Effectiveness of restricted diet with a plate in patients with type 2 diabetes: A randomized controlled trial.

BACKGROUND In view of the complexity of dietary and nutritional education for most patients with type 2 diabetes mellitus (T2DM), a simplified approach called the "restricted diet with a plate" or "plate model" is recommended. PURPOSE To evaluate whether the plate model can effectively improve glycemic control and cardiovascular risk markers in type 2 diabetes mellitus (T2DM), while reducing the time devoted to education and avoiding weight gain. METHODS The study was a randomized, multicenter, controlled study, conducted between October 2018 and October 2019, among patients with T2DM living in Nanjing. The study included 419 participants who were randomly divided into a plate group and a counting group. The plate model included three components: a low-literacy, color leaflet containing the explanation and composition of the plate model, health education, and medical visits. Patients in the counting group received health education, group medical visits, and a paper booklet containing traditional carbohydrate counting education. Primary outcomes were glycemic control and weight. RESULTS Participants in the plate model reduced HbA1c by 0.7% in the first three months, and reduced to a greater extent at six months (1.44%), but this was not sustained, and HbA1c increased slightly over the following six months. Fasting plasma glucose (FPG) and 2-h postprandial glucose (2hPG) values were significantly reduced at the endpoint in the plate model (9.25 ± 1.72% vs. 7.44 ± 0.88%, P = 0.008; 12.07 ± 2.94 vs. 8.35 ± 1.99%; P = 0.004); however, the 2hPG values decreased most significantly. Total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) levels decreased significantly in the plate group, which occurred at six months and lasted for 12 months. In the first three months, the average weight loss in the plate group was 1.2 kg/month (95% CI 0.92-1.48), and in the fourth to twelfth months, the average weight gain was 0.21 kg/month (95% CI 0.08-0.34). There was significant difference in education time between the groups (17.3 ± 4.42 vs. 38.6 ± 12.63; P < 0.001). CONCLUSIONS The plate model is at least as effective as the counting model over the short term for glycemic control and perhaps even better for weight and lipid control. Plate model has the potential to improve education of those with low health literacy by reducing reading demands.