CHRONIC DISEASE

MANAGEMENT
•
Chronic Disease
SPECIAL EDITOR

Edward H. Wagner, MD, MPH

Director, MacColl Institute for
Management:
Healthcare Innovation What Will It Take To Improve
Group Health Cooperative of
Puget Sound Care for Chronic Illness?
Seattle, Wash

Effective Clinical Practice.

eeting the complex needs of patients with chronic illness or impairment is
1998;1:2-4.
M the single greatest challenge facing organized medical practice. Usual care is
not doing the job; dozens of surveys and audits have revealed that sizable proportions
of chronically ill patients are not receiving effective therapy, have poor disease con-
trol, and are unhappy with their care (1). Results of randomized trials also show that
effective disease management programs can achieve substantially better outcomes
than usual care, the control intervention. These trials, along with the ideas and
efforts for improvement discussed in this issue, show that we can improve care and
outcomes. As the articles suggest, these improvements will not come easily.
If we are to improve care for most patients with chronic illness, the evidence
strongly suggests that we reshape our ambulatory care systems for this purpose. Pri-
mary care practice was largely designed to provide ready access and care to patients
with acute, varied problems, with an emphasis on triage and patient flow; short
appointments; diagnosis and treatment of symptoms and signs; reliance on laborato-
ry investigations and prescriptions; brief, didactic patient education; and patient-
initiated follow-up. Patients and families struggling with chronic illness have differ-
ent needs, and these needs are unlikely to be met by an acute care organization and
culture. They require planned, regular interactions with their caregivers, with a
focus on function and prevention of exacerbations and complications. This interac-
tion includes systematic assessments, attention to treatment guidelines, and behav-
iorally sophisticated support for the patient’s role as self-manager. These interactions
must be linked through time by clinically relevant information systems and continu-
ing follow-up initiated by the medical practice.

Comprehensive System Change

Many health plans and provider groups are beginning to understand the need to
improve their care of patients with chronic illness. Chronic disease management
programs tend to fall into two groups: targeting and case management (the larger
group) and comprehensive system change (the smaller group). The articles in this
issue argue that real improvement in outcomes will occur only when clinical sys-
tems reconfigure themselves specifically to address the needs and concerns of
chronically ill patients. Friedman and McCulloch and their colleagues describe two
important system improvement efforts to improve diabetes care, and Alderman
outlines the changes required to provide effective blood pressure treatment.
Although these articles differ somewhat, they share common features. On the basis
of our work at Group Health Cooperative and reviews of the literature, we devel-
oped a model for improving chronic illness care that incorporates these and other
successful interventions (2, 3) (Figure 1). The model suggests that the
patient–provider interactions resulting in care that improves outcomes are found
in health systems that:
• have well-developed processes and incentives for making changes in the care
delivery system

2 © 1998 American College of Physicians–American Society of Internal Medicine

•
 • assure behaviorally sophisticated self-man-
agement support that gives priority to
increasing patients’ confidence and skills so
that they can be the ultimate manager of their
illness (4) Community Health System
Resources and Policies Organization of Health Care
• reorganize team function and practice systems
Self-management Decision Delivery Clinical
(e.g., appointments and follow-up) to meet the Support Support System Information
needs of chronically ill patients Design Systems

• develop and implement evidence-based

guidelines and support those guidelines
Informed, Prepared,
through provider education, reminders, and Productive Proactive
Activated Interactions
increased interaction between generalists and Patient Practice Team
specialists
Functional and Clinical Outcomes
• enhance information systems to facilitate the
development of disease registries, tracking sys- FIGURE 1. Model for improvement of chronic illness care.
tems, and reminders and to give feedback on
performance.
Outcomes of Care Improvement
Because of the heavily preventive nature of high-
quality care for chronic illness, the system changes for Although improved chronic illness care can save money
this care shown in Figure 1 are similar to the 10 process- by reducing exacerbations and institutionalization,
es that improve prevention performance, as discussed by emphasizing cost reduction rather than quality im-
Solberg and colleagues in this issue. provement may be dangerous if it reduces access to
The model assumes that the locus of care remains effective services. This may be the case with several
with the personal physician, supported by an integrat- chronic illnesses. The best approach to cost savings is to
ed (and perhaps expanded) practice team. Targeting improve health status: that is, to ensure access to services
and case management activities, sometimes known as that are proven to improve outcomes. This may require
carve-ins or carve-outs, do not always make this an initially higher outlay of funds. Stroke care may illus-
assumption. In fact, many seem to operate on four trate this point: Evidence is mounting that admission to
major premises: special stroke inpatient units (5, 6) followed by intensive
rehabilitation results in much better outcomes. Such
1. Reduction in the cost of chronic illness is the
approaches seem to run counter to many case manage-
major goal and is assumed to be associated with
ment approaches, which encourage brief hospitaliza-
improvements in health.
tions and lower-intensity follow-up care (7). If case man-
2. The best way to achieve cost reduction is to focus agers were clinically sophisticated advocates for
on the highest-cost patients in the chronically ill evidence-based care, outcomes might improve. How-
population. ever, a recent review of case management programs (8)
suggests that most case managers do little more than
3. Primary care is not up to the task of chronic ill- review utilization.
ness care. Targeting is the other pillar of modern disease
4. Patients will do better if their chronic disease management; this approach correctly assumes that a
management is largely removed from primary care small percentage of the population accounts for most
and is delegated to a case manager. health care costs. But does it make sense to concentrate
on a small subset of a chronically ill population? It may
These premises need to be examined and con- not, for two reasons. First, health status changes fre-
firmed rigorously and quickly, because disease manage- quently among older and chronically ill adults, so that
ment programs based on these conceptual underpin- today’s high utilizers may not be next year’s. For exam-
nings are becoming the accepted standard. Regardless of ple, we have shown that a complex risk equation
whether these premises are true, such programs have involving past utilization and current health status
potentially serious adverse consequences and have not accounts for less than 70% of the variation in future
received sufficient attention. hospitalization rates among elderly persons (9). Thus,

