Skip to main content

Table 3 Factors explaining placebo and intervention effects

From: Do randomized clinical trials with inadequate blinding report enhanced placebo effects for intervention groups and nocebo effects for placebo groups?

Univariate No. of RCTs ß (P-value) on placebo effect* ß (P-value) on intervention effect*
Adverse events:
    % Headache in placebo group 48 -0.11 (0.469) 0.02 (0.918)
    % Headache in intervention group 48 -0.28 (0.052) -0.11 (0.453)
    Risk ratio for headache 48 -0.08 (0.569) -0.11 (0.472)
    % Flushing in placebo group 37 0.24 (0.135) 0.02 (0.917)
    % Flushing in intervention group 37 0.10 (0.559) 0.30 (0.071)
    Risk ratio for flushing 37 -0.17 (0.278) 0.21 (0.207)
Study related:
    Many follow-ups (≥4 vs 1 or 2) 21 0.05 (0.823) -0.04 (0.867)
    Sample size 61 -0.012 (0.928) 0.21 (0.098)
    Long duration of double blind (>12 weeks versus ≤4 weeks) 15 0.56 (0.005) 0.52 (0.025)
    % Randomized not analyzed 43 0.43 (0.002) 0.25 (0.102)
    ITT analysis (yes vs no) 51 -0.07 (0.638) -0.23 (0.094)
    Publication year 61 0.02 (0.881) -0.42 (<0.001)
    Parallel study design (yes vs no) 61 -0.10 (0.451) -0.07 (0.589)
    Study run-in placebo 2 NA NA
    Commercial funding (yes vs no) 36 -0.09 (0.594) -0.04 (834)
    Continent (North America = ref)** 47   
    Across continents 19 0.36 (0.038) 0.16 (0.392)
    Asia 6 0.42 (0.005) -0.07 (0.667)
    Europe 9 0.25 (0.137) 0.21 (0.249)
    Single center study (yes vs no) 56 0.14 (0.308) -0.07 (0.623)
    Type of PDE-5 inhibitor (tadalafil = ref) 59   
    Sildenafil 30 0.24 (0.105) 0.11 (0.474)
    Vardenafil 10 0.17 (0.259) 0.19 (0.199)
    % Prior experience with intervention 14 -0.55 (0.013) -0.05 (0.874)
Patient related:    
    Prior experience with intervention (yes vs no) 9 -0.86 (<0.001) -0.10 (0.772)
    % Psychogenic etiology 40 0.11 (0.485) -0.23 (0.150)
    Prostate cancer or spinal cord injury (yes vs no) 19 0.04 (0.854) -0.59 (0.003)
    Baseline disease severity*** 58 -0.31 (<0.001) -0.44 (<0.001)
    Disease duration 29 0.06 (0.761) -0.07 (0.714)
Multivariate No. of RCTs ß ( P -value)  
Model on placebo effect (76% explained) 10****   
Naïve to intervention (yes vs no)   0.52 (0.012)  
% Randomized not analyzed in placebo groups   0.44 (0.033)  
Model on intervention effect (40% explained) 107   
Placebo effect   0.32 (<0.001)  
Baseline disease severity***   -0.33 (<0.001)  
Year of publication   -0.26 (0.002)  
Sample size   0.23 (0.003)  
  1. *Negative values represent lowering effects. **Analyses shown only for continents with more than two studies. ***Higher values on the International Index of Erectile Functioning-Erectile Functioning domain score questionnaire represent less severe cases of erectile dysfunction. ****Ten studies reported investigation of participants with prior experience or naïve to intervention. ß, standardized regression coëfficient; ITT, intention-to-treat; NA, not assessable or no data; PDE-5, phosphodiesterase-5; RCT, randomized controlled trial.