The Emerging Challenge of Chronic Care

THE URBAN INSTITUTE

The Emerging Challenge of Chronic Care

Robert A. Berenson, M.D.

Senior Fellow, The Urban Institute

27 September, 2007

THE URBAN INSTITUTE

Chronic Condition

• An illness, functional limitation or cognitive impairment that lasts (or is expected to last) at least one year

• Limits what a person can do

• Requires ongoing care

Source: National Academy of Social Insurance, “Medicare in the 21 st Century: Building a Better Chronic Care System,” January 2003.

THE URBAN INSTITUTE

Projected Total Number of People

With Chronic Conditions

118125

133141

149

157164

171

100

120

140

160

180

1995 2000 2005 2010 2015 2020 2025 2030

(in millions)

Sources: Partnership for Solutions. “Multiple Chronic Conditions: Complications in Care and Treatment”; RAND Corporation, 2000.

THE URBAN INSTITUTE

Chronic Conditions by Age Group

24%

38%

62%

84%

5%

13%

62%

35%

0%

20%

40%

60%

80%

100%

0-19 20-44 45-64 65+Ages

Per

cent

of

Pop

ulat

ion

One or MoreChronicConditions

Two or MoreChronicConditions

Source: Partnership for Solutions. “Disease Management and Multiple Chronic Conditions”; Agency for Healthcare Research and Quality, MEPS, 1998.

THE URBAN INSTITUTE

Chronic Condition Prevalence By Race (Total Population)

65.7%70.6%

57.8%

20.6%21.4%24.2%

5.3%7.1%10.2%

3.6%5.9%7.8%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Caucasian African-American Hispanic

0 Conditions1 Condition2 Conditions3+ Conditions

Source: Hwang, W., et al., “Out-of-Pocket Medical Spending for Care of Chronic Conditions,” Health Affairs, December 2001.

THE URBAN INSTITUTE

Proportion of Adults 50+ with Chronic Conditions, by Race

42

64

68

77

0 10 20 30 40 50 60 70 80 90

Asian American

White

Latino

Africa-American

Source: “Cultural Competence in Health Care,” Center on an Aging Society, Georgetown University. No. 5, February 2004.; K. Collins, et al., “Diverse Communities, Common Concerns; Assessing Health Care Quality for Minority Americans,” New York: The Commonwealth Fund, 2002.

THE URBAN INSTITUTE

Chronic Conditions for Children

65

70

65

60

64

31

19

26

34

31

4

8

7

5

5

0% 20% 40% 60% 80% 100%

Disorders of Teeth and Jaw

Eye Disorders

Preadult Disorders

Asthma

Upper Respiratory Disease

Single Condition Condition +1 Condition +2

Source: G. Anderson, “Hospitals and Chronic Care”, PowerPoint Presentation to the American Hospital Association. Partnership for Solutions. 16 June 2004.

THE URBAN INSTITUTE

Chronic Conditions for Adults

26

30

30

46

42

25

29

30

26

28

20

18

16

14

14

13

11

11

7

8

16

13

8

7

13

0% 20% 40% 60% 80% 100%

Arthritis

Hypertension

Mental Conditions

Upper RespiratoryDisease

Chronic RespiratoryInfection

Single Condition Condition +1 Condition +2 Condition +3 Condition +4+


THE URBAN INSTITUTE

Chronic Conditions in Seniors

11

17

10

9

8

22

24

21

23

22

23

23

25

25

25

22

20

24

22

22

21

16

19

19

19

0% 20% 40% 60% 80% 100%

Arthritis

Hypertension

Heart Disease

Eye Disorders

Diabetes

Single Condition Condition +1 Condition +2 Condition +3 Condition +4+


THE URBAN INSTITUTE

Multiple Chronic Conditions and Medical Service Usage

82%

69%

55%50%

0%

20%

40%

60%

80%

100%

Home HealthVisits

PrescriptionDrugs

InpatientStays

PhysicianVisitsP

erce

nt o

f Se

rvic

es U

sed

by P

eopl

e w

ith

Mul

tipl

e C

hron

ic C

ondi

tion

s

Source: G. Anderson, “Hospitals and Chronic Care”, PowerPoint Presentation to the American Hospital Association. Partnership for Solutions. 16 June 2004.; MEPS 2000.

