Effectiveness and acceptability of non-pharmacological interventions in people with mild cognitive impairment: Overview of systematic reviews and network meta-analysis

Background: To explore effectiveness and acceptability of non-pharmacological interventions in mild cognitive impairment (MCI). Methods: Overview of systematic reviews and network meta-analysis were conducted. Systematic reviews (SRs) were searched via seven databases from June 2015 to June 2020. Randomized controlled trials (RCTs) were retrieved. The Methodological quality was assessed by AMSTAR 2 and RoB 2. Outcomes were effectiveness and acceptability measured by standardized mean differences (SMDs) and odd ratios (ORs) with 95% confidence interval (CI). Pairwise meta-analysis was first conducted, followed by network meta-analysis. Results: A total of 22 SRs and 42 RCTs with 4401 participants were included. The methodological quality of included SRs and RCTs were moderate. There were four interventions, with three types of physical activity (aerobic, muscle-strengthening, and mind-body), three types of cognitive (rehearsal-based, compensatory, and mixed), multicomponent (physical and cognitive component), and nutrition intervention. No significant inconsistency was identified. Regarding intervention effectiveness, muscle-strengthening (SMDs 0.87, 95% CI 0.31–1.43; rank 1), mind-body (0.76, 0.38–1.14; rank 2) and aerobic (0.34, 0.13–0.50; rank 3) were significantly better than the control group and there was no significant difference among these types of intervention . Cognitive intervention of rehearsal-based (1.33, 0.30–2.35; rank 1) and mixed (0.55, 0.00–1.11; rank 2) were significantly better than the control group and there was no significant difference among these types of intervention. Multicomponent intervention (0.32, 0.02–0.62) were significantly better than the control group but not better than the single component group. Regarding acceptability, there was no significant difference among types of intervention. Conclusion: Physical activity, cognitive, and multicomponent intervention could be provided regardless of their types and acceptability due to their effectiveness on improved cognitive function for people with MCI.