+C plans were paid 100 percent oforiginal Medicare ESRD expenditures,adjusted for age, original cause of ESRD, andtype of therapy (e.g., dialysis, transplant, etc.)

Preliminary evaluation findings suggest thatdropping the barrier to ESRD enrollmentresults in the same kind of adverse selectionfound in non-ESRD enrollment in M+C,with healthier ESRD beneficiaries choosingto enroll. Those who enroll say they do so tosave out-of-pocket costs and gain extra bene-fits, and are influenced by their physicians interms of whether or not to make this choice.Preliminary findings also indicate that

demonstration enrollees, like ESRD benefi-ciaries in the evaluation’s comparison groups,scored high on tests of patient satisfactionwith care. Their scores on quality of life mea-sures were significantly better than those inthe comparison group and improved signifi-cantly in the year following enrollment(DHHS, 2002b).

The 3-year period of mandated demo opera-tions concluded in January 2001. One of thethree participating M+C plans continuedunder a renewed demonstration waiver. BIPAof 2000 mandated risk-adjusted M+C ESRDcapitation rates. In January 2002, CMSintroduced M+C ESRD capitation paymentsthat were adjusted for age and sex. CMS is currently developing a more comprehen-sive risk adjustment methodology that isscheduled to be implemented in 2004. This methodology is expected to also incor-porate diabetic status and other comorbid conditions.

D. LESSONS FROM PAST INITIATIVES

Most of these initiatives found that chroniccare requires specialized training and thecoordination of providers. PACE and KaiserPermanente’s heart failure program, forexample, showed that well-integrated careteams improve patient care. Conversely, theS/HMOs and the case management demon-strations showed that a lack of coordinationbetween the care manager and physician isdetrimental to chronic care efforts. Evaluationof the case management demonstration alsofound that nurse case managers and socialworkers with insufficient training in casemanagement and clinical care were not effec-tive as case managers. Evaluations have notshown, however, what type of training andparticipation of which provider type is need-ed to ensure their effectiveness.

These initiatives also show that financialincentives that align with program goals maybe helpful. For example, CMS did not pro-vide the case management demonstrationsites with financial incentives for improvingcare or for reducing utilization. The capitatedinitiatives (PACE, Kaiser Permanente’s heart failure program, and the M+C ESRDdemonstration), provided that the capitationpayment is sufficient, have built-in incentiveto manage high-cost utilization. Neverthe-less, there is concern that capitation willinfluence providers to withhold needed careor enroll healthier beneficiaries (as shown in the S/HMOs and the M+C ESRD demonstration.)

The above initiatives require the ability totrack patients, diagnoses, and utilization. Asmost chronic care programs focus on benefi-ciaries with serious or high-cost conditions,information systems that help identify benefi-ciaries who will benefit most are crucial toprogram success. A major reason for the lackof impact of the case coordination demon-stration was that the project participantscould not effectively choose beneficiaries whowere most appropriate for the program(instead, they depended largely on physicianreferrals). Once beneficiaries are selected,tracking patient utilization is necessary, as allof the programs described above are depen-dent on reducing high-cost utilization (e.g.,inpatient costs) so that they can provideother services crucial to beneficiaries’ care.Some utilize more rudimentary systems totrack utilization (which is easier to do with asmall number of beneficiaries); others (suchas some Breakthrough Series Best PracticeCollaborative participants) incorporate moreadvanced computerized information systems.

Experience also shows that sustainedimprovement requires comprehensive systemchange (e.g., community linkages, self-management support, delivery system design,decision support, and clinical informationsystems). PACE and Kaiser Permanente’sheart failure program are successful largelybecause they encompass broad organizationalchange. Major changes to the care system,however, may be difficult to accomplish. Asin the case of the S/HMO I, for example,they were unable to develop an integratedacute, chronic, and long-term care system.Efforts more limited in scope (such as assign-ing a nurse or social worker to manage bene-ficiaries’ care) may be less likely to besuccessful and to sustain any improvementsover time than those that encompass com-prehensive system change.

Lastly, it may not be possible to vastlyimprove systems of care on a budget-neutralbasis. As with Kaiser Permanente’s heart fail-ure program, large scale and comprehensiveefforts may significantly improve the qualityof care, though the short-run savings due toreduced inpatient and other high-cost utiliza-tion may not be as large as the cost of theimproved care system. As Kaiser Permanentefound, the additional staff needed to improvepatient monitoring, provide coordinatedcare, address behavioral, social and functionalissues, and provide self-management educa-tion, as well as the start-up costs for redesign-ing care systems and re-training providers,cost more than the savings earned bydecreasing unnecessary high-cost utilization.Nonetheless, quality of care and patient satis-faction increased as a result of its new systemof care.


The study panel’s recommendations includeits long-term vision for Medicare and sixshort- to mid-term recommendations. Thepanel considered its original principles as itformed its recommendations. (Those princi-ples are: focus on beneficiaries’ needs andpreferences; reimbursement to support rec-ommended models of care delivery; and pro-mote efficiency. (See page 12.)) It found thatit is relatively easy to articulate Medicare’sbarriers to chronic care, but far more difficultto propose solutions.

One barrier to finding solutions is theabsence of evidence to support many changesthat seem, on their face, compelling. This ledthe panel to modify one of its initial princi-ples: reliance on evidence-based change.Evidence to support policy making in healthcare is in its infancy. In many cases, the datato conclusively support recommendationsthat the panel views as important are simplynot available. The panel also recognizes thatcollecting conclusive evidence may take aconsiderable amount of time. Thus, it hasused what evidence is available to composeits recommendations and supplemented thiswith its health care and policymaking knowl-edge and experience.

In principle, the panel believes that its rec-ommendations should help beneficiariesmaintain function and quality of life. Incorp-oration of function and quality of life intoMedicare is hampered by the ability to defineand measure them. Nevertheless, progress intheir definition and measurement has beenmade. Information on function based onsome variant of the ADLs and IADLs is now

routinely collected for all patients receivingnursing home care, home health care, andrehabilitation. The International Classifica-tion of Function, Disability and Health(ICF), an indicator of functional loss andbarriers to access to care, was recentlyreleased (2001) by the World Health Org-anization and may in the future be useful toMedicare. Systematically collecting quality oflife information is more difficult and the sys-tems to do so are more rudimentary, thoughwork in this area is progressing.

