Carbohydrate-counting smartphone apps may be a useful adjunct for estimating carbohydrate content, according to a study published in Diabetes Therapy. Investigators conducted a study to evaluate the accuracy of two carbohydratecounting apps designed to help patients accurately count their carbohydrate intake in order to optimize prandial insulin dose matching. Using the two apps—Foodvisor (which uses automatic food photo recognition technology) and Glucicheck (which requires manual entry of carbohydrates with the help of a photo gallery), medical students filling the role of mock patients evaluated their meals. App-obtained macronutrient quantifications were compared with a reference quantification. Foodvisor underestimated the carbohydrate content of entire meals (Foodvisor quantification minus gold standard quantification = – 7.2 ± 17.3 g), whereas Glucicheck reasonably accurately estimated carbohydrate content (Glucicheck quantification minus gold standard quantification = 1.4 ± 13.4 g). The percentages of meals with an absolute error in carbohydrate quantification above 20 g were 30% for Foodvisor and 14% for Glucicheck. “However, both apps provided a lower mean absolute carb counting error than that usually made by T1D patients in everyday life, suggesting that such apps may be a useful adjunct for estimating carbohydrate content,” write the study authors.
