Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Nursing homes care for two distinct populations: custodial, predominantly Medicaid-financed residents and post-acute, predominantly Medicare-financed residents. Despite important differences in the needs of these two populations, quality of care within a facility has characteristics of a "public good" shared across all nursing home residents. The central implication of this observation is that policies directed at improving care for one group of residents may spillover to other residents. Indeed, recent work has found that the adoption of Medicare prospective payment for skilled nursing home care was associated with lower quality of care for long-stay (i.e., predominantly Medicaid) residents. Other work has shown that the generosity of Medicaid payments has implications for the quality of care received by all nursing home residents, including Medicare residents. The objective of this paper is to directly analyze the effects of Medicaid payment generosity and bed-hold policies (whether the state pays nursing homes to hold a Medicaid resident's bed during acute hospital admission) on the re-hospitalization of Medicare post-acute nursing home residents.

Based on Medicare inpatient claims for 2000, 2002, and 2004, we identified all discharges to nursing homes during each calendar year, as indicated by an available Minimum Data Set (MDS) assessment or a skilled nursing facility (SNF) Medicare claim within 30 days of hospital discharge. We tracked all SNF and MDS records to define a cohort of newly admitted nursing home residents within 30 days of hospital discharge in each year, with one record per person. The total number of observations (base discharges) annually was 927,858 in 2000, 897,288 in 2002, and 857,400 in 2004. We identified 30 day re- hospitalizations following discharge from the initial hospital stay from whence they entered a nursing home. We examined the pre-post difference in Medicare re-hospitalizations in states that changed their Medicaid policies relative to those states that did not undergo a change in these policies. Specifically, we used differences-in-differences regression model to control for potential selection biases.

Our preliminary (descriptive) analyses indicate that Medicaid payment policies are related to Medicare re- hospitalizations. In particular, there appeared to be significant variation in re-hospitalization rates in each year by whether the states paid nursing homes to hold the bed for a Medicaid resident who required admission to an acute hospital. In 2004, for instance, states with a bed hold policy (N=36) had a re- hospitalization rate of 18.9%, on average, compared to 16.7% in states without such policy (N=12).

These results suggest a connection between Medicaid payment for nursing home care and the outcomes of care for Medicare residents. However, neither Medicaid nor Medicare has an incentive to enact payment policies that recognize the welfare of residents covered by the other program. This lack of coordination may lead to increased program costs, a lack of care management, and poor quality of care.