Patient Expectancy as a Mediator of Placebo Effects in Antidepressant Clinical Trials

• Published in: The American journal of psychiatry Vol. 174; no. 2; pp. 135 - 142
• Main Authors: Rutherford, Bret R, Wall, Melanie M, Brown, Patrick J, Choo, Tse-Hwei, Wager, Tor D, Peterson, Bradley S, Chung, Sarah, Kirsch, Irving, Roose, Steven P
• Format: Journal Article
• Language: English
• Published: United States American Psychiatric Association 01.02.2017
• Subjects: Adult; Antidepressants; Antidepressive Agents - therapeutic use; Citalopram - therapeutic use; Clinical outcomes; Clinical trials; Depressive Disorder, Major - diagnosis; Depressive Disorder, Major - drug therapy; Depressive Disorder, Major - psychology; Expectations; Humans; Mental depression; Middle Aged; Placebo Effect; Prospective Studies; Psychiatric Status Rating Scales - statistics & numerical data; Psychometrics; Set (Psychology); Treatment Outcome; Mood Disorders-Unipolar; Antidepressants
• ISSN: 0002-953X; 1535-7228
• DOI: 10.1176/appi.ajp.2016.16020225

Objective:Causes of placebo effects in antidepressant trials have been inferred from observational studies and meta-analyses, but their mechanisms have not been directly established. The goal of this study was to examine in a prospective, randomized controlled trial whether patient expectancy mediates placebo effects in antidepressant studies.Method:Adult outpatients with major depressive disorder were randomly assigned to open or placebo-controlled citalopram treatment. Following measurement of pre- and postrandomization expectancy, participants were treated with citalopram or placebo for 8 weeks. Independent samples t tests determined whether patient expectancy differed between the open and placebo-controlled groups, and mixed-effects models assessed group effects on Hamilton Depression Rating Scale (HAM-D) scores over time while controlling for treatment assignment. Finally, mediation analyses tested whether between-group differences in patient expectancy mediated the group effect on HAM-D scores.Results:Postrandomization expectancy scores were significantly higher in the open group (mean=12.1 [SD=2.1]) compared with the placebo-controlled group (mean=11.0 [SD=2.0]). Mixed-effects modeling revealed a significant week-by-group interaction, indicating that HAM-D scores for citalopram-treated participants declined at a faster rate in the open group compared with the placebo-controlled group. Patient expectations postrandomization partially mediated group effects on week 8 HAM-D.Conclusions:Patient expectancy is a significant mediator of placebo effects in antidepressant trials. Expectancy-related interventions should be investigated as a means of controlling placebo responses in antidepressant clinical trials and improving patient outcome in clinical treatment.