Network meta-analysis of the effects of different cognitive trainings on the cognitive function of patients with mild cognitive impairment

Examining the relationship between the responses of a number of different cognitive trainings on cognitive functioning in middle-aged and elderly patients with mild cognitive impairment. Randomized controlled experimental studies published publicly from the time of inception to October 30, 2023 were searched through Web of Science, PubMed, Embase, and Cochrane library databases. Traditional and network meta-analyses were performed using Stata 17.0 software. Fifty papers on 4 types of cognitive training were included. Traditional meta-analysis showed that virtual reality training (SMD = 0.53, 95%CI:[0.36,0.70], P = 0.00), neuropsychological training (SMD = 0.44, 95%CI:[0.18,0.70], P = 0.00), cognitive strategy training (SMD = 0.26, 95%CI:[0.16,0.36], P = 0.00), and cognitive behavioral therapy (SMD = 0.25, 95%CI:[0.08,0.41], P = 0.00) all had significant improvement effects on the cognitive function of middle-aged and elderly patients with mild cognitive impairment. Network meta-analysis revealed neuropsychological training as the best cognitive training, and subgroup analysis of cognitive function subdimensions showed that neuropsychological training had the best effects on working memory, lobal cognitive function, memory, and cognitive flexibility improvement. Meanwhile, virtual reality training had the best effects on processing speed, verbal ability, overall executive function, spatial cognitive ability, and attention improvement. Cognitive training can significantly improve the cognitive function of middle-aged and elderly patients with mild cognitive impairment, and neuropsychological training is the best intervention, most effective in interventions lasting more than 8 weeks.