A. AVENUES TO CHANGE

Medicare policies that support or constrainchronic care are established through myriadprocesses: law (also known as statute), regu-lations, and a host of sub-regulatory mecha-nisms, including national coverage decisions,contractor policy manuals, local medicalreview policies, program memoranda, opera-tional policy letters, and program issuances.

Only Congress has the power to establish orchange a law. Regulations are usually neededto implement laws. CMS must issue regula-tions in accordance with the AdministrativeProcedures Act (APA). The APA requiresfederal agencies to give the public notice ofproposed actions, with opportunity to com-ment. When making sub-regulatory changes,CMS and its contractors are generallyrequired to adhere to the intent of the law or regulation. Thus, CMS could not issue a policy that contravenes existing law or regulation.

If a change in law or regulation is not need-ed, sub-regulatory avenues may be used to


Chapter 5:Conclusions and Recommendations


change policy. For example, to stipulatewhether a service is covered, CMS may issuea national coverage decision, or Medicarecontractors may issue a local medical reviewpolicy. However, relatively few national cov-erage decisions are issued and the process islengthy. (In 2001, ten national coveragedecisions were issued, with an average timefrom submission to issuance of 383 days.)Instead, most coverage policies are made byMedicare contractors (intermediaries and car-riers) through local medical review policies.CMS can guide contractor policies and deci-sionmaking through program memoranda,operational policy letters, and contractormanuals.

The panel’s recommendations addresschanges across the spectrum of Medicare policy sources.

B. LONG-TERM VISION

NASI’s Chronic Care study panel proposes along-term vision for Medicare. In the panel’svision, Medicare would provide beneficiarieswith access to needed services and financialprotection from health care costs that posebarriers to needed care. This would involveadding coverage for chronic care services notpresently included in Medicare’s benefitpackage, including function and quality oflife-related services. Changes to the benefitpackage would be designed to meet theneeds of beneficiaries. Medicare would alsoset reasonable limits for beneficiaries’ health-related out-of-pocket expenditures.

The panel’s vision entails a dramatic shift toinclude a chronic care focus in Medicare.Evidence-based guidelines would be availablefor all chronic conditions (and for multiplechronic conditions) and providers’ practiceswould be based on these guidelines. Concern

for function and quality of life would be inte-grated into the care system. There would bea seamless continuum across acute, chronic,long-term and end-of-life care. All providerswould be required to use computerizedinformation systems. These systems wouldsupport sharing of electronic medical recordsamong providers, medication order checks,and patient-specific protocols, all of whichwould vastly improve the quality of care, par-ticularly given the high percentage of chroni-cally ill beneficiaries with multiple providers.

As the largest health care purchaser in thecountry, Medicare would actively work toimprove the quality of chronic care. It wouldsurpass the quality standards set by thebroader health care system. Quality of carewould be measured and reported to the pub-lic. Medicare would make additional pay-ments to providers who offer high qualitycare. Measures of quality of care would besensitive to the unique conditions, issues, anddiversity of concerns of beneficiaries withchronic conditions.

Reimbursement methods would cease to bean obstacle to chronic care services. Insteadsuch methods would align incentives, adjustfor risk factors, and provide the flexibilitynecessary for quality chronic care delivery.Variations on prepayment and salaries wouldbe tested to foster a team-based approach tocare, emails and telephone contacts with ben-eficiaries, an emphasis on prevention and self-management education, and incorporation offamily members into the care process. Mostproviders would be affiliated with a providernetwork organization, a health plan, or anintegrated delivery system that would providethem organizational support for chronic care(for computer systems, health educationclasses, etc.)

The study panel also envisions that the rec-ommendations of the IOM’s Crossing theQuality Chasm report are adopted. Thoughthe IOM’s recommendations encompass thebroad U.S. health care system (non-Medicareand Medicare), as stated earlier, majorimprovements to Medicare beneficiaries’ caresystems will require improvements in the broadhealth care system. Medicare should take thelead in chronic care system improvements.

C. SHORT- TO MID-TERMRECOMMENDATIONS

The following are the panel’s short- to mid-term recommendations, some of which couldbe implemented immediately; others maytake five to ten years, though work on eachof them should begin immediately. Some willhave considerably more impact than others;some will cost the Medicare program moreto implement than others; and some will bemore difficult to gain political support thanothers. Nevertheless, the panel believes thateach of its proposed recommendations areimportant for improving the health andfinancial security of beneficiaries with chronicconditions.

Changes to Medicare statutes will facilitateenactment of these recommendations.Statutes that emphasize its acute-care biaswould be revised—including those that basecoverage on the acute care term, “spell of ill-ness.” The obsolete statute that prohibitsMedicare from “exercising any control overthe practice of medicine or the manner inwhich medical services are provided” wouldbe repealed so that CMS would have greaterauthority to improve quality. As stated inRecommendation 2, the statute that pro-hibits services that are “not medically neces-sary,” and exclude “personal comfort

items…eye examinations…hearing aids…custodial care,” would be rewritten.

Appropriate implementation of these recom-mendations requires that CMS be givenadditional administrative resources. Evenwithout any additional responsibilities in itsportfolio, CMS’ present administrativeresources are insufficient. Despite a dramaticincrease in its responsibilities over the pastdecade, CMS program management appro-priations have increased only 26 percent andthe number of full-time employees 12 per-cent (King, et al., 2002). In an open letter toCongress and the Executive more than threeyears ago, experts and advocates warned ofan impending management crisis that threat-ened the agency. The panel believes that poli-cymakers must provide CMS with adequateadministrative and management resources sothat it can operate effectively, as well asimplement changes to improve the chroniccare provided to its beneficiaries.

Recommendation 1: Provide beneficiaries with financial protection from chronic conditions.

