Effects of exercise intervention on exercise behavior in community-dwelling elderly subjects: A randomized controlled trial

Aim: The objective of the present study was to evaluate the long-term effectiveness of an exercise program in modifying the exercise behavior of the community-dwelling elderly subjects. Methods: This study was a single-blinded randomized controlled trial. The subjects included 52 males and 65 females 65 years of age or over who were randomly assigned to an exercise-intervention group or a health-education group. The stages of change in exercise behavior were evaluated before and one-year after the intervention period. The subjects' physical function (muscle strength, balance, walking speed) and self-efficacy in each domain of the physical function were measured during the intervention period. Results: There were no significant differences in the stages of change before the intervention between the two groups. Significant differences in the stages of change were observed in "relapse" of stages at two points in time between the two groups (p<.01). A logistic regression analysis showed that "progression" of stages was associated with improvements in the timed up and go test (AOR 2.7; 95% CI 1.3-5.8) and sit and reach (AOR 1.14; 95%CI 1.0-1.3), while "relapse" of stages was associated with the group allocation (AOR 4.6; 95%CI 1.1-18.8), self-efficacy in "Walking" (AOR 1.54; 95%CI 1.0-2.3) and "Stair climbing" (AOR 0.68; 95%CI 0.5-0.9) with respect to physical activity during the intervention period. Conclusions: The results suggest that exercise intervention in community-dwelling elderly subjects is effective in preventing "relapse" of exercise behavior over long periods.