Effects of different exercise modalities and doses on glycated hemoglobin in patients with type 2 diabetes mellitus and overweight or obesity: a systematic review and Bayesian multilevel network meta-analysis.

This study aimed to compare the effects of different exercise intervention categories and dose characteristics on glycated hemoglobin in adults with type 2 diabetes mellitus and overweight or obesity. Randomized controlled trials were identified from PubMed, Web of Science, Embase, and the Cochrane Library from inception to February 2026. Exercise interventions were classified as high-intensity interval training, combined training, high-intensity aerobic training, moderate-intensity aerobic training, resistance training, and mind-body training. Bayesian network meta-analysis and meta-regression were used to estimate comparative effects and dose-related patterns. Sixty-seven studies involving 5,022 participants were included. Compared with control, high-intensity interval training produced the largest HbA1c reduction (MD = -0.73%, 95% CrI: -0.96 to -0.49), followed by combined training (-0.54%, 95% CrI: -0.70 to -0.38). Other modalities also favored exercise, but with smaller estimated effects. Exploratory study-level meta-regression did not identify a single optimal dose, with greater uncertainty at exposure extremes. Structured exercise interventions were associated with reductions in HbA1c in this population. HIIT and combined training showed the most favorable estimates, but the low to very low certainty of evidence and moderate heterogeneity require cautious interpretation.