Evaluation of Coronary Heart Disease Risk Factors Using Cox and Stratified Cox Proportional Hazards Models Before and After Propensity Score Matching

Coronary Heart Disease (CHD) remains a major cause of mortality globally and in Indonesia, with rising prevalence and healthcare burden. This study investigates the relationship between cardiovascular risk factors and CHD incidence using a Cox Proportional Hazards model, complemented by Propensity Score Matching (PSM) to address confounding. A retrospective cohort design utilized data from the Indonesian Ministry of Health’s Non-Communicable Disease (NCD) Risk Factor Cohort Study (2011–2019), involving 2,984 participants from Bogor City. Cox regression estimated hazard ratios (HRs) for various risk factors, while Kaplan-Meier curves and log-rank tests were applied for survival analysis. PSM was conducted for sex and smoking status to improve group comparability. Before matching, high blood pressure (HR ≈ 47.7), LDL (HR ≈ 33.3), cholesterol (HR ≈ 19.9), blood sugar (HR ≈ 13.4), and obesity (HR ≈ 11.0) were dominant predictors of CHD. After matching, sex (HR ≈ 1.43, p = 0.03) and smoking (p < 0.001) also showed significant associations. Non-controllable factors such as age (HR ≈ 1.60) and family history contributed notably. The proportional hazard assumption was violated for age and smoking, necessitating stratified Cox models. The findings underscore the significant influence of both controllable such as obesity, cholesterol, smoking and un-controllable such as age,and sex risk factors on CHD. The use of PSM enhanced the validity of risk estimates, highlighting the importance of robust analytical methods in epidemiological research. Targeted prevention focusing on controllable risks is essential to curb CHD incidence.