Objective: To compare two-year changes in health and functional status among a national sample of elderly Medicare beneficiaries in risk-based HMOs and in fee-for-service.
Data source: Medicare Current Beneficiary Survey (MCBS) for 1995 to 1998. MCBS is a longitudinal, in-person survey of a nation ally representative sample of the Medicare population. Respondents are asked once a year about health and functional status.
Study design: Noninstitutionalized respondents in the fall 1995 and fall 1996 rounds of MCBS for whom follow-up data were available two years later were included. Of the sample, 1,126 members were in HMOs and 3,449 were in fee-for-service at baseline. Ordered logistic regression and binary logistic regression were used to examine the association of HMO membership at baseline with each measure of health or functional status at follow-up, controlling for baseline health and functioning and other variables.
Principal findings: No significant differences were found between HMO and fee-for-service respondents with respect to two-year changes in health and functional status, either for the full sample or for subsets of chronically ill respondents. there was little ef f ct on the findings when the analysis was restricted to persons who remained in the same care system (HMO or fee-for-service) between baseline and follow-up. Separate analyses of the 1995-97 and 1996-98 cohorts revealed some in consistencies between the cohorts with respect to patterns of change in health and functioning.
Conclusion: There was no consistent evidence of an "HMO effect " on the health and functioning of enrollees over a two-year period.