In accordance with Medicare’s original goalto provide beneficiaries with financial securi-ty, and the study panel’s principle to focus onbeneficiaries’ needs and preferences, thepanel recommends that services vital tochronic care be covered, and that beneficia-ries’ out-of-pocket health care expendituresfor covered services be limited. This wouldreduce an excessive financial burden on bene-ficiaries with chronic conditions. It will alsohelp ensure that beneficiaries seek and haveaccess to appropriate care.

■ Limit cost-sharing requirements byadding an annual cap on out-of-pocket expenditures for covered ser-vices. An annual limit is needed to


protect beneficiaries with chronic condi-tions from high out-of-pocket costs.

■ Cover services necessary for benefi-ciaries’ chronic care needs. Such ser-vices not presently covered by Medicareare addressed in Recommendation 2.

Recommendation 2: Support the continuum of care beyondthose services presently covered byMedicare.

Medicare statute prohibits payment for ser-vices that “…are not reasonable and neces-sary for the diagnosis or treatment of illnessor injury or to improve the functioning of amalformed body member” (1862(a)). Thepanel recommends changing this statute to:“…are not reasonable and necessary for theprevention, diagnosis, treatment of illness orinjury, or to improve, maintain, or slow thedecline of function.” This change wouldexplicitly allow for coverage of preventive andrehabilitative services that provide opportuni-ties for beneficiaries to function to the best oftheir ability.

■ Address gaps in Medicare’s benefitstructure. Two significant gaps are pre-scription drugs and preventive healthservices.

– Prescription drugs. The panel views aprescription drug benefit as the singlemost important addition for manage-ment of chronic conditions. Suchcoverage should be meaningful andconsistent with the need to limit ben-eficiaries’ cost-sharing. To best sup-port chronic care, prescription drugsshould be added to the Medicarebenefit package. The panel recom-mends that drugs be an integral partof Medicare so that their coordina-tion with medical care will beenhanced and so that all beneficiaries

receive the benefit. Coverage formedication management should beincluded in the drug benefit, and itshould offer beneficiary education,compliance monitoring, error checks,and facilitate coordination of pharma-cists and physicians.

– Preventive health services. Healthpromotion and disease preventionservices that assist beneficiaries inmaximizing health status, functionalindependence, and quality of life,should be included in the benefitpackage. The statutory exclusion forpreventive benefits should berepealed, and Congress shouldrewrite the statute to include preven-tive benefits. The decision aboutwhich preventive health services toadd should be based on the recom-mendations of the U.S. PreventiveServices Task Force (USPSTF). Thestatute should be revised so that theUSPSTF’s recommendations arereviewed and approved through theMedicare coverage process.

■ Strive to include services related tofunction and health-related quality oflife. Medicare should help beneficiariesmaintain or restore function and qualityof life.

– Relax the “homebound” requirementfor home care. Though the panelcommends CMS for its July 2002carrier manual changes (i.e., its provi-sion of more examples of when occa-sional absences from the home areacceptable, its statement that a per-son’s homebound determinationshould be made over a period oftime, and for adding ALS and otherneurodegenerative disabilities to thelist of those conditions that may indi-cate a patient cannot leave home),the panel recommends that CMS andCongress go further. For example, it


could expand the homeboundrequirement to any individual whomeets the current homebound defini-tion; or any individual who could notleave the home without the provisionof Medicare home care services, andwithout such services would be con-fined to the home.

Such terminology would not restricta beneficiary to home, nor would itprovide for home health services tothose who are able to receive careregularly at outpatient facilities. Itwould move Medicare further alongthe long-term care continuum,though beneficiaries who requirehome health aide services to leavehome, but who do not need skilledservices, would still be excluded.

– Cover durable medical equipment(DME) with the specific intent ofmaintaining or restoring function.The requirement that DME be usedprimarily in the home should beeliminated because it prevents benefi-ciaries from being as independent aspossible. The panel recognizes, how-ever, that elimination of the “primari-ly in the home” clause will lead toother coverage issues. CMS shouldinterpret DME coverage policies tomaximize beneficiaries’ independencein a way that is reasonable.

– Provide for assistive devices thatcompensate for sensory or neurologi-cal deficits. Eyeglasses and hearingaid benefits, for example, which arenow specifically excluded by statute(though generally one pair of eye-glasses following cataract surgery iscovered), should be covered. Thepanel believes that vision and hearingaids for those with sensory deficits areas important as wheelchairs for bene-ficiaries with mobility limitations.

– Support rehabilitation as a tool toimprove, maintain, or slow thedecline of function. The panel rec-ommends that CMS and its carriersadopt policies for Part B outpatientservices comparable to current Part Aregulations for home health andnursing facility services. (The Part AManual states that the restorationpotential of a patient should not be adeciding factor in determiningwhether therapeutic services are rea-sonable and necessary; the Part Bmanual states that rehabilitation cov-erage, with the possible exception ofthe design of and instructions for amaintenance program, be discontin-ued when a beneficiary reacheshis/her full rehabilitation potential.)

Because the panel believes that con-tractors many times limit coverage,particularly for rehabilitation services,CMS should increase its oversight ofcarriers. It should review and analyzelocal medical review policies pro-posed by contractors, as well as thecomments and testimony submittedby providers, beneficiaries and thegeneral public, prior to implementa-tion. CMS should require contractorsto identify the research, practiceguidelines, or community standardson which the local medical reviewpolicy is based.

■ Involve families. Family care is essentialto the quality of care for beneficiarieswith chronic conditions. Medicare’spolicies should provide information andeducation about choices of health plans,providers, and paperwork requirementsnot only to beneficiaries, but also tofamily members and caregivers. Medi-care should also extend coverage forpatient education to practitioners otherthan physicians and provide for an


explicit patient-family education benefit.Providers should be adequately com-pensated for their time spent on familyconsultations. Evaluation & Manage-ment coding should be modified to rec-ognize this. Research and demon-strations should explore other paymentoptions, including a specific payment forfamily consultations.

Recommendation 3: Promote new models of care.

As the IOM’s Quality of Care panel stated,“Incremental changes in policy are not suffi-cient to significantly improve quality of care.Instead, a new system of care must be devel-oped and implemented” (2001). Medicareshould play a leading role in the developmentof a new care system.

■ Foster delivery system change.

– Encourage improved practice orga-nization and care delivery. A com-prehensive approach is needed, asindividual interventions will notresult in substantial improvements inchronic care. An example of such anapproach is the Chronic Care Model(described on page 43), which sup-ports system change through deliverysystem redesign, decision support,clinical information systems, and self-management support, among others.This model promotes management ofchronic conditions within a primarycare-based setting and coordinateswith specialists as necessary. Such anapproach is most appropriate forthose with multiple chronic condi-tions.

– Support geriatric assessment andmanagement. Many Medicare bene-ficiaries would benefit from an assess-ment that aids in the diagnosis offunctional ability, physical health,

cognitive and mental health, and thesocioenvironmental situation. CMSshould evaluate how to define stan-dards for geriatric assessment andmanagement, ensure that there is sufficient evidence of its effective-ness, and then establish reimburse-ment codes for this service. It shouldalso determine if such services aremost appropriate for a subset of beneficiaries.

– Integrate services for those duallyeligible for Medicare and Medicaid.Systems of care for dual eligiblesshould be better integrated, as thereis a disconnect between the servicesthat Medicare covers (e.g., medicalcare) and those that Medicaid covers(e.g., community-based care). Inte-gration will depend, in part, on bet-ter integration of funding, as well asthe integration of CMS’ Medicareand Medicaid divisions.

■ Increase providers’ knowledge ofchronic and geriatric care.

– Use Graduate Medical Educationfunding to support chronic caretraining. The panel recommendsthat academic medical centers receiv-ing GME funding be required todemonstrate that some portion oftheir DGME funds is being used totrain residents and students in themanagement of chronic conditions.As an initial measure, these academicmedical centers should report thepercentage of funds spent for trainingin outpatient facilities, nursing facili-ties, and home care. CMS shouldassess these data and then pursueadditional collection of data and/or policies to promote outpatient and community-based training forresidents.

The panel recognizes the role thatMedicare GME funds play in sustain-


ing the missions of academic medicalcenters, but also believes that GMEfunds should encourage training tomeet the needs of beneficiaries.Therefore, it recommends that Medi-care continue its development andimplementation of a GME paymentadjustment to promote patient carein outpatient and other care settings.Medicare should also attempt toinfluence the number of geriatriciansby increasing payments to those aca-demic medical centers that train geri-atricians in larger numbers.

– Support geriatric training for allphysicians and train more academicgeriatricians. Because of the severeshortage of physicians trained in geri-atrics, Medicare should ensure that allphysicians, including both primarycare and specialists (with the excep-tion of pediatricians), receive basiceducation and training in geriatricsduring their residencies. This willrequire an increased number of acad-emic geriatricians. (In 1999, only 0.4percent of all residencies were in geri-atrics.) GME funds currently supporta one-year geriatrics fellowship,though they previously supported atwo-year fellowship. The panel rec-ommends that Medicare pay for asecond year of geriatric fellowshiptraining because a two-year fellow-ship is the accepted standard forthose who pursue an academic careerin geriatric research or teaching.

■ Payment should support new modelsof care.

– Risk-adjust Evaluation andManagement (E&M) codes. As perthe panel’s guiding principle thatreimbursement should support rec-ommended models of care delivery,the panel believes that the care deliv-ery system described in this report

should be supported by risk-adjustedE&M codes. Though E&M codesare controversial and actively underdiscussion, the present coding systemdoes not allow for the extendedlength of time a provider needs tooffer quality care to those withchronic conditions. By more appro-priately reimbursing providers for the additional time spent with thesepatients, quality of care shouldincrease.

Add-on payments for beneficiarieswith specified chronic conditionsand/or multiple chronic conditionsare one approach to risk adjustmentthat should be considered. An add-on payment would compensateproviders of beneficiaries with seriouschronic conditions through longervisits, additional care, and coordina-tion with other providers. (Paymentto providers of beneficiaries who donot meet the criteria would not beaffected.) The add-on paymentwould be paid for all E&M servicesprovided to beneficiaries who meetspecified criteria. For example, CMScould establish criteria that include asubset or cluster of those with chron-ic conditions associated with highercosts, higher morbidity, higher ratesof complications, and visits to a largenumber of providers (Berenson,2002). CMS should begin by assign-ing researchers to select appropriatedisease clusters and then adjust itspayment systems to allow beneficia-ries to be “tagged.” Once the incre-mental payment is tested for a smallsubset of beneficiaries, it can beincorporated directly into theMedicare program.

– Improve models for risk-adjustingprepaid arrangements. Sucharrangements are crucial to beneficia-ries with chronic conditions. The



panel believes that its implementationshould be expedited and that adjust-ments be improved to account forchronic conditions.

– Test alternative payment modelswithin original Medicare. The studypanel supports consideration ofbundling and prospective paymentfor some services within originalMedicare, particularly for primarycare services. It believes that, in theo-ry, such payment models help bypassmany of the constraints of fee-for-service Medicare and offer flexibilityand incentives to support well-coordinated care systems.

Though the panel does not have aspecific plan to recommend, itencourages CMS to design and test avariety of payment models for differ-ent subpopulations. For example,PACE reimbursement methodologycould be applied to other chroniccondition or dual-eligible popula-tions. Other potential payment mod-els are the use of partial capitationreimbursement to physicians and/orphysician groups under originalMedicare and full capitation to physi-cian groups under original Medicare.

Recommendation 4: Strengthen CMS’ role as a purchaser of care.

In principle, the panel supports the use ofextra payments to providers who providehigh quality chronic care, though it recog-nizes implementation and political complexi-ties. Until systems to assess quality of carebecome more advanced, CMS should workto enhance quality of care reporting andimplement demonstrations such as theMedicare Partnerships for Quality ServicesDemonstration (formerly called the Centersof Excellence Demonstration) for select

chronic conditions. As per the panel’s princi-ple of efficiency, such initiatives will helpCMS ensure that the federal government andbeneficiaries receive the best possible care forthe dollars spent.

