Survival Times of Breast Cancer Patients in Nigeria: Application of Cox and Parametric Survival Models

Abstract

In this study, we modeled the survival time of breast cancer patients in Nigeria using five survival models, namely the Cox model, the exponential model, the lognormal model, the logistic model, and the Weibull model. The Akaike information criterion (AIC) and Bayesian information criterion (BIC) were used as performance metrics for the selection of the best-fit model. The Cox proportional hazard (CPH) model was the best model for the cancer data. We also noted that the median patient survival time was 295 days. The Kaplan-Meier test was used to compare the survival curves. The CPH model was used to model the data. We observed that the neoadjuvant therapy covariate had a significant effect on the survival time of the breast cancer patients (p < 0.05). This suggests that it has a considerable impact on Nigerian breast cancer patients' survival rates. This study could result in more efficient cancer treatments and has substantial implications for the management and care of breast cancer patients in Nigeria. It further extends the work of Awodutire et al. (2017).