The Burns Evaluation and Mortality Study (BEAMS): predicting deaths in Australian and New Zealand burn patients admitted to intensive care with burns.

BACKGROUND An understanding of prognosis following burns is important. It alleviates patient and familial stress, provides a framework for better resource use, and facilitates benchmarking of performance between specialist centers. METHODS Data were collected from eight tertiary referral burns centers in Australia and New Zealand. Our aim was to identify factors independently associated with mortality to develop a mortality prediction model, which accurately quantifies the risk of death among adults with burns who require intensive care. RESULTS Between January 2005 and December 2011, 1,715 patients were admitted to intensive care unit with acute thermal burns. The mean (SD) age was 41.1 (18.0) years, and 20.3% (n = 348) were female. Median percentage of total body surface area was 17% (6-35%) and percent full-thickness surface area was 4% (0-20%). Inhalational injury was documented as present in 36.2%. Accidental injury was the most common etiology of burn (70.4%) and most frequently via a flame (68.3%). Overall hospital mortality was 10.9% (n = 187). Independent risk factors for death were age (odds ratio [OR], 1.08; 95% confidence interval [CI], 1.06-1.10; p < 0.001), percentage of full-thickness surface area (OR, 1.07; 95% CI 1.06-1.08; p < 0.001), and Acute Physiology and Chronic Health Evaluation II score excluding age (OR, 1.11; 95% CI, 1.07-1.15; p < 0.001) and female sex (OR, 3.35; 95% CI, 1.84-6.11; p = 0.001). There was no association between inhalational injury or deliberate self-harm and death, as well as etiology or type of burn. CONCLUSION A highly discriminatory mortality prediction model was developed using logistic regression. Risk of death following major burns can be predicted from a combination of physiologic and burns specific parameters. Female sex is a highly significant risk factor.