Back

Genetic liability to ADHD and migraine risk: Mendelian randomization evidence for behavioural and psychiatric pathways

Luo, Y.; Dardani, C.; Wootton, R. E.; Sanderson, E.; Stergiakouli, E.

2026-09-10 epidemiology
10.64898/2026.09.08.26362527 medRxiv
Show abstract

Background: Attention-deficit/hyperactivity disorder (ADHD) and migraine frequently co-occur, but the mechanisms underlying this association remain unclear. Examining migraine with aura and without aura separately may help clarify whether ADHD-related pathways are shared across migraine subtypes or differ by aura status. Methods: Two-sample Mendelian randomization (MR) was used to examine the effect of genetic liability to ADHD on overall migraine and on migraine subtypes defined by aura status. MR-based mediation was then used to assess whether behavioural, psychiatric, sleep-related, education, and cardiometabolic traits contribute to these effects. Results: MR provided evidence for an effect of ADHD genetic liability on overall migraine risk (OR = 1.48, 95% CI =1.2, 1.82), with similar effect sizes for migraine with aura (MA) (OR=1.52, 95% CI = 1.18, 1.96) and migraine without aura (MO) (OR = 1.65, 95% CI = 1.19, 2.31). Depression, alcohol intake frequency, lifetime smoking index, ever smoke, risk taking behaviours, insomnia, and college education showed evidence for partial mediation, explaining approximately 34.17% (95% CI = 6.17%, 61.63%), 14.44% (95% CI = 0%, 28.87%), 37.25% (95% CI = 11.86%, 63.64%), and 29.24% (95% CI = 6.12%, 52.37%) of the total effect, individually. Subtype analyses suggested broadly similar patterns across MA and MO, with no clear subtype-specific differences. Conclusion: Our findings provide evidence that ADHD common variant genetic liability increases migraine risk, including both MA and MO. The association appears unlikely to be explained by a single pathway and may involve the contribution of depression-related, lifestyle-related, and education-related processes.

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.