■ Measure and report on the quality ofchronic care. To the extent technicallypossible, Medicare should monitor thequality of chronic care provided by indi-vidual providers, physician groups, andhealth plans and make this informationavailable to beneficiaries. Quality of carereporting requirements similar to thosenow required by M+C plans and nurs-ing facilities should eventually berequired of other providers. CMSshould support the addition of morechronic care measures (including multi-ple chronic condition measures) becausemany of the quality measures assess carefor acute illness. When feasible, mea-surement should attempt to take intoaccount beneficiaries’ preferences forcare. As quality measurements improve,information on individual providers andgroup practices should be made avail-able to the public. CMS’ oversight inthe provision of such information is astep towards acting as a purchaser ofquality care.

■ Designate Medicare Partnerships forQuality Services demonstration (for-merly called Centers for Excellence)for select chronic conditions. Thoughthe Partnerships for Quality Servicesdemonstration has targeted acute condi-tions (including congestive heart failureand hip and knee replacements), thismodel could also applied to single andmultiple chronic conditions.

Recommendation 5: Support enhanced information systems.

Medicare should support implementation ofinformation systems that track beneficiariesacross multiple providers and care settings.Enhanced information systems will supportseamless care for beneficiaries, reduce medicalerrors, and make a wealth of population-based information available to providers andCMS. They also offer the opportunity to col-lect and standardize health and functionalassessment data.

■ Implement electronic informationsystems. CMS should support thereplacement of handwritten medicalrecords with longitudinal electronicmedical records. Electronic recordsshould include all information frombeneficiaries’ multiple providers, includ-ing reports from the primary careprovider, specialists, hospitals, rehabilita-tion services, as well as a medical andsocial history, lab results, and drug aller-gies. Information systems should alsoproactively facilitate care. Systemsshould prompt providers about medicalprotocol (e.g., a physician of a diabeticcould be reminded to do an eye or footexam) and alert clinicians and pharma-cists of a prescription error. They shouldbe capable of analyzing sub-populationsby chronic condition, age, geographicarea, racial and ethnic group, etc. Suchinformation would help CMS assessbeneficiaries’ quality of care, highlightpotential areas for improvement in care,and provide evidence of which practicesoptimize quality of care.

The panel believes Medicare should playan influential role in the creation of aninformation infrastructure system for

medical records. (It should be acknowl-edged that Medicare led the health caresystem in its creation of the DRG sys-tem and electronic billing.) At issue iswho should pay these costs, as a majorobstacle to its implementation is thecost to providers for choosing,installing, and maintaining such systems.Though private insurers, and providersshould all be expected to contribute toits costs, Medicare should pay its fairshare. This could be in the form of sup-porting a pilot program in one state orarea, grants to public hospitals, an add-on to DRG payments, or bonus pay-ments to providers who implement suchsystems.

Because it may be many years beforeelectronic medical records will beinstalled in the majority of providerpractices (particularly smaller practices),CMS should also support the dissemi-nation of flexible, patient-centered reg-istries as an interim measure. Suchregistries are a list of all patients with acondition. Registries provide practicesand practitioners with reminders, feed-back, and care planning tools, as well asthe systematic collection of clinical dataessential for quality monitoring.Though they need not be computer-ized, very simple computer systems canproduce, hold, and make use of reg-istries. Registries are a highly effectiveinformation system enhancement untilthe time that affordable, effective elec-tronic medical records are relevant tosmall practices.

■ Promote the collection and standard-ization of health and functionalassessment data. Because CMS hasmandated separate measures and infor-


mation systems for different providersettings (e.g., MDS, OASIS, IRF-PAI),beneficiaries’ health, function, and qual-ity of life cannot be tracked across caresettings. BIPA required CMS to designstandard instruments such that commonelements may be readily comparable and are statistically compatible. It alsorequired that CMS submit a report onthe development of such instruments by2005. Though a considerable amountof work has been done by CMS, otheragencies, and the provider community,CMS is delaying implementationbecause of lack of financial resources.Congress should allocate the necessaryfunds to implement the development ofthese instruments, beginning with post-acute care settings. These common sys-tems should then be expanded to othercare settings.

Recommendation 6:Implement and support funding for researchand demonstration projects.

Information about what initiatives would (orwould not) improve quality of chronic careand control chronic care expenditures willhelp guide the Medicare program.

■ Sponsor a wide variety of chroniccare research and demonstration projects and readily incorporate successful elements into the Medicareprogram. Research should be specificto the Medicare population. TheMedicare population differs from therest of the population because more ofits beneficiaries are aged and disabledand more have chronic illnesses andfunctional and cognitive limitations.

The panel suggests that research anddemonstrations be used as prototypedevelopment and findings used to expe-dite implementation of prototype mod-els. Projects should aim to identify andsubstantiate solutions to problems.Once solutions are demonstrated, theyshould be incorporated into Medicare.

Along with the projects mentioned ear-lier in this report, the panel believes thatfurther research and demonstrationshould:

– demonstrate whether it would beworthwhile for specific subpopula-tions (e.g., those with five or morechronic conditions or three or moreADLs) to be provided with alterna-tive models of care management services;

– expand its present disease manage-ment projects to address multiple andserious chronic conditions;

– test alternative physician payment sys-tems to accommodate non-visit basedpayments (e.g., emails and telephonecalls);

– research options to promote familyparticipation in care, including pay-ment for family consultations;

– continue research on risk adjustment.

■ Redefine budget neutrality for thepurpose of approving proposeddemonstrations. Change or providegreater flexibility to OMB’s test of bud-get neutrality (which requires that indi-vidual Medicare and Medicaid programcosts, as well as combined Medicare andMedicaid program costs, not exceed theprojected spending in the absence ofthe demonstration.) Flexibility could beenhanced by increasing the number ofyears of the demonstrations or by allow-


ing additional federal cost categories tobe included in the calculation. Thestudy panel also believes that the con-cept of budget neutrality does not allowfor the testing of new policies that mayincrease expenditures but are worth-while because they improve quality ofcare.

