Effects and mediating mechanisms of a structured limbs-exercise program on general cognitive function in older adults with mild cognitive impairment: A randomized controlled trial

Exercise is known to prevent cognitive decline. Sleep quality and depression symptoms, which are associated with processing speed, are considered as common mediators in the exercise-cognition putative model. However, these mediating mechanisms have not been empirically tested in an intervention study. The aim of this study was to evaluate the effects of a structured limbs-exercise program on general cognitive function, and to test the mediating effects and mediating pathways of depressive symptoms, sleep quality and processing speed in the relationship of exercise-induced cognitive benefits. A two-arm and assessor-blinded randomized controlled trial. Community-dwelling older adults with mild cognitive impairment living in an urban area in Chifeng, China. Participants (N=116) were randomly allocated to one of the two arms: (1) a 24-week structured limbs-exercise program (3 supervised limb exercise sessions /week, 60 min /session for the first 12 weeks and 3 unsupervised practice sessions /week, 60 min /session for the following 12 weeks) or (2) health promotion classes alone. Measures of depressive symptoms, sleep quality, processing speed, and general cognitive function were collected at baseline, 12-week, and 24-week. Multivariate analysis of variance and structural equation modeling was used to test the effectiveness and mechanisms of structured limbs-exercise-induced cognitive improvement respectively. The structured limbs-exercise program was beneficial for maintaining general cognitive function at 12 weeks (mean difference = 1.20, 95% CI [0.354, 2.054], p = 0.006) and at 24 weeks (mean difference = 1.59, 95% CI [0.722, 2.458], p = 0.001) in the intervention group. The results from the goodness-of-fit indices of structural equation modeling show as following: (1) The effect of structured limbs-exercise program on cognitive function was partially mediated by depressive symptoms, sleep quality, and processing speed, with 69.22% of joint mediation proportion; (2) Relative to the combined Z values of depressive symptoms and processing speed, sleep quality was more strongly related to cognitive function in the structured limbs-exercise program (Z= 9.294, p<0.01); (3) Processing speed was affected by depressive symptoms, sleep quality, and in turn, yielding a significant effect on cognitive function; and (4) Five potential mediating pathways for improvement in general cognitive function in the structured limbs-exercise intervention were identified. This study shows that this exercise program can maintain general cognitive function for older adults with mild cognitive impairment. Mediating variables include depressive symptoms, sleep quality and processing speed. Future research should continue to incorporate path-oriented intervention strategies in the exercise intervention to maximize improvements in cognitive function. Registration number: ChiCTR1800016299