Posthospitalization home health care use and changes in functional status in a Medicare population.

Abstract

The objective of this work was to estimate the effect of Medicare beneficiaries' use of home health care (HHC) for 6 months after hospital discharge on the change in functional status over a 1-year period beginning before hospitalization. Data came from the Medicare Current Beneficiary Survey, which is a nationally representative sample of Medicare beneficiaries, in-person interview data, and Medicare claims for 1991 through 1994 for 2,127 nondisabled, community-dwelling, elderly Medicare beneficiaries who were hospitalized within 6 months of their annual in-person interviews. Econometric estimation with the instrumental variable method was used to correct for observational data bias, ie, the nonrandom allocation of discharged beneficiaries to the use of posthospitalization HHC. The analysis estimates a first-stage model of HHC use from which an instrumental variable estimate is constructed to estimate the effect on change in functional status. The instrumental variable estimates suggest that HHC users experienced greater improvements in functional status than nonusers as measured by the change in a continuous scale based on the number and mix of activities of daily living and instrumental activities of daily living before and after hospitalization. The estimated improvement in functional status could be as large as 13% for a 10% increase in HHC use. In contrast, estimation with the observational data on HHC use implies that HHC users had poorer health outcomes. Adjusting for potential observational data bias is critical to obtaining estimates of the relationship between the use of posthospitalization HHC and the change in health before and after hospitalization. After adjustment, the results suggest that efforts to constrain Medicare's spending for HHC, as required by the Balanced Budget Act of 1997, may lead to poorer health outcomes for some beneficiaries.