Cox Proportional Hazards Analysis of Clinical and Demographic Determinants of Liver Cancer Survival
Abstract
Liver cancer, particularly hepatocellular carcinoma (HCC), remains one of the leading causes of cancer-related mortality worldwide, with survival outcomes strongly influenced by demographic, clinical and tumor-related factors. This study applied the Cox Proportional Hazards (Cox PH) model to examine prognostic determinants of survival among liver cancer patients. A dataset consisting of 40 patients was analyzed, with survival time as the dependent variable and age, gender, blood level and tumor size as predictors. Results indicated that tumor size and gender had relatively strong associations with hazard, while age and blood level showed weaker, non- significant effects. The estimated hazard function demonstrated that older patients (80 years) faced higher risks of mortality compared to younger patients, with hazard peaks around 10–20 months and 60 months. Model diagnostics, including Cox-Snell residuals, confirmed that the proportional hazards assumption was satisfied, supporting the adequacy of the model for the dataset. The findings suggest that tumor size and gender may serve as key prognostic indicators for liver cancer survival, while also highlighting the heightened vulnerability of older patients.
