Abstract Background The global prevalence of type 2 diabetes mellitus (T2DM) is rising, significantly increasing the risk of cognitive impairment and dementia. Although exercise improves cognitive function in T2DM, few studies have compared different exercise modalities. This network meta-analysis assessed their effects on global cognitive function in patients with T2DM. Methods This study systematically searched PubMed, Embase, Web of Science, Cochrane Library, China National Knowledge Infrastructure (CNKI), and Wanfang databases from their inception to October 10, 2025, and included randomized controlled trials (RCTs) evaluating the effects of exercise interventions on global cognitive function in patients with T2DM. Standardized mean differences (SMDs) and 95% confidence intervals (CIs) were pooled using a random-effects model, and treatment rankings were estimated using surface under the cumulative ranking curve (SUCRA) values. Subgroup analyses were performed according to cognitive assessment tools, intervention duration, and training frequency. Results Twenty-one RCTs involving 2,118 participants were included. Compared with usual care, multimodal exercise (ME, 7 trials, n = 423, SMD = 1.04, 95% CI: 0.49–1.59), aerobic exercise (AE, 9 trials, n = 399, SMD = 0.85, 95% CI༚0.38–1.33), and mind–body exercise (MBE, 4 trials, n = 202, SMD = 0.93, 95% CI༚0.25–1.62) significantly improved global cognitive function, while resistance exercise (RE, 2 trials, n = 77, SMD = 0.46, 95% CI༚–0.57–1.48) showed no significant effect. SUCRA rankings indicated the highest efficacy for ME (78.1%), followed by MBE (69.0%), AE (61.5%), and RE (36.6%). Subgroup analyses showed that ME was most effective when cognition was assessed with MoCA and in long-term interventions (> 3 months), whereas MBE and AE were more effective with MMSE and in short-term interventions (≤ 3 months). At exercise frequencies ≤ 3 sessions per week, ME, AE, and MBE were effective, while at higher frequencies only ME remained effective. Conclusion This study indicates that ME is the most effective intervention for improving global cognitive function in patients with T2DM, while MBE and AE also provide benefits. In clinical practice, exercise interventions should be tailored to individual patient characteristics to optimize cognitive outcomes.
Gao et al. (2026) studied this question.