Estimating the probability of survival of hospitalized burn patients at a tertiary hospital in Ghana

Abstract

This retrospective study estimated the survival patterns and mortality predictors among burn patients admitted to Komfo Anokye Teaching Hospital in Ghana, where burn-related mortality remains high due to challenges in providing comprehensive care. Data from 457 burn patients admitted from January 2021 to December 2023 were extracted from the Lightwave Health Information Management System. Descriptive analyses were conducted for continuous and categorical variables. Survival probabilities were estimated using Lifetable and Kaplan-Meier analyses, while Cox proportional hazard regression model was used to identify mortality predictors. Hazard ratios (HR) were used to estimate mortality risk with 95 % confidence intervals (CI) and a p-value of 0.05 to express the association between significant predictors and the treatment outcome (mortality). Data analysis was performed in Stata/SE Version 17.0. Diagnoses of Systemic Inflammatory Response Syndrome (SIRS) and inhalation injury as well as mechanical ventilation usage were associated with lower survival probabilities. Mortality risk increased with larger total body surface area (TBSA) burned (adjusted hazard ratio [aHR] = 1.05, 95 % CI: 1.04–1.06), full-thickness burns (aHR = 4.86, 95 % CI: 2.70–8.74), and inhalation injury (aHR = 1.66, 95 % CI: 1.05–2.61). Conversely, hemo-transfusion (aHR = 0.50, 95 % CI: 0.31–0.79) and surgical intervention (aHR = 0.15, 95 % CI: 0.06–0.37) were associated with improved survival. TBSA burned, full-thickness burns, and inhalation injuries significantly predicted increased mortality risk, while hemo-transfusion and surgical interventions were associated with improved survival. The findings of this study suggest that early surgical intervention significantly improved survival outcomes.