■ Increase CMS’ budget for researchand demonstrations to improvechronic care. The FY 2002 budget forCMS-wide research and demonstrationprograms is $55 million.18 A significant-ly expanded research and demonstrationprogram to investigate chronic condi-tions (those recommended in this docu-ment, as well as others), however,would require considerably more fund-ing. The study panel suggests that one-half of one percent of the Medicare PartA Trust Fund ($1.25 billion) be allocat-ed to research and demonstrations.

D. PRIORITY AND LOW-COSTPOLICIES

The panel has recommended a number ofways to ensure beneficiaries’ financial securityand improve the quality of care for beneficia-ries with chronic conditions. At issue is thecost to implement these recommendations.While the panel acknowledges that imple-mentation of most of these recommendationswill require additional significant spending, itbelieves that implementation of some or allof its recommendations would result in bet-ter care for beneficiaries and should be seri-ously considered. A recent report of theNASI study panel on Medicare’s long-term

financing addressed ways to pay for increasedcosts. It concluded that additional Medicarerevenues to serve future beneficiaries wouldbe necessary but manageable (Gluck andMoon, 2000).

The panel offers policymakers broad guide-lines for a variety of ways—large and small—to improve Medicare’s care for beneficiarieswith chronic conditions. It recognizes that itis unlikely that all of its recommendationswill be implemented, especially in the short-run. It leaves it up to Congress to decide forwhich recommendations to appropriate fund-ing, how to finance them, and a timeframefor their implementation. Nevertheless, itbelieves it is important to begin to makechanges to improve care for beneficiaries.

To help policymakers prioritize which policiesto support, the panel has listed the top rec-ommendations that it believes would havethe largest impact on chronic care. It alsorecommends the three low-cost recommen-dations that it believes would have substantialimpact if enacted.

Along with a prescription drug benefit, thepanel’s top priority recommendations are:

■ limit cost-sharing requirements byadding an annual cap for out-of-pocketexpenditures;

■ support new models of care by risk-adjusting Evaluation and Management(E&M) codes;

■ implement information systems thattrack beneficiaries across multipleproviders and care settings.


18 CMS’ research budget includes research, demonstrations, and evaluations for the Medicare, Medicaid, and theState Children’s Health Insurance Program.

The three low-cost recommendations thatthe panel believes would significantlyimprove the quality of chronic care to beneficiaries are:

■ use Graduate Medical Education(GME) funding to support chronic caretraining;

■ test alternative payment models;

■ measure and report on the quality ofchronic care.

The panel believes it is important to recog-nize the broader gains for society that cancome from an improved Medicare program.Those gains include improved health status,

better quality of life, reduced financial risk,

and greater peace of mind for Medicare ben-

eficiaries and their families who now cope

with the gaps in Medicare protection for

people with chronic conditions.

The panel believes that Medicare’s chronic

care system and benefit structure is in crucial

need of improvement. It hopes this report

will help policymakers, providers, and benefi-

ciaries better understand the many options to

improve Medicare for people with chronic

conditions and the need to adopt such

changes.



Achman, Lori, and Marsha Gold, Medicare+Choice 1999–2001: An Analysis of Managed CarePlan Withdrawals and Trends in Benefits and Premiums (Washington, DC: The CommonwealthFund, February 2002).

Anderson, Gerard, “Characteristics of Medicare Beneficiaries with Chronic Conditions,”PowerPoint presentation at the NASI Study Panel on Medicare and Chronic Care in the 21stCentury, Washington, DC, October 19, 2001.

Berenson, Robert A., and Jane Horvath, “Clinical Characteristics of Medicare Beneficiaries andImplications for Medicare Reform,” paper presented at Medicare Coordinated Care Conference,sponsored by the Center for Medicare Advocacy, Inc., Washington, DC, March 21–22, 2002.

Berenson, Robert A., “Confronting the Barriers to Chronic Care Management in Medicare,”unpublished contractor report prepared for the NASI Study Panel on Medicare and ChronicCare in the 21st Century, Washington, DC, April 2002.

Bernstein, Jill, and Rosemary A. Stevens, eds., Medicare and the American Social Contract,Final Report of the Study Panel on Medicare’s Larger Social Role (Washington, DC: NationalAcademy of Social Insurance, February 1999).

Bernstein, Jill, and Robert D. Reischauer, eds., Restructuring Medicare: Next Steps, Report ofthe Medicare Steering Committee (Washington, DC: National Academy of Social Insurance,January 2000).

Blumenthal, David, Mark Schlesinger, and Pamela B. Drumheller, Renewing the Promise:Medicare and Its Reform (New York, NY: Oxford University Press, 1988).

The Board of Trustees, Federal Hospital Insurance and Federal Supplementary MedicalInsurance Trust Funds, 2002 Annual Report of the Board of Trustees of the Federal HospitalInsurance and Federal Supplementary Medical Insurance Trust Funds (Washington, DC: 2002).

Bodenheimer, Thomas, Edward H. Wagner, and Kevin Grumbach, “Improving Primary Carefor Patients with Chronic Illness, The Chronic Care Model, Part 2,” Journal of the AmericanMedical Association 288(15):1909–1914, October 14, 2002.

Bodenheimer, Thomas, “Disease Management: Promises and Pitfalls,” New England Journal ofMedicine 15(340):1202–1205, April 15, 1999.

Cassel, Christine K., Richard W. Besdine, and Lydia C. Siegel, “Restructuring Medicare for theNext Century: What Will Beneficiaries Really Need?” Health Affairs 18(1):118–131,January–February 1999.

References


Cohen, Harvey J., et al., “A Controlled Trial of Inpatient and Outpatient Geriatric Evaluationand Management,” New England Journal of Medicine 346(12):905–912, March 21, 2002.

