Expenditures on services for persons with acquired immunodeficiency syndrome under a Medicaid home and community-based waiver program. Are selection effects important?

OBJECTIVES: In 1990, the state of Florida implemented an acquired immunodeficiency syndrome (AIDS)-specific Medicaid waiver program to provide home and community-based services to AIDS patients as an alternative to institutional care. The program is available to Medicaid beneficiaries with AIDS who are at risk of institutionalization. This study examines whether the waiver option was effective in reducing Medicaid expenditures per beneficiary during its first 2 years of operation.
METHODS: The authors used Medicaid claims data and county information on the availability of health services to model the selection of the waiver option by AIDS patients and then to estimate the effect of the waiver on expenditures controlling for nonrandom program selection.
RESULTS: The results indicate that the selection model is highly significant, but that the influence of nonrandom selection on the estimation of the program effects is negligible. More importantly, the regression results indicate that persons with AIDS who use waiver services incur monthly Medicaid expenditures that are on average 22% to 27% lower than otherwise similar nonparticipants.
CONCLUSIONS: These results, based on the first 2 years that Project AIDS Care was operational, suggest that home and community-based care for AIDS patients results in lower expenditures per beneficiary.