To investigate the effects of remote exercise rehabilitation support based on behavior change techniques on exercise self-efficacy and glycosylated hemoglobin A_1c (HbA_1c) in patients with type 2 diabetes mellitus (T2DM).

A total of 381 patients with T2DM who were followed up for more than 6 months in Peking University First Hospital Diabetes Shared Care Clinic from January to December 2018 were enrolled. The patients who voluntarily received a 3-month behavior change techniques based remote exercise intervention were assigned to exercise intervention group [45 cases, (50.4±12.8) year-old, female 24 cases (53.3%), diabetes duration was 4.6 (0.8, 12.0) year] and the rest were in non-exercise intervention group [336 cases, (57.8±11.6) year-old, female 155 cases (46.1%), diabetes duration was 10.3 (2.8, 16.0) year]. After evaluated by exercise coach, patients in exercise intervention group with higher baseline self-efficacy scale scores only received exercise reminders and electronic exercise prescription, and those with lower self-efficacy received individualized video exercise prescription based on behavior change techniques. Including age, disease duration, gender, body mass index, and baseline HbA₁c were used as covariates, the propensity score analysis was used to compare the change of HbA_1c at six months for patients with uncontrolled baseline HbA_1c between exercise intervention group and non-exercise intervention group. The t test and Wilcoxoh test were used for data analysis.

Compared with baseline, both exercise self-efficacy score [40 (31, 54) vs 55 (26, 83) points, Z=-2.229, P=0.026] and exercise self-management ability score [2 (2, 4) vs 4 (3, 5) points, Z=-2.401, P=0.016] were significantly improved after six month in low exercise self-efficacy group. HbA_1c were significantly lower in the exercise intervention group compared with baseline (8.9%±2.0% vs 6.9%±0.8%, t=5.723, P<0.000 01).

In the context of effective multidisciplinary management of diabetes, implementation of behavioral change technique-based remote exercise support can improve exercise self-efficacy and HbA_1c in patients with poor controlled T2DM.

Diabetes mellitus; Hemoglobin A, glycosylated; Diabetes shared care; Exercise support; Behavior change technique; Exercise self-efficacy