Table 4 Performance of prediction models with maternal polygenic risk score, baseline characteristics, and blood pressure in early pregnancy in the maternal cohorts.

From: Preeclampsia prediction with maternal and paternal polygenic risk scores: the TMM BirThree Cohort Study

 

No PRS

M-PRS

C-statistics

  

Apparent

  

 Reference model

0.561 (0.531 to 0.591)

0.579 (0.545 to 0.613),

P-value = 0.171

 Family-history model

0.564 (0.534 to 0.594)

0.579 (0.545 to 0.613),

P-value = 0.224

 At-pregnancy-confirmation model

0.736 (0.702 to 0.769)

0.742 (0.709 to 0.773),

P-value = 0.301

 In-early-pregnancy model

0.782 (0.751 to 0.810)

0.783 (0.753 to 0.812),

P-value = 0.514

Internal validation

  

 Reference model

0.561 (0.561 to 0.561)

0.570 (0.554 to 0.579),

P-value = 0.214

 Family-history model

0.564 (0.559 to 0.565)

0.571 (0.556 to 0.580),

P-value = 0.252

 At-pregnancy-confirmation model

0.730 (0.715 to 0.737)

0.729 (0.713 to 0.739),

P-value = 0.986

 In-early-pregnancy model

0.777 (0.765 to 0.782)

0.776 (0.764 to 0.782),

P-value = 0.86

External validation

  

 Reference model

0.577 (0.543 to 0.612)

0.592 (0.551 to 0.632),

P-value = 0.317

 Family-history model

0.572 (0.538 to 0.607)

0.585 (0.544 to 0.625),

P-value = 0.352

 At-pregnancy-confirmation model

0.777 (0.741 to 0.811)

0.784 (0.750 to 0.818),

P-value = 0.242

 In-early-pregnancy model

0.849 (0.820 to 0.875)

0.851 (0.821 to 0.876),

P-value = 0.520

Calibration slope

  

Apparent

  

 Reference model

0.998 (0.529 to 1.424)

1.000 (0.597 to 1.371)

 Family-history model

0.996 (0.551 to 1.406)

0.997 (0.610 to 1.358)

 At-pregnancy-confirmation model

0.999 (0.894 to 1.111)

0.999 (0.895 to 1.109)

 In-early-pregnancy model

0.999 (0.903 to 1.101)

0.999 (0.903 to 1.101)

Internal validation

  

 Reference model

1.002 (0.701 to 1.883)

0.834 (0.619 to 1.246)

 Family-history model

0.964 (0.694 to 1.684)

0.835 (0.631 to 1.212)

 At-pregnancy-confirmation model

0.953 (0.717 to 1.071)

0.942 (0.723 to 1.056)

 In-early-pregnancy model

0.956 (0.793 to 1.067)

0.946 (0.784 to 1.054)

External validation

  

 Reference model

1.129 (0.613 to 1.598)

1.116 (0.645 to 1.541)

 Family-history model

0.895 (0.421 to 1.324)

0.924 (0.492 to 1.317)

 At-pregnancy-confirmation model

1.198 (1.090 to 1.309)

1.188 (1.081 to 1.298)

 In-early-pregnancy model

1.177 (1.074 to 1.284)

1.156 (1.053 to 1.263)

  1. PRS: polygenic risk score. “No PRS” indicates models without PRS and “M-PRS” indicates models with maternal PRS for diastolic blood pressure. Reference models included maternal age. Family-history models included maternal family history of hypertensive disorders of pregnancy and maternal age. At-pregnancy-confirmation models included variables available at pregnancy confirmation: maternal age, pre-pregnancy body mass index, chronic hypertension, systemic lupus erythematosus, diabetes mellites, maternal family history of hypertensive disorders of pregnancy, conception via in vitro fertilization, parity, gestational age at the previous delivery, and the inter-birth interval. At-early-pregnancy models added blood pressure in early pregnancy to at-pregnancy-confirmation models. Models with M-PRS also included maternal four genetic principal components. Model training and internal validation were conducted using the internal validation cohort. Apparent performance, which reflects the raw performance of the model trained on the internal validation cohort and may be overestimated, was also reported. External validation was conducted using the external validation cohort. P-values and 95% confidence intervals were obtained by the bootstrap method with 5,000 resamplings.
Back to article page