Related Experiment Video
Updated: Aug 6, 2026
12:08
Improving IV Insulin Administration in a Community Hospital
Published on: June 11, 2012
Retrospective Validation of a Stage 5 Decision Support Algorithm Using Clinical Trial Data From Stage 4 Smart Insulin
Jee Hee Yoo1, Seohyun Kim2,3, Younghoon Kim4
1Division of Endocrinology and Metabolism, Department of Internal Medicine, Chung-Ang University Gwangmyeong Hospital, Chung-Ang University College of Medicine, Republic of Korea.
Journal of Diabetes Science and Technology
|August 3, 2026
Summary
Algorithm-generated coaching events from a smart insulin pen (SIP) decision support system (DSS) effectively indicate glycemic control in individuals with diabetes on multiple daily injections (MDI). Higher coaching frequencies correlate with poorer glycemic outcomes, validating the DSS utility.
Area of Science:
- Diabetes Technology
- Artificial Intelligence in Medicine
- Glycemic Control Management
Background:
- Evaluating the relationship between algorithm-generated coaching events from a decision support system (DSS) integrated with a smart insulin pen (SIP) and glycemic outcomes.
- Establishing the algorithmic foundation for DSS coaching in diabetes management.
Purpose of the Study:
- To determine if the frequency of algorithm-generated coaching events differs based on glycemic outcomes.
- To validate the utility of a SIP-integrated DSS for identifying glycemic issues in individuals using multiple daily injections (MDI).
Main Methods:
- Retrospective analysis of continuous glucose monitoring (CGM) and SIP data from a 12-week randomized controlled trial (NCT06406439).
- Analysis of 338 biweekly CGM profiles from 58 participants with diabetes on MDI.
- Calculation of coaching event frequency across 9 categories for each 14-day period.
Main Results:
- Hyperglycemia and missed bolus coaching frequencies were significantly higher in participants not meeting glycemic targets (P < .001).
- Coaching categories demonstrated good discrimination for time in range (TIR) and time above range (TAR) achievement (AUC > 0.7).
- Hypoglycemia coaching effectively identified absence of nocturnal hypoglycemia, and basal insulin titration coaching correlated with nocturnal glucose changes.
Conclusions:
- Algorithm-based coaching event frequency is a robust indicator of glycemic control.
- Findings validate the Stage 5 SIP-integrated DSS.
- The DSS shows potential utility in identifying unresolved glycemic issues for individuals on MDI therapy.