Survival Time and Predictors of Breast Cancer Treatment Outcome at St. Paul’s Hospital Millennium Medical College, Oncology Unit, Addis Ababa, Ethiopia, 2024

Abstract

Background: The incidence and mortality rate of breast cancer is highest in Africa, especially in the Sub-Saharan African countries. Despite significant advances in early detection, treatment technologies, and targeted therapies, breast cancer treatment outcomes remain poor. The median survival time was 38.3 months. Different socioeconomic and cultural settings, budgetary constraints, and limited resources have all been demonstrated to affect treatment outcomes. It is especially concerning in low and middle-income countries where there is increasing trend of prevalence and mortality. Although some studies have been conducted, they are not recent and hence do not reflect the current utilization of treatment methods, they had small sample sizes and were conducted in different areas and set up. Thus, the study will fill these gaps and aim to assess the survival time and predictors of breast cancer treatment outcome. Objective: The main objective of this study is to assess the survival time and predictors of breast cancer treatment outcome at St. Paul’s Hospital Millennium Medical College, Oncology Unit, Addis Ababa, Ethiopia, 2024. Methods: A retrospective cohort study was conducted among 307 female adult patients with breast cancer undergoing treatment from January 2019 to December 2023 in the study setting. They were selected by simple random sampling using computer-generated method. A pre-tested data extraction tool was used to collect data from the patient medical records. Data was coded, cleaned, and entered into Epi info version 7.2 statistical package, checked for completeness and exported to SPSS version 27 software for analysis. Kaplan–Meier survival analysis was used to estimate the median survival time across different variables. A log-rank test was used to compare the observed differences in survival time among different groups. The Cox proportional hazards regression model was used to identify predictors of mortality. Results were presented using adjusted hazard ratios along their corresponding 95% confidence intervals. A significant relationship was declared at p-value of <0.05. The assumption of a Cox regression proportional hazard model was checked using Log-Log survival plot test.