THE URBAN INSTITUTE

Hospitalizations by Number of Chronic Conditions

4%8%

12%17%

22%

32%

0%

10%

20%

30%

40%

50%

0 1 2 3 4 5+

Number of Chronic Conditions

Per

cen

t of

Peo

ple

wit

h I

np

atie

nt

Hos

pit

al S

tays

Source: G. Anderson, “Hospitals and Chronic Care”, PowerPoint Presentation to the American Hospital Association. Partnership for Solutions. 16 June 2004.; MEPS 2000.

THE URBAN INSTITUTE

Hospitalizations for Ambulatory Care Sensitive Conditions

261236

219

9562

361870

169

131

0

50

100

150

200

250

300

0 1 2 3 4 5 6 7 8 9 10+


Hos

pit

aliz

atio

ns

per

100

0 M

edic

are

Ben

efic

iari

es

Sources: Partnership for Solutions. “Multiple Chronic Conditions: Complications in Care and Treatment,” May 2002; Medicare Standard Analytic File, 1999.

THE URBAN INSTITUTE

Activity Limitations by Number of Chronic Conditions

67%

52%

43%

28%

15%

4%

0%

20%

40%

60%

80%

0 1 2 3 4 5+

Chronic Conditions

Per

cen

t w

ith

Act

ivit

y L

imit

atio

ns


THE URBAN INSTITUTE

Annual Prescriptions by Number of Chronic Conditions

0

10

20

30

40

50

0 1 2 3 4 5


Ave

rage

An

nu

al

Pre

scri

pti

ons*

*Includes Refills

Sources: Partnership for Solutions, “Multiple Chronic Conditions: Complications in Care and Treatment,” May 2002; MEPS, 1996.

3.7

10.4

17.9

24.1

33.3

49.2

THE URBAN INSTITUTE

Utilization of Physician Services by Number of Chronic Conditions

7.811.3

14.9

19.5

37.1

13.8

8.16.55.24.01.3 2.0

0 1 2 3 4 5+


Unique Physicians

Physician Visits

Sources: R. Berenson and J. Horvath, “The Clinical Characteristics of Medicare Beneficiaries and Implications for Medicare Reform,” prepared for the Partnership for Solutions, March, 2002; Medicare SAF 1999.

THE URBAN INSTITUTE

Breakdown of Total Health Care Spending

78% Health Care Spending for People with Chronic Conditions

22% Health Care Spending for

People without Chronic Conditions

Sources: Partnership For Solutions, “Chronic Conditions: Making the Case for Ongoing Care,” December 2002; MEPS, 1998.

THE URBAN INSTITUTE

Health Care Spending by Number of Chronic Conditions

$11,500

$8,900

$5,600

$3,400

$1,900

$800

$0 $2,000 $4,000 $6,000 $8,000 $10,000 $12,000 $14,000

5+

4

3

2

1

0

Nu

mb

er o

f C

hro

nic

Con

dit

ion

s

Average Per Capita Health Care SpendingSources: Partnership For Solutions. “Disease Management and Multiple Chronic Conditions”; Agency for Healthcare Research and Quality, MEPS 1998.

THE URBAN INSTITUTE

Medicare Spending Related to Chronic Conditions

22.1%

0.9%

15.1%

3.5%

16.3%

6.8%

14.8%

10.3%

11.3%

12.7%

20.3% 65.8%

Percent of MedicarePopulation

Percent of Medicare Spending

5+ Conditions

4 Conditions

3 Conditions

2 Conditions

1 Condition

0 Conditions

Source: Partnership for Solutions, “Medicare: Cost and Prevalence of Chronic Conditions,” July 2002; Medicare Standard Analytic File, 1999.

THE URBAN INSTITUTE

Medicare Spending on Beneficiaries with Chronic Conditions

4 Chronic Conditions

12%

5+ Chronic Conditions

68%


10%

1 Chronic Condition

3%0 Chronic Conditions

1%


6%


THE URBAN INSTITUTE

Growth of Medicaid Spending

$73

$168

$49 $54$60

$34

$142

$124$120

$91

$0

$50

$100

$150

$200

1992 1995 1997 1998 2000

In B

illi

ons

Disabled Beneficiaries All Beneficiaries

Sources: J. Crowley and R. Elias. “Medicaid’s Role for People with Disabilities,” The Kaiser Commission on Medicaid and the Uninsured, August 2003; Urban Institute estimated based on HCFA-2082 and HCFA-64 Reports.

