Extent, time course, and moderators of antipsychotic treatment in youth with mood disorders: Results of a meta-analysis and meta-regression analyses

Objective: To meta-analytically examine the trends and correlates of antipsychotic use in youth with mood disorders. Methods: Systematic literature search without language restriction in PubMed/ MEDLINE/PsycINFO from database inception through March 2015 using the following search terms: (antipsychotic∗OR neuroleptic∗OR "dopamine blocker∗" OR antidopaminergic) AND (child∗OR adolescen∗OR pediatric OR youth) AND (prescription∗OR prescrib∗OR use OR utilization OR database OR pharmacoepidemiolog∗OR frequency OR rate OR rates). Random effects metaanalysis and meta-regression analyses were conducted. Study Selection: Included were studies reporting on the frequency of (1) mood disorders in antipsychotic-treated youth (≤ 19 years) and (2) antipsychotic use in youth with mood disorders. Data Abstraction: Two independent investigators abstracted data on study, patient, and treatment characteristics. Results: Forty-one studies were meta-analyzed (N = 518,919, mean ± SD age = 12.8 ± 1.8 years, males = 65.7%). Altogether, 24.2% of antipsychotic-treated youth had a mood disorder diagnosis (studies = 34, depression spectrum disorder = 10.9%, bipolar spectrum disorder = 13.6%). In longitudinal studies, the overall proportion increased significantly from 17.3% in 2000 (range, 1996-2009) to 24.5% in 2006 (range, 2004-2011) (odds ratio [OR] = 1.50; 95% confidence interval [CI], 1.26-1.79; P < .0001). This increase was driven entirely by bipolar spectrum diagnoses (2001 = 11.1%, 2006 = 16.3%, P < .0001), rather than depression spectrum diagnoses (2001 = 9.1%, 2007 = 9.2%, P = .77). Among youth with mood disorders (8 studies), 24.0% received antipsychotics (depression spectrum disorder = 4.6%; bipolar spectrum disorder = 44.0%). Conclusions: The proportion of youth with mood disorder diagnoses increased significantly among antipsychotic-treated youth, driven entirely by an increase in youth with bipolar spectrum disorders. Progress in understanding the reasons for these trends and for an evaluation of the appropriateness of the observed antipsychotic prescribing requires more detailed information than is available in traditional pharmacoepidemiologic databases.