Late Breaking Original Research
Sunday, October 13, 2024
12:45 PM–02:15 PM
Abstract
Introduction: The ability of the recently released American Heart Association PREVENT
TM calculator to accurately assign 10-year atherosclerotic cardiovascular disease (ASCVD) risk in a modern cohort of older individuals, including those > 79 years, is unknown.
Research Question or Hypothesis: Does PREVENTTM have better predictive performance than PCE for ASCVD events in a contemporary cohort of older adults?
Study Design: Post-hoc analysis of 18,297 adults from Australia and the U.S., aged =65 years, enrolled in the ASPirin in Reducing Events in the Elderly randomized preventive trial of aspirin, including post-trial observational follow-up to 2022.
Methods: ASCVD events (non-fatal myocardial infarction, fatal coronary heart disease, stroke), occurring in-trial and during follow-up were adjudicated by expert panel. The discriminative ability of both risk calculators was assessed by Harell’s c-statistic following Cox regression. For calibration, predicted event numbers for PREVENTTM and PCE were calculated for 15,510 participants aged 65-79 years by multiplying 10-year predicted risk scaled for the actual length of follow-up of each participant by the total population, and compared with the number of observed events in-trial and follow-up. The analysis was repeated in 2,787 participants aged >79 years.
Results: PREVENTTM showed superior discriminative performance compared to PCE (PREVENTTM vs. PCE, c-statistic: 0.793 vs 0.740 p<0.001 in participants aged 65-79 years; 0.854 vs. 0.799, p<0.001 in those aged >79 years). Among participants aged 65-79 years, 1084 ASCVD events occurred; PCE predicted 3102 events (13.0% overestimate) while PREVENTTM predicted 1290 events (1.3% overestimate). For those >79 years, 355 ASCVD events occurred; PCE predicted 1067 events (25.5% overestimate) while PREVENTTM predicted 350 events (0.16% underestimate). PREVENT’s discrimination and calibration were superior to PCE when subgroups of sex, country, and race were examined separately.
Conclusion: PREVENTTM is superior to PCE in predicting ASCVD events in older adults from the US and Australia, including those aged >79 years.
Presenting Author
Michelle Fravel PharmD, FCCP, BCPSUniversity of Iowa
Authors
Michael Ernst PharmD, FCCP, BCGP
Mark Nelson PhD
Monash University
Kevan Polkinghorne PhD
Monash University
Christopher Reid PhD
Monash University
Rory Wolfe PhD
Monash University
Robyn Woods PhD
Monash University
Zhen Zhou PhD
Monash University