To evaluate emergency department use and outcomes of neuroimaging for headache in a free-standing children's hospital system.Study design
We prospectively enrolled children aged 6-18 years who presented to the emergency department with a chief complaint of headache from September 2015 to September 2016. Standardized data collection was performed in real time, including telephone follow-up as needed, and imaging outcome was determined through a chart review. Using multivariable logistic regression, we estimated the associations between clinically important patient characteristics and neuroimaging.Results
Of 294 enrolled patients, 53 (18%) underwent neuroimaging (computed tomography or magnetic resonance imaging) and 2 (0.7%) had clinically important intracranial findings. Presenting with abnormal neurologic examination findings (OR, 11.55; 95% CI, 3.24-41.22), no history of similar headaches (OR, 2.13; 95% CI, 1.08-4.18), and white race (OR, 3.04; 95% CI, 1.51-6.12) were significantly associated with an increased odds of undergoing imaging in multivariable regression models.Conclusions
Our observed emergency department imaging rate was 26.5 times higher than our positive result rate, suggesting there is room to decrease unnecessary neuroimaging. Associations for abnormal examination and new headache type are consistent with the American Academy of Neurology clinical imaging recommendations. The increased odds of imaging white patients suggests bias that should be addressed. The low rate of positive findings supports the need for an evidence-based clinical decision tool for neuroimaging in the acute care setting.