To Örebro University

Background and Aims: Smart pen injection data can provide unique insight into routine diabetes treatment. NovoPen 6 users may consent to share their injection and continuous glucose monitoring (CGM) data anonymously for research purposes. Here, based on real-world data from people with type 1 diabetes (T1D), we explored the association between the number of daily bolus doses, and smart pen engagement, with time-in-range (TIR).

Methods: We included adults with T1D in Sweden administering Fiasp with a NovoPen 6 device for days with bolus injections and CGM data (‡70% coverage). CGM parameters were used as glycaemic control measures. Smart pen engagement was characterised by the number of injection data uploads over the previous 14 days. A linear mixed model was used to determine relationship between TIR and covariates, adjusted for known confounders.

Results: Overall, data from 224 patients were analysed. The number of daily bolus doses was significantly associated with improved TIR (p <0.0001). Patients with an average£3 daily bolus had <10% chance of reaching the target TIR >70% (Fig-ure). The number of uploads was also significantly associated with improved TIR (p <0.0001). Days where uploads were conducted daily over the previous 14 days had 5% greater TIR than days without any uploads during the previous 14 days.

Conclusions: Daily bolus dose number and smart pen engagement are strong predictors of TIR. Most patients take too few bolus doses to reach treatment targets.