RESUMO
AIM: To evaluate the predictive factors of glycemic control in children and adolescents with type 1 diabetes mellitus (T1DM). METHODS: Cross-sectional study at a referral service in Rio de Janeiro, Brazil. Sociodemographic, anthropometric, clinical, and dietary factors were evaluated. Food consumption was evaluated by 24â¯h dietary recall and the NOVA system was adopted for classifying the foods according to the extent and purpose of industrial processing. The predictive factors were evaluated by multivariate linear regression, adopting pâ¯<â¯0.05. RESULTS: One hundred and twenty children and adolescents participated, with a mean age of 11.74â¯years (±2.88) and HbA1c of 8.13% (±1.26). The mean diabetes duration was 6.68â¯years (±3.33) and the insulin used was 1.05 units per kilogram of ideal weight (IU/kg of ideal weight; ±0.46) About 80% (nâ¯=â¯96) used carbohydrate counting and it was verified that 24.27% (±17.89) of the participants' total calories came from ultraprocessed foods. For each year of diagnosis with T1DM and for each IU/kg of weight used, HbA1c increased by 0.087% (ßâ¯=â¯0.087, pâ¯=â¯0.007) and 0.651%, respectively (ßâ¯=â¯0.651; pâ¯=â¯<0.001). Use of carbohydrate counting was associated with a 1.058% reduction in HbA1c (ßâ¯=â¯-1.058; pâ¯=â¯0.001). CONCLUSION: Disease duration and insulin dose were directly reflected in HbA1c concentrations, while carbohydrate counting showed an inverse association.