UC Berkeley News
Press Release

UC Berkeley Press Release

Nationwide survey of medical groups defines high and low performers

– Not all medical groups are created equal, and relatively few perform well on all measures of quality performance for care of chronic illness, according to a new nationwide study of nearly 700 medical groups led by researchers at the University of California, Berkeley.

The study results, published in the August 2005 issue of the journal Medical Care Research and Review, scored medical groups with at least 20 practicing physicians on such parameters as clinical quality of care, financial performance and organizational learning.

For instance, researchers analyzed the medical group's use of evidence-based care management practices, use of electronic patient records, physician turnover and profitability.

"This is the first systematic review of medical group performance on a nationwide scale," said Stephen M. Shortell, UC Berkeley professor of health policy and management, dean of the School of Public Health and lead investigator of the study. "Prior to this study, relatively little empirical information was available comparing medical group performance on a number of key factors."

The researchers only studied chronic illness, noting that care for chronically ill patients make up the bulk of health care expenditures -- 75 to 80 percent -- in the United States. "Chronic illness, such as diabetes, heart disease, asthma and depression, exact a huge cost to the economy, not only in direct patient care, but in the loss of work days," said Shortell. "Finding ways to improve the management of chronic illness will go a long way in reducing those costs and improving the nation's health."

Data were obtained from interviews with chief executive officers, presidents or medical directors of physician organizations. Among the study's findings are:

.Only 11 percent, or 75 medical groups, scored in the top quartile on at least four out of six overall performance dimensions involving care management, health promotion, disease prevention, physician turnover, use of electronic health records and financial viability.

.High performing groups averaged 20 out of a possible 30 points on overall quality measures compared with 2.9 for the low performing groups.

.High performing groups averaged 10.6 out of 16 possible points for the use of disease registries, clinical guidelines, care management systems, feedback to physicians and patient self-management programs versus 1.0 for low performing groups.

.High performing groups averaged 9.8 out of 13 possible points for the use of health promotion and disease prevention services compared with only 1.1 for the low performing groups.

"We found a significant variance between the high and low performers," said Robin Gillies, UC Berkeley health policy researcher and co-author of the paper. "Medical groups that performed well scored approximately seven times higher than those that performed poorly. Moreover, only a small number of groups did well on all dimensions."

Similar disparities were found between high and low scoring groups for measures of physician turnover, the use of electronic health records and financial viability.

"We identified some key characteristics that separated the high performing from the low performing groups," said Gillies. "Groups that encouraged formal involvement in quality improvement, such as implementing a systematic method of measuring patient satisfaction and a requirement to report quality results to outside bodies were more likely to rank in the top 25 percent of almost all performance measures."

The researchers point out that the study findings are consistent with recent Institute of Medicine recommendations emphasizing the importance of organized approaches to quality improvement, and with the growing demand for external accountability and transparency in medicine. They say the results highlight the need for incentive policies to help medical groups improve.

"Pilot programs funded by the federal Centers for Medicare and Medicaid Services to reward medical groups and hospitals for higher quality of care are probably on the right track," said Shortell. "The problem is not the competence of individual doctors or nurses, but the system within which they practice. Medical care will improve when public policies address the underlying deficiencies in the system. This is beginning to occur, but we need to do more."

Other co-authors of the paper are Thomas G. Rundall, UC Berkeley professor of health policy and management; Lawrence Casalino, assistant professor of health studies at the University of Chicago; Thomas Bodenheimer, clinical professor of family and community medicine at UC San Francisco; and Julie Schmittdiel, Margaret C. Wang and Rui Li, former research assistants at UC Berkeley's School of Public Health.

This study was supported by the Robert Wood Johnson Foundation and the California Healthcare Foundation.