843-P: Multinational Analysis of Factors Associated with Missed Bolus Insulin Injections Using Smart Pen Data

Introduction & Objective: Missing doses in people living with diabetes (PLWD) who use multiple daily insulin injections may lead to inadequate glycemic control. This real-world study assessed the probability of and influencing factors for missed bolus insulin injections in PLWD.

Methods: Data from 21 countries were collected from 22 964 adults with T1D or T2D using CGM and a smart pen (NovoPen 6 or NovoPen Echo Plus) to deliver bolus insulin (2 559 118 CGM and injection days [111 days/adult]). The daily probability of missing ≥ 1 bolus dose (no bolus dose within −15 to +60 mins from meal onset detected by the GRID algorithm) was estimated by a generalized linear mixed model based on logistic regression. Data upload frequency (pen to app) was used as a proxy for smart pen engagement.

Results: The mean probability of having a day with ≥ 1 missed bolus dose was 44.5% (95% CI: 44.1-45.0%), which decreased with increasing age (Figure A) and smart pen data upload frequency (not shown). The probability of having a day with ≥ 1 missed bolus dose was highest on Fridays and Saturdays (Figure B). On any given day, the probability of missing a bolus dose at breakfast, lunch, dinner or snack was 7.5%, 12.2%, 11.1% and 19.4%, respectively.

Conclusion: Bolus insulin adherence increased with age and smart pen engagement and decreased on weekends. Smart insulin pen data can provide novel insights into individuals’ real-world injection behaviors.

View largeDownload slide

View largeDownload slide

Close modal