Respondent and item level patterns of response of aripiprazole in the acute treatment of pediatric bipolar I disorder

Abstract Background Few studies have evaluated the value of a parent- and subject-rated scale in detecting symptom change in response to pharmacologic treatment. Methods This was a post-hoc analysis of data from a 4-week, randomized, double-blind, placebo-controlled study to evaluate which informants detect response to treatment with aripiprazole in pediatric subjects experiencing a mixed or manic episode associated with bipolar I disorder. Efficacy assessments included clinician-rated scales and the parent- and subject-rated 10-item General Behavior Inventory Mania (GBI-M10) and Depression (GBI-D10) scales. Cohen's d quantified effect sizes for total scale scores and individual line items. Results Parent-GBI-M10 total, clinician-rated Young Mania Rating Scale (YMRS) total, and Clinical Global Impression–Bipolar Disorder (CGI-BP) Mania scores produced similar effect sizes, suggesting that the parent-GBI-M10 is sensitive to treatment-related improvements in manic symptoms. Aripiprazole improved a broad spectrum of parent-rated mania symptoms; six parent-GBI-M10 line item effect sizes were moderate (>0.5) in at least one of the two aripiprazole treatment arms (10 or 30 mg/day). Subject-completed GBI-M10 line item effect sizes were consistently smaller, indicating that the subjects' experience of treatment effects were less pronounced. Limitations Study inclusion/exclusion criteria may limit generalizability of these findings. Conclusions Parent ratings of mania severity were in agreement with clinician ratings, indicating that parent-rated assessments can be valuable in detecting symptom change over the course of treatment. These data support the use of the parent-GBI-M10 as an outcome measure in research and clinical settings.