Background.— Puberty is assumed to influence the occurrence of headache, increasing the risk for recurrent episodes, especially in girls. The increase of headache, in particular recurrent headache, in girls from around the age of 12 on, is often ascribed to the occurrence of menarche as the most stringent indicator of completed puberty.
Objectives.— We examined the hypothesis that the occurrence of menarche in girls is predictive of recurrent headache in reference to no or rare headache in the past 6 months. Furthermore, the assumption was tested that headache episodes increase after onset of menarche but remain unchanged in girls not having experienced their first menstruation. We also expected a higher probability of migraine type of headache after menarche. In a further analysis girls with and without menarche were compared with boys, and a difference between gender only for girls with menarche was predicted.
Methods.— In an epidemiological study recruiting 8800 families with children 7-14 years in Southern Lower Saxony (Germany), we conducted 3 yearly panels based on postal questionnaires. Headache data are based on self-report of children from 9 years on. Parents were asked for information regarding the onset of menarche. Approximately 1100 girls and about the same number of boys constitute the sample for our analyses.
Results.— The results of a logistic regression analysis demonstrate that in 11- to 16-year-old girls having experienced menarche in either the year of the assessment of headache or 2 years before that, the risk for recurrent headache is increased when compared with girls without menarche. Intraindividual longitudinal analyses, however, do not support our hypotheses that after the onset of menarche headaches become more frequent. The expected differences in headache between girls with menarche and boys were found. The risk for migraine-like headaches is not significantly higher after onset of menarche than before.
Conclusions.— Thus, results do not consistently support the hypothesis that puberty is a moderator of headache frequency. Even in cases where the outcome of regression analyses is supportive of our expectations, explained variance is diminutive (maximum 2.2%). Thus, the influence of menarche on headache seems to be only marginal.