Table 2

Model calibration

ModelCalibration measures*APO predicted risk†
Brier‡Reliability‡P value§1 (low)234 (high)
Regression modelsActual APO rate*
 Stepwise-selection (LR-S)0.140.013<0.0010.110.350.380.64
 Penalised (LASSO)0.130.0070.130.110.400.390.74
Neural networks (NN)
 One hidden layer (NN-1)0.160.033<0.0010.120.270.400.49
 Two hidden layers (NN-2)0.180.054<0.0010.110.290.320.43
 Random forest (RF)0.130.0040.550.090.350.471.00¶
 Gradient boosting (GB)0.140.0090.190.130.330.380.83
Support vector machine (SVM)
 SuperLearner (SL)0.120.0030.820.090.400.600.75
  • *Average across five independent, 10-fold cross-validations.

  • †APO predicted risk: 1: ≤25%, 2: 26%–50%, 3: 51%–75%, 4: >75%.

  • ‡Agreement between predicted and observed APO; low scores indicate better agreement.

  • §P<0.05 indicates lack of fit using Spiegelhalter goodness-of-fit test.

  • ¶Only one individual (who experienced an APO) had a prediction >75%.

  • APO, adverse pregnancy outcome; LASSO, least absolute shrinkage and selection operator; LR-S, logistic regression wih stepwise selection.