International Psychogeriatrics · Nov 7, 2026

Effects of virtual reality and exergame interventions on cognitive function in Chinese older adults with cognitive impairment: A three-level meta-analysis

By Shen Y, Zhong Y, Lin A

Objectives Technology-based cognitive interventions, such as virtual reality (VR) and exergames, have emerged as promising tools for cognitive rehabilitation. This meta-analysis evaluated their effects on cognitive outcomes among Chinese older adults with cognitive impairment and examined potential moderators. Methods A systematic search of English- and Chinese-language databases identified eligible studies on VR and exergame interventions in Chinese older adults with cognitive impairment. To address data dependency, effect sizes were synthesized using a three-level random-effects meta-analysis estimated by restricted maximum likelihood estimation, with study-level cluster-robust variance estimation incorporating CR2 small-sample correction and Satterthwaite approximation. Sensitivity analyses, subgroup analyses, publication bias analyses, and GRADE assessment were also conducted. Results Thirty-eight studies involving 2873 participants and 202 effect sizes were included, of which 34 were randomized controlled trials (RCTs). In the primary RCT-only analysis, the pooled effect was significant (Hedges’ g = 0.817, 95% CI [0.602, 1.032], p < 0.001). The pooled effect across all studies remained significant (Hedges’ g = 0.738, 95% CI [0.530, 0.946], p < 0.001). Significant positive effects were observed across all four cognitive domains, but between-group differences in subgroup analyses were not statistically significant. Heterogeneity was substantial, and the prediction interval crossed zero. Leave-one-out analyses supported the stability of the findings. Modified Egger’s regression suggested possible small-study effects, although publication bias could not be determined conclusively. The certainty of evidence was rated as low. Conclusions VR and exergame interventions were associated with a significant positive average effect on cognitive outcomes among Chinese older adults with cognitive impairment. However, substantial heterogeneity, a prediction interval crossing zero, possible small-study effects, and low certainty of evidence limit firm conclusions regarding the consistency and clinical importance of this effect. More rigorously designed randomized trials with longer follow-up are needed.

doi.org/10.1016/j.inpsyc.2026.100313PubMed 42998936Interventions