Related Experiment Video
Updated: Jun 14, 2025
An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
PREVENT Risk Score vs the Pooled Cohort Equations in MESA
Brittany Saldivar Murphy1, M Sims Hershey2, Shi Huang1
1Division of Cardiovascular Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA.
Insights
The new PREVENT equations offer more accurate atherosclerotic cardiovascular disease (ASCVD) risk prediction than current guidelines. PREVENT-ASCVD equations show improved performance across diverse populations, aiding better cardiovascular risk assessment.
Area of Science:
- Cardiovascular Medicine
- Epidemiology
- Risk Prediction Modeling
Background:
- The American Heart Association introduced the PREVENT (Predicting Risk of CVD Events) equations in 2023.
- These equations aim to estimate the risk of atherosclerotic cardiovascular disease (ASCVD) and heart failure (HF).
Purpose of the Study:
- To compare the predictive performance of the PREVENT-ASCVD equations against the current Pooled Cohort Equations (PCE).
- To evaluate the accuracy of the PREVENT-HF risk algorithm.
Main Methods:
- Utilized the Multi-Ethnic Study of Atherosclerosis (MESA) cohort of 6,098 individuals.
- Calculated baseline PCE and PREVENT-predicted 10-year ASCVD event percentages.
- Assessed observed event rates, prediction-observation discordance, discrimination (C-index), and calibration (MAE).
Main Results:
- Observed ASCVD events (6.0%) aligned more closely with PREVENT predictions (5.7%) than PCE (10.8%).
- PREVENT-ASCVD showed greater accuracy in women, nonsmokers, individuals with CKD stages 3/4, and those with high social deprivation.
- Forty-two percent of the cohort experienced risk re-classification to a lower category using PREVENT vs. PCE.
- PREVENT-HF overestimated HF events by 2.1% (62.6% relative risk overestimation).
Conclusions:
- PREVENT-ASCVD equations provide more accurate ASCVD risk stratification compared to PCE.
- PREVENT demonstrates superior performance in women, nonsmokers, individuals with renal dysfunction, social deprivation, and Black individuals.
- PREVENT-HF tends to overestimate the risk of incident heart failure within the MESA cohort.
Background:
In 2023, the American Heart Association developed the PREVENT (Predicting Risk of CVD Events) equations to estimate risk of atherosclerotic cardiovascular disease (ASCVD) and heart failure (HF).
Objectives:
Assess the comparative performance of PREVENT-ASCVD vs current guideline-recommended Pooled Cohort Equations (PCE). Evaluate the performance of the PREVENT-HF risk algorithm.
Methods:
In 6,098 individuals from the MESA (Multi-Ethnic Study of Atherosclerosis) cohort, we calculated baseline PCE, and PREVENT predicted 10-year ASCVD event percentages, observed event percentages at 10 years, discordance between observed and expected percentages, discrimination using Harrell's C index, and calibration using mean absolute error.
Results:
Observed ASCVD event rate (6.0%) was closer to the predicted PREVENT event rate (5.7%) than the PCE (10.8%). PREVENT was more accurate in women than men (3.3% vs -11.6% discordance between observed and PREVENT predicted ASCVD), nonsmokers compared to smokers (2.4% vs -37.0% discordance), chronic kidney disease stages 3/4 (discordance 3.2%), and those with high social deprivation scores (discordance -5.0%). Forty-two percent of this cohort would be re-classified to a lower ASCVD risk category using the PREVENT equation vs the PCE. PREVENT-HF overestimates HF events by 2.1%, a relative risk overestimation of 62.6%.
Conclusions:
PREVENT-ASCVD equations demonstrated a more accurate ASCVD risk-prediction stratification than the PCE. PREVENT performs best in women, nonsmokers, those with a greater degree of renal dysfunction, social deprivation, and Black individuals. PREVENT-HF overestimates risk of incident HF in a Multi-Ethnic Study of Atherosclerosis.
More Related Videos
Related Concept Videos
01:12Relative Risk
01:23Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
01:12Hazard Ratio
For example, in a clinical trial...
01:23Mechanistic Models: Compartment Models in Individual and Population Analysis
01:24Kaplan-Meier Approach
01:29Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...