Bishop score: A poor diagnostic test to predict failed induction versus vaginal delivery

Background. We evaluated the accuracy of the Bishop score in predicting the likelihood of successful labor induction (entry into active phase) in nulliparous and multiparous women. Methods. During an index year, all patients having induction of labor and a preinduction Bishop score were included in a standard protocol for cervical ripening and use of oxytocin. Receiver-operating characteristic (ROC) curves were constructed for Bishop scores (0 to 11) to predict abdominal delivery for failed induction (final cervical dilation <4 cm) versus vaginal delivery. Results. Parturients who had vaginal delivery (n = 253) and those in whom attempted induction failed (n = 38) did not differ significantly with respect to maternal demographics, length of gestation, Bishop score and its distribution, and infant birth weight. The area under the ROC curve did not differ significantly from the area under the nondiagnostic line. Conclusion. The Bishop score appears to be a poor predictor of the outcome of labor induction.