Table 1. Characteristics of Included Studies.
Characteristic | No. (%) | |||
---|---|---|---|---|
Association studies | Migraine outcomes studies | Incidence studies | ||
In meta-analysis | Not in meta-analysis | |||
No. of total studies | 51 | 29 | 18 | 2 |
Study design | ||||
Cross-sectional | 21 (41.2) | 10 (34.5) | 8 (44.4) | 0 |
Case-control | 28 (54.9) | 17 (58.6) | 8 (44.4) | 0 |
Cohort | 2 (3.9) | 2 (6.9) | 2 (11.2) | 2 (100.0) |
Setting | ||||
Clinical | 34 (66.7) | 19 (65.5) | 15 (83.3) | 0 |
Community | 5 (10.0) | 1 (3.5) | 1 (5.6) | 2 (100.0) |
Population-based | 8 (15.7) | 3 (10.3) | 0 | 0 |
Not reported | 4 (7.8) | 6 (20.7) | 2 (11.1) | 0 |
Sample size | ||||
<50 | 4 (7.8) | 5 (17.2) | 0 | 0 |
50-100 | 16 (31.4) | 8 (27.6) | 11 (61.1) | 0 |
>100-1000 | 20 (39.2) | 10 (34.5) | 4 (22.2) | 0 |
>1000 | 11 (21.6) | 6 (20.7) | 3 (16.7) | 2 (100.0) |
Ages | ||||
<12 y | 4 (7.8) | 0 | 0 | 0 |
≥12-18 y | 11 (21.6) | 7 (24.1) | 2 (11.2) | 2 (100.0) |
Mixed ages | 36 (70.6) | 22 (75.9) | 16 (88.8) | 0 |
Sex distribution | ||||
<50% Female | 11 (21.6) | 5 (17.2) | 4 (22.2) | 1 (50.0) |
50%-60% Female | 21 (41.2) | 8 (27.7) | 7 (38.9) | 1 (50) |
>60%-70% Female | 8 (15.7) | 0 | 3 (16.7) | 0 |
>70% Female | 8 (15.7) | 5 (17.2) | 3 (16.7) | 0 |
Not reported | 3 (5.9) | 11 (37.9) | 1 (5.5) | 0 |
Outcome type | ||||
Episodic migraine only | 5 (10.0) | 5 (17.2) | 1 (5.6) | 0 |
Chronic migraine only | 4 (7.8) | 1 (3.5) | 1 (5.6) | 1 (50.0) |
Mixed episodic and chronic migraine | 17 (33.3) | 6 (20.7) | 9 (50) | 1 (50.0) |
Not reported | 25 (49.0) | 17 (58.6) | 7 (38.8) | 0 |
Outcome ascertainment | ||||
ICHD criteria | 43 (84.3) | 14 (48.3) | 15 (83.3) | 1 (50.0) |
Not ICHD criteria | 6 (11.8) | 7 (24.1) | 2 (11.2) | 1 (50.0) |
Not reported | 2 (3.9) | 8 (27.6) | 1 (5.5) | 0 |
Results adjusted for important covariates (eg, age, sex) | ||||
Yes | 13 (25.5) | 5 (17.2) | 7 (38.9) | 2 (100.0) |
No | 38 (74.5) | 24 (82.8) | 11 (61.1) | 0 |
Study quality rating | ||||
High (>80%) | 12 (23.5) | 1 (3.4) | 2 (11.2) | 1 (50.0) |
Moderate (50%-80%) | 31 (60.8) | 14 (48.3) | 12 (66.6) | 1 (50.0) |
Low (<50%) | 8 (15.7) | 14 (48.3) | 4 (22.2) | 0 |
Exposure type(s) measureda | ||||
Anxiety symptoms | 21 | 10 | 9 | 0 |
Anxiety disorders | 19 | 6 | 3 | 0 |
Depressive symptoms | 21 | 15 | 9 | 0 |
Depressive disorders | 20 | 8 | 2 | 1 |
Mixed-internalizing symptoms | 13 | 5 | 2 | 1 |
Mixed-internalizing disorders | 10 | 3 | 2 | 0 |
Exposure assessment toolb | ||||
Symptom level | ||||
Validated questionnaire | ||||
Anxiety symptoms | 20 (95.2) | 10 (100.0) | 8 (88.9) | NA |
Depressive symptoms | 21 (100.0) | 14 (93.3) | 8 (88.9) | NA |
Mixed-internalizing symptoms | 13 (100.0) | 5 (100.0) | 2 (100.0) | 1 (100.0) |
Unvalidated tool | ||||
Anxiety symptoms | 1 (4.8) | 0 | 1 (11.1) | NA |
Depressive symptoms | 0 | 1 (6.7) | 1 (11.1) | NA |
Mixed-internalizing symptoms | 0 | 0 | 0 | 0 |
Disorder level | ||||
Validated diagnostic interview | ||||
Anxiety disorders | 12 (63.1) | 3 (50.0) | 1 (33.3) | NA |
Depressive disorders | 12 (60.0) | 3 (37.5) | 1 (50.0) | 0 |
Mixed-internalizing disorders | 0 | 1 (33.3) | 1 (50.0) | NA |
