King's College London

Objectives: Physical exercise may benefit people with Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, randomised controlled trials (RCTs) of exercise have shown conflicting findings and it is unclear if positive outcomes are comparable to a commonly used cholinesterase inhibitor, donepezil. Methods: Embase, Medline, PsycINFO, PsycARTICLES, SCOPUS were searched for RCTs of physical activity compared to a control condition, and donepezil compared to placebo in people with AD and MCI. Effect sizes were calculated from pre- and post-MMSE and ADAS-Cog scores and pooled using a random effects meta-analysis. Results: Ninteen RCTs were included in the exercise meta-analysis (AD, N = 524; MCI, N = 1269). Physical exercise improved MMSE scores in AD (Hedges' g = 0.46) and MCI groups (g = 0.63). For the MCI group, exercise appeared to have a stronger effect for those with lower MMSE scores at baseline (p = 0.022). 18 RCTs were included in the donepezil meta-analysis (AD, N = 2984, MCI, N = 1559). In people with AD, donepezil improved cognition (MMSE g = 0.23; ADAS-Cog, g = −0.17) but there was no evidence of improved cognition in MCI. Conclusions: Physical exercise improved cognition in both AD and MCI groups. Where comparisons were possible, the effect size for physical exercise was generally comparable to donepezil. These results strengthen the evidence base for exercise as an effective intervention in AD and MCI, and future clinical trials should examine exercise type, intensity and frequency, in addition to cholinesterase inhibitors to determine the most effective interventions for AD and MCI.