Length of Stay Per Total Body Surface Area Burn: A Validation Study Using the National Burn Registry.

Abstract

A length of stay (LOS) of one day per percent total body surface area (TBSA) burn has been generally accepted but not validated in current pediatric burn studies. The primary objective of this study is to validate previous Pediatric Injury Quality Improvement Collaboration (PIQIC) findings by using a national burn registry to evaluate LOS per TBSA burn relative to burn mechanism, sociodemographic characteristics, and clinical factors which influence this ratio. We evaluated patients 0-18 years old who sustained a burn injury and whose demographics were submitted to the National Burn Registry (NBR) dataset from July 2008 through June 2018. Mixed effects generalized additive regression models were performed to identify characteristics associated with the LOS per TBSA burn ratio. Among 51,561 pediatric burn patients, 45% were Non-Hispanic White, 58% were male, and median age was 3.0 years old (IQR: 1.0, 9.0). The most common burn mechanism was scald (55.9%). The median LOS per TBSA burn ratio across all cases was 0.9 (IQR: 0.4, 1.75). In adjusted models, scald burns had a mean predicted LOS per TBSA burn value of 1.2 while chemical burns had the highest ratio (4.8). Non-Hispanic White patients had lower LOS per TBSA burn ratios than all other races and ethnicities (p < .05). These data substantiate evidence on variance in LOS per TBSA burn relative to burn mechanism and race/ethnicity. Knowing these variations can guide expectations in hospital LOS for patients and families and help burn centers benchmark their clinical performance.