LO86: The diagnosis of concussion in pediatric emergency departments: a prospective multicenter study

Introduction: Accurate identification of children with a concussion by emergency department (ED) physicians is important to initiate appropriate anticipatory guidance and management. In children meeting international criteria for concussion, we aimed to determine the proportion who were provided this diagnosis by the ED physician and which variables were associated with a physician-diagnosed concussion. We also compared persistent symptoms in concussion cases versus those with alternative diagnoses. Methods: This was a planned secondary analysis of a prospective, multicenter cohort study. Participants were children aged 5 through 17 years and met Zurich/Berlin International Consensus Statement criteria for concussion. The primary outcome was the proportion of study participants who were assigned a diagnosis of concussion by the treating ED physician. Based on available evidence, between 50% and 90% of children meeting international concussion criteria are also diagnosed by an ED physician as having a concussion. Assuming a worst case scenario that 50% of physicians would diagnose concussion, our anticipated study sample size of 2946 would be accompanied by a +2% margin of error at the 95% confidence level for the primary outcome. Results: Among the 2946 eligible children, 2340 [79.4% (95% CI 78.0, 80.8)] were diagnosed with a concussion by an ED physician. Twelve variables were associated with this ED diagnosis, five of which had an odds ratio (OR) > 1.5: older age (13-17 vs 5-7 years, OR=2.9), longer time to presentation (>16 vs. <16 hours, OR=2.1), nausea (OR=1.7), sport mechanism (OR=1.7), and amnesia (OR=1.6). In those with physician-diagnosed concussion versus no concussion, the frequency of persistent symptoms was 62.5% vs. 38.8% (p<0.0001) at one week, 46.3% vs. 25.8% (p<0.0001) at two weeks and 33.0% vs. 23.0% (p<0.0001) at four weeks. Conclusion: Most children meeting international criteria for concussion were provided this diagnosis by the ED physician. There were five variables which increased the odds of this diagnosis by at least 1.5-fold. Relative to international criteria, the more selective assignment of concussion by ED physicians was associated with a greater frequency of persistent concussion symptoms. Nevertheless, many patients with alternative diagnoses exhibited persistent concussive symptoms at all time points. Clinicians should therefore weigh the benefits and risks of strictly applying the Zurich/Berlin international criteria versus individual discretion.