Prevention & lipidsCohort study
AHA PREVENT Equations and Cardiovascular Disease Risk in Diverse Health Care Populations
- Retrospective cohort study
- multicentre
- external validation
Authors' conclusion
The PREVENT model moderately discriminated ASCVD incidence across 4 geographically distinct academic health care systems in the United States. However, calibration metrics varied widely across health care systems, sociodemographics, and underlying cardiometabolic comorbidities.
Study context
- Population
- Patients aged 30 to 79 years without prior atherosclerotic cardiovascular disease (ASCVD), US integrated health care systems, 2010 to 2014
- Participants
- 270,320 patients · 4 health care systems
- Exposure
- PREVENT predicted vs observed 10-year ASCVD across 4 US health care systems
- MGBn = 136,654
- Mount Sinain = 43,321
- Pennn = 56,889
- Vanderbiltn = 33,456
- Outcome
- Calibration (discordance) and discrimination (C-index) of PREVENT for incident ASCVD
- Follow-up
- 10 years
Main finding
As reported in the abstract
PREVENT calibration and discrimination for 10-year ASCVD
| Group | Value | Discordance | C-index |
|---|---|---|---|
| MGB | mean estimated 10-year ASCVD risk 4.9% (SD 4.7%) | −71.0% | 0.70 [95% CI: 0.70-0.70] |
| Mount Sinai | mean estimated 10-year ASCVD risk 6.0% (SD 5.6%) | −36.2% | 0.74 [95% CI: 0.73-0.75] |
| Penn | mean estimated 10-year ASCVD risk 6.0% (SD 5.2%) | +1.3% | 0.69 [95% CI: 0.68-0.70] |
| Vanderbilt | mean estimated 10-year ASCVD risk 4.8% (SD 1.3%) | −40.0% | 0.73 [95% CI: 0.72-0.74] |
Observational data: associations, not effects.
Further findings
Calibration by sex (subgroup)
- Greater underestimation among women, MGB (discordance)
- −80.7%
- Greater underestimation among women, Vanderbilt (discordance)
- −55.5%
- Greater underestimation among men, Mount Sinai (discordance)
- −40.7%
Calibration by race and ethnicity (subgroup)
- Predictive performance varied across health care systems
Overall calibration, PREVENT vs pooled cohort equations (discordance)
- Penn: PREVENT better (PREVENT vs PCE)
- +2.5% vs +93.6%
- MGB: PREVENT worse (PREVENT vs PCE)
- −70.0% vs −41.6%
- Mount Sinai: PREVENT worse (PREVENT vs PCE)
- −36.4% vs 4.8%
- Comparable discrimination
Discrimination by DM or antihypertensive use (subgroup)
- Better discrimination among patients without DM or antihypertensives
Original publication
AHA PREVENT Equations and Cardiovascular Disease Risk in Diverse Health Care Populations. J Am Coll Cardiol. Published 22 July 2025. doi:10.1016/j.jacc.2025.04.066
Summarised from the published abstract; every number is checked against it