Clinical trials in irritable bowel syndrome are associated with high placebo response rates. We aimed to identify the magnitude of the placebo response and the contributing factors to this occurrence.We did a systematic review and meta-analysis with a search of MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials between April 1, 1959, and April 30, 2020. We included all randomised controlled trials that compared an active pharmacotherapeutic agent with placebo and had a dichotomous outcome of response to therapy (in terms of global improvement or improvement in abdominal pain) in adults (aged ≥18 years) with irritable bowel syndrome. Exclusion criteria were trials reporting on treatment satisfaction as a dichotomous outcome of response to therapy or clinician-reported outcomes and a treatment duration of less than 4 weeks. Our main outcome was identification of the magnitude of the pooled placebo response rate for the following endpoints: global improvement, abdominal pain, and US Food and Drug Administration (FDA) endpoints. We extracted information from published reports and pooled proportions through meta-analysis with random effects. The study was registered with PROSPERO, CRD42020170908.Of the 6863 publications identified, 70 articles describing 73 randomised controlled trials were included in our analysis. The pooled placebo response rate was 27·3% (95% CI 24·3-30·9) using the global improvement endpoint, 34·4% (31·2-37·8) using the abdominal pain endpoint, and 17·9% (15·2-21·0) using the composite FDA endpoint responder definition, all with substantial heterogeneity between the trials. Studies published before 2006, and those done in Europe, with a parallel design, a run-in period of 2 weeks or less, a dose schedule of three times a day or more, or a smaller sample size of the control group were significantly associated with an increased pooled placebo response rate.More than a quarter of patients with irritable bowel syndrome had a placebo response in terms of global improvement, with multiple associated moderators. We recommend future trials apply a run-in period of at least 2 weeks and dose once or twice a day to minimise the placebo response rate.None.
Michelle Bosman, Sigrid Elsenbruch, Maura Corsetti, Jan Tack, Magnus Simrén, Bjorn Winkens, Thimo Boumans, Ad Masclee, Daniel Keszthelyi