Gestational diabetes & diabetes in pregnancy
Prediction of time to insulin initiation in gestational diabetes mellitus: a secondary analysis of the EMERGE trial
Authors' conclusion
The RSF model effectively identified women at high risk of requiring insulin. The decision curve analysis could help clinicians to balance insulin initiation decisions. However, prospective validation is needed to confirm the generalisability of the model.
Study context
- Population
- Women with gestational diabetes mellitus (GDM) from the EMERGE trial; predictors: maternal characteristics and early glucose data (2 weeks after randomisation)
- Participants
- 413 women
- Comparison
- Random survival forest (RSF) prediction model, analysed separately in placebo and metformin groups
- Primary endpoint
- Prediction of time to insulin initiation
Methodological detailsrandom survival forest prediction model, decision curve analysis
- Random survival forest prediction model
- Decision curve analysis
Primary result
As reported in the abstract
RSF model performance for time to insulin initiation
| Group | Concordance index (C-index) | Time-dependent AUC | Brier score |
|---|---|---|---|
| Placebo | 0.71 (95 % CI: 0.64-0.77) | ≥0.70 | ≤0.2 |
| Metformin | 0.72 (95 % CI: 0.64-0.80) | ≥0.75 | ≤0.2 |
Concordance index (C-index)
Placebo0.71 (95 % CI: 0.64-0.77)
Metformin0.72 (95 % CI: 0.64-0.80)
Time-dependent AUC
Placebo≥0.70
Metformin≥0.75
Brier score
Placebo≤0.2
Metformin≤0.2
Further findings
Decision curve analysis: net benefit
- RSF provided a higher net benefit for both groups vs default strategies across clinically relevant threshold probabilities
Original publication
Prediction of time to insulin initiation in gestational diabetes mellitus: a secondary analysis of the EMERGE trial. Diabetes Res Clin Pract. Published 21 December 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →