A comparison of the usage of an open-source automated insulin delivery system and the MiniMed™ 780G system in children and adolescents with type 1 diabetes in real-world settings: the AWeSoMe Study Group

Abstract Purpose: In recent years there has been a noticeable increase in the use of advanced hybrid closed-loop systems (AHCLs) for managing type 1 diabetes (T1D) among youth. However, there is a lack of comparison between the open-source automated insulin delivery (AID) system and the MiniMed™ 780G system (780G). Methods: In this multi-center study, we retrospectively compared selected glycemic ranges of 26individuals who used open-source AID and 20 individuals who used 780G (age 11.3 years [IQR 9.3, 12.9] and 13.4 years [IQR10.9, 16.5], respectively, p = 0.069) from system initiation to the most recent visit. Results: At baseline, the median HbA1c was significantly lower and the TBR< 54mg/dL was significantly higher in the open-source AID group compared to the 780G group (6.8% [IQR 6.4, 7.1] vs. 7.4% [IQR 6.9, 8.6], p = 0.006 and (1.0% [IQR 0.5, 2.8] vs. 0.0 [0.0, 1.0], p = 0.014), respectively; the median time in range (TIR70-180mg/dL) was similar (p = 0.068). After a median duration of 10.9 months on AHCLs the reduction of HbA1c was similar (~ 0.3%). The time spent in the hypoglycemic ranges was longer among users of the open-source AID compared to 780G (TBR54-70mg/dL 4.2% [IQR 2.6, 7.3] vs. 2.0 [1.0, 4.0], p = 0.005) and TBR< 54mg/dL 1.1% [IQR 0.4, 2.3] vs. 0.0 [0.0, 1.0], p = 0.001). Conclusions: Both AHCLs similarly improved HbA1c and TIR70-180mg/dL. The open-source AID youth had better glycemic control but spent longer time in the hypoglycemic range. These findings must be considered when choosing the use of AHCL technologies.