Physical Exercise as a Nonpharmacological Intervention for the Treatment of Neuropsychiatric Symptoms in Persons With Dementia: A Meta-analysis of Randomized Controlled Trials.

Background: Neuropsychiatric symptoms (NPS) are prevalent in dementia and affect both patients and caregivers in multiple ways. Because of the complications of drug treatments, nonpharmacological interventions, such as exercise, are of particular value. This study aimed to investigate the effect of exercise on the NPS of dementia and draw recommendations for the disease management.

Methods: Meta-analyses were conducted on the findings of randomized controlled trials identified during an initial systematic review of the literature, according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The included studies examined the effect of exercise interventions on patients with dementia or mild cognitive impairment using valid assessment tools. The quality of evidence was assessed using Grading of Recommendations, Assessment, Development, and Evaluation analysis.

Results: The meta-analyses revealed that exercise significantly limits NPS [mean difference: -5.28, (95% CI, -9.46, -1.11), P = 0.01] and symptoms of depression [standardized mean difference: -0.16, (95% CI, -0.29, -0.02), P = 0.02], and has a beneficial effect on agitation symptoms. The Grading of Recommendations, Assessment, Development, and Evaluation analysis results showed that exercise has a moderate and high confidence positive effect on NPS and depression, respectively.

Conclusions: Exercise could be an alternative approach for nonpharmacological treatment of NPS in dementia. Therefore, exercise could be utilized as a treatment of choice or to support existing treatment regimens.