Special review article A meta-analysis investigating the prevalence and moderators of migraines among people with bipolar disorder Michele Fornaro a,b,n , Brendon Stubbs c a New York Psychiatric Institute, Columbia University, NYC, USA b Department of Education Sciences, University of Catania, Italy c Faculty of Education and Health, University of Greenwich, Southwood Site, Avery Hill Road, Eltham, London SE9 2UG, UK article info Article history: Received 19 February 2015 Accepted 26 February 2015 Available online 9 March 2015 Keywords: Migraine Bipolar disorder (BD) Comorbidity Prevalence Meta-analysis abstract Background: Uncertainty exists regarding the prevalence and moderators of migraine comorbidity among people with bipolar disorder (BD). We conducted a meta-analysis and meta-regression to investigate the prevalence and moderators of migraine among people with BD. Method: Two authors independently searched major electronic databases from inception till 02/2015. Articles were included that reported the prevalence of migraine in people with BD with or without a control group. A random effects meta-analysis and exploratory meta-regression were conducted. Results: Fourteen studies were included encompassing 3976 individuals with BD (mean age 35.5 years, SD 7.6, 29% male). The overall pooled prevalence of migraine was 34.8% (95% CI ¼25.54–44.69). The prevalence of migraine was higher among people with BD-II (54.17%, 95% CI ¼31.52–75.95, n ¼742) compared to BD-I (32.7%, 95% CI ¼18.16–49.19, n ¼2138, z ¼3.97, p o0.0001). The prevalence of migraine was 33.9% (95% CI ¼26.02–42.44), 39.5% (95% CI ¼18.81–62.39) and 47.11% (95% CI ¼22.24–72.77) in North America, Europe and South America respectively. The prevalence of migraine was higher when classified according to recognized criteria at 47.91% (95% CI ¼32.51–63.5) compared to non-recognized criteria (20.0%, 95% CI ¼12.44–29.06, z ¼À8.40, p o0.0001). Meta regression suggests mean age may be a potential moderator. Conclusion: Migraine is common and burdensome among people with BD. People with BD-II appear to be particularly affected. Nonetheless, future research is required to better understand these relationships, with a special emphasis toward the course specifiers of comorbid migraine cases of either BD-I vs. BD-II. & 2015 Elsevier B.V. All rights reserved. Contents 1. Introduction ......................................................................................................... 89 2. Material and methods ................................................................................................. 89 2.1. Eligibility criteria ............................................................................................... 89 2.2. Information sources and searches.................................................................................. 90 2.3. Study selection................................................................................................. 90 2.4. Data collection process .......................................................................................... 90 2.5. Risk of bias in individual studies .................................................................................. 90 2.6. Meta-analysis .................................................................................................. 90 2.7. Quality assessment and risk of bias across the studies ................................................................. 90 3. Results ............................................................................................................. 90 3.1. Study selection................................................................................................. 90 3.2. Included study and participant characteristics ........................................................................ 91 3.3. Quality score and quality differentiation results ...................................................................... 91 3.4. Meta-analysis of the prevalence of migraine in people with bipolar disorder ............................................... 91 3.5. Sub-group analyses ............................................................................................. 91 Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/jad Journal of Affective Disorders http://dx.doi.org/10.1016/j.jad.2015.02.032 0165-0327/& 2015 Elsevier B.V. All rights reserved. n Correspondence to: New York Psychiatric Institute, Columbia University,1051 Riverside Drive, New York, NY 10032, unit 2712, USA. Tel.: þ1 646 774 7652. E-mail addresses: mf3000@columbia.edu (M. Fornaro), b.stubbs@greenwich.ac.uk (B. Stubbs). Journal of Affective Disorders 178 (2015) 88–97