Biting off more than you can chew; is it possible to precisely count carbohydrate?

Aim: Carbohydrate counting is used to adjust premeal insulin to carbohydrate intake in intensive insulin regimens. The aim of the present study was to determine the potential variability in the carbohydrate content of a slice of bread (one 'exchange') from that reported on the label and hence, the potential variability in carbohydrate intake when consuming a serve. Methods: A cross-sectional survey of 11 different loaves of bread commonly consumed by children with type 1 diabetes was undertaken. All slices in each loaf were weighed to an accuracy of +/- 1 g; and the reported carbohydrate content per 100 g of each loaf was used to determine the carbohydrate content of the mean, minimum and maximum slice in each loaf of bread. Results: There was no difference between the reported and the mean estimated carbohydrate content of a slice. The minimum slice of bread across all loaves was estimated to contain only 10.0 g of carbohydrate, whereas the maximum slice contained an estimated 20.7 g of carbohydrate. The greatest variation in carbohydrate amount within a loaf was 12.3 g. Conclusions: In commercially available loaves of bread in Australia, the carbohydrate content of a slice can vary by up to 45% of that reported on the label, in accordance with Food Standards Australia New Zealand. This highlights a practical limitation inherent with the commonly held view that food labels can facilitate accuracy in carbohydrate counting in 1-g increments.