Nonlinear dose-response of mind-body exercise in older adults with mild cognitive impairment: A systematic review and meta-analysis.

Abstract

This study aimed to quantify domain-specific dose-response relationships between mind-body exercise and cognitive function in older adults with mild cognitive impairment (MCI), and to derive actionable dosing recommendation. Eight major electronic databases were systematically searched up to September 15, 2025. Randomized controlled trials (RCTs) of mind-body exercise intervention in individuals aged ≥60 years with MCI that reported at least one cognitive domain outcome were included. Hedges'g effect sizes were calculated using random-effects models, and dose-response relationships were examined using restricted cubic spline regression. Twenty-seven RCTs (2034 participants) were included. Moderate-certainty evidence supported improvements in global cognition (Hedges' g: 0.67, 95% CI: [0.49, 0.78]) and processing speed (0.45 [0.3, 0.6]). Global cognition exhibited a U-shaped weekly dose-response, with 240 min/week as the relative nadir that still yielded a significant pooled effect (0.37; [0.19, 0.55]); longer intervention durations significantly amplified this overall response (β = 0.012, p < 0.05), and achieving significant benefit at this dose required ≥11 weeks intervention. Memory followed a similar pattern, but with very low certainty. Processing speed showed a U-shaped cumulative dose-response (nadir: 3400 minutes), and this effect was attenuated by weekly dose (β = -0.0011, p < 0.05), yet caution is warranted because the weekly doses in the included studies were heavily clustered. Mind-body exercise produces domain-specific dose-response patterns in older adults with MCI. For global cognition, the observed U-shaped pattern suggests that if 240 min/week dose is adopted clinically, extending intervention to at least 11 weeks is advisable. However, prospective validation of these findings is needed before clinical application.