Assessing the generalizability of a National Burn Database to the United States pediatric burn injury population

Abstract

Background: A national database is used to evaluate pediatric burn survivor outcomes, but the generalizability to the United States pediatric burn injury population is unclear, as only 60% of enrollees are U.S. residents. An increased understanding of population similarities and differences between residents of the United States and residents of Mexico treated in the United States within this database will help determine its ability to extrapolate.

Objective: Assess the generalizability of a national burn database to the U.S. pediatric burn injury population.

Design: This retrospective cohort study analyzed pediatric enrollees in a burn database. Data were stratified into children who were residents of the United States or Mexico at time of injury and grouped by age into <5 years, 5-10 years, and >10 years.

Setting: Not applicable.

Participants: A total of 2043 children enrolled in the database from 1998 to 2020.

Interventions: Not applicable.

Main outcomes measured: Determination of similarities and differences between gender, injury etiology, total body surface area injured, length of stay, and place of injury between the two groups.

Results: Both groups showed an increased incidence of burn injury among males as they aged. In U.S. residents, the most frequent etiology in patients <5 years was scald (292 [53.6%]), followed by fire/flame (157 [28.8%]). In residents of Mexico, fire/flame burns were more frequent in all ages and electric burns were more common among children >5 years. In both groups, outdoor injuries became more common as children aged. Children from Mexico had larger burns and longer lengths of hospital stay than children from the United States across all ages.

Conclusion: Differences between U.S. and Mexico groups were likely attributable to children with more severe burns being transferred to the United States for care, whereas those with low-severity burns were treated locally. This suggests that children from Mexico in the database were not representative of the pediatric burn injury population of the United States or Mexico more broadly and caution should be used before generalizations are made using this database.