Objective: In a previous review, we found that response to placebo in studies of major depression was increasing over time. The purpose of this study was to examine placebo response rates in trials of acute bipolar mania.
Data sources: We searched MEDLINE for placebo-controlled trials in which patients with bipolar mania were randomly assigned to receive medication or placebo. Searches included combinations of the terms acute bipolar mania, placebo-controlled, double-blind, randomized, and common medication names (e.g., lithium, risperidone). In addition, reference lists from identified articles and any reviews of bipolar mania were examined. The search was limited to literature in English, published between January 1980 and November 2005.
Study selection: The review identified 21 studies published between 1991 and 2005. Twenty studies used a response criterion of a 50% or greater decrease on the Young Mania Rating Scale (YMRS) or Mania Rating Scale (MRS), or a designation of much or very much improved on the Clinical Global Impressions-Improvement scale (CGI, score of 1 or 2).
Data extraction: Data were extracted from the articles by one of the authors (R.S.) and all of the data used in the analyses were verified by the other author (B.T.W.).
Data synthesis: Pearson correlation coefficients (2-tailed) and linear regression were used to examine the strength of the relationship between continuous variables. There was a significant association between the year of publication and placebo response rate in studies using the YMRS (N = 14) (r = 0.545, p = .04); however, when data from studies using the CGI and MRS were added, the association was no longer significant (r = 0.374, N = 20, p = .10).
Conclusions: The response rate to placebo in studies of bipolar mania (31.2%) was similar to the rate observed in major depression (29.7%). Over a limited number of years, there was some indication of a change in placebo response on the YMRS in studies of bipolar mania; however, the small number of studies available for analysis limits our ability to draw definitive conclusions.