Comparative Effectiveness of Multicomponent Exercise Interventions on Cognitive Function in People With Cognitive Impairmsent: A Systematic Review and Network Meta-Analysis

Abstract

Background: Although multicomponent exercise is a popular nonpharmacological treatment, its effects on cognition vary across studies because of the diversities in exercise combinations. Identifying the most effective combination is of great importance to the prevention and treatment of cognitive impairment. Aims: To compare and rank the efficacy of various multicomponent exercise interventions on cognition in people with cognitive impairment. Methods: We searched PubMed, Web of Science, Embase, Cochrane, SPORTDiscus and PsycInfo databases up to April 2025 for eligible randomised controlled trials about multicomponent exercise interventions in people with cognitive impairment. Primary outcome was global cognition, with secondary outcomes being executive function and memory. Pairwise and network meta-analyses were performed using random-effects models. Results: Twenty-five trials involving 2298 participants were included. Pairwise meta-analyses showed multicomponent exercise interventions were effective on global cognition (standardised mean difference (SMD) = 0.59; 95% confidence interval (CI): 0.30, 0.89; p < 0.001) and executive function (SMD = 0.28; 95% CI: 0.12, 0.45; p < 0.001). Network meta-analyses revealed that aerobic exercise (AE) + balance & flexibility (BF) training had the highest probability (70.8%) of being the optimal exercise combination for global cognition (SMD = 1.07; 95% CI: 0.23, 1.90; p = 0.016), and AE + resistance exercise (RE) had the highest probability (43.1%) of being the optimal exercise combination for executive function (SMD = 0.56; 95% CI: 0.03, 0.10; p = 0.042). We did not observe significant effects of multicomponent exercise on memory. Conclusion: AE + BF training is likely the most effective multicomponent exercise combination for global cognition, while AE + RE showed the optimal effect on executive function in people with cognitive impairment. Relevance to Clinical Practice: Our study contributes to guiding clinical professionals to design and conduct targeted multicomponent exercise interventions as per individual cognitive impairment characteristics to protect individual cognition. Patient or Public Contribution: No patient or public contribution applies to this work. Study Registration: The study protocol was registered with PROSPERO (CRD42023489517).