Examining Trends in Medicare Advantage Plan Disenrollment Associated With Expanded Supplemental Benefit Adoption.

Abstract

Objective: To evaluate the association between Medicare Advantage (MA) supplemental benefit adoption and plan disenrollment among plans that adopted either the 2019 nonmedical primarily health-related benefits (PHRB) or the 2020 social needs Special Supplemental Benefits for the Chronically Ill (SSBCIs).

Study setting and design: We linked individual-level Medicare administrative data to publicly available, plan-level MA benefit, enrollment, crosswalk, and penetration files from 2017 to 2021. The PHRB benefits included benefits such as caregiver support, adult day care, in-home support services, and so forth. The SSBCI benefits included benefits such as food and produce, nonmedical transportation, pest control, and so forth. We used a difference-in-differences design studying MA enrollees stratified by Medicare-Medicaid dual eligibility status.

Data sources and analytic sample: We included individuals from across the 50 United States and DC enrolled in MA plans that adopted a PHRB in 2019 or SSBCI in 2020 and matched comparator plans from the same counties that did not adopt either benefit. Individuals were excluded if they moved, died, or lacked county-level information during the year.

Principal findings: Our sample includes 8,947,810 unique MA enrollees (27.4% in plans that adopted a PHRB and 1.0% in plans that adopted an SSBCI). For dual-eligible enrollees, neither PHRB adoption (0.2%, 95% CI, -2.7%, 2.8%) nor SSBCI adoption (-1.7%, 95% CI, -6.0%, 2.5%) was significantly associated with the rate of plan disenrollment. For Medicare-only enrollees, neither PHRB adoption (-2.6%, 95% CI, -5.9%, 0.7%) nor SSBCI adoption (-5.4%, 95% CI, -15.8%, 5.1%) was significantly associated with the disenrollment rate.

Conclusion: The promise of these benefits was that MA plans could more directly address enrollees' nonmedical and social needs, leading to better social and health outcomes and reducing costs. We find that adoption did not decrease plan disenrollment, which suggests it may not drive enrollment decisions.