Gestational diabetes & diabetes in pregnancy
Linking Metabolic Subgroups and Adverse Pregnancy Outcomes in Gestational Diabetes Mellitus (GDM) : A Complementary Analytic Approach of Logistic Regressions and Machine Learning
Authors' conclusion
This study confirms the metabolic heterogeneity in GDM women, supporting synergistic effects among metabolic factors and the need for integrated metabolic management.
Study context
- Population
- Women with GDM, medical records 2018–2023, Central South China; subgroups by oral glucose tolerance test (OGTT) 0, 1 and 2 h, uric acid, triglycerides (TG), high-density lipoprotein (HDL) and low-density lipoprotein (LDL)
- Participants
- 2,246 women
- Exposure
- Metabolic subgroups of GDM vs reference subgroup
- Outcome
- Adverse pregnancy outcomes by metabolic subgroup
Methodological detailsunsupervised k-means clustering, logistic regression and machine learning (SHAP)
- Unsupervised k-means clustering
- Logistic regression and machine learning (SHAP)
Main finding
As reported in the abstract
Preterm delivery
Each group vs referenceOdds ratio (95% CI)
Hyperuricemia and hypertriglyceridemia
1.86(1.15–3.01)
Combined metabolic dysregulation
3.45(1.79–6.65)
Observational data: associations, not effects.
Further findings
| Outcome | OR (95% CI) | |
|---|---|---|
| Insulin treatmentCombined metabolic dysregulationvs Reference | 14.67 (8.72–24.66) | |
| Hypertensive disorders of pregnancyCombined metabolic dysregulationvs Reference | 2.72 (1.48–5.01) | |
| 0.050.21520Odds ratio (log scale) |
SHAP feature importance
- Subgroup-defining biomarkers ranked among top 15 predictive features
- 57%-71%
Original publication
Linking Metabolic Subgroups and Adverse Pregnancy Outcomes in GDM : A Complementary Analytic Approach of Logistic Regressions and Machine Learning. Diabetes Obes Metab. Published 18 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →