Non-Pharmacological Interventions for Patients with Mild Cognitive Impairment: A Meta-Analysis of Randomized Controlled Trials of Cognition-Based and Exercise Interventions

Background: Non-pharmacological interventions, including cognition-based intervention and physical exercise, are available for mild cognitive impairment (MCI), but their efficacy remains uncertain. Objective: To evaluate efficacy of cognition-based intervention and physical exercise on cognitive domains in patients with MCI. Methods: We searched MEDLINE, EMBASE, the Cochrane library, and BIOSIS previews to identify randomized controlled trials (RCTs) that involved cognition-based intervention and physical exercise for persons who were diagnosed with MCI. The pooled weight mean difference or standard mean difference (SMD) were calculated using a random-effect model. Results: We found significant effects of cognition-based intervention on global cognitive function (SMD 0.37 [0.07, 0.68] p = 0.02). Besides, cognition-based intervention produced significant effects on executive function reported with TMT-B (SMD 0.8 [0.09, 1.5] p = 0.03) and delayed memory (SMD 0.31 [0.01, 0.61] p = 0.05). A beneficial improvement in global cognitive function was also seen in the exercise group compared to the control group (SMD 0.25 [0.08, 0.41] p = 0.003). Conclusions: Both of cognition-based intervention and physical exercise had the potential to improve global cognitive function. Weak evidence of improvements was also observed for the domains of executive function and delayed memory following cognition-based intervention. Nevertheless, future standard RCTs are still needed to identify the clinical value of our results.