Abstract: Objective: To construct an early warning evaluation model for Postpartum Hemorrhage and verify its validity. Methods: A retrospective analysis of 6790 maternal cases in a tertiary hospital in Beijing was conducted to verify the validity of the Postpartum Hemorrhage warning evaluation model. Results: The area under ROC curve of the evaluation model is 0.632 [95%CI(0.613, 0.651)]. The optimal cutoff value is 5 points, and the corresponding sensitivity is 51.9% [95%CI (48.7%, 55.1%)], the specificity was 66.7% [95%CI (65.5%, 67.9%)]. Conclusion: The early warning model of Postpartum Hemorrhage has certain accuracy, which provides a model for clinical medical staff to predict the occurrence of Postpartum Hemorrhage. Early warning and interventions of Postpartum Hemorrhage should be taken to reduce the occurrence.

Key words: Postpartum Hemorrhage; risk factors; early warning evaluation model; validity; nursing evaluation