Optimizing exercise dosage for executive function in alzheimer's disease: A Bayesian dose-response meta-analysis of...

This study aimed to identify the optimal dose range of physical exercise for improving executive function in individuals with Alzheimer's disease (AD) and to compare the dose-dependent cognitive response profiles across different exercise modalities.

A systematic literature search was conducted across five databases-PubMed, Embase, Cochrane Library, SPORTDiscus, and Web of Science-up to June 20, 2025, with no language restrictions. All analyses were performed in R (version 4.4.2). A Bayesian framework was used to model continuous dose-response curves and to evaluate the relative effects of aerobic training, integrated physical training, and resistance exercise within a network structure.

Exercise interventions demonstrated a nonlinear relationship with executive function, with the strongest effect observed at approximately 1000 MET-min/week. Among exercise types, continuous aerobic training showed significant improvement at 670 MET-min/week, indicating a relatively low-dose threshold for benefit. In contrast, integrated physical training programs required higher doses (around 1200 MET-min/week) to reach peak effect, suggesting greater intensity dependence. Resistance training did not yield consistent or statistically significant outcomes across the tested dose levels.

This study establishes a quantitative framework that integrates dose modeling with intervention structure to guide exercise-based cognitive strategies for AD. The findings emphasize the need to match exercise volume to modality demands and individual capacity, supporting the development of precision-oriented, non-pharmacological interventions. The identified optimal range (670-1200 MET-min/week) aligns with WHO recommendations for older adults, reinforcing both feasibility and translational potential.