Efficacy of technology-based cognitive and exercise interventions for mild cognitive impairment: A systematic review, network meta-analysis, and meta-regression of randomized controlled trials

Background: Technology has been increasingly integrated into controlling the decline of cognitive function. It is unclear whether technology-based cognitive and exercise interventions (T-CEIs) could generate synergistic benefits and what components would optimize this effect. This study aimed to compare the effectiveness of various T-CEIs on cognitive function in individuals with mild cognitive impairment (MCI). Methods: In this study, we searched MEDLINE, Web of Science, Scopus, Embase, and APA PsycInfo from inception to November 4, 2023. We included randomized controlled trials that evaluated the effects of T-CEIs on cognitive function for individuals with MCI. The primary outcome was global cognition. The outcomes were summarized in narrative synthesis and combined using meta-analysis. Pairwise meta-analysis and network meta-analysis were sequentially performed to investigate the effects of each category of interventions and their comparative intervention effectiveness, respectively. Meta-regression was performed to examine the influence of study design and participants' characteristics on the intervention effectiveness. This systematic review protocol was registered in PROSPERO (CRD 42023486359). Results: Twenty-eight studies with 1633 participants were included. The results of pairwise meta-analyses indicated that T-CEIs were superior to active/passive controls in improving global cognition, cognitive shifting, processing speed, working memory, delayed recall, and category fluency (p p < 0.05). The results of network meta-analyses indicated that the optimal components in improving global cognition (SUCRA 77.0 %, SMD 0.85, 95 % CI-0.17 to 1.87) and cognitive shifting (SUCRA 92.4 %, SMD 1.57, 95 % CI 0.88-2.25) were cognitive stimulation (CS) combined with mind-body exercise (MBE), while cognitive training combined with MBE was the most beneficial in developing processing speed (SUCRA 88.5 %, SMD 0.68, 95 % CI 0.14-1.22). Meta- regression further suggested that the effects of the tested interventions were independent of the various factors related to study design and participants' characteristics. Conclusions: T-CEIs are effective in improving global cognition and core subdomains of cognition in individuals with MCI. This review highlights the superior effects of technology-based CS combined with MBE in improving global cognition.