Gestational diabetes & diabetes in pregnancy
Early prediction of insulin requirement in gestational diabetes using a parsimonious LASSO model and a point-based risk score
Authors' conclusion
This LASSO-derived score using readily available variables accurately predicts insulin need in GDM, potentially facilitating risk-stratified management in community settings.
Study context
- Population
- Consecutive women with gestational diabetes mellitus (GDM); underweight (BMI < 18.5 kg/m²) and incomplete cases excluded from model development403 in derivation sample
- Participants
- 490 women
- Outcome
- Insulin therapy initiation based on routine clinical decisions
Methodological detailssingle-center, prediction model development and validation, LASSO logistic regression with 10-fold cross-validation
- Single-center
- Prediction model development and validation
- LASSO logistic regression with 10-fold cross-validation
Main finding
As reported in the abstract
Prediction of insulin therapy initiation (LASSO-derived point score)
- Insulin therapy initiated
- 11.0%
- Derivation sample receiving insulin
- 43 of 403
- Predictors selected: BMI category, ethnic risk background, chronic hypertension, medically assisted reproduction, number of abnormal OGTT values
- 5
- AUC
- 0.7755 (95% CI 0.695-0.853)
- Calibration slope
- 1.02
- Brier score
- 0.082
- Score ≥ 14 identified high-risk patients, LR +
- 11.7
Observational data: associations, not effects.
Original publication
Early prediction of insulin requirement in gestational diabetes using a parsimonious LASSO model and a point-based risk score. Diabetes Res Clin Pract. Published 28 March 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →