A meta-analysis of anxiety disorder comorbidity in pediatric bipolar disorder

Abstract

Objectives: AD are among the most prevalent comorbid conditions in pediatric bipolar disorder (PBD). There are conflicting results in the literature regarding prevalence of AD subtypes, and significant discrepancy with PBD course of illness (episodic or chronic) or diagnostic criteria (narrow or broad). Our aim in conducting meta-analysis is to investigate the prevalence of subtypes of comorbid anxiety disorders and its relations with the onset (childhood or adolescent) and course (episodic or chronic) of PBD. Methods: We have conducted a systematic research of Pubmed by using “bipolar disorder, affective psychosis, generalized anxiety disorder, panic, social phobia, obsessive compulsive disorder and anxiety disorder” as keywords to search in title/abstract until September 2015. Among 3202 articles, a total of 430 abstracts were found to be related; 82 were conducted in pediatric population, which were read in full text by at least two authors and data was extracted for outcome measures. Articles that include the data from the same population sample were excluded. Data was analyzed with random effects model using R statistical program package. Results: Data from 33 studies were included in the final analysis. The prevalence of any AD in PBD was 44 percent (95% CI 0.38–0.50), prevalence of AD subtypes were GAD 25 percent (95% CI 0.18–0.36); Separation Anxiety Disorder (SAD) 22 percent (95% CI 0.14–0.33); OCD 17 percent (95% CI 0.11–0.23); Social Phobia (SP) 15 percent (95% CI 0.08–0.27); Panic Disorder (PD) 10 percent (95% CI 0.05–0.19). When only episodic PBD were concerned, prevalence rates differed, with any AD 38 percent (95% CI 0.28–0.48); GAD 19 percent (95% CI 0.08–0.41); SAD 21 percent (95% CI 0.10–0.40); OCD 11 percent (95% CI 0.03–0.29); SP 11 percent (95% CI 0.04–0.27); PD 9 percent (95% CI 0.03–0.23). Prevalence of any AD (34% (95% CI 0.23-0.48), GAD and SAD were found as lower and OCD, SP and PD were slightly higher in adolescent onset compared to childhood onset PBD. Conclusions: Youth with BD are at increased risk of AD; nearly one in two has an AD. GAD and SAD are among the most prevalent comorbidities. AD are seen less with episodic and adolescent onset PBD. AD should be carefully investigated alongside the mood symptoms in PBD, as comorbidity may change course, treatment and subtyping of the disorder.