Risk factors for mortality among hospitalised adult burn patients in a Malawian tertiary hospital burns unit
Background: Malawi has the highest rates of mortality directly or indirectly associated with burn injuries in Southern Africa. There is however limited published literature on risk factors of mortality among adult patients.
Methods: We conducted a retrospective cross-sectional audit of records of patients admitted at the burns unit of Queen Elizabeth Central Hospital (QECH) between the years 2007 and 2017. Death due to burns was our outcome of interest. We collected patient data including demographic information, details of the burn injury and its management and determined how these factors were associated with the risk of death using Pearson Chi-square tests in a univariate analysis and likelihood ratio tests in a multivariate logistic regression model. We also determined the odds ratios of death within the categories of the risk factors after adjusting for important variables using a logistic regression model.
Results: An analysis of 500 burns patient records showed that 132(26.4%) died during the 10-year period. The lethal area for 50% of burns (LA50) was 28.75% and mortality reached 100% at 40% total burn surface area. The following variables were found to have a significantly higher risk of mortality after controlling for confounders: increasing total burn surface area (p<0.0001) and inhalation burns (OR 5.2; 95% CI 2.0-13.3 p 0.0004). Scalds (OR 0.13; 95% CI 0.05-0.33; <0.0001), time-lapse to hospital presentation between 48 hours and one week (OR 0.27; 95%CI 0.11-0.68; <0.0001) and length of hospital stay greater than two months (OR 0.04 95%, CI 0.01-0.15; P<0.0001) were associated with a significantly lower risk of mortality.
Conclusions: The main predictors of death were burn size, scalds, time to presentation, and length of hospital stay. Patient survival can be improved by paying more attention to these factors.