Table 2 Significant correlates of placebo effect size or response rate identified either through meta-regression or correlation analyses.

From: Placebo effects in randomized trials of pharmacological and neurostimulation interventions for mental disorders: An umbrella review

 

Patient Population

     

Reference

Mental Disorder

Sub-population

k

Correlate

Relationship Measure

Statistic

95% CI

Furukawa et al., 2016 [51]

Major Depressive Disorder

Adults

112

Multicenter vs single center

RR

1·41

1·22, 1·62

Flexible vs Fixed dose

RR

1·17

1·04, 1·31

Trial duration (every 1 week increase)

RR

1·04

1·01, 1·06

Razza et al., 2018 [53]

Major Depressive Disorder

Neurostimulation trials

61

Active effect size

Beta

0·28

0·12, 0·43

Meister et al., 2020 [54]

Major Depressive Disorder

Children & adolescents

24

Number of sites

OR

1·01

1·00, 1·02

Ahmadzad-Asl et al., 2022 [56]

Anxiety Disorders

Panic Disorder

NR

Publication year

Beta

0·02

0·01, 0·04

Number of participants

Beta

0·001

0·000, 0·002

Bandelow et al., 2015 [57]

Anxiety Disorders

GAD, SAD, Panic Disorder

111

Publication year

r

0·41

NR

Agid et al., 2013 [38]

Schizophrenia Spectrum

Adults

61

Publication year

Regression coefficient

0·04

NR

Age

Regression coefficient

−0·05

NR

Illness duration

Regression coefficient

−0·05

NR

Baseline CGI severity

Regression coefficient

0·42

NR

Trial duration

Regression coefficient

−0·08

NR

Leucht et al., 2018 [59]

Schizophrenia Spectrum

Adults

95

Number of participants (10 participants more)

Regression coefficient

0·15

0·07, 0·24

Average age

Regression coefficient

−6·23

−10·46, −2·00

Number of sites

Regression coefficient

1·38

0·71, 2·04

Czobor et al., 2022 [60]

Schizophrenia Spectrum

Negative symptoms

25

Increased number of active treatment arms

Regression coefficient

0·25

NR

Trial duration

Regression coefficient

−0·37

−0·65, −0·08

Faraone et al., 2022 [62]

ADHD

Mixed ages

NR

Active medication effect size (clinician-rated outcomes)

r

0·69

NR

Active medication effect size (parent-rated outcomes)

r

0·44

NR

Active medication effect size (teacher-rated outcomes)

r

0·85

NR

Del Re et al., 2013 [61]

Alcohol Use Disorder

Adults

47

Publication year

Regression coefficient

0·04

0·00, 0·08

36

Number of administrations

Regression coefficient

0·30

0·00, 0·61

47

Baseline severity

Regression coefficient

0·22

0·01, 0·43

Siafis et al., 2020 [63]

Autism Spectrum

Predominantly children & adolescents

51

Active medication effect size on overall core symptoms

Spearman’s ρ

0·54

NR

Active medication effect size on social-communication difficulties

Spearman’s ρ

0·53

NR

Number of sites

Regression coefficient

0·03

0·01, 0·05

Low risk of bias in allocation concealment

Regression coefficient

0·25

0·02, 0·49

Mohamadi et al., 2022 [66]

OCD

Mixed ages

NR

Publication year

Beta

0·01

NR

Follow up duration (weeks)

Beta

−0·01

NR

Mean age (years)

Beta

−0·01

NR

Number of participants

Beta

0·003

NR

Iovieno et al., 2016 [64]

Bipolar Depression

Adults

15

Active medication response rate

Regression coefficient

0·892

NR

Relative risk of responding to active medication vs placebo

Regression coefficient

−1·11

NR

  1. Correlates are included here only if they were reported as significant. Results from meta-regressions were preferred, and results from multivariate versus univariate regressions were preferred where these were available. Positive statistics signify correlates of larger placebo response, while negative statistics signify correlates of smaller placebo response.
  2. k number of component studies, GAD generalised anxiety disorder, SAD social anxiety disorder, ADHD attention-deficit hyperactivity disorder, OCD obsessive-compulsive disorder, RR relative risk, OR odds ratio, r Pearson’s correlation coefficient, NR not reported.
Back to article page