Impact of variability in baseline pain on the placebo response in randomized, placebo-controlled, crossover trials in peripheral neuropathic pain.

Abstract

Large placebo responses often negatively affect randomized controlled trials within the pain area. Understanding different possible factors that influence the placebo response is therefore important. In this retrospective analysis, we hypothesized that a large variability in baseline pain score would predict a greater placebo response and analyzed the impact of the coefficient of variation, SD, and difference between the highest and lowest numeric rating scale (NRS) score at baseline on the placebo response. A total of 160 observations on placebo response from 3 controlled clinical trials with a crossover design were included in this study. In general, the placebo response was low with a mean reduction in pain intensity of 0.5 points (range -5 to 7) measured on a 0 to 10 point NRS, and only 15% were placebo responders as defined by more than 30% reduction in NRS pain score from baseline to the end of the placebo treatment period. We found no significant impact of baseline pain coefficient of variation, SD, or the difference between lowest and highest baseline pain score on the placebo response. Placebo response in one trial did not predict placebo response in another trial. A large placebo response was not associated with a large treatment response. In conclusion, in this retrospective data analysis, there was no impact of baseline pain variability on the placebo response in controlled clinical trials with a crossover design in patients with peripheral neuropathic pain.