From Stability to Variability: Classification of Healthy Individuals, Prediabetes, and Type 2 Diabetes Using Glycemic Variability Indices from Continuous Glucose Monitoring Data
Authors' conclusion
Our findings demonstrate a gradual deterioration of glucose homeostasis and increased glycemic variability across the spectrum from normo- to dysglycemia, as evidenced by CGM metrics. The performance of CGM-based indices in classifying healthy individuals and those with prediabetes and diabetes is promising.
Study context
- Population
- Healthy individuals and those with prediabetes or T2D, each with at least two days of continuous glucose monitoring (CGM), pooled from five studies
- Participants
- 282 / 133 / 432
- Comparison
- CGM indices across healthy, prediabetes and T2D
- Healthyn = 282reference
- Prediabetesn = 133
- T2DMn = 432
- Outcome
- Classification performance (ROC AUC) of XGBoost models on test data set
Methodological detailspooled data from five studies, 70/30 training/test split, XGBoost and logistic regression models
- Pooled data from five studies
- 70/30 training/test split
- XGBoost and logistic regression models
Pooled result
As reported in the abstract
- XGBoost AUC, prediabetes identification
- 0.91 [CI: 0.87-0.95]
- XGBoost AUC, dysglycemia identification
- 0.97 [CI: 0.95-0.98]
- Progressive shift across all CGM indices from healthy to diabetes
- P < 0.001
Further pooled results
| Outcome | P value |
|---|---|
| CGM indices, healthy vs prediabetes | P < 0.01 |
- CGM indices, healthy vs prediabetes: Significant differences in mean glucose, TBR, time above 140 mg/dl, mobility, multiscale complexity index and glycemic risk index
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
From Stability to Variability: Classification of Healthy Individuals, Prediabetes, and Type 2 Diabetes Using Glycemic Variability Indices from Continuous Glucose Monitoring Data. Diabetes Technol Ther. Published 26 August 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →