Abstract 4364156: Impact of Value-Based Care Risk Arrangements on Cardiovascular Clinical Quality Measures Among Medicare Advantage Members of a Nationwide Payor

Abstract

Background: Value-based care (VBC) models aim to improve health care quality and affordability through aligned incentives across payors and providers. Although widely promoted, limited real-world evidence exists on how varying levels of risk influence cardiovascular quality outcomes. Understanding these dynamics is critical to informing strategies aligned with the American Heart Association’s position to advance high-value, whole person-centered, affordable care. Research Question: Do cardiovascular-related clinical quality outcomes differ between VBC and non-VBC models, and across risk sharing levels (2-sided, 1-sided, pay-for-performance [P4P], and non-VBC) in a Medicare Advantage (MA) population? Methods: Retrospective analysis of 2023 CMS Star Ratings data compared clinical quality scores (%, mean ± standard deviation) across MA members with ≥6 mo of enrollment for cardiovascular-related measures: medication adherence for hypertension (MAH), medication adherence for cholesterol (MAC), statin therapy for patients with cardiovascular disease (SPC), and controlling blood pressure (CBP). Propensity score matching and weighting balanced sociodemographic and clinical characteristics between payment models and across risk levels. Generalized linear models with Bonferroni correction tested group differences; post-hoc Tukey tests assessed pairwise comparisons. Results: Between-group standardized mean differences were ≤10.8%, indicating acceptable balance. Among 3,228,872 members (72.4±9.2yr, 56.5% female), VBC consistently outperformed non-VBC (MAH 1.8±0.1, MAC 1.8±0.1%, SPC 3.0±0.2%, and CBP 17.3±0.1%; all P <.001). Higher levels of risk were associated with improved performance (all P <.001). For SPC, 2-sided risk outperformed 1-sided risk (0.6±0.1%); P4P outperformed non-VBC (1.0±0.1%; both P <.001). For MAH, MAC, and CBP, performance improved incrementally with increasing risk sharing (i.e., 2-sided>1-sided>P4P>FFS; range: 0.4±0.1 to 9.3.4±0.1%; all P <.001). Conclusion: This large, real-world study of ~3.2 million MA members found VBC models were associated with superior cardiovascular quality outcomes, with performance improving as financial risk increased. Observed differences in adherence, secondary prevention, and blood pressure control reflect clinically meaningful outcomes with direct relevance to population health. These findings support shared-risk arrangements as a practical lever for advancing high-quality, value-based cardiovascular care at scale.