Importance: Though it is presumed that children and adolescents with migraine are at risk of internalizing symptoms and disorders, high-level summative evidence to support this clinical belief is lacking.
Objective: To determine if there is an association between internalizing symptoms and disorders and migraine in children and adolescents.
Data sources: A librarian-led, peer-reviewed search was performed using MEDLINE, Embase, PsycINFO, and CINAHL databases (inception to March 28, 2022).
Study selection: Case-control, cohort, and cross-sectional studies on the association between internalizing symptoms and disorders and migraine in children and adolescents 18 years or younger were eligible.
Data extraction and synthesis: Two investigators independently completed abstract and full-text screening, data extraction, and quality appraisal using the Newcastle-Ottawa scales. Studies were pooled with random-effects meta-analyses using standardized mean differences (SMD) or odds ratios (OR) with 95% CIs. Where sufficient data for pooling were unavailable, studies were described qualitatively.
Main outcomes and measures: The primary outcome was migraine diagnosis; additional outcomes included migraine outcomes and incidence. Associations between these outcomes and internalizing symptoms and disorders were evaluated.
Results: The study team screened 4946 studies and included 80 studies in the systematic review. Seventy-four studies reported on the association between internalizing symptoms and disorders and migraine, and 51 studies were amenable to pooling. Meta-analyses comparing children and adolescents with migraine with healthy controls showed: (1) an association between migraine and anxiety symptoms (SMD, 1.13; 95% CI, 0.64-1.63); (2) an association between migraine and depressive symptoms (SMD, 0.67; 95% CI, 0.46-0.87); and (3) significantly higher odds of anxiety disorders (OR, 1.93, 95% CI, 1.49-2.50) and depressive disorders (OR, 2.01, 95% CI, 1.46-2.78) in those with, vs without, migraine. Stratification of results did not reveal differences between clinical vs community/population-based samples and there was no evidence of publication bias. Twenty studies assessing the association between internalizing symptoms or disorders and migraine outcomes (n = 18) or incident migraine (n = 2) were summarized descriptively given significant heterogeneity, with minimal conclusions drawn.
Conclusions and relevance: In this study, children and adolescents with migraine were at higher risk of anxiety and depression symptoms and disorders compared with healthy controls. It may be beneficial to routinely screen children and adolescents with migraine for anxiety and depression in clinical practice. It is unclear whether having anxiety and depressive symptoms or disorders has an affect on migraine outcomes or incidence.