status, or risk factors such as cigarette smoking. Furthermore, databases lack information on medical conditions for which the patient does not seek treatment and may not list information on deaths that occur outside the hospital (although this can be overcome by use of links to diverse local vital records offices).
Clinical researchers may also find problems with these types of integrated databases when changes in HMO policies and procedures occur, which might affect patient care, outcomes, data collection, selection bias in the patient population (because of the nature of the HMO membership), disenrollment (which complicates long-term follow-up), and the lack of research infrastructures in HMOs. For example, a proposed research project on a rotavirus vaccine would require laboratory tests that the HMOs themselves would be unlikely to conduct. Other limitations might be posed by the nature of services provided by the HMO; that is, if its emphasis is on primary care, the patient population might be skewed away from those in need of specialty care, thereby limiting the types of data collected on the enrollee population. These limitations might have implications for the development of evidence-based research or guidelines. Concerns about the privacy of the medical record may also pose an impediment to the clinical researcher (although this is not restricted to managed care organizations). For example, the Mayo Clinic reports that privacy laws in Minnesota may prevent managed health care systems from conducting medical record reviews, and laws in other states may erect barriers that would prevent the use of patient records for research. The issue of medical privacy is complex and is being examined and considered by the U.S. Congress and the Secretary of the U.S. Department of Health and Human Services.
Presented by Walter Stature, M.D.
Head, Division of Allergy and Infectious Diseases, University of Washington School of Medicine
The trend toward managed care is often viewed as antithetical to research since managed care priorities often emphasize rapid patient throughput, low-cost care, limited laboratory testing, quotas for numbers of patient consults per physician, and lack of reimbursement for costly procedures. In reality, the managed care environment may be uniquely well suited for patient-oriented clinical research as well as for epidemiological and health services research. Unique advantages provided by large managed care organizations include well-defined, representative, and stable patient populations; the integration of inpatient and outpatient services within one system; centralized pharmacy services; centralized laboratory services; computerized and uniform databases, including databases containing medical, pharmacy, and laboratory records; and control of approaches to the delivery of health care at the individual practitioner level. These