Congressional Research Service (CRS), U.S. Congress, Medicare+Choice, prepared by Hinda R.Chaikind, Madeleine Smith, and Paulette C. Morgan, RL30702 (Washington, DC: January 22,2002).

Eggers, Paul W., “A Quarter Century of Medicare Expenditures for ESRD,” Seminars inNephrology 20(6):516–522, November 2000.

Eichner, June, “Coordination of Health Coverage for Medicare Enrollees: A Case Study ofHIV/AIDS,” National Academy of Social Insurance, Medicare Brief No. 7, 2001.

Ferguson, Jeffery A., and Morris Weinberger, “Case Management Programs in Primary Care,”Journal of General Internal Medicine 13(2):123–126, February 1998.

Foote, Sandra M., and Christopher Hogan, “Disability Profile and Health Care Costs ofMedicare Beneficiaries Under Age Sixty-Five,” Health Affairs 20(6):242–253, November–December 2001.

Fox, Peter D., Lynn Etheredge, and Stanley B. Jones, “Addressing the Needs of Chronically IllPersons Under Medicare,” Health Affairs 17(2):144–151, March–April 1998.

Fox, Peter D., et al., “The Medigap Reform Legislation: A Ten-Year Review,” presentation atthe National Health Policy Forum Session, October 2, 2002.

Fried, Leslie B., Director, Alzheimer’s Association Medicare Advocacy Project, Washington,DC, memorandum to the Secretary’s Advisory Committee on Regulatory Reform, May 9,2000.

General Accounting Office (GAO), U.S. Congress, Medigap: Current Policies Contain CoverageGaps, Undermine Cost Control Incentives, statement prepared by William J. Scanlon, GAO-02-533T (Washington, DC: March 14, 2002).

Gluck, Michael, and Paul Ginsburg, eds., From a Generation Behind to a Generation Ahead:Transforming Traditional Medicare, Final Report of the Study Panel on Fee-for-ServiceMedicare (Washington, DC: National Academy of Social Insurance, January 1998).

Gluck, Michael, and Marilyn Moon, eds., Financing Medicare’s Future, Final Report of theStudy Panel on Medicare’s Long-Term Financing (Washington, DC: National Academy ofSocial Insurance, September 2000).

Health Policy Alternatives, Inc., in collaboration with the National Chronic Care Consortium,Physician Reimbursement Issues (Washington, DC: March 2002).


61B u i l d i n g a B e t t e r C h r o n i c C a r e S y s t e m61

HMO Workgroup on Care Management, Improving the Care of Older Adults with CommonGeriatric Conditions (Washington, DC: AAHP Foundation, February 2002).

Hwang, Wenke, et al., “Out-Of-Pocket Medical Spending For Care of Chronic Conditions,”Health Affairs 20(6):267–268, November–December 2001.

Institute of Medicine (IOM), Crossing the Quality Chasm: A New Health System for the 21stCentury (Washington, DC: National Academy Press, 2001).

International Longevity Center, A National Crisis: The Need for Geriatrics Faculty Trainingand Development (New York, NY: October 2000).

Kaiser Family Foundation, Medicare Chart Book, second edition (Menlo Park, CA: fall 2001).

Kaiser Family Foundation, The Commonwealth Fund, and Health Research and EducationalTrust, Erosion of Private Health Insurance Coverage for Retirees: Findings from the 2000 and2001 Retiree Health and Prescription Drug Coverage Survey (Menlo Park, CA: April 2002).

King, Kathleen, Sheila Burke, and Elizabeth Docteur, eds., Matching Problems with Solutions:Improving Medicare’s Governance and Management, Final Report of the Study Panel onMedicare’s Governance and Management (Washington, DC: National Academy of SocialInsurance, July 2002).

Love, Jeffery, Medicare Prescription Drug Legislation in the U.S. Senate: Public Reaction AmongAmericans Aged 45 and Over (Washington, DC: AARP, August 8, 2002).

Medicare Payment Advisory Commission (MedPAC), Report to the Congress: Applying QualityImprovement Standards in Medicare (Washington, DC: January 2002a).

Medicare Payment Advisory Commission (MedPAC), Report to the Congress: Assessing MedicareBenefits (Washington, DC: June 2002b).

Moon, Marilyn, and Matthew Storeygard, One-Third at Risk: The Special Circumstances ofMedicare Beneficiaries with Health Problems (Washington, DC: The Commonwealth Fund,September 2001).

Moss, D. “Something Old, Something New: Viewing Social Security and Medicare Through aRisk-Management Lens,” unpublished draft, July 15, 1998. Quoted in Medicare and theAmerican Social Contract, National Academy of Social Insurance, 1999.

National Conference of State Legislatures (NCSL) “State Pharmaceutical Assistance Programs,”http://www.ncsl.org/programs/health/drugaid.htm, August 16, 2002, accessed September 5,2002.

Omenn, Gilbert S., “Prevention and the Elderly: Appropriate Policies,” Health Affairs9(2):80–93, summer 1990.


Partnership for Solutions, unpublished data prepared for the NASI Study Panel on Medicareand Chronic Care in the 21st Century, Baltimore, MD, 2001.

Partnership for Solutions, Chronic Conditions: Making the Case for Ongoing Care (Baltimore,MD: December 2002).

Safran, Dana G., et al., “Prescription Drug Coverage and Seniors: How Well Are States Closingthe Gap?” Health Affairs, Web Exclusive, July 31, 2002.

Schore, Jennifer L., Randell S. Brown, and Valerie A. Cheh, “Case Management for High-CostMedicare Beneficiaries,” Health Care Financing Review 20(4):87–101, summer 1999.

Stuck, Andreas E., et al., “Effects of Comprehensive Geriatric Assessment on Survival,Residence and Function: A Meta-analysis of Controlled Trials,” Lancet 342:1032–1036, 1993.

Super, Nora, “Medigap: Prevalence, Premiums, and Opportunities for Reform” National HealthPolicy Forum, Issue Brief No. 782, September 9, 2002.

