How does the use of services vary among hospitals treating patients with severe chronic illnesses at the end of life?

Among chronically ill Medicare beneficiaries who received the majority of their care during 1999–2000 at 77 hospitals ranked as the best in America, there was striking variation in use of resources in the last six months of life, suggesting that where one receives care—more than the nature of one's illness—determines the amount of care that is provided.

Why is this important?

Medicare per capita spending among regions is closely correlated with the use of hospitals, intensive care units, and physician services in managing chronic illness such as congestive heart failure, chronic obstructive pulmonary disease (COPD), and solid tumor cancers (Wennberg et al. 2004a).

Findings

Among chronically ill Medicare beneficiaries receiving most of their care at 77 of America's best-ranked hospitals during 1999–2000, there was extensive variation in the amount of care provided to terminally ill patients in their last six months of life. (Rates of use were case-mix adjusted to control for differences in patients' age, sex, race, and disease comorbidities.) Specifically, rates of service use varied:

• more than 14-fold for cancer patients;
• more than seven-fold for COPD patients; and
• more than six-fold for congestive heart failure patients.

The frequency of services used by patients with one chronic disease was closely correlated with the frequency of services used by patients with other chronic diseases at a given hospital (Wennberg et al. 2004b).

Implications

These findings suggest that the hospital where patients are treated—rather than the nature of their illness—dictates the amount of care they receive. From the standpoint of mortality (all died within six months), more care did not equate to better care among these patients. To preserve Medicare's ability to provide universal access for seniors, delivery of effective care in a more efficient manner is paramount.

Improvement Ideas and Resources

Health care providers require information on their performance over time to assess their efficiency and identify areas where intervention is needed. This study shows how Medicare claims can be used to measure population-based, provider-specific use of resources for patients enrolled in traditional fee-for-service Medicare.

Medicare is testing pay-for-performance initiatives that link payment to quality. The Physician Group Practice Demonstration, for example, will reward physicians for improving the quality and efficiency of health care services delivered to Medicare fee-for-service beneficiaries, especially patients with chronic illness who account for a significant proportion of Medicare expenditures (CMS 2005).

Measure:

This study identified individuals who were enrolled in Medicare Part A and Part B for at least two years before their death and who had at least one nonsurgical hospital admission for one of 11 chronic conditions in the last two years of life. Each patient was assigned to the hospital that the patient most frequently used in the last two years of life; ties were decided in favor of the hospital discharge closest to the date of death. The final analysis included 90,616 patients who died in 1999–2000 and most frequently used one of 77 hospitals listed in U.S. News and World Report's 2001 rankings of the best U.S. hospitals for geriatric care and care of heart and pulmonary diseases. Utilization rates were adjusted to control for differences in patients' age, sex, race, and clinical comorbidities. To control for differences in the severity of illness, the analysis was restricted to care delivered in the last six months of life. Results focused on patient cohorts with solid tumor cancer, chronic obstructive pulmonary disease, or congestive heart failure. Utilization rates were based on "the total experience of the cohort, not just on services provided by the index hospital and associated providers. However, since the percentage of total hospital care provided by the index hospital is high, the variations in illness-adjusted use of care primarily reflect clinical choices made by physicians associated with that hospital" (Wennberg et al. 2004b).

Limitations:

These data do not describe processes of care or patient preferences.

Source:

Data from the Medicare Denominator File and the Medicare Provider Analysis and Review (MedPAR) File were analyzed by researchers at the Dartmouth Medical School, the Veterans Affairs Outcomes Group, and the Institute for Clinical Evaluative Sciences (Wennberg et al. 2004b).

References:

* Indicates source of data used in the chart(s).

CMS (Centers for Medicare and Medicaid Services). 2005. Medicare Physician Group Practice Demonstration. Baltimore, Md.: U.S. Department of Health and Human Services.

Wennberg, J. E., E. S. Fisher, and J. S. Skinner. 2004a. Geography and the Debate over Medicare Reform. Health Affairs (Millwood) Suppl Web Exclusive: W96–114.

* Wennberg, J. E., E. S. Fisher, T. A. Stukel et al. 2004b. Use of Medicare Claims Data to Monitor Provider-Specific Performance Among Patients with Severe Chronic Illness. Health Affairs (Millwood) Suppl Web Exclusive: VAR5–18.