Table 2.

Univariate and multivariate logistic regression analysis of variables studied for the prediction of clinical pregnancy in patients with low AMH and area under the receiver operating characteristic curves (AUC-ROC) analysis.

Variable	R 2	P value	AUC-ROC	95% CI
Univariate continuous models
DHEAS	0.466	0.004	0.726	0.641–0.801
Age	−0.173	0.031	0.662	0.573–0.743
AFC	0.115	0.104	0.619	0.529–0.703
AMH	0.172	0.260	0.592	0.502–0.678
BMI	−0.038	0.626	0.502	0.413–0.592
E 2	0.086	0.969	0.518	0.428–0.607
FSH	−0.020	0.661	0.525	0.435–0.614
LH	−0.024	0.848	0.525	0.435–0.613
T	0.473	0.234	0.587	0.497–0.673
PRL	−0.008	0.843	0.501	0.412–0.590
TSH	−0.044	0.747	0.544	0.454–0.632
Univariate categorical model
Age-c	−1.907	0.015	0.688	0.600–0.766
Multivariate categorical model
DHEAS-age	−1.167	0.001	0.796	0.716–0.862

Note: R ²: regression coefficient; P: P value for R ²; CI: confidence interval; AFC: antral follicle count; AMH: anti-M$\ddot{\text{u}}$llerian hormone; BMI: body mass index; age-c: univariate categorical model using age as categorical variable with cut-off value of 37.5 years; DHEAS-age: multivariate categorical model using DHEAS as categorical variable with cut-off value of 5.7 μmol/L and age as categorical variable with cut-off value of 37.5 years.