Abstract
A significant proportion of readmissions occurs at a different hospital than the index admission, and is thus missed by current quality metrics. No study has examined all-hospital adult 30-day readmission rates, including different hospitals, following burn injury across the United States. The purpose of this study was to evaluate nationwide readmission rates, potential risk factors, and ultimately the burden of burn injury readmission, including readmission to a different hospital. The 2010-2014 Nationwide Readmissions Database was queried for patients admitted for burn. Multivariate logistic regression identified risk factors and associated cost for 30-day readmission at index and different hospitals. There were 94,759 patients admitted during the study period, with 7.4% (n = 7000) readmitted and of those, 29.2% (n = 2047) readmitted to a different hospital. The most common reason for readmission was infection (29.4% [n = 1990]). Risk factors for unplanned 30-day readmission to any hospital included burn of lower limbs (odds ratio [OR] 1.29, [1.21-1.37], P < .01), third degree burns (OR 1.31, [1.22-1.41], P < .01), Charlson Comorbidity Index ≥2 (OR 1.48, [1.37-1.60], P < .01), depression (OR 1.30, [1.19-1.41], P < .01), and psychoses (OR 1.53, [1.40-1.67], P < .01). Risk factors unique to readmission to a different hospital included: length of stay greater than 7 days (OR 2.07, [1.78-2.40], P < 0.01), and initial admission to a metropolitan teaching hospital (OR 1.50, [1.26-1.78], P < .01). Previously unreported, one in three burn readmissions nationally occur at a different hospital, have unique risk factors, and are costlier. Current hospital benchmarking underestimates readmission by failing to capture this unique subpopulation.
