Fig. 1Enrollment of patients with type 2 diabetes mellitus and the study design. Data represent the number (%) of patients.
Fig. 2Changes in the scores of lifestyle behaviors during the study period with respect to the duration of diabetes before education. (A) Before education. (B) After education. Diet habit: Score 1, irregular diet with unlimited snacks; 2, irregular with intermittent snacks; 3, appropriate meal calories, regular diet with some snacks (>2 times/day or >300 kcal/day of excess intake; 4, appropriate meal calories, but a few snacks (≤1 time/day, or 100 to 300 kcal/day of excess calorie); 5, tightly controlled, with no intermittent snacks. Self-monitoring of blood glucose (SMBG): Score 1, never; 2, monthly; 3, weekly; 4, 3 to 4 times/wk; 5, ≥ daily. Physical activity: 1, never; 2, <30 min/wk (weekly); 3, <60 min/wk (1 to 2 times per week); 4, <120 min/wk (3 to 4 times per week); 5, daily, >150 min/wk. The P values denote the differences between the groups (≤1Y vs. ≥3Y) at the given time point.
Fig. 3Mean hemoglobin A1c (A1C) levels in both groups. (A) Mean values during the study period. (B) Changes in A1C levels in both groups. Data are presented as the mean±standard deviation. ≤1Y, diabetes duration of less than 1 year before diabetes education or in recently diagnosed patients; ≥3Y, diabetes duration of more than 3 years before education. aP<0.05 vs. ≤1Y group.
Table 1Study design and follow-up
Table 2Baseline characteristics between groups
Table 3Multiple logistic regression analysis of a mean A1C level higher than 7.0% (53 mmol/mol) during the follow-up period