A small randomized controlled trial found that remote low-carbohydrate nutrition education was associated with greater short-term improvements in glycemic control, body weight, and oral antidiabetic medication use compared with usual primary care among adults with noninsulin-treated type 2 diabetes.
The single-center superiority trial included 58 patients recruited from a municipal primary health care registry in Francisco Beltrão, Paraná, Brazil. Eligible patients were aged 40 to 89 years, had type 2 diabetes, were not using insulin, had baseline glycated hemoglobin A1c of at least 6.5%, could read Portuguese, and had internet access and a smartphone communication application.
The primary endpoint was change in glycated hemoglobin A1c (HbA1c) from baseline to week 16. Secondary endpoints included fasting glucose, body weight, body mass index (BMI), the proportion of patients achieving HbA1c of 6.5% or lower, and changes in oral antidiabetic medication use.
Patients were randomly assigned to a remote intervention group or a usual-care control group, with 29 patients in each group. All randomized patients were included in the final analysis.
The intervention was delivered through WhatsApp and included nine educational videos, a digital nutrition guide, and 48 educational or motivational messages over 16 weeks. The low-carbohydrate approach limited carbohydrate-rich foods to about four tablespoons per meal, using household measures rather than strict daily gram counting. Patients also had remote access to the study nutritionist for questions, feedback, and guidance on potential diet-related adverse effects.
Patients in the control group continued conventional municipal primary health care, which included general guidance on healthy eating, physical activity, blood glucose monitoring, and medication use. The researchers noted that usual care varied across primary care units and did not include structured follow-up meetings or remote nutritional support.
At 16 weeks, HbA1c decreased from 7.90% to 6.99% in the intervention group and increased from 8.62% to 8.87% in the control group. Fasting glucose decreased by 16% in the intervention group and increased by 12% in the control group.
Body weight decreased by 5% in the intervention group and increased by 3% in the control group. BMI decreased by 5% in the intervention group and increased by 3% in the control group. Because the control group worsened on several measures, the between-group differences partly reflect both improvement with the intervention and deterioration with usual care.
Thirty-one percent of patients in the intervention group achieved HbA1c of 6.5% or lower at 16 weeks, compared with no patients in the control group. The researchers described this threshold as compatible with a glycemic target for remission; however, the trial did not establish diabetes remission. Current consensus definitions generally require HbA1c below 6.5% for at least 3 months without glucose-lowering medication, and this study did not use a standardized medication-withdrawal protocol.
Nearly half of patients in the intervention group reduced or discontinued oral antidiabetic medication use, whereas no patients in the control group did so. Medication was added or increased in 34% of control patients and in no patients in the intervention group. Medication decisions were made by treating physicians rather than directed by the study protocol, which reflects real-world practice but may introduce variability or bias in an unblinded trial.
The trial excluded patients with other types of diabetes, prior bariatric surgery, severe cardiovascular disease, chronic kidney disease limiting participation, pregnancy or lactation, eating disorders, or enrollment in weight-loss programs.
The findings may not be generalizable to other primary care settings, including those in the US. The study was conducted in a single Brazilian municipality, required internet and smartphone access, and compared an intensive digital nutrition-support program with usual primary care rather than isolating the independent effect of carbohydrate restriction.
The researchers noted several limitations, including no data on diabetes duration, missing waist circumference data for the control group, no quantitative dietary intake or macronutrient assessment, no direct adherence measurement, no formal assessment of physical activity changes, and only 16 weeks of follow-up. Baseline differences also warrant caution: patients in the intervention group had higher mean weight and BMI and slightly lower HbA1c, although these differences were not statistically significant.
“The online educational intervention conducted by a nutritionist and based on a low-carbohydrate dietary approach proved to be more effective for managing [type 2 diabetes mellitus] in primary health care than conventional care during the same period,” wrote lead study author Gisely Sanagiotto Balbinot, of the Master’s Program in Applied Health Sciences, Health Sciences Center, State University of Western Paraná, and colleagues.
The researchers reported no conflicts of interest.
Source: Diabetology International
