Impact of Anxiety and Depression on Migraine-Related Disability: Results from the Headache Assessment via Digital Platform in the United States (HeAD-US) Study

Background: Anxiety and depression are common psychiatric comorbidities among people with migraine and may contribute to increased disease burden. However, their impact on both disability during attacks and burden between attacks is not well-characterized in real-world studies. We aimed to assess the relationship between symptoms of anxiety and depression and both migraine-related disability and interictal burden in a large real-world sample of individuals with migraine. Methods: We used cross-sectional data from 6267 participants with migraine recruited using the Migraine Buddy app to create the Headache Assessment via Digital Platform in the United States (HeAD-US) registry. Disability was assessed using the Migraine Disability Assessment Scale (MIDAS), and interictal burden was measured using the Migraine Interictal Burden Scale (MIBS). Anxiety and depressive symptoms were evaluated using the 4-item Personal Health Questionnaire (PHQ-4). Participants were categorized into four subgroups: anxiety symptoms only, depressive symptoms only, both, or neither. Negative binomial regression was used to explore associations between subgroups and migraine disability or burden. Results: Participants were 41.5 +/- 13 years old; 90.8% were women. Coexisting anxiety and depression were more common in chronic than episodic migraine (30.3% vs. 20.0%, p<0.001). In adjusted models, anxiety symptoms alone were associated with an 8.7% increase in MIDAS scores and an 11.7% increase in MIBS scores; depressive symptoms alone with 28.4% and 15.5% increases, respectively; and coexisting anxiety and depressive symptoms with 32.5% higher MIDAS scores and 22.6% higher MIBS scores (all p<0.05). Conclusions: In this large real-world study, anxiety and depression were more prevalent in chronic migraine and independently associated with greater migraine-related disability and increased burden between attacks. These findings extend results from population-based and clinic studies to a digital health cohort.