Migraine headache presents simultaneously with early symptoms in most patients with MS, showing association with MS prevalence and incidence, but the pathological significance of headache remains unclear. Self-reporting instruments offer an opportunity to study this relationship.
Cross-sectional study of self-reported headache frequency, MS disability, severity, and disease-modifying therapy (DMT) retrospectively using a de‑identified, archival, database from MS Care ConnectTM, a digital application (InterPRO Bioscience).
Patient survey responses were transformed into number of headache days per month (HD). A generalized linear mixed-effects (GLMM) model was performed to test the association of HD with: sex, age, weight, disease modifying therapy class, patient self-reported history of MS, MS phenotype, DMT, and either Patient-Derived Disability Score (PDDS) or Patient-Derived MS Severity Score (P‑MSSS). Additional post hoc comparisons further tested associations among significant variables.
253 participants self-reported headache (79% CIS/RRMS, 10% PPMS/SPMS, others “unsure”; mean 45.4 y+11.4 SD, median 45.4 y). Of the 219 suffering >1 HD, 78.7% were female and used mainly highly effective IV and oral MS DMTs with a mean 9.0 HD (median 3). Responders self-reported PDDS (mean 2.2+ 2.0 SD) and yearly relapses mean 0.32 ± 0.6 with a calculated, P‑MSSS decile mean 3.67+2.54. No correlations were detected between HD and PDDS or P-MSSS. A small PPMS/SPMS cohort demonstrated higher median PDDS in the high HD group. Daily headache had highest median PDDS and P-MSSS in RRMS phenotype.
A Poisson GLMM revealed significantly lower HD for males, older subjects, and RRMS subjects over females, younger subjects, and progressive MS subjects, respectively. No interaction occurred among these three variables (age, gender, and MS phenotype). However, a skewed frequency distribution of DMT use by males (more likely to be treated) was associated with fewer HD than females; this skewing did not extend to DMT type. Progressive MS patients were less likely to be on any DMT despite having high HD. The model found non-significant contributions of family history of MS, DMT, P-MSSS and weight. When stratifying PDDS and age and comparing high and low HD groups, younger people (<45y) with high HD have lower PDDS than peers with low HD. In contrast, older people (45+y) have more HD with higher PDDS as compared to similarly aged peers with low HD, although this comparison did not reach significance.
A self-efficacy digital tool presented opportunities to study the interaction of migraine and MS. We significantly found that older, CIS/RRMS, or male subjects have fewer HD than younger, PPMS/SPMS, or female subjects, respectively. Age, MS phenotype, and gender predict HD in our best models and must be controlled in future studies; a relationship to PDDS and P-MSSS requires further investigation in a larger cohort.