Placebo Response in Chronic Idiopathic Constipation: A Systematic Review and Meta-Analysis.

Abstract

OBJECTIVES\nChronic idiopathic constipation (CIC), like other functional gastrointestinal disorders, has been associated with a high placebo response rate. However, the placebo response in randomized controlled trials has not been described.\n\n\nMETHODS\nWe conducted a search of the medical literature following the protocol outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis statement using MEDLINE, EMBASE and EMBASE Classic, Web of Science, and the Cochrane Central Register of Controlled Trials for all drugs used for the treatment of CIC. Two independent reviewers performed eligibility assessment and data extraction. The mean response rate was examined for the following 2 responder endpoints: (i) greater than or equal to 3 complete spontaneous bowel movements (CSBMs)/wk (≥3 CSBMs/wk responders) and (ii) mean increase of ≥1 CSBM/wk compared with baseline (increase in ≥1 CSBM/wk responders).\n\n\nRESULTS\nA total of 23 placebo-controlled trials met our inclusion criteria and were included in this meta-analysis. The placebo response in CIC trials ranged from 4% to 44%. The magnitude of the placebo response was 13% (95% confidence interval 11%-16%) with the ≥3 CSBM/wk responder endpoint and 28% (95% confidence interval 21%-30%) with the increase of ≥1 in the CSBM responder endpoint. Higher baseline CSBM, older age, and trials with more male participants were significantly associated with a stronger placebo response for both the ≥3 CSBMs/wk endpoint and increase in the ≥1 CSBM/wk endpoint. Trial characteristics such as location (Europe vs Asia/United States) and laxative class (prokinetic vs secretagogue) revealed key differences in the placebo response for both endpoints. The placebo response was not significantly affected by the number of study visits, study duration, year of publication, number of drop outs, or likelihood of receiving active drug.\n\n\nDISCUSSION\nThe placebo response in CIC trials ranges from 4% to 44% depending on the endpoint. Modifying factors of the placebo response include multiple subject and trial characteristics.