Optimizing exercise prescriptions for older adults with mild cognitive impairment: a dose-response network meta-analysis of modality-dose combinations on executive function.
Mingliang Song, Qiang Xiong, Xiaojin Qin
✦ AI-curated · Sources linked
In 30 seconds
This Bayesian dose-response network meta-analysis synthesized randomized controlled trials of repeated exercise programmes in adults aged 60 years or older with mild cognitive impairment. It examined working memory, cognitive flexibility, and inhibitory control across exercise doses and modalities. Thirty-eight studies were included. Across all three executive-function domains, the dose-response pattern was non-linear: approximately 300-1,000 METs⋅min/week was associated with the most reliable gains, while higher doses showed plateaus or modest declines. The most consistent modality findings differed by domain, with exercise joint cognitive training for working memory and comprehensive physical training or walking aerobic exercise for the other two domains.
Key findings
- Thirty-eight randomized controlled trials met the eligibility criteria.
- Across all three executive-function domains, approximately 300-1,000 METs⋅min/week was associated with the most reliable gains; higher doses showed plateaus or modest declines.
- Exercise joint cognitive training showed the most consistent working-memory improvements; comprehensive physical training and walking aerobic exercise had more stable effects on flexibility and inhibitory control.
- Evidence certainty ranged from low to moderate, and publication bias was minimal.
Why it matters
The findings suggest that exercise dose and modality may both matter when targeting executive function in older adults with mild cognitive impairment. The non-linear pattern cautions against assuming that more exercise necessarily produces greater cognitive benefit. Domain-specific dose windows could inform intervention design, although the evidence certainty limits confidence in precise prescriptions.
What to keep in mind
Evidence certainty ranged from low to moderate. The authors identify a need for future trials using multi-dose gradient designs to validate optimal dose ranges.