Creation of a decision aid for goal setting after geriatric burns: a study from the prognostic assessment of life and limitations after trauma in the elderly [PALLIATE] consortium.

OBJECTIVES:We hypothesized that a decision-support aid to predict index admission mortality and discharge disposition for geriatric burns could be constructed using the well-accepted Baux score (age +total body surface area burned) in a geriatric-specific cohort. METHODS:National Burn Repository version 8.0 (2002-2011) was queried for all subjects aged 65 years or older. Baux scores were calculated and patients grouped into deciles. Three discharge outcomes (death,home, discharge to nonhome setting) were measured per decile. A receiver operating characteristic analysis was used to determine optimal Baux score cutpoints based on the Youden Index. The odds of mortality at various Baux score cutoffs were estimated using logistic regression. RESULTS:The sample was composed of 8,001 subjects. Withdrawal of care was documented in 264 deaths; median time to withdrawal was three days. As Baux score increased, three peaks in disposition were seen. Less than 50% of patients with a Baux score of 80 or greater were discharged home. Patients with a moderate Baux score (80-130) had an increased likelihood of discharge to a nonhome setting. Baux scores of 130 or greater were nearly uniformly fatal (mortality, 94-100%). Baux score of 86.15 or less was predictive of discharge home (area under the curve, 0.698; sensitivity, 75.28%; specificity, 54.64%), and a score greater than 93.3 was predictive of mortality (area under the curve, 0.779; sensitivity, 57.46%; specificity, 87.08%). CONCLUSION:For geriatric patients whose Baux scores exceed 86, return-to-home rates drop drastically; mortality increases at a score greater than 93, and mortality is nearly universal at a score ≥130 or greater. We are piloting a display of these findings as a decision-making aid when setting goals of care with stakeholders after geriatric burns. LEVEL OF EVIDENCE:Epidemiologic/prognostic study, level III; therapeutic/care management, level IV.