Abstract: Examining the Glycemia Risk Index and Its Association with Continuous Glucose Monitor (CGM)-Derived Glycemic Risk Categories in Patients with Type 1 Diabetes
J. Litwin1, B. Lockee1, C. Vandervelden1, M. Barnes1, R. Mcdonough2, S. Patton3, D. Williams4, K. Noland1,M. Clements5
1Children’s Mercy Hospital, Endocrinology, Kansas City, United States of America, 2Children’s Mercy-Kansas City, Endocrinology, Kansas City, United States of America, 3Nemour’s Children’s Health, Center for Healthcare Delivery Science, Jacksonville, United States of America, 4Children’s Mercy Hospital, Health Services And Outcomes Research, Kansas City, United States of America, 5Children’s Mercy Hospital, Endocrinology And Diabetes, Kansas City, United States of America
Background and Aims: Previous work has prioritized patient outreach by analyzing weekly CGM data and assigning threshold-based risk categories [Ferstad et al., Ped Diab, 2021; 22(7): 982-991]. To further explore the use of Glycemia Risk Index (GRI) as a biomarker, this study examined the relationship between GRI and three CGM risk categories: time in range (TIR) <65%, %time extremely hypoglycemic (<54 mg/dL) >1%, and %time extremely hyperglycemic (>250 mg/dL) >4%.
Methods: We used retrospective weekly CGM data from 2016-2022 collected by a Midwest USA pediatric diabetes center. GRI was calculated as (3 · %measures <54 mg/dL) + (2.4 ·%measures 54-70mg/dL) + (1.6 · %measures >250 mg/dL) + (0.8 ·%measures 170-250mg/dL). We ran two-sample t-tests to compare differences in mean GRI between youth who were/were not in each risk category.
Results: We calculated the mean GRI for the population within and outside each risk category. There were GRI score differences between qualifiers and non-qualifiers in the TIR <65% and severe hyperglycemia categories (p < .001); however, no difference was found for the clinically significant hypoglycemia risk category (p = .59).
Conclusions: Preliminary findings suggest that GRI may be useful as a risk assessment tool for diabetes management. Future testing should compare individual components (high and low) of the GRI in relation to these risk categories and explore changes to the category thresholds, such as applying a threshold of extreme lows over two percent or relating the GRI low component to the risk category for >4% time low.