Progression from Prediabetes to Diabetes in a Diverse U.S. Population: A Machine Learning Model
Authors' conclusion
Leveraging readily obtainable laboratory data, our ML risk classifier accurately predicts elevation in A1C associated with progression from prediabetes to diabetes.
Study context
- Population
- Adults with initial HbA1c 5.0%–6.4%, laboratory data from a large U.S. outpatient laboratory network, over a 5-year period
- Participants
- 15,029 adults
- Outcome
- Progression to HbA1c ≥6.5% (indicative of diabetes)
- Follow-up
- Within 5 years
Methodological detailsmachine learning prediction model, random forest survival methods
- Machine learning prediction model
- Random forest survival methods
Main finding
As reported in the abstract
Prediction of A1C (A1C) ≥6.5% within 5 years
| Group | AUC |
|---|---|
| Prediabetes risk classifier model | 0.87 |
Prediabetes risk classifier model
AUC0.87
Observational data: associations, not effects.
Further findings
Most important predictors of progression
- Initial HbA1c, initial serum glucose, HbA1c slope, serum glucose slope, initial HDL, HDL slope, age and sex
Original publication
Progression from Prediabetes to Diabetes in a Diverse U.S. Population: A Machine Learning Model. Diabetes Technol Ther. Published 24 April 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →