459 Identification of Characteristics of Patients Admitted to the Burn Unit for One Day

Abstract

Patients with burns less than 20% total body surface area (TBSA) make up the majority of burn admissions, yet there is minimal data in the literature pertaining to these patients. The purpose of this study was to analyze and identify the characteristics of patients with small burns with hospital length of stay (LOS) of one day in an attempt to reduce inappropriate admissions into the burn center. This was a retrospective cohort study of adult patients with less than 20% TBSA burns who were admitted to the burn unit over three years (2015-2017). Patients were stratified into two cohorts by LOS = 1 and LOS ≥ 2 days. Patients without a thermal injury, died within one day or underwent an operative procedure were excluded. The following variables were recorded and analyzed: demographics, TBSA, etiology and mechanism of injury, burn location, LOS, and mortality. Comorbidities such as hypertension, diabetes mellitus, seizure disorders, and psychiatric disorders were also recorded. The main outcome under study was LOS. Wilcoxon rank sum tests, Fishers’ exact test, and multivariable logistic regression were used to analyze the association of patient or injury characteristics and LOS. During the study period, there were 1007 admissions with 920 (91%) having burns < 20% TBSA of which 324 required an operative procedure and four others were excluded for other reasons. Thus, 592 patients met the inclusion criteria, 254 (42.9%) had a LOS of 1 day and 338 (57.1%) had a LOS ≥ 2 days. The median length of stay for those with LOS ≥ 2 days was 4 days (IQR 2, 7). There were no significant differences in age, BMI, or race between two groups. Males compromised 72% of the patients. In multivariable analysis, males had a lower probability for LOS beyond 1 day compared to females (OR=0.62, CI 95% [0.40-0.94], p=0.025). Psychiatric illness was the comorbidity with the strongest relationship for LOS (OR=1.90, CI 95% [1.20-3.03], p<0.001). Patients with LOS = 1 had a smaller percent TBSA, (2.5 (0, 15.5) vs. 5 (0.1, 19.0), p<0.001). TBSA was the predominant influencing factor on LOS beyond one day (OR=1.19 per 1% increment, CI 95% [1.12-1.25], p<0.001). Location of burn was found not to have a significant effect between the two cohorts. Of all admissions, 25% stayed for only one day. The multivariable analysis identified key factors that influence hospital LOS such as patients with LOS = 1 had a smaller percent TBSA and location of burn was found not to have a significant effect. Application of this study highlights the potential to reduce the health care burden and improve resource utilization by identifying patients requiring one day admission to the burn unit.