Waning light, waxing pain: The lunar cycle's association with migraine headache occurrence

Abstract Objective To examine circalunar rhythms in migraine headache occurrence in a prospective cohort of 98 adults with episodic migraine. Background Migraine is a prevalent neurological disorder characterized by paroxysmal attacks. While time‐of‐day and seasonal rhythmicity in migraine occurrence have been described, little is known about circalunar patterns. Understanding these rhythms may inform headache prediction and guide personalized preventive medication timing. Methods We performed a secondary, post‐hoc analysis using data from a prospective cohort study (March 2016–October 2017). Participants completed twice‐daily electronic diaries, recording various characteristics including headache, and, when applicable, menstrual cycle timing. Participants wore wrist actigraphs for 6 weeks. We tested for a 30‐day circalunar rhythm in headache risk at the population‐level. We then examined for lunar synchronized rhythms in individuals, adjusting our analysis for participant‐specific factors and differences in baseline headache risk. Sleep characteristics and menstrual timing were assessed as potential mediators of the relationship between lunar phase and headache occurrence. Results Ninety‐eight participants were followed for a median length of 43 days (interquartile range [IQR]: 42–45 days) with an average of 24.2% (standard deviation [SD] 13.2) of those days being headache days. Population‐level analysis showed a significant relationship between lunar phase and headache risk (trough‐to‐peak odds ratio: 1.2 [95% confidence interval {CI}: 1.04, 1.49]). Individual‐level analysis, adjusted for age, sex/menopausal status, and preventative medication use, showed 1.34 (95% CI: 1.1, 1.72) times higher headache odds at lunar cycle peak versus trough. Risk peaked 1–2 days before new moon. Neither sleep characteristics nor menstrual timing appeared to mediate the lunar phase‐headache relationship. Conclusions Headache risk varied with the lunar cycle, with 34% higher odds shortly before new moon compared to before full moon. Sleep and menstrual cycle timing did not appear to explain this relationship, suggesting underlying chronobiological mechanisms. Additional studies are needed to confirm these findings and characterize individual variability in rhythmicity. Plain Language Summary image