Objective: The purpose of this study was to examine the effectiveness of the Functional Improvement Program of the Musculoskeletal System among users of Preventive Care Service under Long-Term Care Insurance. Methods: A total of 3,073 subjects were analyzed. We used the prediction formula to estimate the predicted value of the Kihon Checklist after one year, and calculated the measured value minus the predicted value. The subjects were divided into two groups according to the measured value minus predicted value tertiles: the lowest and middle tertile (good-to-fair measured value) and the highest tertile (poor measured value). We used a multiple logistic regression model to calculate the odds ratio (OR) and 95% confidence interval (CI) of the good-to-fair measured values of the Kihon Checklist after one year, according to the Functional Improvement Program of the Musculoskeletal System. Results: In potentially dependent elderly, the multivariate adjusted ORs (95% CI) of the good-to-fair measured values were 2.4 (1.3–4.4) for those who attended the program eight times or more in a month (vs those who attended it three times or less in a month), 1.3 (1.0–1.8) for those who engaged in strength training using machines (vs those who did not train), and 1.4 (1.0–1.9) for those who engaged in endurance training. Conclusions: In this study, among potentially dependent elderly, those who attended the program eight times or more in a month and those who engaged in strength training using machines or endurance training showed a significant improvement of their functional capacity.