Related Experiment Video
Updated: Feb 22, 2026
An R-Based Landscape Validation of a Competing Risk Model
Published on: September 16, 2022
Validation and Recalibration of PCE, China-PAR, and PREVENT Models for Estimating ASCVD Risk in China
Haibin Li1, Shuohua Chen2, Ruolin Zhang3
1Department of Cardiology, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China; Medical Research Center, Beijing Chaoyang Hospital, Capital Medical University, Beijing, China.
Background:
The Pooled Cohort Equations (PCE) and Prediction for Atherosclerotic Cardiovascular Disease (ASCVD) Risk in China (China-PAR) models tend to overestimate risk, whereas the Predicting Risk of Cardiovascular Disease Events (PREVENT) equations may underestimate risk in many contemporary cohorts.
Objectives:
This study aimed to validate and determine if recalibration of these risk scores using contemporary population-level data improves risk stratification for primary prevention in China.
Methods:
These risk scores were first validated and then recalibrated in the Kailuan study. Participants aged 40 to 79 years without ASCVD at baseline were included. Original and recalibrated models were assessed for discrimination and calibration.
Results:
Of 79,497 participants, 4,425 ASCVD events occurred over a median follow-up of 10 years (Q1-Q3: 10-10 years). All 3 original models showed good discrimination in women (Harrell's C-index: PCE 0.735 [95% CI: 0.712-0.757], China-PAR 0.738 [95% CI: 0.715-0.760], PREVENT 0.737 [95% CI: 0.713-0.759]) and moderate discrimination in men (PCE 0.675 [95% CI: 0.667-0.683], China-PAR 0.685 [95% CI: 0.677-0.693], and PREVENT 0.685 [95% CI: 0.677-0.693]). Original models showed differential mean calibration in 10-year ASCVD risk estimation: PCE overestimated by 34.9% (men [95% CI: 32.8%-36.9%]) and 15.1% (women [95% CI: 6.8%-22.7%]); China-PAR by 17.7% (men [95% CI: 15.1%-20.2%]) and 48.2% (women [95% CI: 43.1%-52.8%]); PREVENT underestimated by 29.4% (men [95% CI: 25.4%-33.5%]) and 2.7% (women [95% CI: -6.5% to 12.8%]). After recalibrating to the local population, observed vs predicted risks exhibited improved alignment for the recalibrated models.
Conclusions:
In this Chinese cohort, the original PCE and China-PAR overestimated 10-year ASCVD risk, whereas PREVENT underestimated it. Recalibration mitigated such misestimation in the local population, potentially enhancing risk stratification in primary cardiovascular prevention in China.
