Background - Non-randomised studies assessing intrarectal botulinum toxin type A (BoNTA) injections for faecal incontinence are promising. We aimed to evaluate the efficacy of BoNTA for the treatment of faecal incontinence in a randomised study. Methods - In this randomised, double-blind, placebo-controlled study, we included adult patients who had at least one urgency or faecal incontinence episode per week for at least 3 months and who had experienced a failure of conservative or surgical treatment from eight French specialist hospital units with the skills to manage patients with faecal incontinence. Patients were randomly assigned (1:1) by a central web form to receive intrarectal submucosal injections of either 200 units of BoNTA (Botox; Allergan, Irvine, CA, USA; BoNTA group) or an equivalent volume of saline (placebo group), stratified by Cleveland Clinic Severity scores (CCS score; ≥12 or <12). Patients, investigators, study site staff, and sponsor personnel were masked to treatment allocation up to the 6-month visit. The primary endpoint was the number of episodes of faecal incontinence and urgency per day assessed using 21-day patient bowel diaries 3 months after the treatment. The primary analysis was performed using a modified intention-to-treat (mITT) approach (ie, in all the randomised patients who had received a treatment) with adjustment for baseline faecal incontinence and urgency episodes. After the final data collection at 6 months after injections, patients were unmasked and offered the BoNTA treatment if they were in the placebo group (rescue therapy) without masking, with an additional 6 months of safety follow-up. This trial is registered with ClinicalTrials.gov, number NCT02414425. Findings - Between Nov 25, 2015, and Nov 25, 2020, we randomly assigned 200 patients to receive either BoNTA (n=100) or placebo (n=100) injections. Due to withdrawals before the injections, 96 patients were included in the BoNTA group and 95 patients were included in the placebo group (mITT analysis). The mean number of faecal incontinence and urgency episodes per day in the BoNTA group decreased from 1·9 (SD 2·2) at baseline to 0·8 (1·8) at 3 months after the injections, and from 1·4 (1·1) to 1·0 (1·0) in the placebo group, with a baseline-adjusted mean group difference at 3 months estimated at -0·51 (95% CI -0·80 to -0·21, p=0·0008). No serious treatment-related adverse events were reported in the trial. The most frequently reported non-serious adverse event (treatment related or not) following the BoNTA or placebo injections was constipation (reported in 68 [40%] of 169 patients who received the BoNTA injections and 38 [40%] of 95 patients who received placebo injections). Interpretation - BoNTA injections are an efficacious treatment for urge faecal incontinence. Further research will define the optimum selection criteria, dose, site of injection, re-injection frequency, and long-term results. Funding - General Direction of Healthcare (French Ministry of Health).