Optimal type and dose of exercise for improving executive functions in older adults: a systematic review and Bayesian model-based network meta-analysis of RCTs

OBJECTIVE: To identify the optimal type and dose of exercise for improving executive function in adults aged 65 years and older. METHODS: We systematically searched five databases (Cochrane Library, PubMed, Embase, Web of Science, and China National Knowledge Infrastructure) from inception to 7 December 2024. Following systematic screening, data extraction, and risk-of-bias assessment, a random-effects Bayesian network meta-analysis and dose-response analysis were performed using R software (version 4.2.0). RESULTS: A total of 38 studies involving 2,928 participants were included. The effects of exercise modalities on executive function in adults aged ≥ 65 years differed according to baseline cognitive status. In cognitively healthy older adults, aerobic exercise significantly improved all three domains of executive function, whereas mind-body regulation training and resistance training showed no significant effects. In contrast, among older adults with cognitive impairment, all three exercise types significantly improved working memory, while no significant benefits were observed for inhibitory control or cognitive flexibility. Dose-response analysis identified optimal aerobic exercise doses for cognitively healthy older adults: 1,134 MET-min/week for working memory and inhibitory control, and 530.2 MET-min/week for cognitive flexibility. No minimum threshold for cognitive benefit was identified, suggesting that even low-volume of physical activity may confer advantages. CONCLUSION: Aerobic exercise is the most effective modality for enhancing executive function in cognitively healthy older adults and for improving working memory in those with cognitive impairment. Although the findings support the World Health Organization's physical activity guidelines (600-1,200 MET-min/week) for general cognitive benefits, optimal outcomes require dose personalisation according to the targeted executive function domain.