Effective Clinical Practice ■ August/September 1998 Volume 1 Number 1 3

•
 most targeting schemes exclude a substantial percent- agers, hospital care to hospitalists, and preventive care to
age of persons at risk. Second, and more important, centralized systems.
almost all patients with major chronic illnesses, such as Our primary care system is increasingly being
asthma, AIDS, congestive heart failure, chronic absorbed by the care of older and chronically ill persons.
obstructive pulmonary disease, depression, diabetes, It will either change accordingly or risk being pulled
and epilepsy, benefit from periodic assessment of clini- apart because it cannot do the job well enough.
cal and psychosocial status, effective medical therapy, Everyone may suffer as a consequence.
greater confidence and skills in self-management, and References
sustained follow-up. Thus, improved chronic disease 1. Wagner EH. Managed care and chronic illness: health services
management should benefit most of the population research needs. Health Serv Res. 1997;32:702-14.
with a given chronic illness. As several articles in this 2. Wagner EH, Austin BT, Von Korff M. Improving outcomes
issue suggest, risk stratification may make good sense if in chronic illness. Managed Care Quarterly. 1996;4:12-25.
3. Wagner EH, Austin BT, Von Korff M. Organizing care for
it is accompanied by efforts to improve care for each
patients with chronic illness. Milbank Q. 1996;74:511-44.
stratum of health care utilizers. Systems that carve out 4. Von Korff M, Gruman J, Schaefer J, Curry SJ, Wagner EH.
the highest-utilizing patients for case management Collaborative management of chronic illness. Ann Intern Med.
without making efforts to improve care for the remain- 1997;127:1097-102.
der seem only to be interested in reducing cost, regard- 5. Indredavik B, Slørdahl SA, Bakke F, Rokseth R, Håheim LL.
less of their motivation. Stroke unit treatment. Long-term effects. Stroke. 1997;28:1861-66.
6. Ronning OM, Guldvog B. Stroke units versus general medical
wards, I: twelve- and eighteen-month survival: a randomized, con-
The Primary Care Team
trolled trial. Stroke. 1998;29:58-62.
7. Retchin SM, Brown RS, Yeh SC, Chu D, Moreno L.
How does separating chronic illness care from primary Outcomes of stroke patients in Medicare fee for service and man-
care affect the patient and the primary care team? aged care. JAMA. 1997;278:119-24.
Evidence to answer this question is scanty and urgently 8. Pacala JT, Boult C, Hepburn K, et al. Case management of
needed. The paper by O’Connor and colleagues in this older adults in health maintenance organizations. J Am Geriatr
issue provides some evidence of the importance of pro- Soc. 1995;43:538-42.
9. Coleman EA, Wagner EH, Grothaus LC, Hecht J, Savarino J,
fessional advice in patient decisions about an important Buchner DM. Predicting hospitalization and functional decline in
chronic disease intervention—aspirin treatment. But, older health plan enrollees: are administrative data as accurate as
some very successful chronic disease management pro- self-report? J Am Geriatr Soc. 1998;46:419-25.
grams seem to have involved the patient’s primary care 10. Rich MW, Beckham V, Wittenberg C, Leven CL, Freedland
team only tangentially (10, 11). However, these pro- KE, Carney RM. A multidisciplinary intervention to prevent the
readmission of elderly patients with congestive heart failure. N
grams included only a few hundred patients collected in
Engl J Med. 1995;333:1190-95.
a community or a hospital; therefore, their impact on 11. Stuck AE, Aronow HU, Steiner A, et al. A trial of annual
any particular primary care practice was probably mini- in-home comprehensive geriatric assessments for elderly people
mal. If many chronically ill patients in a practice are living in the community. N Engl J Med. 1995;333:1184-89.
cared for by a separate system, how does this affect the Correspondence
practice’s ability to care for the rest of its patients with Edward H. Wagner, MD, MPH, Director, MacColl Institute for
chronic illness and the intellectual and emotional Healthcare Innovation, Center for Health Studies, Group Health
rewards of practice? If trends continue, primary care Cooperative of Puget Sound, 1730 Minor Avenue, Suite 1290,
practice will increasingly resemble nonurgent emer- Seattle, WA 98101; e-mail: [email protected].
gency care, with chronic illness care left to case man- This paper is available at ecp.acponline.org.

4 Effective Clinical Practice ■ August/September 1998 Volume 1 Number 1

•