Sex differences in clinical characteristics of migraine and its burden: A population-based study.

Abstract

Understanding migraine in a sex-specific manner is crucial for improving clinical care, diagnosis, and therapy for both females and males. Here, we provide data on sex differences in the presentation of migraine in a large European-based population cohort, which is representative of the general population. We performed a population-based study of 62,672 Danish blood donors (both present and previous donors), of whom 12,658 had migraine. All participants completed a 105-item diagnostic migraine questionnaire send via electronic mailing system (e-Boks) between May 2020 and August 2020. The questionnaire allowed for correct diagnosis of migraine according to the International Classification of Headache Disorders third edition. The migraine questionnaire was in-cohort validated and had a positive predictive value of 97% for any migraine, a specificity of 93%, and a sensitivity of 93%. There were 9,184 females (mean age=45.1) and 3,434 males (mean age=48.0). The three-months prevalence of migraine without aura was 11% in females and 3.59% in males. The three-months prevalence of migraine with aura was 1.72% in females and 1.58% in males. In females, the age-related three-months prevalence of migraine without aura increased markedly during the childbearing age. In males, both migraine with- and without aura showed less age variation. Females had a higher frequency of migraine attacks (OR=1.22), but a lower frequency of non-migraine headaches (OR=0.35). Females also had a greater intensity of pain, more unilateral and pulsatile pain, and exacerbation by physical activity (OR=1.40-1.49) as well as more associated symptoms (OR=1.26-1.98). Females carried 79% of the total migraine disease burden, which was almost exclusively driven by migraine without aura (77%), while there was no sex difference in the disease burden of migraine with aura. Females have more severe disease, resulting in a much higher migraine disease burden than indicated by prevalence alone.