Impact of the Medicare Advantage Hospice Carve-In on End-of-life Care Outcomes among Beneficiaries with Alzheimer's Disease and Related Dementia

REVISED PROJECT SUMMARY An increasing proportion of Medicare beneficiaries are enrolled in Medicare Advantage (MA), representing half of enrollees in 2023. Until recently, the Medicare Hospice Benefit was carved out of the MA program, so MA enrollees who needed hospice would receive that care through traditional Medicare. As of 2021, the Centers for Medicare & Medicaid Services (CMS) started testing the inclusion of the Medicare Hospice Benefit through the MA’s Value-Based Insurance Design (VBID) Model. By 2023, 119 MA organizations were participating in the Hospice Benefit Component across 806 counties. Also referred to as the MA hospice “carve-in”, this constitutes a major change for the MA programs and the implications are not yet known. While the MA carve-in will likely impact all beneficiaries, especially in areas with high MA penetration, there are particular concerns for the quality of end-of-life care among decedents with Alzheimer’s disease and related dementia (ADRD). Prior work has found that these beneficiaries are more likely to disenroll from hospice after long enrollment periods than those without this diagnosis. If we are to improve the quality of end- of-life care provided to Medicare beneficiaries with dementia, there is a critical need to assess the impact of the MA hospice carve-in on end-of-life care outcomes. Our long-term goal is to improve the quality of end-of-life care provided to persons with dementia. The overall objective of this application is to explore the implementation of the hospice carve-in and assess its impact on various aspects of care delivery and end-of- life care outcomes. Our central hypothesis is that the hospice carve-in represents a profound change in incentives for providers and may worsen end-of-life care outcomes among beneficiaries with dementia. We plan to accomplish the overall research objective by pursuing the following specific aims: 1) explore how MA organizations and hospice providers have adapted to the MA hospice carve-in by gathering primary qualitative data from carefully selected informants from specific markets, 2) determine the effect of being enrolled in an MA plan participating in the VBID Hospice Benefit Component on end-of-life care outcomes for Medicare beneficiaries with dementia, and finally 3) examine heterogeneity in the impact of the MA hospice carve-in on hospice utilization and other indicators of quality of end-of-life care among MA beneficiaries with ADRD across different cohorts of health care users and MA plan and market characteristics. The expected outcomes from this proposal include a stronger evidence base to inform policy decisions and oversight of the MA Hospice Benefit Component, including a better understanding of national variation in how MA organizations have been adapting to this new care model.