CLINICAL, GENETIC, AND LIFESTYLE PREDICTORS OF CORONARY ARTERY DISEASE IN MIDDLE-AGED POPULATIONS

Abstract

Background: Coronary artery disease remains a leading cause of global morbidity and mortality, particularly in middle-aged adults, where subclinical disease often precedes overt clinical events. Conventional risk prediction models frequently fail to capture the combined influence of genetic susceptibility and lifestyle behaviors, limiting their precision.Objective: To evaluate the integrated and independent contributions of clinical, genetic, and lifestyle predictors to coronary artery disease risk in middle-aged populations using a multimodal analytical framework.Methods: A mixed-methods, multicenter cohort study was conducted among adults aged 40–60 years. Clinical parameters, polygenic risk scores derived from validated CAD-associated loci, and comprehensive lifestyle assessments were integrated using multivariable and machine learning–based models. Model performance, interaction effects, and stability were systematically evaluated.Results: Integrated models demonstrated markedly improved predictive performance compared with clinical-only approaches, with significant gains in discrimination, stability, and risk convergence. Polygenic risk scores independently contributed to CAD prediction and identified high-risk individuals not captured by traditional models. Lifestyle factors significantly modified genetic risk, with favorable behaviors attenuating CAD susceptibility even in genetically predisposed individuals. Non-linear interactions between clinical burden, genetic risk, and lifestyle exposures were consistently observed, underscoring the complexity of CAD pathogenesis.Conclusion: Multimodal integration of clinical, genetic, and lifestyle data substantially enhances coronary artery disease risk prediction and enables personalized risk stratification. This approach supports the advancement of precision prevention strategies aimed at reducing CAD burden in middle-aged populations.