ABSTRACT
Background: Frequent intervention has been strongly recommended for diabetic patients undergoing exercise therapy. However, high-frequency interventions for all patients are inefficient. The purpose of this study is to examine the characteristics of groups divided based on changes in exercise behaviors. Methods: The participants comprised 42 diabetic patients who completed a two-week program to improve their glycemic control and receive diabetes education. Their change in exercise behavior, self-efficacy of exercise, and diabetes and social statuses were collected at the time of discharge and 12 and 24 weeks after discharge. Based on such data, the participants were divided into five groups: (1) the IW12 group whose exercise habits were interrupted within 12 weeks; (2) the IW24 group whose exercise habits were interrupted within 24 weeks; (3) the CO24 group who continued exercising after 24 weeks; (4) the Action group that actively exercised at the time of discharge; and (5) the Maintenance group that maintained same level of exercise at the time of discharge. Results: The total score of self-efficacy of exercise at the time of discharge was 13.7 ± 2.1 (the IW12 group), 11.3 ± 2.5 (the IW24 group), 16.2 ± 3.2 (the CO24 group), 16.7 ± 3.1 (the Action group), and 15.9 ± 2.3 (the Maintenance group). The scores for the IW24 group were significantly lower compared with the Action and Maintenance Patients groups (p < 0.01). Conclusion: Our findings suggest that intervention should be performed frequently, especially in a diabetic patient who is the preparation stage under low self-efficacy of exercise.