Primary Care/Managed Care

Outpatient resource use varies substantially by provider and region in one state

The case mix of Medicaid patients seeking outpatient care is the major determinant of the medical resources (tests, medications, etc.) used to treat them. However, the type of health care provider and the geographic area also influence the use of medical services, according to a study supported by the Agency for Health Care Policy and Research (HS06170).

Jonathan P. Weiner, Dr.P.H., Barbara H. Starfield, M.D., M.P.H., and their colleagues at Johns Hopkins University and the Maryland Department of Health and Mental Hygiene used Maryland Medicaid claims files to study the practices of health care providers (office-based physicians, community health centers, hospital outpatient departments), who billed the State's Medicaid program for outpatient services during fiscal year 1988. At that time, Medicaid patients were treated on a fee-for-service basis and had no out-of-pocket expenses for deductibles or copayments.

After the researchers adjusted for differences in case mix, source of care alone was associated with variation in outpatient visits of about 20 percent, ancillary testing (for example, lab tests and x-rays) and prescription variation in the 50 to 60 percent range, and variation in hospital admission rates of about 80 percent. Patients whose usual source of care was a hospital outpatient department used $228 more in outpatient services and $375 more in overall services per year than those using a private doctor.

The researchers found considerable variation in resource use across Maryland's 24 counties. Visit rates were highest in the suburban counties, next highest in rural counties, and lowest in the central city of Baltimore. Counties with a greater supply of physicians had a somewhat higher visit rate. For example, in a county with a population of 50,000 persons, two additional physicians were associated with 0.2 more visits per year per patient. On the other hand, the greater the availability of hospital resources, the lower the visit rate, perhaps due to a substitution effect, according to the researchers.

At the individual practice level, however, geographic area had a negligible effect (less than 1 percent); patient characteristics explained up to 60 percent of the variation in outpatient medical resource use, and physician characteristics accounted for up to 17 percent. Family/general practitioners were the most efficient users of medical resources, using 7 percent less resources than general internists and 10 percent less than pediatricians.

See "Ambulatory care practice variation within a Medicaid program" by Dr. Weiner, Dr. Starfield, Neil R. Powe, M.D., M.P.H., M.B.A., and others, in the February 1996 issue of HSR: Health Services Research 30(6), pp. 751-770.

Papers published from HMO/health services research conference

Health maintenance organizations (HMOs) and government and academic health researchers often have different and sometimes opposing incentives to conduct and disseminate the results of health services research and apply its findings. Academic researchers focus on methodological rigor, and time is not usually a serious constraint. For HMOs, impeccable methodology may not be possible because of the need for timely results that may improve practice or reduce costs. The government, on the other hand, conducts and/or supports research for the "public good."

In an attempt to find common ground where industry and researchers can jointly improve health services research in the managed care setting, the Agency for Health Care Policy and Research and the Group Health Association of America (GHAA) have sponsored two conferences on "Building Bridges Between the HMO and Health Services Research Communities." The conferences brought together representatives of HMOs, the government, universities, and other organizations interested in conducting or funding health services research or using data from HMOs. The goals were to publicize the importance of producing generalizable, high-quality health services research for providers, payers, consumers, and policymakers and to create opportunities for collaborative activities among HMOs and the academic, government, and funding communities.

A recent supplement to Medical Care Research and Review contains nearly all of the papers presented at the first conference, held in San Diego, CA, in March 1995. The supplement was edited by Amy B. Bernstein, Ph.D., formerly of GHAA, and Jill Bernstein, Ph.D., and Terry Shannon, of AHCPR. It includes an overview by the three editors, a paper by AHCPR's Administrator, Clifton R. Gaus, Sc.D., on the benefits to be derived from collaboration between HMOs and health services research, a review article by Amy Bernstein and Jill Bernstein, and other papers on issues such as the internal economics of HMOs, perspectives on HMO research from key HMO groups, and research perspectives from academia.

The second conference was held in March 1996, also in San Diego, and focused on the use of health services research to improve health plan performance. A summary of the meeting is in preparation.

For more details on the first conference, see "Building bridges between the HMO and health services research communities," a special supplemental issue to Medical Care Research and Review 53, 1996, guest edited by Drs. Amy and Jill Bernstein and Terry Shannon.

Return to Contents
Proceed to Next Section