ABSTRACT
ABSTRACT Objective: The objective of this study was to verify the impact of carbohydrate counting (CC) on glycemic control and body weight variation (primary and secondary outcomes, respectively) between consultations in patients with diabetes mellitus (T1D) followed at a tertiary hospital in southern Brazil in a public health system environment. We also sought to investigate CC adherence. Materials and methods: This retrospective cohort study included 232 patients with T1D who underwent nutritional monitoring at a referral hospital for diabetes care between 2014 and 2018. To assess primary and secondary outcomes, data from 229 patients, 49 of whom underwent CC during this period and 180 individuals who used fixed doses of insulin, were analyzed. The impact of CC on glycemic control was assessed with the mean glycated hemoglobin (HbA1c) level at all consultations during the follow-up period. Results: In the model adjusted for the most confounders (except pregnancy), the mean HbA1c was better in the CC group (8.66 ± 0.4% vs. 9.36 ± 0.39%; p = 0.016), and body weight variation was lower (0.13 ± 0.28 kg vs. 0.53 ± 0.24 kg; p = 0.024). Adherence to CC was reported in 69.2% of consultations. Conclusion: CC optimized the glycemic control of individuals with T1D, resulting in less weight variation than in the fixed insulin dose group, which indicates that CC is an important care strategy for these patients.