Continuous Glucose Monitoring–Based Machine Learning Identification of Diurnal Glycemic Patterns and Diabetes Distress in Type 2 Diabetes
Authors' conclusion
CGM-based machine learning identified physiologically distinct glycemic phenotypes that were also associated with psychosocial burden. This work demonstrates the added value of integrating CGM-derived profiles with patient-reported outcomes. These findings highlight the potential of CGM phenotyping to support precision diabetes care by enabling early identification of high-risk subgroups, guiding tailored behavioral and psychosocial interventions, and informing technology-enabled decision tools that connect physiological monitoring with emotional well-being in T2D management.
Study context
- Population
- Adults with T2D wearing blinded continuous glucose monitoring (CGM); 1657 days of data1,657 CGM days
- Participants
- 137 adults
- Exposure
- CGM-derived glycemic clusters compared for diabetes distress
- Outcome
- Diabetes distress (17-item Diabetes Distress Scale) across CGM-derived diurnal glycemic patterns
Methodological detailsblinded CGM (FreeStyle Libre Pro), unsupervised machine learning (Gaussian mixture modeling), analysis of covariance (ANCOVA) adjusted for age, sex, BMI, diabetes duration and GMI
- Blinded CGM (FreeStyle Libre Pro)
- Unsupervised machine learning (Gaussian mixture modeling)
- ANCOVA adjusted for age, sex, BMI, diabetes duration and GMI
Main finding
As reported in the abstract
Cluster 3
2.37
mean · 95% CI 1.99–2.76
Cluster 4
1.67
mean · 95% CI 1.48–1.86
P = 0.03
Observational data: associations, not effects.
Original publication
Composed by AI from the published abstract and checked against it. How Briefs are made →