Use of CT for Head Trauma: 2007–2015

Abstract

International efforts have been focused on identifying children at low risk of clinically important traumatic brain injury in whom computed tomography (CT) neuroimaging can be avoided. We sought to determine if CT use for pediatric head trauma has decreased among US emergency departments (EDs). This was a cross-sectional analysis of the National Hospital Ambulatory Care Medical Survey database of nationally representative ED visits from 2007 to 2015. We included children <18 years of age evaluated in the ED for head injury. Survey weighting procedures were used to estimate the annual proportion of children who underwent CT neuroimaging and to perform multivariable logistic regression. There were an estimated 14.3 million pediatric head trauma visits during the 9-year study period. Overall, 32% (95% confidence interval [CI]: 29%-35%) of children underwent CT neuroimaging with no significant annual linear trend (P trend = .50). Multivariate analysis similarly revealed no difference by year (adjusted odds ratio [aOR]: 1.02; 95% CI: 0.97-1.07) after adjustment for patient- and ED-level covariates. CT use was associated with age ≥2 years (aOR: 1.51; 95% CI: 1.13-2.01), white race (aOR: 1.43; 95% CI: 1.10-1.86), highest triage acuity (aOR: 8.24 [95% CI: 4.00-16.95]; P < .001), and presentation to a nonteaching (aOR: 1.47; 95% CI: 1.05-2.06) or nonpediatric (aOR: 1.53; 95% CI: 1.05-2.23) hospital. CT neuroimaging did not decrease from 2007 to 2015. Findings suggest an important need for quality improvement initiatives to decrease CT use among children with head injuries.