# The effects of traditional mind-body exercises on cognitive function in neurodegenerative diseases or prodromal cognitive decline: a meta-analysis

## TL;DR

This study finds that traditional mind-body exercises like tai chi and yoga can improve cognitive function in people with early signs of neurodegenerative diseases.

## Contribution

This is the first meta-analysis to systematically evaluate the cognitive benefits of traditional mind-body exercises in neurodegenerative and prodromal cognitive decline populations.

## Key findings

- TMBE significantly improved overall cognitive functioning as measured by MMSE and MoCA.
- Improvements were observed in executive function, verbal fluency, and memory, particularly in mild dementia subgroups.
- Attention improved, though results varied by assessment tool, with moderate to high heterogeneity in some outcomes.

## Abstract

The meta-analysis aimed to systematically assess the influence of traditional mind–body exercise (TMBE), including taichi, baduanjin, wuqinxi, and yoga, on cognitive functioning and related cognitive decline for patients with neurodegenerative diseases (NDDs) or Prodromal Cognitive Decline.

Randomized controlled trials (RCTs) were published until October 19, 2025, as determined by searching PubMed, Cochrane Library, Web of Science, and Embase. The study population included adults with AD and PD, including those at prodromal stages, such as MCI and SCD. The interventions were a TMBE group and a control group. The primary outcome was overall cognitive functioning score, which was measured using the MMSE or MoCA, and secondary outcomes included executive function, memory, attention and language. Data were analyzed using random effects models and quality assessed with the Cochrane Risk of Bias Tool.

Twenty-one RCTs were included, with the total sample size varying according to outcome metrics. TMBE showed significant improvement in overall cognitive functioning (MMSE: MD = 0.65, 95% CI: 0.20 to 1.09, p = 0.004; MoCA: MD = 0.87, 95% CI: 0.46 to 1.29, p = 0.001). Significant differences were seen in executive function (e.g., digit reversal: MD = 0.24, 95% CI: 0.05 to 0.44, p = 0.013; TMT-B: MD = −1.18, 95% CI: −1.70 to −0.67, p < 0.001), verbal fluency (MD = 0.36, 95% CI: 0.14–0.57, p = 0.001), and significant benefits were also observed in specific subgroups with respect to long-term delayed recall (e.g., mild dementia: MD = 1.35, 95% CI: 0.81–1.88, p < 0.001). Attention improvement effects were generally positive but varied by assessment tool. Moderate to high heterogeneity existed for some of the outcome indicators, but this tended to be resolved after sensitivity analyses. The degree of publication bias was low.

TMBE has demonstrated the ability to contribute to improvement of overall and specific cognitive functions in individuals diagnosed with NDDs or Prodromal Cognitive Decline. These trainings offer a promising non-pharmacological intervention strategy that is safe, reliable, and multi-targeted to improve cognitive impairment in this population.

PROSPERO – International prospective register of systematic reviews Unique identifier: CRD420251106629, https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD420251106629.