Key result
Non-laboratory P-EOCVD model predicts early-onset cardiovascular disease and mortality with ~0.69 AUROC and ~0.72 C-index.
Why the study?
A questionnaire-based model for predicting early-onset cardiovascular disease and subsequent mortality in adults younger than 65 years was needed to enable risk stratification without laboratory testing.
Does a non-laboratory, questionnaire-based machine learning model accurately predict incident early-onset cardiovascular disease risk and subsequent all-cause mortality in adults younger than 65 years?
Cohort (n=56,560)
Yes
Does a non-laboratory, questionnaire-based machine learning model accurately predict incident early-onset cardiovascular disease risk and subsequent all-cause mortality in adults younger than 65 years?
A non-laboratory, questionnaire-based machine learning model demonstrates moderate discrimination for early-onset cardiovascular disease risk and mortality prediction, but its low precision limits it to first-step screening rather than diagnostic use.
Answers are grounded in this paper and cite their sources.
May limit non-laboratory models to preliminary screening; leaves open added value for EOCVD risk stratification in practice.
Background and aims This study aimed to develop and externally validate a questionnaire-based model for two related tasks in adults younger than 65 years: classification of incident early-onset cardiovascular disease (EOCVD) among participants without cardiovascular disease at baseline, and time-to-event prediction of all-cause mortality among participants with EOCVD at baseline. Methods Longitudinal data were obtained from CHARLS, HRS, KLoSA, ELSA, and SHARE. The incidence classification analysis included 37,082 participants, and the prognostic survival analysis included 19,478. Eighteen questionnaire or routinely obtainable predictors were used. No laboratory biomarker or imaging measurement was entered into either model, although reported diagnoses may have depended on earlier clinical testing. Results CatBoost achieved an AUROC of 0.6869, an AUPRC of 0.1438, a precision of 0.1105, and a recall of 0.6512 in the independent SHARE cohort. The Cox proportional hazards model achieved a test-set C-index of 0.7175; its 3-, 5-, and 7-year AUROCs were 0.7314, 0.7348, and 0.7350. Conclusions P-EOCVD provides questionnaire-based risk stratification for incident EOCVD and subsequent mortality. Its low precision precludes diagnostic use. The model may support first-step screening where laboratory testing is not immediately available, but prospective validation and recalibration are required before use in under-resourced populations.
No takes yet. Share an insight, caveat, or question.
Shanjie Luan (2026) conducted a cohort in Early-onset cardiovascular disease (EOCVD) (n=56,560). Non-laboratory prediction model (P-EOCVD) was evaluated on Incident early-onset cardiovascular disease (EOCVD) and time-to-event all-cause mortality (AUROC 0.6869 / C-index 0.7175). The non-laboratory prediction model classified incident early-onset cardiovascular disease with an AUROC of 0.6869 and predicted all-cause mortality with a C-index of 0.7175.
Synapse has enriched 10 related papers. Consider them for comparative context: