State Policies and Hospitalizations from Nursing Homes

Brown University, Providence RI

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Every 6 months, nearly 25 percent of nursing home residents are hospitalized in the U.S. This figure varies considerably within and between states. In spite of the high cost and iatrogenic problems associated with hospitalizing nursing home residents, and observed inter-state variation, there has been little systematic study of the influence of state policy on these rates and whether this influence may be differential for subgroups of vulnerable residents. Preliminary evidence suggests that states with low Medicaid nursing home payment rates tend to have higher hospitalization rates. A more complete analysis of how state policies affect the strategic clinical and management investment choices nursing homes make should inform the development of more coherent and equitable state and federal policies affecting this highly vulnerable population. Using MDS data and matched Medicare hospital claims for all long stay nursing home residents, in all non-hospital based facilities in the 48 contiguous states merged with facility-level Online Survey Certification Automated Records (OSCAR) data, market-level information from the Area Resource File, and data on state policies, we propose examining the effect of state policies on hospitalization as mediated by nursing homes? investments in medical and managerial resources. The specific aims are: (1) To characterize inter and intra state variation in the long-stay nursing home population, particularly the dually eligible population, in terms of patients? clinical conditions and their concentration. (2) To examine the relationship between state Medicaid nursing home policies and facilities? investment in medically relevant clinical and managerial infrastructure to care for long-stay Medicaid residents. (3) To model the unique association of facility and state-level factors with hospitalization events among long-stay nursing home residents. (4) Using the model developed in (3), to summarize the moderating effects of state Medicaid payment rates and policies on the relationship between facility context and hospitalization for specific sub-populations of long-stay residents: (4a) prevalence of cognitively impaired residents and/or availability of special dementia unit among cognitively impaired residents; (4b) prevalence of African Americans; and (4c) prevalence of dually eligible residents. The results of the proposed study should inform extant theories about how long term care providers respond to exogenous policy shocks, the relative competitiveness of the market and local resource constraints. These theoretical insights will help shape the policy implications emerging from the study.