• X Tang, D D Zhang, L He, Y Cao, J W Wang, N Li, S P Huang, H D Dou, P Gao, and Y H Hu.
• Department of Epidemiology & Biostatistics, Peking University School of Public Health, Beijing 100191, China.
• Beijing Da Xue Xue Bao. 2017 Jun 18; 49 (3): 439-445.

ObjectiveTo validate five-year risk prediction models for atherosclerotic cardiovascular disease (ASCVD) in a contemporary rural Northern Chinese population.MethodsTotally 6 489 rural adults aged 40 to 79 years without clinical ASCVD were enrolled at baseline between June and August 2010, and followed up through January 2017. Expected prediction risk using the China-PAR (prediction for ASCVD risk in China) model was compared with the pooled cohort equations (PCE) reported in the American College of Cardiology/American Heart Association guideline. Kaplan-Meier analysis was used to obtain the observed ASCVD event (including nonfatal myocardial infarction, coronary heart disease death, nonfatal or fatal stroke) rate at 5 years, and the expected-observed ratios were calculated to eva-luate overestimation or underestimation in the cohort. The participants in the cohort were divided into 4 categories (<5.0%, 5.0%-7.4%, 7.5%-9.9%, and ≥10.0%) for comparisons based on ASCVD prediction risk. The models were assessed by discrimination C statistic, calibration χ2, and calibration charts and plots for illustration as well.ResultsOver an average 5.82 years of follow-up in this validation cohort with 6 489 rural Chinese participants, 955 subjects developed a first ASCVD event. Recalibrated China-PAR model overestimated ASCVD events by 22.2% in men and 33.1% in women, while the overestimations were much higher for recalibrated PCE as 67.3% in men and 53.1% in women. Gender-specific China-PAR model had C statistics of 0.696 (95%CI, 0.669-0.723) for men and 0.709 (95%CI, 0.690-0.728) for women, which were similar to those of 0.702 (95%CI, 0.675-0.730) for men and 0.714 (95%CI, 0.695-0.733) for women in the PCE. Calibration χ2 values in China-PAR were 17.2 and 54.2 for men and women, respectively; however, the PCE showed poorer ca-libration (χ2=192.0 for men and χ2=181.2 for women). In addition, the calibration charts and plots illustrated good agreement between the observations and the predictions only in the China-PAR model, especially for men.ConclusionIn this validation cohort of rural Northern Chinese adults, the China-PAR model had better performance of five-year ASCVD risk prediction than the PCE, indicating that recalibrated China-PAR model might be an appropriate tool for risk assessment and primary prevention of ASCVD in China.

20 keywords...

hide…