Impact of Omnipod 5 automated insulin delivery on continuous glucose monitoring metrics and predictors of improvement in time in range

Abstract

Aims

This study aimed to assess the impact of the Omnipod 5 automated insulin delivery (AID) system on continuous glucose monitoring (CGM) metrics, HbA1c, and weight in a real-world setting. Additionally, independent predictors of glycaemic response were assessed.

Methods

Observational analysis of adults with type 1 diabetes using Omnipod 5 (n = 353). Paired data on CGM metrics (n = 268), HbA1c (n = 193), and weight (n = 173) were collected at baseline and compared after median of 191, 120, and 221 days, respectively. Independent predictors of TIR response (≥5%) and HbA1c (≥5 mmol/mol) were assessed.

Results

Omnipod 5 use was associated with improved TIR (+16%, p < 0.001) and a reduction in HbA1c (−3 mmol/mol, p < 0.001). The greatest improvements (−7 mmol/mol, p < 0.001) were observed in individuals with elevated baseline HbA1c (≥58 mmol/mol). Sensor choice (Dexcom G6 vs. Freestyle Libre 2 Plus) influenced time in full auto mode (94% vs. 96%, p < 0.001) but did not affect the likelihood of improved TIR or HbA1c. Logistic regression identified baseline HbA1c (OR 1.24 per mmol/mol, p < 0.001) as the main association with improved HbA1c. Similarly, baseline TIR was associated with improvement in TIR (OR 0.83 per %, p < 0.001). Greater time in automation and using the lowest glucose target were also associated with improved outcomes.

Conclusions

Omnipod 5 is associated with significant and sustained improvements in CGM metrics and HbA1c, particularly in individuals with higher baseline HbA1c. The results suggest the potential benefits of prioritizing AID for individuals at greatest risk of complications.