While age-adjusted mortality rates in rural and urban counties in the United States were similar during the 1970s and 1980s, over the last two decades mortality rates declined at a much faster rate in urban counties, resulting in a widening rural-urban survival gap. The factors contributing to this growing disparity remain largely unidentified, despite the fact that more than 23% of the fee-for-service (FFS) Medicare population resides in non-metropolitan, rural counties. This application explores how much of this gap can be explained by differences in how post-acute care (PAC) is delivered. Very little is known about discrepancies in the types of PAC resources available to Medicare beneficiaries living in urban and rural areas, such as the type of PAC setting in which they receive their care, the quality of that PAC provider, or drivers of variations in care. The absence of this information complicates efficient and informed hospital discharge planning for rural Medicare beneficiaries and policy efforts to ameliorate this growing disparity in health outcomes. The objective of this application is to assemble and analyze national data to assess rural-urban disparities in the availability, selection, utilization, and outcomes of PAC. The central hypothesis of this proposal is that rural PAC markets have fewer and lower quality PAC choices, and that these restricted options result in worse health outcomes, contributing to the widening rural-urban health gap. This hypothesis is based on preliminary work revealing that: 1) there is no rural-urban discrepancy in hospital mortality rates; 2) there is no rural-urban discrepancy in the predicted likelihood of mortality following live hospital discharge; 3) yet, a substantially lower proportion of rural beneficiaries are discharged with PAC, and 4) consistent with other studies, post-hospital-discharge mortality is higher in rural areas. To achieve this objective, we propose three aims. First, we will compare trends in the availability of different types of PAC in rural and urban markets and determine the relationship of these trends in PAC availability to age-adjusted mortality. Second, we will compare the selection of PAC settings between beneficiaries hospitalized in urban versus rural hospitals. Third, we will compare PAC outcomes, including mortality, hospital readmission, emergency department visits, and length of PAC use between beneficiaries discharged from urban versus rural hospitals. We propose to study fee-for-service Medicare beneficiaries with an acute hospitalization in 2012-14 for one of five tracer conditions: hip fracture, lower extremity joint replacement, congestive heart failure, COPD, and stroke, integrating Medicare enrollment, claims, and PAC resident assessment data. This study lays important groundwork for future studies on developing efficient ways to deliver home-based care for beneficiaries living in remote areas, improving care coordination between rural hospitals and PAC providers, and designing appropriate payment schemes for care providers and risk-bearing health insurers to deliver optimal care in rural markets.
The proposed study examines rural-urban disparities in the availability, selection, utilization, and outcomes of post-acute care (PAC). Lower access to and quality of PAC may play a key role in the rapidly widening rural-urban gap in health outcomes. Our results will help hospitals and patients to make informed decisions about PAC selection and improve coordination between acute and post-acute care.