Aims/hypothesisThe aim of this study was to evaluate the impact on metabolic control of the periodic use of a 5-day fasting-mimicking diet (FMD) program as an adjunct to usual care in people with type 2 diabetes under regular primary care surveillance.MethodsIn this randomised, controlled, assessor-blinded trial, people with type 2 diabetes using metformin only and/or diet alone for glycaemic control were randomised to receive 5-day cycles of FMD monthly as adjunct to regular care by their general practitioner or regular care only. Primary outcomes were changes in glucose-lowering medication and HbA1c levels after 12 months. Moreover, changes in use of glucose-lowering medication and/or HbA1c levels in individual participants were combined to yield a clinically relevant primary outcome measure (‘glycaemic management’), categorized as improved, stable or deteriorated after one year of follow-up.Results100 individuals with type 2 diabetes, age 18-75 years, and BMI > 27 kg/m2, were randomised to the FMD (n=51) or control group (n=49). Eight FMD participants and ten controls were lost to follow-up. In complete case intention-to-treat analyses, the mean medication effect score (MES) significantly declined in patients receiving FMD as compared to controls (FMD −0.2 ± 0.3 vs controls +0.2 ± 0.4, p<0.0001) in the face of similar changes of HbA1c adjusted for MES (FMD −0.4 ± 0.8 % vs controls +0.2 ± 0.8 %, p=0.0021). Glycaemic management improved in 53% of participants using FMD vs 8% of controls, remained stable in 23% vs 33%, and deteriorated in 23% vs 59% (p<0.0001).Conclusions/interpretationIntegration of a monthly FMD program in regular primary care for people with type 2 diabetes who use metformin only and/or diet alone for glycaemic control reduces the need for glucose-lowering medication and appears to be safe in routine clinical practice.Trial registrationClinicalTrials.gov:NCT03811587