more satisfied with costs in managed care plans compared to indemnity plans (Miller and Luft, 1994). Take the following example of joint pain and chronic arthritis, which are common conditions for many Medicare beneficiaries. A study comparing Medicare beneficiaries with chest and joint pain in risk HMOs with their counterparts in fee-for-service plans found that those in the HMO were less likely to be referred to a specialist and less likely to receive follow-up care. Although the outcomes were similar, the HMO enrollees experienced less alleviation of joint pain (Clement et al., 1994).

Other studies indicate that HMOs are not as proficient in some areas. A study conducted by Shaughnessy et al. in 1994 found that most home health care outcomes for individuals in fee-for-service plans were better than those for individuals in HMOs. HMO costs for home health care were significantly lower than the home health care costs incurred by fee-for-service plans. The approach taken by many HMOs was one of maintenance as opposed to rehabilitation or restoration.

Several studies, however, indicate that HMOs do some things very well. For example, the Group Health Cooperative of Puget Sound practices population-based medicine, an approach to providing clinical care, especially for patients with chronic conditions. The plan identifies enrollees by such characteristics as age, sex, health status, health complaints, and disease diagnoses. Once the subgroups have been identified, specific services and programs are developed for them.

Most of the studies assessing how Medicare enrollees have fared in managed care plans have involved staff and group model HMOs, which are different from the current and emerging independent practice associations, PPOs, and provider service networks. Several key studies have looked at a variety of HMOs, however (Miller and Luft, 1994; Brown et al., 1993b). The managed care environment and health care delivery models will continue to change and evolve. Given this, the fact that managed care has had little experience with the elderly, and the fact that there is little conclusive evidence on how managed care organizations manage and coordinate care for the frail elderly, there may be a need for studies to determine areas in which managed care does result in real improvements for the elderly