Abstract / Summary
Objective: To identify demographic and daily cognitive, affective, and contextual predictors of headache-related interference with activity and concentration in a smartphone-based migraine cohort, using statistical models appropriate to the ordinal structure of patient-reported interference. Background: Migraine burden extends beyond headache to encompass cognitive and affective disturbance, yet most disability instruments summarize functioning retrospectively and cannot capture day-to-day variability in interference. Ecological momentary assessment addresses this limitation, yet few studies have examined cognitive and affective correlates of interference independent of pain intensity or have modeled interference using methods that respect its ordinal, non-equal-interval structure. Methods: From May 2025 to April 2026, adult participants were enrolled in this prospective, decentralized observational study via specialty clinics and social media campaigns. Eligibility required fulfilling ICHD-3 diagnostic criteria for migraine using a validated electronic survey and experiencing between 3 and 20 monthly headache days. Enrolled individuals recorded daily smartphone diary entries and performed three brief cognitive tests over a 30-day monitoring period. Daily interference with activities and concentration were captured on 5-point ordinal scales and analyzed via cumulative link mixed-effects models incorporating participant-specific random intercepts. Predictor selection across nested specifications relied on likelihood ratio testing and Akaike Information Criterion (AIC) metrics, prioritizing the more parsimonious model when goodness-of-fit was comparable. Results: The analytic sample comprised 1,267 headache-day observations from 143 participants. In the final models, brain fog (OR = 2.49 for activity, 3.19 for concentration), forgetfulness (OR = 1.48, 1.65), and mood disturbance (OR = 1.19, 1.31) were each independently associated with both outcomes (all p < 0.01), while age, sex, and employment status showed no significant association with either. Sleep quality (OR = 0.86) and anticipated next-day enjoyment (OR = 0.81) provided additional explanatory value for activity interference (p < 0.05) but not for concentration interference, for which the three-predictor model of brain fog, forgetfulness, and mood alone was retained as the most parsimonious final model. Conclusions: Cognitive and affective symptoms, rather than demographic or socioeconomic characteristics, are the primary correlates of headache-related functional interference in this cohort. Activity and concentration interference are explained by partially distinct sets of daily influences, supporting the clinical value of assessing cognitive and affective symptoms alongside conventional pain-focused evaluation.