Disease's type
GDM
Experimental grouping
GDM(n=264),Normal Glucose Tolerance(n=528)
GPT's summary
This study aimed to develop a predictive model for early identification of gestational diabetes mellitus (GDM) based on clinical factors and biomarkers. Using data from a cohort of 7929 pregnant women, a nested case-control analysis was conducted, including 264 women with GDM and matched controls. HbA1c and hsCRP were elevated, while SHBG was reduced in women who developed GDM (p<0.001). The predictive model for GDM, incorporating HbA1c, SHBG, BMI, history of GDM, family history of diabetes, and pre-pregnancy soft drink intake, achieved an AUC of 0.79. For predicting GDM requiring insulin therapy, the model had an AUC of 0.88 and a sensitivity of 68.9% at a 10% false-positive rate. This simple model could enable early identification of women at risk of GDM, particularly those requiring insulin therapy, facilitating timely interventions.
RF's name
Hemoglobin A1c
Sample's type
Whole Blood
Gestational weeks
14th to 17th gestational weeks
Experiemental methods
Urbidimetric Inhibition Immunoassay
Machine learning algorithms
Logistic Regression
Title
Early prediction of gestational diabetes: a practical model combining clinical and biochemical markers
Evidence's type
Risk factor
Year
2016
Journal
Clinical Chemistry And Laboratory Medicine
PMID