Project summary Post-acute care (PAC) is common, and costly, and may not lead to optimal health outcomes for older adults. However, it is unknown how to improve outcomes and/or lower costs, or value, of PAC. More than 40% of Medicare fee-for-service (FFS) beneficiaries receive PAC after hospitalization, predominantly in skilled nursing facilities (SNFs) and by home health (HH) agencies, at a cost of more than $60 billion annually. Unfortunately, around 1 in 4 is readmitted to the hospital within 30 days of discharge to PAC, and nearly half of beneficiaries in SNF fail to return to the community within 100 days of hospital discharge. The rapid expansion of Medicare Advantage (MA) provides an opportunity to evaluate a different approach to PAC utilization. More than one- third of all Medicare beneficiaries are now enrolled in Medicare Advantage (MA) plans, which receive capitated payments and take financial risk for the care needs of beneficiaries. MA plan directors confirm in interviews that PAC is a major focus of efforts to reduce care utilization and costs through four mechanisms: limiting use of PAC overall; steering beneficiaries to less expensive forms (HH) instead of more expensive forms (SNF); restricting choice of providers; and limiting PAC length of stay. Early reports suggest MA plans may strongly influence PAC utilization. Whether reductions in PAC utilization in MA improve PAC value is unknown. The limited existing literature has two main gaps: first, it does not adequately account for the substantial underlying differences in the MA and FFS populations. Second, it has focused on short-term outcomes, while PAC likely has a substantial impact on longer-term functional status and independence. Our overall goal is to inform patients, providers, and policymakers about ways to improve the value of PAC for all Medicare beneficiaries. To achieve this goal, we propose innovative analytic strategies that address limitations in the prior literature and allow accurate assessment of the use and outcomes of PAC in similar MA and FFS beneficiaries. Our aims are to: 1) Compare use of SNF and HH in similar MA and FFS beneficiaries after hospital discharge, and the impact of different mechanisms for limiting PAC utilization; 2) Compare PAC outcomes (community days, long-term institutionalization, rehospitalization, mortality) at 100 days and 1 year after hospital discharge in MA and FFS; and 3) Evaluate the effects of MA versus FFS enrollment on specific subpopulations of patients known to be at high risk for poor PAC outcomes. These results will provide novel insights into the effect of MA plans on PAC use and outcomes, identifying potential benefits or unintended consequences that can shape policy.