997-P: Empowering CGM Users with Multi-horizon Glucose Forecasting Capabilities—The Accu-Chek SmartGuide Predict App

Introduction & Objective: Despite the significant increase in the utilization of CGM technology, the prevalence of diabetes distress and fear of hypoglycemia remains high, and many users continue to face challenges in achieving their glycemic targets. To tackle this issue, we present scientific data on the Accu-Chek SmartGuide Predict app, a CGM digital companion that provides glucose forecasts and notifies users about upcoming glycemic events.

Methods:The app's functionalities, powered by three machine learning-enabled algorithms, include a 2-hour glucose forecast, a low glucose detection within 30 min, and nighttime low glucose prediction for bedtime interventions. The evaluation of the algorithms encompassed three datasets, which included people with T1D on MDI (n=21) and pump therapy (n=201), as well as real-world data from people with T2D (n=59). In total, the evaluation covered ~73K subject days.The development of the app relied on collecting objective user feedback through usability studies.

Results: Across the three datasets, the 2-hour forecast algorithm, at a prediction horizon of 30, 60, and 120 min, achieved a Consensus Error Grid A&B of 99.8%, 98.8%, and 96.6%, respectively. The low glucose detection algorithm achieved a sensitivity, specificity, and mean lead time of 94.5%, 97.5% and 19.3 min, respectively. The nighttime low glucose detection algorithm achieved an accuracy, sensitivity, and specificity of 89.1%, 53.1%, and 94.8%, respectively.Usability studies showed understanding of the app's functionalities and willingness to use its features. Task completion rate was 98.1%, indicating high success.

Conclusion: The consistent performance of the predictive algorithms on different cohorts offers reassurance for real-life efficacy. Usability studies suggest high user satisfaction, engagement, and retention with the app. However, verification is needed once the app is deployed and used over an extended duration.