Importance: In 2023, the Centers for Medicare & Medicaid Services terminated dual-eligible special needs plan look-alikes-Medicare Advantage plans with beneficiary panels composed of more than 80% dual-eligible individuals but lacking Medicaid integration. Understanding whether this policy promoted dual-eligible enrollment in integrated care plans, particularly those attaining high-level integration, is critical.
Objective: To describe dual-eligible enrollment transitions after the look-alike plan termination and evaluate whether the policy was associated with increased enrollment in highly integrated plans.
Design, setting, and participants: This repeated cross-sectional study analyzed US Medicare administrative data from January 2017 to January 2023. Samples were limited to full-benefit dual-eligible beneficiaries.
Main outcomes and measures: First, a beneficiary-level analysis was conducted on 2023 enrollment patterns among full-benefit dual-eligible individuals whose 2022 plans were terminated, including factors associated with enrollment in highly integrated plans in 2023. Next, a county-year-level difference-in-differences design was used to compare changes in full-benefit dual-eligible enrollment before (2017-2022) and after (2023) the termination policy between counties with vs without terminated look-alike plans. A difference-in-differences design was used to evaluate whether the look-alike termination policy was associated with the proportion of full-benefit dual-eligible individuals enrolled in highly integrated care plans.
Results: Between 2017 and 2022, 482 of 2576 counties had full-benefit dual-eligible individuals enrolled in look-alike plans for at least 1 year. Of the 170 399 full-benefit dual-eligible individuals enrolled in look-alike plans in 2022 (58.9% female; 20.6% Asian, 44.8% Hispanic, 11.3% non-Hispanic Black, 21.4% non-Hispanic White, and 2% other) and remained dual-eligible in 2023, only 5.4% transitioned to highly integrated plans, while 55.6% moved to nonintegrated plans. Dual-eligible individuals transitioning to highly integrated plans were more likely to be older (65-74 years: adjusted difference, 3.4 percentage points [pp]; 95% CI, 2.8-4.1 pp; 75-84 years: adjusted difference, 4.1 pp; 95% CI, 3.3-4.8 pp; ≥85 years: adjusted difference, 5.0 pp; 95% CI, 4.0-5.9 pp), female (adjusted difference: 0.6 pp; 95% CI, 0.2-0.9 pp), without disabilities (adjusted difference, -0.7 pp; 95% CI, -1.2 to -0.2 pp), and less likely to be Asian (adjusted difference, -5.0 pp; 95% CI, -5.6 to -4.4 pp) or Black (adjusted difference, -0.9 pp; 95% CI, -1.6 to -0.2 pp). The termination policy was not associated with a significant differential increase in enrollment into highly integrated plans in counties with look-alike plans compared with those without (0.6 pp; 95% CI, -0.4 to 1.6 pp). However, there was a 2.6-pp differential increase (95% CI, 0.01-5.1 pp) in enrollment into plans offering some integration, primarily driven by enrollment growth in plans with lower levels of integration. Enrollment also increased in conventional Medicare Advantage plans with fewer than 80% of dual-eligible enrollees (2.6 pp; 95% CI, 0.7-4.5 pp) after the termination policy.
Conclusions and relevance: In this study, the termination of look-alike plans was insufficient to significantly shift dual-eligible individuals toward highly integrated plans. Complementary strategies are necessary to ensure that dual-eligible individuals enroll into highly integrated care models that may improve outcomes.