Usability Testing of A Carbohydrate Counting Application designed For and With Young Adults with Type 1 Diabetes (Preprint)

BACKGROUND Carbohydrate counting assists people with type 1 diabetes (T1D) adjust mealtime insulin doses, however, it is often burdensome. Mobile applications can simplify this process by automating carb estimation and insulin calculations, yet no comprehensive solution currently combines photo-based carb recognition with an integrated bolus calculator. OBJECTIVE This study aimed to optimize a novel app designed to support young adults with T1D in carbohydrate counting and insulin dosing by incorporating feedback from usability testing. METHODS We used a think-aloud protocol and conducted four rounds of interviews, each with 3–5 participants, to assess effectiveness of the app and identify areas for improvement. RESULTS A total of 18 completed the usability testing. Thematic analysis revealed seven key insights; 1) A person-centered design is important for an individualized experience, addressing the individual as a whole, not just their diabetes; 2) An intuitive user interface is essential to maintain engagement, with clear information presentation and easy interaction; 3) The relevance of information, should be presented in familiar language to enhance identification and inclusion; 4) Personalized features for a tailored user experience; 5) Robust data verification mechanisms and override abilities to avoid human and technological errors; 6) A comprehensive application to improve patient-practitioner communication and reduce manual tracking or the use of multiple apps; 7) Linkage of various factors (exercise, diet, time of injection) to glucose levels could improve self-efficacy and promote personalized learning. CONCLUSIONS This will be the first Canadian app to combine carbohydrate counting and insulin dosing while involving end users in its development.