BACKGROUND: Current standard management of severe acute malnutrition uses ready-to-use therapeutic food (RUTF) at a single weight-based calculation resulting in an increasing amount of RUTF provided to the family as the child's weight increases during recovery. Using RUTF at a gradually reduced dosage as the child recovers could reduce costs while achieving similar growth response. METHODS: We conducted an open-label, non-inferiority, randomised controlled trial in the Democratic Republic of the Congo. Children aged 6-59 months with a mid-upper-arm circumference (MUAC) of less than 115 mm or a weight-for-height z-score (WHZ) of less than -3 or bipedal oedema and without medical complication were randomly assigned (1:1 ratio) using a specially developed software and random blocks (size was kept confidential), to either the current standard treatment (increasing the RUTF amount with increasing weight) or the OptiMA strategy (decreasing the RUTF dose with increasing weight and MUAC). The main endpoint was proportion of children who achieved recovery over the 6 months follow up period, as defined as meeting the following criteria for two consecutive weeks after a minimum of 4 weeks' treatment: axillary temperature less than 37.5 °C, no bipedal oedema, and anthropometric improvement (either MUAC 125 mm or greater or WHZ -1.5 or higher). We performed analyses on the intention-to-treat (ITT) (all children) and per-protocol populations (participants who had a minimum prescription of 4 weeks' RUTF, received at least 90% of the total amount of RUTF they were supposed to receive as per the protocol, and had a maximum interval of 6 weeks between any two visits in the 6-month follow-up). The non-inferiority margin was 10%. This trial is registered at ClinicalTrials.gov, and is now closed NCT03751475. FINDINGS: Between July 22, 2019, and January 20, 2020, 491 children were randomly assigned, of whom 482 were analysed (240 in the standard group and 242 in the OptiMA group). In the ITT analysis, 234 (98%) children in the standard group and 231 (96%) children in OptiMA recovered (difference 2.0%, 95% CI -2.0% to 6.4%). In the PP analysis, 234 (98%) children in the standard group and 228 (97%) in OptiMA recovered (difference 1.3%, 95% CI -2.3% to 5.1%). Sensitivity analyses applying the same anthropometric recovery criteria to each group also showed non-inferiority of the OptiMA strategy in ITT and PP analysis. INTERPRETATION: This non-inferiority trial treating uncomplicated children with MUAC of less than 115 mm or a WHZ of less than -3 or bipedal oedema with decreasing RUTF dose as MUAC and weight increase demonstrated non-inferiority compared to the standard protocol in a highly food-insecure context in the Democratic Republic of the Congo. These findings add evidence on the safety of RUTF dose reduction with significant RUTF cost savings. FUNDING: Innocent Foundation and European Civil Protection and Humanitarian Aid Operations. TRANSLATION: For the French translation of the abstract see Supplementary Materials section.Read less <