External Validation of a Machine Learning Model to Predict Postpartum Hemorrhage in a US Northeastern Healthcare System

Introduction

Postpartum hemorrhage (PPH) is a major cause of maternal morbidity and mortality. Timely prediction may prevent adverse maternal outcomes, and efforts are needed to develop accurate predictive tools. A high-performing machine learning model to predict PPH using data from the US Consortium for Safe Labor (CSL) remains to be widely validated in contemporary clinical settings using electronic health record (EHR) data. Our goal was to evaluate the performance of the CSL PPH predictive model using EHR data across a large healthcare system in the Northeastern US.

Methods

We conducted a retrospective cohort study across eight hospitals in the Northeast US between 05/2015 and 05/2024. We used the same sociodemographic, clinical diagnoses, family history, laboratory, and vital signs available on labor and delivery admission in the EHR that were used to train the original CSL model. The binary outcome was PPH, defined as estimated blood loss of 1000 ml or more at delivery or blood transfusion within 24 hours postpartum. We then refit a new model using the original features to assess whether model performance could be further improved in our study population using the best-performing machine learning approach (XGBoost) from the original CSL model. We evaluated model discrimination as measured using the area under the curve (AUC), feature importance, calibration, and decision analysis curves of both the original CSL model with external validation and the further refit model.

Results

Among 87,662 deliveries, the incidence of PPH was 7.7%. The original CSL model demonstrated modest discrimination for predicting PPH with an AUC of 0.60 (95% CI, 0.58– 0.61). Refitting a new model with XGBoost resulted in improved discrimination with an AUC of 0.75 (95% CI, 0.74–0.76). Calibration analyses demonstrated that the refit model overestimated PPH risk across a range of predicted probabilities.

Conclusion

A promising PPH predictive model had substantially reduced performance with external validation using contemporary EHR data across an eight-hospital health system in the Northeastern US. These findings highlight the importance of external validation, local adaptation, and ongoing surveillance for assessing model performance in an era of evolving prevention, management, and treatment strategies for PPH.

Key Points

• This study aimed to externally validate a previously published machine learning model for predicting postpartum hemorrhage (PPH) and assess its portability across eight hospitals using electronic health record data on labor and delivery admission.

• We found that the original model demonstrated modest discrimination (area under the curve, AUC: 0.60) with external validation.

• A refit model achieved improved discrimination (AUC: 0.75) but remained poorly calibrated and overestimated the risk of PPH across a range of predicted probabilities.

• These findings underscore the importance of local validation and adaptation of external models, and ongoing performance monitoring before clinical deployment of PPH prediction models in an era of evolving prevention, management, and treatment strategies for PPH.