Hospital care varies widely for chronically ill Medicare patients

Washington -- Striking differences exist in the level and quality of care for chronically ill Medicare patients among some of the top U.S. hospitals and academic medical centers, according to two recently released Dartmouth Medical School studies.

The studies, conducted by researchers John Wennberg, MD, MPH, and Elliott Fisher, MD, MPH, utilize fee-for-service Medicare claims data.

Medicare beneficiaries with chronic illness who sought care in their last six months of life at Mount Sinai Medical Center in New York, for instance, spent nearly twice as many days in the hospital as those who went to a Mayo Clinic facility in Rochester, Minn., Dr. Wennberg said.

U.S. News and World Report ranked Mount Sinai the third best geriatric hospital and Mayo Clinic the sixth best in 2001, a period covered by the first of the two Dartmouth studies, both of which were published online Oct. 7 by the journal Health Affairs.

In terms of physician visits, chronic care seniors at first-ranked University of California at Los Angeles Medical Center experienced twice as many as those who went to fourth-ranked Duke University Medical Center in Durham, N.C., Dr. Wennberg reported.

Dr. Fisher's companion study, which, like the other report, adjusts for severity of illness, suggests that such increased use of hospital admittance and physician visits is not necessarily a good thing for geriatric patients, at least when it comes to those in the nation's nearly 300 academic medical centers. His investigation into initial hospitalizations of Medicare heart attack and colorectal cancer patients finds that the seniors exhibited a small but "statistically significant" increase in long-term mortality rates at the higher-intensity facilities.

The Dartmouth group, which for years has pointed to an inverse relationship between amount of utilization and health care quality at the community level, offered results at the hospital level for the first time. The researchers also contend that doctors' behavior plays an important role and merits further study.

"Practice styles among physicians, even those practicing at the same hospital, can differ, and knowledge of those differences can be useful for efforts to improve quality and efficiency in a given hospital," Dr. Wennberg said.

But such investigations might not prove useful if the main thrust of the original research is misdirected, said Albert Siu, MD, head of geriatrics at Mount Sinai. Quality assessments based on Medicare claims reports alone cannot take into account factors such as individual patient preferences, physician referral arrangements and availability of alternative treatment types -- all of which influence the intensity of care in a given setting, he explained.

"This was a good study looking at variations in care, but to draw conclusions about quality and efficiency was probably going beyond what the data showed," Dr. Siu said. His comments were echoed by the American Hospital Assn.

Dr. Siu suggested that the strategy employed in the Wennberg study is analogous to assessing the quality of obstetrical care by looking at the number of cesarean sections recorded in administrative reports, which do not say whether the patients had received prior C-sections or whether a breech presentation was involved.

At least one of the hospitals listed in the Dartmouth reports nevertheless plans to use them to look into its care utilization levels. Tom Rosenthal, MD, chief medical officer at UCLA, said analysis there would start with eliciting feedback on the studies from medical specialists who might have unique insight into the issue.

Such follow-ups could help determine whether anything about the high-use centers of care needs to be changed, Dr. Rosenthal said. If the Wennberg study has any shortcoming, he said, it is that focusing only on Medicare seniors who have died does not address the potential benefits of more intense care for other patients.

"The bigger criticism to me is that it's essentially all retrospective, and what you really don't know is what the outcomes of treatment are for the people who survived and what percentage of people who go into these systems survive," he said.

Slowing down excessive chronic care

The federal government already has begun formulating alternative treatment and payment models for what it views as inefficient and excessive care that is costing Medicare billions of extra dollars and lowering outcomes.

The Dartmouth studies are the clearest indication yet that the Medicare chronic care system needs revision, said Centers for Medicare & Medicaid Services Administrator Mark McClellan, MD, PhD. He cited disease management and pay-for-performance models as particularly promising avenues for change.

"We need to move from paying more for more complications and more services and higher costs to paying more for higher quality and lower overall costs," he said.

Disease management programs utilize private organizations to coordinate patients' care between primary care doctors, specialists and hospitals.

Demonstration projects utilizing the model already have shown their ability to save Medicare hundreds of dollars per beneficiary each year, Dr. McClellan said.

The Congressional Budget Office, however, recently analyzed peer-reviewed studies on existing disease management programs and concluded that the promise of better care at a lower price might not materialize.

"There is insufficient evidence to conclude that disease management programs can generally reduce overall health spending," the budget office states in a letter to Sen. Don Nickles (R, Okla.). "It is important to note that such programs could be worthwhile even if they did not reduce costs."

Several existing and upcoming demonstrations that provide financial incentives for health care professionals to coordinate more efficient and higher quality care show great potential, Dr. McClellan said. He stressed, however, that the growth of such incentives within Medicare would follow their success only on a smaller scale and likely would never account for more than 30% of total federal payments to doctors.

"I want to be very clear: Pay-for-performance isn't the answer to everything," the CMS chief said. "There are a lot of aspects of care that we don't have good performance measures for yet. There are a lot of features of care that we need to probably continue to reimburse directly."