A predictive model for successfully inducing active labor among pregnant women: Combining cervical status assessment and clinical characteristics

Abstract

Objective: To develop a predictive model for successfully inducing active labor by using a combination of cervical status and maternal and fetal characteristics. Study design: A retrospective cohort study was conducted among pregnant women who underwent labor induction between January 2015 and December 2019. Successfully inducing active labor was defined as achieving a cervical dilation > 4 cm within 10 h after adequate uterine contractions. The medical data were extracted from the hospital database; statistical analyses were performed using a logistic regression model to identify the predictors associated with the successful induction of labor. The receiver operating characteristic (ROC) curve and area under the curve (AUC) were used to assess the accuracy of the model. Results: In total, 1448 pregnant women were enrolled; 960 (66.3 %) achieved successful induction of active labor. Multivariate analysis revealed that maternal age, parity, body mass index, oligohydramnios, premature rupture of membranes, fetal sex, dilation, station, and consistency were significant factors associated with successful labor induction. The ROC curve of the logistic regression model had an AUC of 0.7736. For the validated score system to predict the probability of success, we found that a total score > 60 has a 73.0 % (95 % CI 59.0–83.5) probability of successful induction of labor into the active phase stage within 10 h. Conclusions: The predictive model for successfully achieving active labor using the combination of cervical status and maternal and fetal characteristics had good predictive ability.

Keywords

• Pregnancy
• Active phase of labor
• Induction of labor
• Predictive model