摘要:Background: The prevention of severe hypoglycaemia is based in art on accurately anticipating its occurrence. The aim of the study was to analyse the effect of a feedback procedure on the glycaemic state of children with type 1 diabetes, aswell as the associated symptoms and estimated BGL. Method : For 7 days, 17 participants performed both scheduled capillary blood tests, and an estimate and self-analysis whenever they thought that there were changes in their BGL. Results : A total of 454 trials were analysed with a mean of 178 mg/dl (SD= 36.4). Only 5.1% of the estimates corresponded to hypoglycaemia states, 36.77% were in areas with clinically dangerous consequences and were more accurate when the BGL reached the levels of 240 mg/dl. The children reported symptoms at random, made mistakes when linking them with the BGL and demanded opportunity feedback believing they were in state of hypoglycaemia, while being in hyperglycaemia. Conclusions : The BGL was highly variable, certain glycaemic profiles peaked, with a greater average of hypoglycaemia than in the children’s usual daily surroundings and did not present a profile of reliable signs indicative of BGL. The tendency to underestimate values when glycaemia was in hyperglycaemic levels is confirmed.