Migraine and depression risk in women: integrated evidence from the global burden of disease, UK Biobank, and Mendelian randomization
Abstract
Migraine and depression disproportionately affect women. However, their global trends, prospective associations, and genetic causal mechanisms remain largely uninvestigated. This study aimed to integrate multi-level evidence to comprehensively dissect the relationship between these two conditions in women. We analyzed temporal trends in incidence and DALY rates utilizing data from the Global Burden of Disease (GBD) Study 1990–2023. Within the UK Biobank (UKB) prospective cohort ( N = 249,322), Cox proportional hazards models were employed to examine the association between migraine and incident depression. Two-sample Mendelian randomization (MR) analyses, with supplementary bidirectional analysis using FinnGen data, were performed to evaluate genetic causality between the two disorders. Between 1990 and 2023, age-specific incidence and DALY rates of depression in women rose across all age groups, whereas those of migraine remained essentially stable, with slight increases only at middle and older ages. In 2023, migraine DALY rates remained high across 35–59 years, while depression rates declined steadily with age. In the UKB, migraine was associated with a significantly increased risk of incident depression(HR = 1.46, 95% CI 1.35–1.58), with a stronger effect observed in individuals aged under 60 years (HR = 1.62, 95%CI 1.46–1.78). MR analysis provided no statistical evidence for a direct causal effect of migraine on depression using current female-specific GWAS (OR = 1.03, 95% CI 0.97–1.08); however, limited power due to a small number of instrumental variables precludes a definitive exclusion of a subtle causal link. Conversely, genetically predicted depression significantly increased the risk of migraine in the sex-unspecific FinnGen cohort, which may reflect a shared cross-sex liability. Migraine was associated with an increased risk of incident depression in women, but female-specific MR analysis, with limited power, found no statistical evidence of a direct causal effect. Future studies should integrate multi-omics and longitudinal designs in larger female-specific cohorts. Not applicable.