Retrospective Real-world Analysis of Insulin Pump and Continuous Glucose Monitoring Technology Utilization in a Large...

We performed a retrospective, real-world analysis of a large outpatient adult type 1 diabetes clinic to assess the patterns of use of diabetes technologies and their impact on glycemic metrics.

Data from the most recent clinic visit between January 2024 and April 2025 were used to capture diabetes technologies used, glycated hemoglobin (A1C), continuous glucose monitoring (CGM) metrics, and incidence of diabetic ketoacidosis and level 3 hypoglycemia during the prior year.

A total of 794 people were identified: 56.7% were females, the mean age was 45.6 years (standard deviation [SD]=16.8), the mean diabetes duration was 25.5 years (SD=14.9), basal bolus insulin (BBI) use was 24%, and pump use was 76%. CGM was used by 81.2% of BBI users and 93.4% of pump users. The mean pump duration was 11.3 years (SD=6.9). Automated insulin delivery (AID) was used by 58.7% of pump users: 78% used Tandem Control IQ, 19.2% used Medtronic SmartGuard, and 2.8% used Omnipod with Do-It-Yourself looping. Level 3 hypoglycemia/diabetic ketoacidosis rates were low and insufficient for comparison. Of pump users using CGM, 35.2% were using a CGM not compatible with their pump AID. A1C level did not differ between BBI users without CGM and pump users without CGM or between BBI users with CGM and open-loop pump users with CGM. The use of AID was associated with lower A1C level, more time in range, less time below range and time above range, and less glycemic variability.

In this large, real-world Canadian study, the use of AID systems was associated with significantly better glycemic management compared with both BBI and open-loop insulin pump therapy.