Effects of mind-body exercise on cognitive function in patients with mild cognitive impairment: A systematic review and network meta-analysis

Objective This investigation carried out a systematic review and network meta-analysis to assess how different forms of mind–body exercise influence overall cognitive performance in patients with MCI. Methods A systematic search was conducted across eight Chinese and English databases (up to April 2026). A frequency-based network meta-analysis was performed to estimate the standardized mean differences and 95% confidence intervals, with interventions ranked by SUCRA values. Results A total of 44 RCTs (3,189 patients with MCI) were included. The network meta-analysis showed that, for the MoCA outcome, Baduanjin, Tai Chi, Qigong, and dance were all significantly superior to the control group ( p < 0.05). For the MMSE outcome, all three exercise types except Qigong were significantly superior to the control group. SUCRA rankings indicated that Baduanjin ranked first for both the MoCA (91.6%) and MMSE (88.3%). Subgroup analyses suggested that Baduanjin was most effective for short-term (≤ 12 weeks), high-frequency (> 3 times/week), or short-duration (< 60 min) interventions, while Tai Chi was more effective for long-term (> 12 weeks), low-frequency (≤ 3 times/week), or long-duration (≥ 60 min) interventions. The sensitivity analysis results were robust, and the risk of publication bias was low; however, the certainty of the evidence was generally low. Conclusion Baduanjin, Tai Chi, dance, and Qigong may be beneficial for global cognition in individuals with MCI. Baduanjin ranked highest in the SUCRA analysis. Tai Chi showed a potential advantage in longer-duration interventions. Given the low certainty of evidence and reliance on indirect comparisons, these findings should be interpreted cautiously.