Optimizing exercise prescriptions for improving neuropsychiatric symptoms in older adults with MCI or dementia: a dose-response meta-analysis

Background Mild cognitive impairment (MCI) and dementia are common among older adults and are often accompanied by neuropsychiatric symptoms, which negatively affect quality of life and functional independence. Although exercise is a promising non-pharmacological intervention, the effective dose ranges for different exercise modalities in improving neuropsychiatric symptoms remain unclear. This study aimed to examine the dose-response relationships between different exercise modalities and neuropsychiatric symptoms among older adults with MCI or dementia. Methods Exercise dose was calculated as the product of duration, frequency, and intensity, expressed as metabolic equivalent of task minutes per week (MET-min/week). Bayesian network and dose-response meta-analyses were conducted to compare the effects of different exercise modalities and characterize their dose–response relationships. Results Nineteen studies involving 1587 older adults with MCI or dementia were included. Overall exercise showed a U-shaped dose-response association with improvements in neuropsychiatric symptoms, with the greatest effect observed at approximately 490 MET-min/week (SMD = −0.51, 95% CrI: −0.91, −0.08). When stratified by exercise modality, aerobic exercise (8 arms), resistance training (4 arms), and mind-body exercise (7 arms) were significantly associated with improvements in neuropsychiatric symptoms, although their effective and optimal doses varied across modalities. The doses associated with the model-identified greatest effects were 460 MET-min/week for aerobic exercise (SMD = −0.53, 95% CrI: −0.76, −0.24; very low certainty evidence), 470 MET-min/week for resistance training (SMD = −0.49, 95% CrI: −0.76, −0.17; very low certainty evidence), and 490 MET-min/week for mind–body exercise (SMD = −0.63, 95% CrI: −0.94, −0.28; low certainty evidence). No significant moderating effects were identified in network meta-regression analyses. Conclusion Aerobic exercise, resistance training, and mind–body exercise showed distinct dose–response relationships for improving neuropsychiatric symptoms in older adults with MCI or dementia. However, given the very low to low certainty of evidence, these findings should be interpreted cautiously. Further high-quality randomized controlled trials are warranted to confirm these findings. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420261486763 , identifier CRD420261486763.