Toward a Deeper Understanding of PREVENT for 10‐Year Atherosclerotic Cardiovascular Risk: Subgroup Fairness and Predictive Value of Social Determinants of Health

This seminar will present findings from a large-scale evaluation of the American Heart Association’s PREVENT model for predicting 10-year atherosclerotic cardiovascular disease (ASCVD) risk. Using electronic health records from over 550,000 adults in the Truveta data platform, we examined the model’s fairness across demographic and social determinants of health (SDOH) subgroups and assessed whether adding SDOH predictors improves performance. Results showed that PREVENT demonstrated consistent calibration and fairness across most subgroups, with notable disparities in race, education, and insurance. Incorporating SDOH predictors or recalibrating the model provided minimal incremental benefit, supporting the robustness and practical utility of the original PREVENT equations in diverse real-world populations.