Evidence-based ranking of diabetes prevention in prediabetes: A comprehensive network meta-analysis of 378 randomised controlled trials

Why clinicians should know about it

• Picked for Epidemiology (paper of the day, 10 September 2026): Network meta‑analysis of diabetes‑prevention interventions in prediabetes

Abstract

BACKGROUND: Ten percent of the global population presents impaired glucose tolerance. Although numerous interventions have been investigated to prevent type 2 diabetes (T2D), their comparative effectiveness remains uncertain. In this study, we assessed and ranked current pharmacological and non-pharmacological interventions for preventing diabetes in individuals with prediabetes. METHODS: The study protocol was registered on PROSPERO (CRD42023473875). Randomised controlled trials (RCTs) reporting on adult patients with prediabetes undergoing any intervention addressing dysglycaemia were eligible. Our systematic search was conducted from database inception to 19 December 2025 in three databases (MEDLINE, Embase and Cochrane Library [CENTRAL]). Network meta-analysis was performed using the random-effect frequentist method. Our primary outcomes were the incidence of T2D, haemoglobin A1c (HbA1c), fasting plasma glucose (FPG) and 2-h glucose after the glucose oral tolerance test (2hPG) expressed as relative risk (RR) and mean difference (MD), respectively, with the corresponding 95% confidence intervals (CIs). RESULTS: We included 378 RCTs comprising 113 122 patients. The highest risk reduction for T2D when compared with control was obtained with dual incretin-based therapy (glucagon-like peptide 1 [GLP-1] and gastric inhibitory polypeptide (GIP) receptor agonists [GLP-1 + GIP RAs]) (RR = .1, CI [.05;  .21]), with GLP-1 receptor agonists (GLP-1 RAs) ranking second (RR = .26, CI [.19;  .4]) followed closely by metformin plus intensive lifestyle intervention (ILI) (RR = .27, CI [.12;  .64]). Only curcumin (MD = -.49, CI [-.70; -.28]), dual GLP-1 + GIP RA (MD = -.46, CI [-.63; -.29]), single GLP-1 RA (MD = -.41, CI [-.50; -.31]) and physical activity (MD = -.35, CI [-.48; -.23]) achieved >.3% HbA1c reductions after 1 year. Sulphonylureas and physical activity were most effective for the 1-year FPG decrease, while the greatest 2hPG reduction was obtained with zinc supplementation and sulphonylureas. CONCLUSION: Incretin-based therapies, particularly dual GLP-1 + GIP RA and single GLP-1 RA, followed by metformin together with ILI, are the most effective interventions to prevent T2D in patients with prediabetes.

Abstract as published, via PubMed.