PREVENT Equations Accurately Predict 10-Year CVD Risk Across Diverse Groups

Cardiovascular disease (CVD) remains a leading cause of morbidity and mortality worldwide, yet risk prediction models often overlook diversity within racial and ethnic populations, according to new study findings.¹ In 2023, the Predicting Risk of CVD Events (PREVENT) equations were introduced to estimate 10-year CVD risk. While validated broadly, their accuracy among non-Hispanic Asian and Native Hawaiian and Other Pacific Islander populations was previously unclear.

This retrospective cohort study is published in JAMA Network Open.

“The current study found notable differences in calibration between non-Hispanic Asian and non-Hispanic Native Hawaiian and Other Pacific Islander adults when applying the PREVENT equations,” wrote the researchers of the study. “These findings suggest that the consolidation of both groups into a single race category may obscure meaningful differences in risk prediction accuracy.”

The PREVENT calculator represents a major advancement in CVD risk prediction by incorporating cardio-kidney-metabolic factors and socioeconomic metrics.²

“In an era of increasing demand for shared decision-making in health care, risk prediction models should contribute to patient-physician communication,” according to an invited commentary author. “Importantly, although it is well-known that risk prediction models were not designed to perform perfectly for each individual patient, they may serve as a starting point for conversation. From there, discussion and consideration of patient-specific characteristics can refine clinician judgment.”

The cohort included adults aged 30 to 79 years without prior CVD.¹ Participants self-reported as non-Hispanic White, non-Hispanic Asian, or non-Hispanic Native Hawaiian and Other Pacific Islander. Asian participants were further disaggregated into ethnic groups. The PREVENT base equation included age, total and high-density lipoprotein cholesterol, systolic blood pressure, body mass index, estimated glomerular filtration rate, diabetes, smoking status, and lipid and antihypertensive medication use.

The full equation added hemoglobin A_1c, urine albumin-creatinine ratio, and the Social Deprivation Index. Predicted risks were compared to observed events using the Harrell C index and mean calibration.

Participants were followed through 2019 to assess 10-year incidence of total CVD, atherosclerotic CVD, and heart failure.

The cohort included 424,277 non-Hispanic White adults (mean [SD] age, 55.6 [11.8] years; 55.6% female), 110,855 non-Hispanic Asian adults (mean age, 52.5 [11.9] years; 59.8% female), and 7716 non-Hispanic Native Hawaiian and Other Pacific Islander adults (mean age, 51.4 [11.9] years; 57.0% female). Over 10 years, 31,556 CVD events occurred.

The PREVENT base equation demonstrated strong discrimination for total CVD:

• 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)
• Non-Hispanic Native Hawaiian and Other Pacific Islander: C index, 0.757 (95% CI, 0.733-0.780)

Disaggregated Asian groups showed heterogeneity, with C indexes ranging 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 full PREVENT equations produced similar results. Mean calibration revealed slight overestimation of CVD risk in non-Hispanic Asian adults (0.96-1.33) and underestimation in Native Hawaiian and Other Pacific Islander adults (0.74-0.96) and non-Hispanic White adults (0.63-1.03).

However, the researchers noted some limitations. The study relied on data from a single health care system and may not generalize to all populations. Although the cohort included disaggregated Asian ethnicities, smaller sample sizes for some groups limit precision. Furthermore, risk prediction may also be affected by unmeasured lifestyle, genetic, or socioeconomic factors.

Despite these limitations, the researchers believe the study finds the PREVENT base and full equations show overall strong performance in predicting 10-year CVD risk across non-Hispanic Asian, Native Hawaiian and Other Pacific Islander, and non-Hispanic White populations. However, performance variation among disaggregated Asian ethnic groups underscores the need for tailored risk prediction models that account for racial and ethnic diversity. Clinicians should recognize this heterogeneity when applying PREVENT equations to guide prevention and treatment strategies.

References

1. Au M, Zhang Y, Zhou MM, et al. PREVENT equation performance in Asian and Native Hawaiian and Other Pacific Islander groups. JAMA Netw Open. 2026;9(2):e2556915. doi:10.1001/jamanetworkopen

2. Scheuermann B. Targeting disparities in cardiovascular disease risk prediction in Asian and Pacific Islander populations. JAMA Netw Open. 2026;9(2):e2556865. doi:10.1001/jamanetworkopen