- 2017;20;387-396Randomized Placebo-Controlled Placebo Trial to Determine the Placebo Effect Size
Ludger Gerdesmeyer, MD, Tim Klueter, MD, Volker W. Rahlfs, PhD, CStat, Munjed Al Muderis, MD, Amol Saxena, PDM, Hans Gollwitzer, MD, PhD, Norbert Harrasser, MD, Martin Stukenberg, MD, and Alexander Prehn-Kristensen, PhD.
BACKGROUND: It is the gold standard to use a placebo treatment as the control group in prospective randomized controlled trials (RCTs). Although placebo-controlled trials can reveal an effect of an active treatment, the pure effect of a placebo treatment alone has never been presented or evaluated. No evidence-based, placebo-therapeutic options are currently available, and no placebo-controlled trials have been performed to elucidate the pure placebo effect.
OBJECTIVES: To analyze the pure placebo effect on clinical, chronic pain through a blinded RCT.
STUDY DESIGN: A prospective, randomized, placebo-controlled trial.
SETTING: Medical University centers.
METHODS: One-hundred eighty-two patients suffering from chronic plantar heel pain for over 6 months,who failed to respond to conservative treatments, were screened and 106 of these patients were enrolled into this study. The patients were randomly assigned to receive either a blinded placebo shockwave treatment or an unblinded placebo shockwave treatment. The primary outcome measure was the differences in percentage change of visual analogue scale (VAS) scores 6 weeks after the intervention. The secondary outcome measure was the differences in Roles and Maudsley pain score (RMS) 6 weeks after intervention. As an exploratory outcome, 2-sided group comparisons for baseline characteristics between active treatment and controls were done using the Mann-Whitney-U tests for group comparisons; treatment efficiency was calculated by the effect size coefficient and benchmarks for the Mann-Whitney estimator according to the t-test of 2 independent samples for quantitative data, as well as the Fisher’s exact test for binary data.
RESULTS: Patients from both groups did not differ with respect to heel pain ratings at baseline, for both the VAS (P = .476) and RMS (P = .810) scores. After 6 weeks, patients receiving the blinded placebo treatment reported less heel pain on both scales (VAS: P = .031; RMS: P = .004). Change scores of pain ratings were significantly higher in the blinded placebo group than in the un-blinded placebo group (VAS: P = .002; RMS: P = .002).
LIMITATIONS: As the study represents the first to use an inverse placebo RCT (IPRCT), further conceptual and methodological issues need to be addressed to describe detailed, underlying mechanisms. Specific contextual, intrapersonal, and interpersonal factors modulating the placebo effects should be addressed in future IPRCTs.
CONCLUSIONS: The present study indicated that true placebo effect sizes can be analyzed through a proper IPRCT design. Instead of treating high numbers of patients with placebos in a RCT, which increases the risk for subjects not receiving the active treatment, the IPRCT technique seems to be much more appropriate to analyze the effect sizes of any active treatment, in accordance with the Good Clinical Practice guidelines and Declarations of Helsinki.
KEY WORDS: Pain, randomized controlled trial, RCT, placebo, effect size, inverse placebo, study, pain therapy