Post-acute care (PAC) following a hospitalization has experienced a rapid growth in the last decade. However, there is a dearth of knowledge about the comparative effectiveness and quality of alternative PAC services needed to make informed clinical and policy decisions. Existing research indicates that PAC patients afflicted with Alzheimer?s Disease and Related Dementias (ADRDs) are at an increased risk of experiencing poor function rehabilitation, not being able to return to the community, or returning to the community but requiring institutionalization soon thereafter. The diversity in PAC settings , including inpatient rehabilitation facilities (IRFs), long term care hospitals (LTCHs), skilled nursing facilities (SNFs), and home with home health agencies care (HHAs), each with its own set of regulatory and payment incentives and its own treatment approach, and the heterogeneity in the response of patients to specific treatments hinder the comparisons across settings. Of particular relevance is the lack of knowledge about which PAC settings are best suited for ADRD patients, and to what degree new PAC policies, such as bundled payments, have differential and unintended impact on patients with ADRD. The overarching goal is to examine PAC utilization and patient outcomes across different conditions and PAC settings to address important gaps in knowledge that hinder the design of current policy efforts. To achieve this goal, we intend to do a comparative effectiveness study to determine what PAC setting (in isolation or combined with others) delivers the best outcomes. Our underlying hypothesis is that specialization in PAC markets with a larger variety of PAC settings, and value based payment models, like PAC bundled payments, will lead to improved outcomes for some patient subgroups but they may induce segregation of costlier ADRD patients into lower quality providers. The proposed study will link Medicare claims data and SNF, IRF and HHA assessment data across the 2007-2017 period to examine PAC outcomes for patients with five tracer conditions: hip fracture, lower extremity joint replacement, CHF, COPD, and stroke, and compare the results for similar patients with and without ADRD. The expected outcomes of our analyses are a deeper understanding of how availability, market and provider characteristics impact PAC utilization patterns and patient outcomes among ADRD and non-ADRD patients. This knowledge has the potential for informing current and future PAC policies to increase PAC value for ADRD and non-ADRD patients while controlling for unintended negative consequences.
Post-acute care is widely used and costly. While patients have options for PAC services between different post-acute settings, there is little information on how these alternative post-acute care settings impact outcomes for different patient groups. ADRD patients are a particularly vulnerable group with different PAC needs and at higher risk of long-term institutionalization. The results of our study will help inform the referral of patients in need of PAC, and help guide policy aimed at improving PAC patient outcomes.
