JAMA Netw Open. 2026 Feb 2;9(2):e2556915. doi: 10.1001/jamanetworkopen.2025.56915.
ABSTRACT
IMPORTANCE: In 2023, the PREVENT (Predicting Risk of CVD Events) equations were introduced to estimate 10-year risk of total cardiovascular disease (CVD). However, their accuracy in individual Asian or Native Hawaiian and Other Pacific Islander ethnic groups remains unknown.
OBJECTIVE: To evaluate the risk prediction accuracy of the PREVENT base and full equations in Asian and Native Hawaiian and Other Pacific Islander ethnic groups.
DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study was conducted among adults aged 30 to 79 years without CVD who self-reported as being non-Hispanic White, Asian, or Native Hawaiian and Other Pacific Islander and were active Kaiser Permanente Southern California members as of September 30, 2009. Participants were followed up through 2019. Non-Hispanic Asian adults were further disaggregated into ethnic groups. Analysis was performed between February and June 2025.
MAIN OUTCOMES AND MEASURES: The main measures were the PREVENT base equation (age, total and high-density lipoprotein cholesterol, systolic blood pressure, body mass index, estimated glomerular filtration rate, diabetes, smoking, and lipid and antihypertensive medication) and the PREVENT full equation (base plus hemoglobin A1c, urine albumin-creatinine ratio, and Social Deprivation Index). The main outcome was the 10-year incidence of total CVD, atherosclerotic CVD, and heart failure. Estimated risks were compared with observed events using the Harrell C index and mean calibration (predicted to observed event ratios).
RESULTS: The study cohort consisted of 542 848 adults, including 424 277 non-Hispanic White adults (mean [SD] age, 55.6 [11.8] years; 235 722 [55.6%] female), 110 855 non-Hispanic Asian adults (mean [SD] age, 52.5 [11.9] years; 66 292 [59.8%] female), and 7716 non-Hispanic Native Hawaiian and Other Pacific Islander adults (mean [SD] age, 51.4 [11.9] years; 4 398 [57.0%] female). A total of 31 556 CVD events occurred during 10 years. For total CVD, the PREVENT base equation demonstrated good discrimination across non-Hispanic White (C index, 0.764; 95% CI, 0.761-0.767), non-Hispanic Asian (C index, 0.773; 95% CI, 0.765-0.779) and non-Hispanic Native Hawaiian and Other Pacific Islander (C index, 0.757; 95% CI, 0.733-0.780) groups. Among Asian ethnic groups, C indexes for the PREVENT base equation ranged from 0.738 (95% CI, 0.701-0.774) in Vietnamese adults to 0.806 (95% CI, 0.787-0.826) in Chinese adults. The PREVENT full equations showed consistent results. The PREVENT base and full equations generally overestimated total CVD, atherosclerotic CVD, and heart failure risk in non-Hispanic Asian (mean calibration, 0.96-1.33) and underestimated risk in non-Hispanic Native Hawaiian and Other Pacific Islander (mean calibration, 0.74-0.96) and non-Hispanic White (mean calibration, 0.63-1.03) populations.
CONCLUSIONS AND RELEVANCE: In this retrospective cohort study, the PREVENT base and full equations demonstrated overall strong performance in predicting 10-year cardiovascular risk. However, notable differences were observed across disaggregated Asian and Native Hawaiian and Other Pacific Islander ethnic groups, underscoring the importance of recognizing heterogeneity within these populations when applying risk prediction models.
PMID:41678194 | DOI:10.1001/jamanetworkopen.2025.56915