Why the study?
The study aimed to describe the proportion of community-screened adults in Nanjing identified at high CVD risk, compare classifications between China-PAR and ICVD-10 among those without prior CVD, and characterize the overlap among four prespecified high-risk identification sources.
How well do the China-PAR and ICVD-10 risk assessment tools agree, and how much do the four prespecified cardiovascular high-risk identification pathways overlap among community-screened adults aged 35–79 years in Nanjing, China?
How well do the China-PAR and ICVD-10 risk assessment tools agree, and how much do the four prespecified cardiovascular high-risk identification pathways overlap among community-screened adults aged 35–79 years in Nanjing, China?
Although China-PAR and ICVD-10 identify nearly identical overall proportions of community-screened Chinese adults as high cardiovascular risk (~11%), they disagree on nearly half (46.29%) of individual high-risk classifications, underscoring the need for screening registries to record the specific triggering criterion and risk tool used.
Answers are grounded in this paper and cite their sources.
Screening proportions and tool comparisons in Nanjing adults should not yet change practice; leaves open optimal China-PAR versus ICVD-10 selection for Chinese populations.
Objective: To describe the proportion of screened adults aged 35–79 years identified as being at high risk of cardiovascular disease (CVD) in five selected areas of Nanjing; compare classifications by the Prediction for ASCVD Risk in China (China-PAR) and the 10-year ischemic cardiovascular disease risk assessment for Chinese adults (ICVD-10) among participants without prior CVD; characterize the four prespecified identification sources and their overlap; and explore factors associated with multiple-source identification. Methods: We analyzed cross-sectional screening data collected in 2020–2021. Of 38,488 eligible residents invited, 38,427 completed screening (response rate, 99.84%). High-risk identification was based on prior CVD, severe dyslipidemia, China-PAR risk ≥10%, or ICVD-10 risk ≥10%. Paired classifications and exact intersections were described. Confidence intervals, group comparisons, and an exploratory logistic regression of multiple-source identification used community-cluster robust variance. Results: Among 38,427 screened participants, 8,905 were identified as being at high risk (23.17%; cluster-robust 95% confidence interval [CI] = 21.80%–24.61%). China-PAR classified 3,957 of 35,814 participants without prior CVD as high risk (11.05%), and ICVD-10 classified 3,966 of 35,815 as high risk (11.07%). Among 35,814 participants with both classifications, 30,660 were non-high risk by both tools, 2,768 were high risk by both, 1,189 were high risk only by China-PAR, and 1,197 were high risk only by ICVD-10; discordant classifications accounted for 46.29% of participants classified as high risk by either tool. Among 8,904 high-risk participants with complete source data, 2,980 (33.47%) met at least two criteria. In exploratory cluster-robust regression, male sex, older age, study area, and harmful alcohol use were associated with multiple-source identification; insufficient leisure-time physical activity was not statistically significant (adjusted odds ratio [aOR] = 1.14; 95% CI = 0.996–1.32; P = 0.057). Conclusions: The four prespecified identification sources overlapped substantially in this screened sample. China-PAR and ICVD-10 produced similar overall high-risk proportions, but discordant classifications for nearly half of participants classified as high risk by either tool. Screening registries should retain the specific triggering source and the risk tool used to support appropriate verification and follow-up. Lay Summary Community screening may identify cardiovascular high risk from prior disease, very high cholesterol, or one of two Chinese risk tools. In five selected areas of Nanjing, 23.17% of screened adults were high risk; China-PAR and ICVD-10 gave similar totals, but nearly half of those high risk by either tool were classified differently. Recording why a person was labeled high risk—and which tool was used—can direct appropriate review and follow-up.
No takes yet. Share an insight, caveat, or question.
Ma et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: