Background Carbohydrate counting has become a principal strategy for children with type1 diabetes. Precision with carbohydrate counting are associated with lower HbA1C in children with type 1 diabetes (Mehta, 2009). Traditionally, children were not taught this during their first presentation with diabetes. In Gloucester and our network, carbohydrate counting is being taught from diagnosis to embed this practice early.

Aims With regards early carbohydrate counting education, we asked: is it practical? is it acceptable to families? is it effective?

Methods We compared two 12 month time periods 1/3/12–28/2/13 (education delayed) and 1/3/13–28/2/14 (education early). A sample of 15 patients from both years had their notes and biochemistry results examined. A questionnaire about patient’s experience of early carbohydrate counting was sent.

Results Comparing the two time periods (delayed vs. early); the mean age at diagnosis was 9.07 vs. 9.33 years, 36% vs. 47% of patients presented in DKA. The majority started on basal bolus regime but 14% vs. 7% of patients were started on a pump during first admission. There was no significant difference in length of stay (4.21 vs. 5.13 days, p = 0.056). Falls in HbA1C appear sustained across the early education group. 81% learnt CHO counting within a week, 64% preferred to learn CHO counting in hospital and 100% found the education useful.

Conclusion Early education about carbohydrate counting appears both practical and effective. There was no increase in length of stay between the two groups. There was generally positive feedback from those families that had early CHO counting education.