THE URBAN INSTITUTE

Projected Total Medicaid Spending Per Enrollee

$16,300$17,200

$1,400$2,300

$11,200$12,300

$2,000$3,200

Children Adults Disabled Elderly

FY 2001FY 2006

Note: Includes federal and state spending on benefits.

Sources: J. Crowley and R. Elias. “Medicaid’s Role for People with Disabilities,” The Kaiser Commission on Medicaid and the Uninsured, August 2003; KCMU analysis based on CBO baseline for Jan. 02.

THE URBAN INSTITUTE

Private Health Insurance Spending on Individuals with Chronic Conditions


13%


15%


14%

5+ Chronic Conditions

31%


13%

1 Chronic Condition

14%Source: G. Anderson, “Hospitals and Chronic Care”, PowerPoint Presentation to the American Hospital Association. Partnership for Solutions. 16 June 2004.; MEPS 2000.

THE URBAN INSTITUTE

Incidents in the Past 12 Months

1. Been told about a possibly harmful drug interaction

2. Sent for duplicate tests or procedures

3. Received different diagnoses from different clinicians

4. Received contradictory medical information

Sometimes or often

54%

54%

52%

45%

Among persons with serious chronic conditions, how often has the following happened in the past 12 months?

THE URBAN INSTITUTE

Barriers to Improvement

THE URBAN INSTITUTE

Barriers to Implementing Change in Most of Medicare

• The nature of medical education and the resultant professional culture and orientation of clinical practices

• Traditional Medicare is based in traditional indemnity insurance

• Major benefit limitations and restrictions in the Medicare statute

THE URBAN INSTITUTE

Professional Issues

• Hard to influence by public policy• Based on an orientation to identifying and caring

for acute illnesses and injuries, not chronic conditions

– “find it and fix it” – solve, rather than manage problems– “the tyranny of the urgent”– Failure to find the unusual and the life-threatening is

worse than overlooking the common and considering quality of life

THE URBAN INSTITUTE

Professional Issues (cont.)

• Oriented to those who present for care, rather than to populations who inhabit their chronic conditions

• Little division of labor – M.D. as captain of the ship

• Underuse of information management and decision support tools

• Resistance to change, even in the face of demonstrable failures

THE URBAN INSTITUTE

Specific Structural and Organizational Deficiencies

• Residency training takes place in hospitals• Shortage of geriatricians• Guidelines (even when followed) usually ignore co-

morbidities – may conflict or produce overwhelming compliance burden

• Disease management and primary/principal care are not well coordinated

• Lack of integrated care orientation (also fostered by siloed payment systems)

THE URBAN INSTITUTE

Medicare Statute Based on Indemnity Insurance of the ’60s

• Kenneth Arrow in 1963: for people with chronic illness, “insurance in the strict sense is probably pointless.”

• Why? Moral hazard• Yet, 80% of beneficiaries have one or more

chronic condition and 20% have 5 or more and account for two-thirds of program spending

THE URBAN INSTITUTE

Example of the Problem: Should Medicare Pay for E-mails?

• Why not phone calls, while you’re asking?

• In a fee-for-service payment system, there are a number of concerns:

– Relatively high transaction costs relative to the value of the underlying service

– Substantial program integrity concerns

– “Nuclear force” moral hazard

THE URBAN INSTITUTE

Problems in How Traditional Medicare Pays for MD Services

• Many Medicare payment systems have evolved from FFS to prepayment for episodes of care – physician payments is the main exception

• Physician payment is for discrete, narrowly defined services or transactions

• Partly fails to account for complexity• Pays based on resources expended, whether serve a

useful purpose or not• And doesn’t pay differently for quality

THE URBAN INSTITUTE

Medicare Benefits Need to Be Improved and Upgraded

• Now, reasonable coverage for prescription drugs (although still 4 million not in)

• Sensory loss support devices not covered (eyeglasses, hearing aids)

• DME and home health limitations, e.g., the “homebound” definition

• Program interpretation that rehabilitation services require prognosis of improvement, and not maintenance or slowed deterioration

