1065-OR: Negative Premeal NetIOB and Large Meals Are Associated with Higher Postprandial Excursions in AID Users with Type 1 Diabetes

Introduction and Objective: Many insulin-on-board (IOB) analyses overlook basal insulin delivery with automated insulin delivery (AID) systems. Net insulin on board (netIOB) integrates bolus insulin with basal deviations from the scheduled rate, providing a more accurate physiologic measure of insulin exposure at meal onset. Negative netIOB can occur when a pump is suspended or insulin delivery is reduced below scheduled basal. NetIOB has not been evaluated in relation to postprandial glycemic outcomes at scale. Methods: Using the T1DEXI dataset, meals (n=8,031) and CGM data were analyzed from 141 adults using AID. Meals were categorized by size based on grams of carbs (small, 0-21g; medium, 21-49g; or large, ≥50g) and pre-meal netIOB units of insulin (negative, <−0.5u; neutral, −0.5-0.5u; or positive, >0.5u). Primary outcomes were postprandial glucose excursions (peak minus pre-meal), time to peak glucose, and pre-defined adverse postprandial patterns (such as glucose spike >80 mg/dL; maximum postprandial glucose >250 mg/dL). Results: Negative pre-meal netIOB was associated with higher postprandial glucose excursions and less favorable glycemic outcomes than positive netIOB. The effect strengthened with meal size: for large meals, all pairwise comparisons of glucose spikes were significant (p<0.001). Medium meals showed a similar but attenuated pattern: negative vs positive (p<0.001) and negative vs neutral (p=0.027) were statistically significant; neutral vs positive was not (p=0.062). Small meal spike distributions were not significant across most netIOB categories (negative vs positive, p=0.405; negative vs neutral, p=0.169; neutral vs positive, p=0.049). Conclusion: Negative netIOB at meals is associated with larger postprandial glycemic excursions for AID users, particularly for medium and large meals. Incorporating netIOB rather than bolus-only IOB into commercial insulin delivery systems may support more physiologically grounded insulin dosing decisions and improve post-meal glycemic outcomes. Disclosure A. Zafar: None. A. Shahid: None. R.S. Kingman: None. D. Zaharieva: Speaker's Bureau; Ended; Dexcom, Inc., Medtronic. Speaker's Bureau; Current; Insulet Corporation. Research Support; Current; Leona M. and Harry B. Helmsley Charitable Trust. M. Riddell: Research Support; Ended; Dexcom, Inc. Speaker's Bureau; Current; Dexcom, Inc., Novo Nordisk. Consultant; Current; Jaeb Center for Health Research. Advisory Panel; Current; Zucara Therapeutics. R. Lal: Consultant; Current; Abbott Diabetes, Adaptyx Biosciences. Consultant; Ended; Biolinq, Capillary Biomedical, Deep Valley Labs. Consultant; Current; Gluroo, Portal Diabetes, Tidepool. Advisory Panel; Ended; ProventionBio, Lilly, Sanofi, Rezolute. R. Johari: None. R. Guo: Other - Summer intern; Ended; Apple. L. Jacobsohn: None. A. Cortes: None. L. Figg: None. A.A. Solanke: None. D.M. Lewis: None. Funding The Leona M. and Harry B. Helmsley Charitable Trust (Grant #: 2407-07175)