Background: Non-pharmacological interventions (NPIs) are important in cognitive decline prevention in individuals with mild cognitive impairment (MCI). However, the dose-response relationship remains unclear.
Design: Systematic review and meta-analysis of randomised controlled trials.
Methods: Seven databases were searched until April 2020. RCTs of NPIs in individuals with MCI were eligible for inclusion. Hedge's g was used to calculate the effect size. A random-effect meta-analysis was used to explore the impact of NPIs on cognition. Subgroup analysis was used to investigate the moderates. The dose was measured by prescription (frequency, intensity, type, time and volume) and intervention characteristics (period, energy expenditure, delivery mode and setting) in NPIs.
Results: Forty-two studies with 4401 participants were included. In the NPIs, cognitive intervention (g = 0.167), physical exercise (g = 0.536) and multicomponent intervention (g = 0.386) had significant effect on cognition in individuals with MCI. Dose-response results showed cognition could be significantly improved in 1-2 times/week (p < .05), 60-120 min/session (p < .05), ≥12 weeks (p < .05), supervised (p < .05), clinical setting (p < .05) in cognitive intervention. In physical exercise, cognition could be improved in ≥3 times/week (p < .05), vigorous-intensity (p < .05), muscle-strengthening activity (p < .05), 30-60 min/session (p < .05), 6-12 weeks (p < .05), unsupervised (p < .05), community setting (p < .05). In the multicomponent intervention, cognition could be improved in 1-2 times/week (p < .05), 30-60 min/session (p < .05), 8-16 weeks (p < .05), clinical (p < .05). In nutrition intervention, cognition could be better improved DHA (p < .05), >1000 mg/day (p < .05).
Conclusions: The effectiveness of cognitive intervention is significantly influenced by frequency, time, period, delivery mode and setting. The effectiveness of physical exercise is significantly influenced by frequency, intensity, type, time, period, delivery mode and setting. The effectiveness of multicomponent intervention is significantly influenced by frequency, time, period and setting. The effectiveness of nutrition intervention is significantly influenced by dose and type.
Relevance to clinical practice: The research summarised the evidence to guide the best prescription of NPIs and helped clinicians design more effective interventions in individuals with MCI.
Keywords: cognitive function; meta-analysis; mild cognitive impairment; non-pharmacological interventions; randomised controlled trials.
© 2022 John Wiley & Sons Ltd.