Performance of the American Heart Association PREVENT Cardiovascular Risk Equations in Older Adults
journal contribution
posted on 2025-05-04, 23:51authored byMichelle A Fravel, Michael E Ernst, Robyn L Woods, Suzanne G Orchard, Shiva GanjaliShiva Ganjali, James B Wetmore, Christopher Reid, Joanne Ryan, Kevan R Polkinghorne, Rory Wolfe, Mark R Nelson, Sophia Zoungas, Zhen Zhou
BACKGROUND:
The ability of the American Heart Association Predicting Risk of Cardiovascular Disease Events (PREVENT) calculator to accurately assign 10-year atherosclerotic cardiovascular disease (ASCVD) risk in older individuals, including those aged ≥80 years, is unknown. This study compares PREVENT with the 2013 Pooled Cohort Equation (PCE) calculator for predicting 10-year ASCVD risk in a large cohort of older adults.
METHODS:
This was a prospective cohort study of adults without CVD from Australia and the United States aged ≥70 years (≥65 years, if US minorities). They were enrolled from 2010 to 2014 in the ASPREE trial (Aspirin in Reducing Events in the Elderly), a 5-year randomized trial of low-dose aspirin in community-dwelling older adults with posttrial observational follow-up extending to 2022. ASCVD events were adjudicated by expert panels. The discriminative ability of the 2 risk calculators was assessed by Harell C statistic following Cox regression in the 65- to 79-year age group and >80-year age group, separately. For calibration, predicted event numbers were calculated using PREVENT and PCE, scaled for the actual length of follow-up, and compared with the number of observed events in-trial and during extended follow-up.
RESULTS:
Among the 15 510 participants aged 65 to 79 years (median age, 73.2 years; 56.1% women), 1084 ASCVD events occurred (median follow-up, 8.3 years); PCE predicted 3102 events while PREVENT predicted 1290 events. For the 2787 participants ≥80 years (median age, 82.6 years; 59.2% women), 355 ASCVD events occurred (median follow-up, 7.4 years); PCE predicted 1067 events while PREVENT predicted 350 events. PREVENT showed superior discriminative performance compared with PCE (PREVENT versus PCE, C statistic, 0.793 versus 0.740;
P
<0.001 in participants aged 65 –79 years; 0.854 versus 0.799;
P
<0.001 in those aged ≥80 years).
CONCLUSIONS:
The PREVENT risk calculator is superior to the PCE calculator in predicting ASCVD events in older adults from the United States and Australia, including those aged ≥80 years.
REGISTRATION:
URL:
https://www.clinicaltrials.gov
; Unique identifier: NCT01038583. URL:
https://www.isrctn.com
; Unique identifier: ISRCTN83772183.