Effects of telemedicine interventions on cognitive function in post-stroke cognitive impairment: A systematic review and meta-analysis

Background: Post-stroke cognitive impairment is characterized by cognitive dysfunction occurring within 6 months post-stroke. Telemedicine uses communication technologies to deliver healthcare remotely and has shown efficacy in improving cognitive impairment. However, a systematic review specifically evaluating telemedicine's effects on cognitive outcomes in post-stroke cognitive impairment is lacking. Objectives: This systematic review aimed to examine the effectiveness of telemedicine interventions for cognitive function in post-stroke cognitive impairment. Methods: A comprehensive search was performed across 10 electronic databases, including PubMed, Web of Science, CINAHL, EMBASE, Cochran library, Scopus, and ProQuest Dissertations and three Chinese-language databases (CNKI, Wan Fang, and Vip) from their respective inception dates to May 2025. This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Meta-analysis was performed by the use of Review Manager 5.3 and Stata 15. Results: The initial search yielded 10,365 articles, with 19 studies included in the systematic review. The results showed that telemedicine interventions had a significant moderate effect on global cognition (standardized mean difference (SMD = 0.69, Z = 4.23, P < 0.01) and significantly improved visuospatial function (SMD = 0.57, Z = 2.76, P < 0.05) and language (SMD = 0.62, Z = 2.59, P < 0.05). Sub-group analyses showed significant effects for both online tools or computer software and mobile apps, though high heterogeneity was noted. Additionally, telemedicine interventions had a significant effect on activities of daily living (SMD = 0.64, Z = 5.16, P < 0.01). Conclusion: Considering the obstacles and limitations of traditional face-to-face rehabilitation, telemedicine is an effective approach for treating post-stroke cognitive impairment that can significantly improve cognitive function. Future studies should address heterogeneity through rigorous designs, long-term follow-ups, neuroimaging, and biomarker integration to elucidate underlying mechanisms.