Factors Affecting Emergency Department Computed Tomography Use in Children

J Surg Res. 2019 Sep:241:294-301. doi: 10.1016/j.jss.2019.04.014. Epub 2019 Apr 29.

Abstract

Background: Variability in computed tomography (CT) use during pediatric emergency department (ED) visits has been reported. Our objective was to identify patient and hospital characteristics associated with CT use during pediatric ED visits.

Methods: Patients <18 y treated and released from EDs in the 2006-2012 Nationwide Emergency Department Sample were included. Associations were evaluated between pediatric CT scan rate and patient/hospital factors using logistic mixed effects models. Independent predictors of being a high outlier (having a pediatric CT scan rate in the top 10%) were also evaluated using logistic regression models.

Results: There were 1543 EDs and 20,703,273 visits included. CT scans were prescribed in 4.7% of pediatric ED visits; the highest 10% of EDs prescribed CT scans in >7.63% of all pediatric visits. In multivariable analysis, older age, male gender, private insurance, higher zip code level median income, and higher injury severity were all associated with an increased probability of receiving a CT scan (all P < 0.001). The chance of receiving a CT scan also varied by diagnosis and was independently associated with geographic location and annual pediatric ED volume. Rates of CT use increased with increasing pediatric volume up to approximately 5400 annual pediatric visits, and then decreased with volume >5400 annual visits.

Conclusions: Several patient-level and ED-level characteristics, including annual pediatric volume, are associated with the probability of a child having a CT scan during an ED visit. Future work should focus on determining drivers behind these associations to develop intervention strategies to decrease pediatric CT use.

Keywords: Computed tomography; Emergency medicine; NEDS; Pediatric; Radiation; Utilization.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Child
  • Child, Preschool
  • Emergency Service, Hospital / statistics & numerical data*
  • Female
  • Humans
  • Infant
  • Logistic Models
  • Male
  • Retrospective Studies
  • Tomography, X-Ray Computed / statistics & numerical data*
  • United States
  • Workload / statistics & numerical data