Abstract
Background: Virtual reality (VR) intervention is an innovative and efficient rehabilitative tool for patients affected by stroke, Parkinson's disease, and other neurological disorders. This meta-analysis aims to evaluate the effects of VR intervention on cognition and motor function in older adults with mild cognitive impairment or dementia. Methods: Seven databases were systematically searched for relevant articles published from inception to April 2020. Randomized controlled trials examining VR intervention in adults with mild cognitive impairment or dementia aged >60 years were included. The primary outcome of the study was cognitive function, including overall cognition, global cognition, attention, executive function, memory, and visuospatial ability. The secondary outcome was motor function, consisting of overall motor function, balance, and gait. A subgroup analysis was also performed based on study characteristics to identify the potential factors for heterogeneity. Results: Eleven studies including 359 participants were included for final analysis. Primary analysis showed a significant moderate positive effect size (ES) of VR on overall cognition (g = 0.45; 95% confidence interval (CI) = 0.31-0.59; P < 0.001), attention/execution (g = 0.49; 95% CI = 0.26-0.72; P < 0.001), memory (g = 0.57; 95% CI = 0.29-0.85; P < 0.001), and global cognition (g = 0.32; 95% CI = 0.06-0.58; P = 0.02). Secondary analysis showed a significant small positive ES on overall motor function (g = 0.28; 95% CI = 0.05-0.51; P = 0.018). The ES on balance (g = 0.43; 95% CI = 0.06-0.80; P = 0.02) was significant and moderate. The ES on visuospatial ability and gait was not significant. In the subgroup analysis, heterogeneity was detected in type of immersion and population diagnosis. Conclusions: VR intervention is a beneficial non-pharmacological approach to improve cognitive and motor function in older adults with mild cognitive impairment or dementia, especially in attention/execution, memory, global cognition, and balance. VR intervention does not show superiority on visuospatial ability and gait performance.