Placebo effects in mental health disorders: protocol for an umbrella review.

INTRODUCTION: Given the high prevalence of mental health disorders and their significant socioeconomic burden, there is a need to develop improved treatments, and to evaluate them through placebo-controlled trials. However, the magnitude of the placebo response in randomised controlled trials to test medications may be substantial, affecting their interpretation. Therefore, improved understanding of the patient, trial and mental disorder factors that influence placebo responses would inform clinical trial design to better detect active treatment effects. There is a growing literature exploring the placebo response within specific mental health disorders, but no overarching synthesis of this research has been produced to date. We present a protocol for an umbrella review of systematic reviews and/or meta-analyses in which we aim to understand the effect size and potential predictors of placebo response within, and across, mental health disorders. METHODS AND ANALYSIS: We will systematically search databases (Medline, PsycINFO, EMBASE+EMBASE Classic, Web of Knowledge) for systematic reviews and/or meta-analyses that report placebo effect size in clinical trials in patients with mental health disorders (initial search date 23 October 2022). Screening of abstracts and full texts will be done in pairs. We will extract data to qualitatively examine how placebo effect size varies across mental health disorders. We also plan to qualitatively summarise predictors of increased placebo response identified either quantitatively (eg, through meta-regression) or qualitatively. Risk of bias will be assessed using the AMSTAR-2 tool. We aim to not only summarise the current literature but also to identify gaps in knowledge and generate further hypotheses. ETHICS AND DISSEMINATION: We do not believe there are any specific ethical considerations relevant to this study. We will publish the results in a peer-reviewed journal.

No evidence for an acute placebo effect on emotional processing in healthy volunteers.

Placebo-controlled trials are the gold standard measure of efficacy in the development of new treatments for depression. However, the large placebo effects associated with standard measures of subjective symptoms reduce the sensitivity of such trials to detect antidepressant effects. There is a need to develop novel efficacy markers that are resistant to placebo effects. Measures of emotional processing, known to be sensitive to antidepressant treatment, may be such a marker, although the effect of an acute placebo treatment on these measures remains unclear. We assessed the influence of placebo on a validated battery of emotional processing tasks, the Emotional Test Battery (ETB), in healthy participants. Participants were informed they might receive the antidepressant drug bupropion, placebo or no treatment, with placebo effect being estimated as the difference between the placebo and no treatment groups. We found no significant difference between these groups on measures of emotional processing. There was also no effect of subjective treatment expectancy on performance in the tasks. This suggests that the ETB might be a useful tool for Phase I trials assessing novel antidepressant agents against placebo.