Tapay, Nicole, and Gerry Smolka, Disabled Medicare Beneficiaries Under Age 65: A Review ofState Efforts to Provide Access to Medicare Supplemental Insurance (Washington, DC: AARP,1999).

Urban Institute, unpublished data prepared for the NASI Study Panel on Medicare and ChronicCare in the 21st Century, Washington, DC, 2002.

U.S. Department of Health and Human Services (DHHS), the Centers for Medicare &Medicaid Services (CMS), Challenges and Accomplishments of the Second Generation SocialHealth Maintenance Organization, prepared by Robert J. Newcomer, Charlene Harrington,and Robert Kane, Contract #500-93-0033 (Washington, DC: 2002a).

U.S. Department of Health and Human Services (DHHS), the Centers for Medicare &Medicaid Services (CMS), “End Stage Renal Disease (ESRD) Program,” website material,http://www.cms.gov/esrd/6.asp, August 31, 2002b.

U.S. Department of Health and Human Services (DHHS), the Centers for Medicare & Medi-caid Services (CMS), “Healthy Aging” website material, http://www.cms.gov/healthyaging,January 9, 2003.

U.S. Department of Health and Human Services (DHHS), Health Care FinancingAdministration (HCFA), A Profile of QMB-Eligible and SLMB-Eligible Medicare Beneficiaries,prepared by the Barents Group LLC, Contract #500-95-0057/Task Order 2 (Washington, DC:Barents Group LLC, April 7, 1999).


U.S. Department of Health and Human Services (DHHS), Health Care Financing Administra-tion (HCFA), Evaluation of the Program of All-Inclusive Care for the Elderly (PACE) Demonstra-tion: The Impact of PACE on Participant Outcomes, prepared by Abt Associates Inc., Contract#500-96-003/TO4 (Cambridge, MA: Abt Associates Inc., July 1998).

Vladeck, Bruce, “Round Pegs and Square Holes: Medicare and Chronic Care,” unpublishedcontractor report prepared for the NASI Study Panel on Medicare and Chronic Care in the 21stCentury, Washington, DC, April 2002.

Wagner, Edward H., Brian T. Austin, and Michael Von Korff, “Organizing Care for Patientswith Chronic Illness,” Milbank Quarterly 74(3):511–544, 1996.

Wagner, Edward H., et al., “Improving Chronic Illness Care: Translating Evidence IntoAction,” Health Affairs 20(6):64–78, November–December 2001.

Wolff, Jennifer L., Barbara Starfield, and Gerard Anderson, “Prevalence, Expenditures, andComplications of Multiple Chronic Conditions in the Elderly,” Archives of Internal Medicine162:2269–2276, November 11, 2002.


To order additional copies of this report or other Medicarereports, or to be included on our mailing list, please use thefollowing form:

Matching Problems with Solutions: Improving Medicare’s Governance and Management, July 2002, 79 pages, $15.00This report focuses on two tasks: examining whether a different governance structure might help the feder-al agency that runs Medicare, the Centers for Medicare & Medicaid Services (CMS), be more effective andidentifying ways in which current Medicare management could be improved. In part because some panelmembers had different views on whether Medicare should be restructured, the panel concentrated itsfocus on making the current Medicare program work better.

Reflections on Implementing Medicare (Second Edition), January 2001, 61 pages, $15.00 This report, originally printed in the spring of 1993, provides a historical reflection on the early days of aprogram enacted in 1965 to provide health care coverage to 19 million elderly Americans. This second edi-tion was released by the study panel on Medicare Management and Governance as part of theAcademy’s Restructuring Medicare for the Long Term Project. The report provides useful insights into theintentions of Medicare’s founders and an historical benchmark against which to gauge the program’s evo-lution over the last three-and-a-half decades.

Financing Medicare’s Future, September 2000, 101 pages, $15.00This report describes options for financing Medicare beneficiaries’ health care under several possible approachesfor changing the program’s structure and benefits. It is the final report of a nonpartisan study panel convened bythe National Academy of Social Insurance. The 12 members of the study panel represented a broad diversity ofphilosophical perspectives, disciplinary training, and professional experience.

Title Price Quantity Total

Medicare in the 21st Century: Building a Better Chronic Care System $15.00Matching Problems with Solutions: Improving Medicare’s Governance and Management $15.00Reflections on Implementing Medicare (Second Edition) $15.00Financing Medicare’s Future $15.00Restructuring Medicare: Next Steps $15.00

Medicare and the American Social Contract $15.00Structuring Medicare Choices $15.00

From a Generation Behind to a Generation Ahead: $15.00Transforming Traditional Medicare

Securing Medicare’s Future: What Are the Issues? $5.00

Subtotal

Shipping($3.00 per copy)

Grand Total

★ Please add me to your mailing list. Area(s) of interest:❒ Medicare ❒ Social Security ❒ Social Insurance Update (Academy Newsletter)❒ Workers’ Compensation Newsletter

Name___________________________________________________________________________________________________

Organization _____________________________________________________________________________________________

Address_________________________________________________________________________________________________

________________________________________________________________________________________________________

City/State/Zip ____________________________________________________________________________________________

Telephone _______________________________________________Fax ____________________________________________

❐ Check or money order enclosed, please make payable to the National Academy of Social Insurance

❐ Purchase Order # ______________________________________________________________________________________

❐ Credit Card ❐ VISA ❐ Mastercard Account #________________________________________Exp. ______________

Signature________________________________________________________________________________________________

Return this form with payment to: National Academy of Social Insurance 1776 Massachusetts Avenue, NW Suite 615Washington, DC 20036-1904 202/452-8097 ■ 202/452-8111 Fax ■ www.nasi.org

NATIONAL ACADEMY OF SOCIAL INSURANCE1776 Massachusetts Avenue, NW 202/452-8097

Suite 615 Fax: 202/452-8111

Washington, DC 20036 www.nasi.org

NA

SI M

edicare in the 21st Century: B

uilding a Better C

hronic Care System

■

January 2003

insert UNION BUG

building a better chronic care - national academy of ... · building a better chronic care system i...

Documents