Longitudinal assessment of comorbidities and co‐occurring conditions in adolescents with migraine: A secondary analysis of the National Longitudinal Study of Adolescent to Adult Health (Add Health)

OBJECTIVE: To investigate the links between adolescent migraine and comorbid and co-occurring conditions using a large, nationally representative longitudinal study. BACKGROUND: Comorbidities and co-occurring conditions play an important role in the clinical treatment of individuals with migraine. Research in this area has focused largely on the adult population using cross-sectional data, but less is known about adolescents and how conditions may co-occur over time from a broader developmental perspective. The goals of this manuscript were to empirically evaluate the associations between adolescent migraine and several linked conditions and explore the relative timings of onset of these conditions from adolescence to adulthood. METHODS: Data came from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a school-based study of the health-related behaviors and conditions of adolescents. The present study examined data from Wave 1 (W1, study years: 1994-1995), Wave 4 (W4, study years: 2008-2009), and Wave 5 (W5, study years: 2016-2018). Analyses and visual plots were used to evaluate potential links between parent-reported adolescent migraine status (PR-AdMig) at W1 and 15 medical conditions identified based on self-reported medical diagnoses (SR-MDs) at W4 and W5. Based on prior literature in adults, we identified 11 conditions predicted to be associated with PR-AdMig and four conditions predicted not to be associated with PR-AdMig. The analyses were exploratory and post hoc. RESULTS: The total sample size pooling over all analyses was n = 13,786, but the wave-specific sample sizes differed due to missing data (W4 analyses, n = 12,692; W5 analyses, n = 10,340); 7243/13,786 (unweighted: 52.5%; weighted: 50.5%) of participants were female, 7640/13,786 (unweighted: 55.4%; weighted: 68.6%) were White, and 1580/13,786 (unweighted: 11.5%; weighted: 12.0%) had PR-AdMig. The average ages were 15.8 years at W1, 28.7 years at W4, and 37.8 years at W5. Findings showed that PR-AdMig was associated with anxiety/panic disorder (W4: PR-AdMig vs. Control weighted %: 17.1% vs. 12.6%, unadjusted odds ratio [OR] = 1.43, 95% confidence interval [CI] 1.18-1.74, p = 0.0003; W5: 31.6% vs. 22.4%, OR = 1.60, 95% CI 1.28-2.02, p < 0.0001), asthma/chronic bronchitis/emphysema (W4: 20.0% vs. 14.7%, OR = 1.45, 95% CI 1.20-1.76, p < 0.001; W5: 21.0% vs. 14.6%, OR = 1.55, 95% CI 1.25-1.94, p < 0.001), attention deficit hyperactivity disorder (W4: 8.3% vs. 5.4%, OR = 1.58, 95% CI 1.18-2.10, p = 0.002), depression (W4: 23.7% vs. 15.4%, OR = 1.71, 95% CI 1.43-2.04, p < 0.0001; W5: 33.8% vs. 25.1%, OR = 1.53, 95% CI 1.22-1.90, p < 0.001), epilepsy/seizure disorder (W4: 2.2% vs. 1.2%, OR = 1.84, 95% CI 1.23-2.76, p = 0.004), migraine (W4: 38.8% vs. 11.9%, OR = 4.7, 95% CI 4.1-5.5, p < 0.001), post-traumatic stress disorder (W4: 4.1% vs. 2.8%, OR = 1.45, 95% CI 1.01-2.08, p = 0.042; W5: 11.3% vs. 7.1%, OR = 1.67, 95% CI 1.27-2.20, p < 0.001), and sleep apnea (W5: 11.0% vs. 7.6%, OR = 1.51, 95% CI 1.15-1.98, p = 0.003). Among theoretically unlinked conditions, only hepatitis C at W4 was shown to have a relationship with adolescent onset migraine (0.7% vs. 0.2%, OR = 3.63, 95% CI 1.32-10.0, p = 0.013). Visual plots suggested that the retrospective, self-report timing of onset of specific subsets of co-occurring conditions tended to group together over time. CONCLUSIONS: Consistent with the existing headache literature, results showed that adolescent migraine was associated with other medical and psychological conditions and visual plots suggested that there may be developmental patterns in the occurrence of migraine with other related conditions.