Skip to main content

Table 2 The correlations between the confounding variables and live birth probabilities in this dataset

From: Associations between the artificial intelligence scoring system and live birth outcomes in preimplantation genetic testing for aneuploidy cycles

Variables

Univariate

Multivariate

OR

95% CI

p

aOR

95% CI

p

Lower

Upper

Lower

Upper

Female age

1.005

0.971

1.040

0.794

AMH

0.997

0.955

1.040

0.878

BMI

1.014

0.963

1.067

0.599

Oocyte sources

(autologous vs. donor*)

0.867

0.511

1.473

0.598

Endometrial preparation protocols

(artificial vs. natural or modified natural *)

0.856

0.599

1.221

0.390

Endometrial thickness

1.047

0.961

1.141

0.293

Pulsatility index level

(< 3 vs. ≥ 3*)

1.389

0.839

2.300

0.202

1.625

0.943

2.800

0.081

Serum E2 level on the ET day

1.000

1.000

1.000

0.642

Serum P4 level on the ET day

1.000

0.994

1.006

0.983

Ploidy status

(euploidy vs. mosaicism*)

1.136

0.741

1.742

0.558

Abnormal chromosome site numbers

(0 vs. > 2*)

1.967

0.865

4.475

0.107

Abnormal chromosome site numbers

(1 vs. > 2*)

2.711

1.059

6.944

0.038

Abnormal chromosome site numbers

(2 vs. > 2*)

0.783

0.241

2.551

0.685

Types of chromosomal abnormalities

(none vs. whole*)

8.500

1.855

38.946

0.006

6.632

1.382

31.817

0.018

Types of chromosomal abnormalities (segmental vs. whole*)

9.308

1.937

44.728

0.005

9.206

1.817

46.649

0.007

AI scores of iDAScore

2.002

1.607

2.495

< 0.001

2.037

1.632

2.542

< 0.001

  1. The generalized estimating equation (GEE) analysis was used for statistical analysis. The abbreviations “OR”, “aOR”,“CI”, “p”, “AMH”, “BMI”, “E2”, “P4”, and “AI” denoted odds ratio, adjusted odds ratio, confidence interval, p -value, anti-mullerian hormone, body mass index, estradiol, progesterone, and artificial intelligence, respectively. *Indicated of a reference group in the GEE model. The backward stepwise selection was applied to identify the confounders (P < 0.2)