Age and Sex-Specific Percentiles of 30-Year Cardiovascular Disease Risk Based on the PREVENT Equations.

BACKGROUND: Current primary prevention guidelines recommend estimation of long-term cardiovascular disease (CVD) risk among younger adults to enable preventive efforts earlier in the life course. However, conceptualizing absolute risk estimates over this time horizon is challenging for both clinicians and patients. Framing risk relative to peers (ie, "population-based percentiles") as complementary information may have utility for risk communication. OBJECTIVES: We sought to develop population-based age- and sex-specific percentiles of 30-year absolute risk estimates for CVD, atherosclerotic cardiovascular disease (ASCVD), and heart failure (HF) with the PREVENT (Predicting Risk of CVD EVENTs) equations. METHODS: We used data from a nationally representative sample of U.S. adults aged 30 to 59 years without prevalent CVD in the National Health and Nutrition Examination Survey (NHANES) 2011 to March 2020. We calculated 30-year risk estimates with PREVENT for eligible participants and derived age- and sex-specific percentiles at each age (using rolling windows of ±5 years) for each outcome of CVD, ASCVD, and HF. RESULTS: Among 8,686 NHANES participants, representing approximately 91 million U.S. adults aged 30 to 59 years in the final sample, mean age was 44.8 ± 0.2 years among female participants and 44.2 ± 0.2 years among male participants. The median 30-year absolute CVD risk in the overall sample was 13.1%. Then, corresponding percentile values were developed and evaluated for each age among women and men aged 30 to 59 years for 30-year risk of CVD, ASCVD, and HF. For example, at age 45 years, the 50th percentile of 30-year CVD risk for was 9.9% for women and 16.2% for men, and the 75th percentile was 14.7% for women and 21.2% for men at that age. Similar patterns were observed for ASCVD and HF, with higher 30-year absolute risk in men compared with women. An online tool was developed to present risk percentiles alongside absolute 30-year risk for CVD, ASCVD, and HF. CONCLUSIONS: Population-based age- and sex-specific percentiles for 30-year CVD risk with PREVENT may offer a complementary tool for clinicians and patients in addition to communicating absolute CVD risk estimates. Prospective studies should test whether this approach improves risk perception or decision-making in younger adults.

Duke Scholars

Author Michael J Pencina Biostatistics & Bioinformatics, Division of Biostatistics