Predictive modeling in hospital admission decision-making for obstetric emergency department patients at 18 weeks or more of pregnancy: A retrospective study.

BACKGROUND

This study developed a predictive model for hospital admission decision-making among obstetric emergency department (ED) patients at 18 weeks or more of pregnancy.

METHODS

The clinical data of 44,421 patients (22,211 in the training group and 22,210 in the validation group) at 18 weeks or more of pregnancy who visited the ED of a hospital in China in 2024, were included. Least absolute shrinkage and selection operator regression was used for the initial screening of 14 risk factors. Binary logistic regression was used to build a nomogram model.

RESULTS

Vaginal fluid leakage, gestational age ≥ 37 weeks, multiple pregnancy, and intrahepatic cholestasis of pregnancy significantly increased hospital admission risk; abnormal fetal movements and re-examination of fetal monitoring were protective factors. The area under the curve values of the nomogram model in the training and validation groups were 0.842 (95% confidence interval (CI): 0.837-0.848) and 0.830 (95% CI: 0.824-0.835), respectively. The calibration curve showed high consistency between predicted values and actual hospital admission rates. Brier scores in the training and validation groups were 0.147 and 0.154, respectively. Clinical decision curve analysis confirmed a significant net benefit when the threshold probability was >15%.

CONCLUSIONS

The predictive model facilitates rapid decision-making.