Comparison of Short‐ and Long‐Term Cardiovascular Disease Risk Assessment Tools in US Young Adults

Background In 2023, the American Heart Association published the Predicting Risk of Cardiovascular Disease Events (PREVENT) equations for estimating atherosclerotic cardiovascular disease (ASCVD) risk in adults aged 30 to 79 years. We compared PREVENT's performance with existing US guideline recommended models—Pooled Cohort Equations for 10‐year ASCVD risk and FHS (Framingham Heart Study) equations for 30‐year ASCVD risk—among young adults. Methods We analyzed adults aged 20 to 39 years without baseline ASCVD from 2 sources: (1) pooled data from 2 large epidemiologic cohorts (CARDIA Coronary Artery Risk Development in Young Adults and FHS, n=7763), and (2) electronic health records from Kaiser Permanente Southern California (n=266 378). Incident ASCVD events were defined as nonfatal myocardial infarction, coronary heart disease death, and fatal/nonfatal stroke at 10 and 30 years. Results PREVENT improved 10‐year risk discrimination over Pooled Cohort Equations in both the epidemiologic cohorts (∆Harrell's C, 0.057 95% CI, 0.013–0.101) and Kaiser Permanente Southern California (∆Harrell's C=0.041 95% CI, 0.034–0.049). The Pooled Cohort Equations overestimated 10‐year risk (mean calibration 3.26 in the epidemiologic cohorts; 1.73 in Kaiser Permanente Southern California), whereas PREVENT was well calibrated in the epidemiologic cohorts (1.10 95% CI, 0.83–1.73) but underestimated risk in KPSC (0.91 [95% CI, 0.86–0.96), particularly among Black individuals (0.54 95% CI, 0.48–0.61). For 30‐year risk, PREVENT and FHS had similar discrimination, but PREVENT underestimated 30‐year risk (mean calibration 0.63) whereas FHS had good calibration (mean calibration 0.90 to 0.99]). Conclusions PREVENT may be a better tool for short‐term ASCVD risk assessment in young adults than the PCEs, whereas the FHS equations may be better for long‐term risk assessment than PREVENT in this age group.