Validated questionnaire cutoff score | ||||
Anxiety disorders | 6 (31.6) | 1 (16.7) | 0 | NA |
Depressive disorders | 7 (35.0) | 2 (25.0) | 0 | 1 (100.0) |
Mixed-internalizing disorders | 10 (100.0) | 2 (66.7) | 1 (50.0) | NA |
Other | ||||
Anxiety disorders | 1 (5.3) | 2 (33.3) | 2 (66.7) | NA |
Depressive disorders | 1 (5.0) | 2 (25.0) | 1 (50.0) | 0 |
Mixed-internalizing disorders | 0 | 0 | 0 | NA |
Exposure-outcome association (most adjusted results) c | ||||
Positive association | ||||
Anxiety symptoms | 13 (61.9) | 4 (40.0) | 3 (33.3) | NA |
Anxiety disorders | 8 (42.1) | 2 (33.3) | 2 (66.7) | NA |
Depressive symptoms | 12 (57.1) | 11 (73.3) | 7 (87.5) | NA |
Depressive disorders | 4 (20.0) | 3 (37.5) | 0 | 1 (100.0) |
Mixed-internalizing symptoms | 12 (92.3) | 5 (100.0) | 2 (100.0) | 0 |
Mixed-internalizing disorders | 8 (80.0) | 0 | 1 (50.0) | NA |
No association | ||||
Anxiety symptoms | 5 (23.9) | 6 (60.0) | 4 (44.4) | NA |
Anxiety disorders | 2 (10.5) | 1 (16.7) | 0 | NA |
Depressive symptoms | 6 (28.6) | 3 (20.0) | 1 (12.5) | NA |
Depressive disorders | 4 (20.0) | 1 (12.5) | 1 (50.0) | 0 |
Mixed-internalizing symptoms | 0 | 0 | 0 | 1 (100.0) |
Mixed-internalizing disorders | 0 | 1 (33.3) | 1 (50.0) | NA |
Negative association | ||||
Anxiety symptoms | 0 | 0 | 1 (11.1) | NA |
Anxiety disorders | 0 | 0 | 0 | NA |
Depressive symptoms | 0 | 0 | 0 | NA |
Depressive disorders | 0 | 0 | 0 | 0 |
Mixed-internalizing symptoms | 0 | 0 | 0 | 0 |
Mixed-internalizing disorders | 0 | 0 | 0 | NA |
Measured, association not reported | ||||
Anxiety symptoms | 3 (14.2) | 0 | 1 (22.2) | NA |
Anxiety disorders | 9 (47.4) | 3 (50.0) | 1 (33.3) | NA |
Depressive symptoms | 3 (14.3) | 1 (6.7) | 0 | NA |
Depressive disorders | 12 (60.0) | 4 (50.0) | 1 (50.0) | 0 |
Mixed-internalizing symptoms | 1 (7.7) | 0 | 0 | 0 |
Mixed-internalizing disorders | 2 (20.0) | 2 (66.7) | 0 | NA |
Abbreviations: ICHD, International Classification of Headache Disorders; NA, not applicable.
Because many studies reported on multiple exposure types, percentage not given.
Percentage reflects proportion of studies measuring a given exposure type using a given type of exposure assessment tool over the total number of studies measuring that exposure type.
Percentage reflects proportion of studies reporting a given association direction for a given type of exposure over the total number of studies measuring that exposure type. Here, a positive association denotes: (1) for association studies, that higher exposure to internalizing symptoms/disorders was associated with migraine diagnosis; (2) for migraine outcomes studies, that higher exposure to internalizing symptoms/disorders was associated with worse outcomes; and (3) for incidence studies, that higher exposure to internalizing symptoms/disorders was associated with higher incidence of migraine. Associations that are marked as not reported were either not reported because hypothesis testing was not done, or the migraine group was not compared with a healthy control group (eg, compared with other headache control), or to any control group.