Abstract
PURPOSE/BACKGROUND: For a drug to acquire Food and Drug Administration approval, it must significantly outperform placebo treatment. In recent years, the placebo effect seems to be increasing in neuropsychiatric conditions. Here, we examine placebo effects across self-reported, clinically rated, and performance-based data from a trial using a corticotropin-releasing hormone receptor type 1 (CRHR1) antagonist for treatment of post-traumatic stress disorder (PTSD). METHODS/PROCEDURES: Women with chronic PTSD were randomized to treatment with either GSK561679, a CRHR1 antagonist, or placebo. Before randomization, participants completed self-report scales, clinician-rated measures of PTSD and depression symptoms, and objective tests of cognition and functioning. Differences in change scores on measures were compared between GSK561679 and placebo-treated participants. FINDINGS/RESULTS: GSK561679 failed to produce any significant improvement in the participants. A substantial placebo effect was observed in both self-report and clinical rating scales, with effect sizes up to 1.5 SD. No single variable predicted placebo-related changes. Notably, there was an improvement on objective performance measures of cognition that exceeded previous standards for practice effects. IMPLICATIONS/CONCLUSIONS: Participants in this trial manifested retest effects on performance-based measures of cognition. Notably, they had minimal prior experience with performance-based assessments. Experiencing the structure and support of a clinical trial may have contributed to significant reductions in subject-reported and clinician-rated PTSD symptom levels. The improvement seen across all assessment domains was consistent with that seen in previous studies where the active treatments separated from placebo. Investigators conducting clinical trials treating PTSD patients should expect placebo effects and design studies accordingly.