Development and validation of a novel coronary artery disease risk prediction model

This study aims to develop a novel risk assessment tool for coronary artery disease (CAD) based on data of patients with chest pain in outpatient and emergency department, thereby facilitating the effective identification and management of high-risk patients.

A retrospective analysis was conducted on patients hospitalized for chest pain. Patients were divided into a control group and a CAD group based on angiographic results. Logistic regression was used to identify factors associated with CAD, and R-Studio was utilized to construct the CAD risk prediction model.

Multivariate logistic regression analysis indicated that age, gender, diabetes, ECG (electrocardiogram) ST-T changes, neutrophils (NE), coronary artery calcification (CAC), and typical chest pain were independent factors associated with CAD. Based on the results of multifactorial logistic analysis, the CAD risk prediction model built with R-Studio had a highest C-index of 0.909, and a validation cohort C-index of 0.897, demonstrating excellent predictive ability. Decision Curve Analysis showed that the model significantly outperformed others in terms of clinical net benefit.

The present study successfully developed a CAD risk assessment model based on Chinese population. This novel model could be used to assess CAD risk in patients with chest pain, optimize clinical decision making, and improve patient outcomes, regardless of whether it is applied in large hospitals or resource-limited Community Healthcare Center.