Comparison of Cox Proportional Hazard Models with Interaction and Without Interaction in Heart Failure Patients

One of the leading causes of morbidity and mortality worldwide is heart failure. The prevalence of heart failure in Indonesia reached 5% in 2020. One survival analysis method to identify factors affecting survival time is the Cox Proportional Hazard model. In this study, models with and without variable interactions were compared to model the survival time of heart failure patients using the Cox Proportional Hazard method. The research data consisted of 299 heart failure patients, including variables of high blood pressure, anemia, and age. The analysis results showed that the variables of high blood pressure and age significantly affected the survival time of patients, while anemia status and interactions between independent variables were not significant. The model without interaction had a lower AIC value than the model with interaction, making it the best model. Based on the best model, patients with high blood pressure were 1.52 times more likely to die than patients without high blood pressure. Additionally, the probability of death increased by 4.33% for every one-year increase in age. This study concludes that the model without interaction is more suitable for describing the relationship between independent variables and survival time in heart failurepatients.