Placebo and nocebo reactions in randomized trials of pharmacological treatments for persistent depressive disorder. A meta-regression analysis

TY - JOUR

T1 - Placebo and nocebo reactions in randomized trials of pharmacological treatments for persistent depressive disorder. A meta-regression analysis

AU - Meister, Ramona

AU - Jansen, Alessa

AU - Härter, Martin

AU - Nestoriuc, Yvonne

AU - Kriston, Levente

N1 - Copyright © 2017 Elsevier B.V. All rights reserved.

PY - 2017/6

Y1 - 2017/6

N2 - BACKGROUND: We aimed to investigate placebo and nocebo reactions in randomized controlled trials (RCT) of pharmacological treatments for persistent depressive disorder (PDD).METHODS: We conducted a systematic electronic search and included RCTs investigating antidepressants for the treatment of PDD. Outcomes were the number of patients experiencing response and remission in placebo arms (=placebo reaction). Additional outcomes were the incidence of patients experiencing adverse events and related discontinuations in placebo arms (=nocebo reaction). A priori defined effect modifiers were analyzed using a series of meta-regression analyses.RESULTS: Twenty-three trials were included in the analyses. We found a pooled placebo response rate of 31% and a placebo remission rate of 22%. The pooled adverse event rate and related discontinuations were 57% and 4%, respectively. All placebo arm outcomes were positively associated with the corresponding medication arm outcomes. Placebo response rate was associated with a greater proportion of patients with early onset depression, a smaller chance to receive placebo and a larger sample size. The adverse event rate in placebo arms was associated with a greater proportion of patients with early onset depression, a smaller proportion of females and a more recent publication.CONCLUSIONS: Pooled placebo and nocebo reaction rates in PDD were comparable to those in episodic depression. The identified effect modifiers should be considered to assess unbiased effects in RCTs, to influence placebo and nocebo reactions in practice.LIMITATIONS: Limitations result from the methodology applied, the fact that we conducted only univariate analyses, and the number and quality of included trials.

AB - BACKGROUND: We aimed to investigate placebo and nocebo reactions in randomized controlled trials (RCT) of pharmacological treatments for persistent depressive disorder (PDD).METHODS: We conducted a systematic electronic search and included RCTs investigating antidepressants for the treatment of PDD. Outcomes were the number of patients experiencing response and remission in placebo arms (=placebo reaction). Additional outcomes were the incidence of patients experiencing adverse events and related discontinuations in placebo arms (=nocebo reaction). A priori defined effect modifiers were analyzed using a series of meta-regression analyses.RESULTS: Twenty-three trials were included in the analyses. We found a pooled placebo response rate of 31% and a placebo remission rate of 22%. The pooled adverse event rate and related discontinuations were 57% and 4%, respectively. All placebo arm outcomes were positively associated with the corresponding medication arm outcomes. Placebo response rate was associated with a greater proportion of patients with early onset depression, a smaller chance to receive placebo and a larger sample size. The adverse event rate in placebo arms was associated with a greater proportion of patients with early onset depression, a smaller proportion of females and a more recent publication.CONCLUSIONS: Pooled placebo and nocebo reaction rates in PDD were comparable to those in episodic depression. The identified effect modifiers should be considered to assess unbiased effects in RCTs, to influence placebo and nocebo reactions in practice.LIMITATIONS: Limitations result from the methodology applied, the fact that we conducted only univariate analyses, and the number and quality of included trials.

KW - Antidepressive Agents

KW - Depression

KW - Depressive Disorder

KW - Humans

KW - Nocebo Effect

KW - Placebo Effect

KW - Randomized Controlled Trials as Topic

KW - Research Design

KW - Journal Article

KW - Meta-Analysis

KW - Review

U2 - http://doi.org/10.1016/j.jad.2017.03.024

DO - http://doi.org/10.1016/j.jad.2017.03.024

M3 - SCORING: Journal article

C2 - 28363152

VL - 215

SP - 288

EP - 298

JO - J AFFECT DISORDERS

JF - J AFFECT DISORDERS

SN - 0165-0327

ER -