MON-589 - Battle of the Insulin Pumps: Six-Months Real-World comparison of Automated Insulin Delivery (AID) Systems

Automated insulin delivery (AID) systems using carb counting, sensitivity factor, insulin carb ratio presets for bolus dosing require more user input compared to weight input-initiated Automated Insulin Delivery system which use AI based adaptive algorithms to simplify mealtime bolusing and automate correction bolusing with no manual correction bolus option. We compared the two systems with regards to time in range (TIR), glucose management indicator (GMI), and HbA1c.

Material and

Methods:
Trends in GMI and TIR were assessed for 6 months after AID initiation in a community endocrine clinic. AID and CGM (continuous glucose monitoring) device data downloads were analyzed at 1 week, 4 weeks and 6 months. 6 patients on conventional AID and 7 patients on AID with autonomous bolusing option were studied. Independent and paired t- test analysis used.

Results:
Baseline characteristics including gender, type of diabetes (Type 1 or type 2) and age were comparable between the two groups. There was no significant difference in GMI at baseline (7.79 ± 1.34 vs. 8.22 ± 0.76, p value = 0.438), at 4 weeks (7.32 ± 0.75 vs. 7 ± 0.12, p value = 0.265) and at 6 months (7.12 ± 0.37 vs. 7.14 ± 0.36, p value = 0.9) between the conventional AID and AID with autonomous bolus dosing. In intragroup comparison, GMI showed no significant change in the conventional AID group at 4 weeks (p value = 0.164) and 6 months (p value = 0.596), whereas AID system with autonomous bolus dosing group showed a significant reduction from baseline (8.22 ± 0.76) to 4 weeks (7 ± 0.12, p value = 0.002) and 6 months (7.14 ± 0.36, p value = 0.025).

There was no significant difference between the two groups in terms of HbA1c at baseline (7.48 ± 1.02% vs. 8.34 ± 0.63%, p value = 0.061) and at 6 months (6.77 ± 0.57% vs. 6.81 ± 0.32%, p value = 0.854). On intragroup comparison, the conventional AID did not show any significant difference in HbA1c (p value = 0.080) while there was significant reduction in HbA1c from the baseline (8.34 ± 0.63%) to 6 months (6.81 ± 0.32%, p value = 0.0002) in the group on AID system with autonomous bolus dosing.

Conclusion:
In this real-world comparison, AID systems utilizing AI-driven autonomous correction bolusing demonstrated comparable improvement in glycemic outcomes with reduced user burden and less health care provider intervention compared to the AID systems requiring more user and provider input. This comparable efficacy of AID systems utilizing AI based autonomous / simplified bolusing has the potential to improve quality of life and increased acceptability of these AID systems. The small sample size limits the strength of these findings; larger, head-to-head real-world studies are needed to confirm these observations and draw more definitive conclusions.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*