Back

Statin Eligibility Disparities with Transition from the Pooled Cohort Equations to the AHA PREVENT

Yan, X.; Huang, Q.; Li, J.; Husby, H.; Jose, P. O.; Kenkare, P.; Solomon, M. D.; Rodriguez, F.; Bacong, A. M.

2026-02-01 epidemiology
10.64898/2026.01.29.26345173 medRxiv
Show abstract

BackgroundThe 2023 AHA PREVENT (Predicting Risk of Cardiovascular Disease Events) equations were expected to replace the 2013 ACC/AHA Pooled Cohort Equations (PCE) for estimating atherosclerotic cardiovascular disease (ASCVD) risk. The real-world implications of this transition on statin eligibility and disparity are unknown. ObjectivesTo evaluate how transitioning from PCE to AHA PREVENT alters statin eligibility across risk thresholds and racial and ethnic subgroups. Design, Setting, and ParticipantsRetrospective cohort analyses of adults aged 40-75 years without diabetes, LDL-C [≥]190 mg/dL, or prior statin use from Sutter Health (2010-2024) and NHANES (2011-2020). Ten-year ASCVD risk was estimated using both equations. Weighted analyses were applied to NHANES data. Main Outcomes and MeasuresStatin eligibility at PREVENT-ASCVD thresholds (3%, 4%, 5%, 6%, 7.5%) compared with PCE [≥]7.5%, and the proportion of individuals reclassified below PREVENT-ASCVD thresholds. ResultsAmong 229,839 Sutter Health patients (mean age 53.7 years; 53.8% women), 22.3% had PCE risk [≥]7.5%. Among individuals above PREVENT-ASCVD 5% threshold level (18.0% in the cohort) 94.7% also met PCE criteria. However, 6.3% (11,866) would lose eligibility, disproportionately affecting non-Hispanic Black adults (18.7%) compared with non-Hispanic Asian adults (3.3%) among individuals who were below PREVENT-ASCVD 5% threshold. In NHANES (n=3,226; representing 32.7 million adults), 9.4% overall and 21.7% of non-Hispanic Black adults with PCE [≥]7.5% lost eligibility at PREVENT-ASCVD 5% threshold level. ConclusionsTransitioning from PCE to PREVENT recalibrates statin eligibility and may disproportionately affect non-Hispanic Black adults. Disparity-focused monitoring is essential for clinical implementation of this new model.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.