A Study on Factors Associated with Mortality of Patients Admitted with Burns at the University Teaching Hospitals, Lusaka, Zambia

Abstract

The mortality rate following burn injury is an important outcome parameter. Burns more than 30% are associated with high mortality rate. In Zambia, there is no comprehensive model to predict mortality in hospitalized burn patients. In the surgical practice at UTH, there is high mortality of patients with less than thirty percent burns. The main goal of this study was to determine factors associated with mortality in patients with burns less than 30% treated at UTH. This was an observational prospective cross-sectional study performed during a 12-months period in the Department of Surgery at UTH, Lusaka, Zambia. During the study period, 269 patients who met the criterion of the study were enrolled. Multivariate logistic regression analyses were performed to identify factors associated with mortality. The mean age of the patients was 6.79 years (SD=11.1). Mean percent burned total body surface area (TBSA) was 13.7% (SD=6.9). There was no statistically significant association between age, sex and mortality in my study (p=0.48). The study found no statistically significant association between medical history and mortality (p=0.861). The mortality rate in this study was 13.8%. From the logistic regression, patients with patients with anaemia had a 5-fold likelihood of mortality; those with sepsis 3-fold likelihood of mortality; and those with difficulties in breathing were 12 times more likely to die than those without. The model calibrated well, as the predictions did not significantly differ suggesting a good model fit (χ2=14.544, p = 0.069).The present study showed that clinical characteristics such as fever, difficulties in breathing, anaemia, no weekly haemoglobin check-ups, sepsis, and pneumonia were found to be factors associated with mortality in patients with burns less than 30% TBSA Early detection of these factors leading to morbidity would reduce mortality in burn patients.