Evidence-based guidelines for the treatment of children with migraine are limited given the paucity of randomized controlled trials, especially in the emergency department (ED). Our objectives were to: (1) characterize the treatment of children with migraine in the ED; (2) determine whether treatment varies in pediatric versus mixed (pediatric and adult) EDs. Children aged 5 to 17 years presenting to 4 regional emergency departments in Edmonton, Alberta, Canada during the 2003/2004 fiscal year with a diagnostic code of headache or migraine were selected. A standardized retrospective chart abstraction was performed and migraine or probable migraine cases were classified based on the International Classification of Headache Disorders II. Three hundred and eighty-two cases were identified of which 48.7% (n = 186/382) met sufficient criteria for migraine. No treatment was given in 44.2% (n = 169/382). Simple oral analgesics (23.3%; n = 89/182) and dopamine antagonists (metoclopramide and prochlorperazine; 20.7%; n = 79/182) were prescribed first-line most commonly. Opiate medications (5.5%), ketorolac (4.7%), dihydroergotamine (1%) were prescribed first-line infrequently. There was a significant difference in the management choices between pediatric and mixed adult/pediatric EDs (chi(2)= 19.695; df = 5; P= .001). The pediatric ED was more likely to prescribe a dopamine antagonist (12.9 vs 6.8%; P= .044) while the mixed adult/pediatric EDs were more likely to prescribe an opiate (28.1% vs 18.4%; P= .031). Children with migraine in all EDs were significantly more likely to receive drug therapy (68.3% vs 42.9%; P < .001) or a dopamine antagonist (32.3% vs 9.7%; P < .001). Polypharmacy (31.2%; n = 119/382) and neuroimaging (29.1%; n = 111/382) were common. Outcome was poorly documented overall. No adverse events were recorded. Significant variation in practice in the management of acute headaches in children was observed between mixed population and pediatric-only emergency physicians in the same city. Most children do not receive any drug therapy. Children presenting to the pediatric ED were significantly more likely to receive a dopamine antagonist while opiates were prescribed more commonly in the mixed ED. More clinical trials are required in children to clarify areas of clinical uncertainty on which evidence-based practice guidelines can be formed.