Validation of a migraine-specific questionnaire for use in family studies.

The availability of valid migraine-specific questionnaires is important when large numbers of migraine patients have to be analysed. The Finnish Migraine-Specific Questionnaire has been validated in two stages. In the first, a clinical diagnosis of migraine was reached, using International Headache Society criteria, in 100 consecutive patients. Migraine was then diagnosed independently on the basis of responses to the Finnish Migraine-Specific Questionnaire. In the second stage, responses to 100 questionnaires returned consecutively in a family study in progress were analysed, and respondents were contacted by telephone for interview and diagnosis of migraine. Contact proved impossible in six cases. The sensitivity of the questionnaire for migraine was 0.99 (167 out of 168; validation stages 1 and 2 combined) and specificity was 0.96 (25 out of 26 cases; validation stage 2). It also proved possible to differentiate between migraine with and without aura on the basis of responses to the Finnish Migraine-Specific Questionnaire: chance-corrected agreement (Cohen's kappa) was 0.804 in relation to diagnoses reached on the basis of responses to the Finnish Migraine-Specific Questionnaire and clinically was 0.858 in relation to diagnoses reached on the basis of responses to the Finnish Migraine-Specific Questionnaire combined with the results of the telephone interviews. A value for Cohen's kappa > 0.75 indicates good agreement. Therefore, use of the Finnish Migraine-Specific Questionnaire in research into migraine genetics is justified.