Consistency of atypical antipsychotic superiority to placebo in recent clinical trials

Background: The use of control placebos in clinical trials of new antipsychotic medications is increasingly under examination. The active controlled equivalence study could offer a potential alternative design. First, however, it must be clear that any proposed standard control agent has been consistently superior to placebo in previous studies. Methods: Through a Freedom of Information Act request, we identified nine placebo-controlled trials of risperidone, olanzapine, or quetiapine. Results: Meta-analysis indicated that the pooled estimate of the true population effect size ± SE was 0.46 ± 0.06 for categorical response rates and >0.53 ± 0.07 for the continuous Brief Psychiatric Rating Scale change score outcome measure. If the desired detectable effect size is set very conservatively at a 95% confidence lower bound for the estimate of true effect size, statistical power for random samples of 80 per group drawn from a population of subjects similar to that of the nine meta-analyzed studies is .67 for categorical response rates and >.82 for the continuous measure, based on one-sided α = .05. Conclusions: These data suggest substantial confidence that a therapeutic dose of an atypical antipsychotic will be statistically superior to placebo in an adequately sized randomized trial, when reporting a continuous measure as the principal outcome.