Development and validation of prediction models for gestational diabetes treatment modality using supervised machine learning: a population-based cohort study

Table 3 Prediction results using final simplified logistic regression models with predictors at varied stages of pregnancy

	Cross-validated AUC (95% CI)		Integrated calibration index	Calibrated AUC (95% CI)
	Discovery set	Validation set	Integrated calibration index	Calibrated AUC (95% CI)
Level 1^a	0.632 (0.623–0.640)	0.609 (0.587–0.632)	0.073	0.609 (0.587–0.632)
Levels 1–2^b	0.648 (0.640–0.656)	0.621 (0.599–0.643)	0.075	0.621 (0.599–0.643)
Levels 1–3^c	0.770 (0.764–0.775)	0.746 (0.730–0.763)	0.072	0.752 (0.734–0.77)
Levels 1–4^d	0.825 (0.820–0.830)	0.798 (0.783–0.813)	0.038	0.802 (0.786–0.818)

AUC, area under the receiver operating characteristic curve; CI, confidence interval; GDM, gestational diabetes
^aPredictors included history of GDM, pre-pregnancy obesity, and prediabetes before pregnancy
^bPredictors included history of GDM, pre-pregnancy obesity, glucose levels at 50-g, 1-h glucose challenge test for GDM screening (≥ 200 mg/dL), and prediabetes before pregnancy, in addition to three pairwise interactions between the first three predictors
^cPredictors included fasting glucose value at 100-g, 3-h oral glucose tolerance test, gestational week at GDM diagnosis (continuous), and GDM diagnosis by Carpenter-Coustan criteria (versus by fasting hyperglycemia)
^dPredictors included gestational week at GDM diagnosis (continuous), fasting glucose value at 100-g, 3-h oral glucose tolerance test, self-monitored glycemic control status at fasting, number of fasting self-monitored blood glucose measurements, and an interaction term between last two variables

ISSN: 1741-7015