THE URBAN INSTITUTE

Various Models of Enhanced Chronic Care Management

THE URBAN INSTITUTE

Disease Management

• I use the term to refer to third parties attempt to influence patients directly, bypassing physicians

• Relies on predictive modeling, decision-support software, and remote monitoring devises to complement core nurse-patient communication, which focuses on patient self-management (diabetes) and early detection of clinical deterioration (CHF)

THE URBAN INSTITUTE

Case Management

• Targeted to a subset of patients who are typically the most complex – with a combination of health, functional, and social problems

• Approach is more customized to needs of particular patients

• Relies mostly on telephonic interventions

THE URBAN INSTITUTE

The Wagner Chronic Care Model

• Pioneered by Wagner and associates at Group Health Cooperative of Puget Sound and The MacColl Institute

• Offers a multidimensional approach to a complex problem

• Identifies 6 essential elements: community resources, health care organization, self-management support, delivery system redesign, decision support, clinical information systems

THE URBAN INSTITUTE

Delivery System Redesign

• Specialized assessment tools to identify patients at risk

• Multi-professional team responsibility and delineation of roles

• Active promotion of patient self-management

• Proactive follow-up/communication, outside of the anachronistic office visit

THE URBAN INSTITUTE

Chronic Care Strategies That Bypass Physicians Make No Sense

• From 30 years of Medicare demos -- approaches

that are supplemental to the patient/physician relationship have had little impact – the MMA disease management demo seems to be failing; in commercial and Medicaid settings D.M. may have some, but limited, usefulness.

• In contrast, CMS just announced modest positive results from the Medicare physician group practice demo, which incentivizes, rather than bypasses, practices – mostly, but not only, large groups

THE URBAN INSTITUTE

Challenging the Status Quo in Chronic Disease Care: Seven Case

Studies Robert A. Berenson, M.D.

September, 2006

Available on California Health Care Foundation website

THE URBAN INSTITUTE

Seven Case Studies

• Sutter Health Sacramento Sierra Region• Park Nicollet Health Services• Integrated Resources for Middlesex Area (Ct.) • Billings Clinic• Care Level Management• Washington Hospital Center Medical House Call • MDxL

THE URBAN INSTITUTE

Case Study Finding 1

• Physicians and hospitals can do much more to manage patients with chronic conditions

• Physicians and hospitals do not think third-party disease and case management has worked because of the absence of physician engagement

THE URBAN INSTITUTE

Finding 2

• Viable models of chronic care management fall between the Chronic Care Model and third-party approaches

• Case study sites do not attempt to redesign traditional practice of frontline primary care physicians

THE URBAN INSTITUTE

Finding 3

• Although third-part D.M. remains the dominant framework for chonic care improvement, some health plans also support innovative approaches that more closely relate to patients’ regular sources of care

THE URBAN INSTITUTE

Finding 4

• Provider-based programs carefully distinguish among patients based on their specific clinical conditions and other assessments

• Differentiators include: whether patient home-bound, have limitations in activities of daily living, and specific conditions, e.g. CHF vs. diabetes vs others

THE URBAN INSTITUTE

Finding 5

• Approaches to case management for medically complex patients vary more than do disease management programs for patients with one or more specific chronic conditions

• For the former, programs rely more on point of care decision-making by clinicians

THE URBAN INSTITUTE

Finding 6

• Capitation is more compatible with chronic care programs and their populations than fee-for-service reimbursement

• Capitation provides greater flexibility and organizations can benefit from reduced expenditures

• The Medicare “shared savings” approach used in the PGP demo also may be a practical approach

THE URBAN INSTITUTE

Finding 7

• Current Medicare payment rules greatly influence the configuration of chronic care programs, e.g., how to get reimbursed for diabetes education or the “incident to” rules.

THE URBAN INSTITUTE

Finding 8

• The negative business case for hospitals to support chronic care management does limit the robustness of programs

• However, in some circumstances, there are offsets to the negative ROI

THE URBAN INSTITUTE

Finding 9

• Communications, monitoring, and data-sharing technologies enhance chronic care programs but, state-of-the-art, “high tech” technologies are not essentail.

• EMRs, disease registries, PDAs, yes

• Sophisticated telemonitoring devices, not really

THE URBAN INSTITUTE

Some Final Thoughts on Physician Payments to Support All of This

THE URBAN INSTITUTE

We Should Not Expect Pay-for-Performance to Solve the Problem

• It focuses on marginal dollars and ignores the incentives in the basic payment system -- which drive behavior

• A lot of what we want physicians to do is not easily measurable. Are we looking under the light for the keys lost in the bushes?

• P4P can’t easily address “overuse” and “misuse” quality dimensions, much less cost.

• We are still learning about P4P. Don’t overload it.

THE URBAN INSTITUTE

The Bottom Line

• A one-size fits all, RBRVS fee schedule no longer makes sense as physicians increasingly do very different things

– Perhaps, PCPs need mixed FFS and prospective monthly payments (with a dash of P4P)

– Surgeons could be paid for episodes (but addressing the bias to inappropriate surgical episodes)

– Other specialists who perform one-time, discrete services might still be paid FFS for their services

• The payment system should promote integrated care, including multi-specialty groups, but not single specialty consolidation

THE URBAN INSTITUTE

Continuum of Approaches for Paying for “Medical Home” Services

• Aggressive, politically difficult RBRVS/fee schedule revaluations

• New CPT codes for targeted medical home activities

• A new payment, i.e. pmpm or pppm, for chronic care management activities to the practice on top of FFS payments

• Bundled payment for medical services and medical home activities – either a more improved pmpm or a hybrid FFS/bundled payment approach

THE URBAN INSTITUTE

FFS Revaluations

• Hope that better payment for E&M services cross-subsidizes medical home activities (as some are already included in pre and post service work, according to the RBRVS methodology

• Avoid difficult design issues of a formal medical home --

• Who qualifies for payment, e.g. primary care or principal care?

• The physician or the practice? • Is there a formal patient lock-in – hard or soft?• No obligation to hold any one accountable and all that

that entails

THE URBAN INSTITUTE

FFS Revaluations -- Cons

• No obligation to hold any one accountable and all that that entails – in a FFS system, it might be putting good money after bad

• Politically difficult to redistribute within a fee schedule context

• A CPT code based payment system that pays for specific services cannot really accommodate the set of “soft” activities we want to promote

THE URBAN INSTITUTE

New CPT Codes for Particular Medical Home Activities

• Or particular services in the Chronic Care Model• As examples, palliative care family conferences,

“email consultations,” geriatric health assessment• These should be included in CPT and paid for,

but can’t really include most medical home or care coordination activities on a FFS payment basis, as discussed before

• Even here, face political obstacles to adoption from vested interests who are involved in CPT

THE URBAN INSTITUTE

PPPM Payment for Medical Home and/or Chronic Care Management

• Assumes there is a definable and designated subpopulation that “qualifies” for additional activities supported with additional payment

• Would small practices reengineer their processes for a small subset of patients which may make up a highly disproportionate share of health spending but not a relatively small share of their time and attention?

• Compounded if not an all-payer approach

THE URBAN INSTITUTE

An Add-on PPPM Payment (cont.)

• Which raises the fundamental question, do all patients benefit from a medical home or should the approach be targeted to only some, for efficiency?

• How would eligible patients be selected – physician referral (then self-referral issues), history of high costs, data mining re conditions and co-morbidities – the issues that are relevant to eligibility for case management?

THE URBAN INSTITUTE

Bundled (“Capitated”) Payments for All

Services and All Patients or a FFS Hybrid

• The advantage is that all patients are included, so no practice dissonance for different patients and risk adjustment handles the fact that different patients have different needs for chronic care management

• But should medical home services be provided to everyone? Do they all want a home? Is this efficient? (But some of us think FFS sends wrong signals for all patients)

• Can we correct the execution errors of 1990s capitation approaches related to: insurance risk, absence of risk adjustment, mechanical actuarial conversion of pmpms under FFS to a situation when more is expected of the practice?

THE URBAN INSTITUTE

A FFS/Bundled Payment Hybrid

• Some very smart people, e.g., Joe Newhouse, have recommended a mixed approach to soften the effects of capitation and FFS payment incentives

• Some European primary care payment models, e.g. Denmark, is a hybrid

• But surely more complex operationally for the payer and maybe the practice and may negate some of the appeal of bundled/“capitated” payments

Documents

The Emerging Challenge of Chronic Care