proached a level of significance that indicated a relationship may exist" (Langwell and Hadley, 1989, p. 71). Also looking at the Medicare Competition Demonstrations, Retchin and associates concluded that "although disenrollees were more likely to experience functional declines (in one or more ADLs), at follow-up, these declines were of marginal statistical significance and most likely due to baseline differences" (Retchin et al., 1992, p. 665).
Several years ago, Fox and colleagues (1991) recognized that there was a dearth of research about the delivery of services to older persons in managed care. They and others have proposed research agendas to fill this gap (e.g., Davis et al., 1994; Lewin/ICF, 1990). Although interest in managed care research has grown considerably, results and conclusions remain sparse. Managed care is still evolving. Without firm and consistent evidence, it would be premature to make judgments about the effectiveness of current practices. The term "best practices" should be reserved for those that have been tested and evaluated on the basis of the quality and cost-effectiveness of the care provided.
Nevertheless, many managed care plans are actively involved in demonstrating different approaches to the provision of care of their vulnerable populations (see, e.g., Group Health Association of America, 1995a; Health Care Financing Administration, Division of Policy and Evaluation, Office of Managed Care, 1994). These might be called "promising interventions" or "models of care" (Teresa Fama, Deputy Director, Chronic Care Program, Washington, D.C., personal communication, November 1995). As Medicare enrollment continues to increase, plans that have heretofore had few older, disabled, or chronically ill members will be confronted with the greater diversity of needs of such individuals. Undoubtedly, they will look to current practices, as well as experiment with other interventions to care for their Medicare enrollees.
Pawlson (1994) has suggested that the "acute disease paradigm" (an isolated event with a single, proximate cause) that has been the hallmark of the U.S. health care system is no