Sarcouncil Journal of Internal Medicine and Public Health

Abstract: Background: Mild cognitive impairment (MCI) and dementia are marked by important transition phases, which are optimal periods to influence the progression of cognitive decline. Exercise training is promoted worldwide, but mixed data regarding efficiency are found regarding randomized clinical trials. Objective: To quantify the effects of structured exercise training on cognitive function in older adults with MCI or early dementia, and to identify intervention- and participant-level moderators influencing outcomes. Methods: A systematic literature search of PubMed, Embase, and the Cochrane CENTRAL database (from inception to November 2025) identified RCTs assessing the effect of multi-session exercise interventions on adults [?] 60 years with MCI and dementia. Data was extracted independently by two authors. Hedges' g was calculated by using models of random effects. Heterogeneity, subgroup, meta-regression, sensitivity, and bias analysis (Egger's test, Trim and Fill) were conducted. Findings: A total of eighteen studies (n = 1,380 participants) were included. Exercise training resulted in a significant, large improvement on global cognitive performance (Hedges' g = 0.79, 95% CI: 0.28 to 1.30). Significant benefits were found regarding global cognitive performance (g = 0.82), executive functions (g = 0.67), and memory (g = 0.61). Combined aerobic and strength training programs achieved the largest effects (g = 1.42), and supervised exercise training significantly outperfomed unsupervised training (b = 0.68, p = 0.032). Heterogeneity was found to be very high (I2 = 94.9%). These findings were verified to be robust by sensitivity analysis. No indication of publication bias could be found. Conclusion: Structured exercise produces significant cognitive benefits to older adults with mild cognitive impairment and dementia, and multimodal (aerobic and resistance) and supervisor-led interventions are associated with more favorable cognitive outcomes. These results indicate the value of adding supervised multimodal exercise programs to cognitive management interventions.