P-34: Validity of the mealtime BOLUS score in youth with type 1 diabetes using automated insulin delivery
S Patton, B Lockee, M ClementsAbstract
Introduction
Most contemporary automated insulin delivery (AID) systems primarily automate basal insulin delivery, while mealtime insulin dosing or meal announcements require user input. The BOLUS is an objective measure of mealtime insulin use previously validated in youth with type 1 diabetes (T1D) using open pumps, but its relevance in youth using AID remains unclear.
Objectives
To evaluate the construct validity of the mealtime BOLUS in youth with T1D using AID.
Methods
We analyzed baseline HbA1c, BOLUS scores, and youth demographics drawn from a randomized trial involving youth using AID. We examined construct validity using bivariate correlations, quartile-based analyses, and multivariable regression models predicting HbA1c.
Results
Among youth (n=118; [mean±SD] age: 13.7±1.9 years; T1D duration: 4.8±3.4 years) mean BOLUS score was 1.89±0.86 (possible range: 0-3). BOLUS scores were significantly negatively correlated with HbA1c (r=-.35, p<.001), with a trend-level association with age (r=-.15, p=.09). Quartile analyses indicated significant differences in HbA1c across BOLUS quartiles (p<.001). Post-hoc contrasts indicated that HbA1c was significantly higher in the lowest BOLUS quartile compared with the third and fourth quartiles, whereas differences between the first and second quartiles and between the third and fourth quartiles were not statistically significant. In a multivariable model, BOLUS and longer T1D duration were independent predictors of HbA1c, whereas age was not.
Conclusions
In youth using AID, BOLUS scores demonstrated meaningful variability, suggesting that AID does not eliminate individual differences in mealtime insulin use. Associations with HbA1c, independent of age and alongside T1D duration, support the construct validity of the BOLUS as an objective measure of mealtime insulin use in youth with T1D using AID.