Objectives.—To determine the prevalence of migraine and its association with age, gender, and social class and to find out whether or not the headache and nonheadache characteristics differ between children with migraine, with and without aura, using the diagnostic criteria of the International Headache Society for childhood migraine.
Design.—Population-based study in two stages comprising an initial screening questionnaire followed by telephone interviews of children with symptoms.
Setting.—Eighteen kindergarten and 39 primary and secondary schools in Thessaloniki and its semiurban areas.
Subjects.—Four thousand children, aged 4 to 15 years, representing a random sample of 5% of schoolchildren in Thessaloniki and its semiurban areas.
Main Outcome Measures.—(1) The prevalence of migraine, (2) the connection of migraine with social class, (3) differences in the occurrence of individual symptoms between migraine with and without aura.
Results.—The results of the present study show that migraine prevalence was 6.2% (95% confidence interval [CI], 5.4 to 7.0). The estimated prevalences of migraine with and without aura were 2.8% (95% CI, 2.3 to 3.4) and 3.4% (CI, 2.8 to 4.0), respectively. The prevalence of migraine increased with age and it was found to be almost equal in boys and girls aged 7 to 9 years or younger, but in older age groups the prevalence was higher in girls than in boys. The data showed no evidence that connected migraine with social class. It also showed that except for the aura, the headache (eg, frequency, duration, location, quality, and severity) and nonheadache (eg, nausea, vomiting, phonophobia, and photophobia) characteristics were no different between children with migraine, with and without aura. In conclusion, our findings indicate that migraine is a common underdiagnosed cause of severe recurrent headache in children. The findings show that childhood migraine is not connected with social class and varies with age and gender, and that except for the aura, both migraine with and without aura are so similar in their headache and nonheadache clinical characteristics that a common pathogenesis is plausible.