Objective To establish and verify the a prediction model for the influencing factors of postpartum hemorrhage (PPH) in elderly parturient, and provide a scientific basis for early identification and intervention of PPH.

Methods A total of 470 elderly parturients were selected as the research subjects, and divided into the training set (329 cases) and validation set (141 cases) in the optimal ratio of 7∶3. The important variables affecting PPH in elderly parturients were selected based on the minimum absolute contraction, selection operator (LASSO) regression and multivariate logistic screening, and a nomogram prediction model was constructed. The predictive ability and clinical value of the model were verified by the receiver operating characteristic (ROC) curve, decision analysis curve (DCA), and calibration curve.

Results Seven variables affecting postpartum hemorrhage in elderly parturients were screened by LASSO regression and multivariate logistic regression: age, placenta previa, placental abruption, placental implantation, gestational hypertension, macrosomia and D-D. The nomogram prediction model of PPH was successfully constructed based on seven variables. The AUC values of the nomogram in the training set and the validation set were 0.899 (95%CI = 0.850–0.947, P ＜ 0.05) and 0.903 ( 95%CI = 0.850–0.957, P ＜ 0.05), respectively. The goodness-of-fit of the calibration curve was consisted with Hosmer Lemeshow, which indicated that the model had a relatively high calibration degree (χ² = 8.09, P ＞ 0.05). The results of the clinical decision curve in the training set and test set showed that the prediction model hds a good clinical net rate of return.

Conclusions The age, placenta previa, placental abruption, placental implantation, gestational hypertension, macrosomia and D-D are important variables affecting PPH in elderly parturients, which can provide a scientific basis for early prediction and intervention of PPH in elderly parturients.