Interpretable Survival Modeling for Mortality Risk Stratification in Heart Failure Using Cox Proportional Hazard Regression.

BACKGROUND: Heart failure remains a major contributor to morbidity and mortality, highlighting the need for prognostic models that can accurately characterize survival risk while remaining interpretable for clinical use. Statistical survival models are well suited for this task, as they explicitly address time-to-event outcomes and censoring. METHODOLOGY: A retrospective survival analysis was conducted on a cohort of 299 patients diagnosed with heart failure. Time to all-cause mortality was analyzed using Cox proportional hazards regression, with right-censoring appropriately handled. The model incorporated routinely collected demographic, clinical, and laboratory variables. Internal validation was performed using bootstrap resampling to assess model stability and discriminative performance. RESULTS: During the follow-up period, 96 patients (32.1%) experienced the event of interest. The Cox model showed stable, moderate discriminative ability under resampling, with a concordance index close to 0.70. Renal function, anemia, age, hypertension, ejection fraction, and serum sodium were identified as independent predictors of mortality, with serum creatinine exhibiting the strongest association with adverse outcomes. CONCLUSIONS: Cox proportional hazards regression offers a statistically robust and clinically interpretable approach for mortality risk prediction in heart failure. Using routinely available clinical variables, the model provides reproducible prognostic insights and supports practical risk stratification in cardiovascular research.