Predictive validity of Bishop and Burnett Scores for vaginal delivery modified by parity

Abstract

Validating the predictive capacity on the outcome of labour of the Bishop Score (BS) and the simplified Burnett Score (BRS) compared to their modified versions, in which parity is incorporated. Historical cohorts out of a total of 728 inductions during the years 2012-2013 in the "La Mancha-Centro" Hospital of Alcázar de San Juan. We evaluated the predictive characteristics by areas under the (AUC) ROC curve for each parameter of BS and BRS and for parity, as well as for each of the 4 indices. Parity and all the parameters of BS and BRS, except for foetal station and cervical position, were associated with the outcome of labour. Two modified scales were defined on the basis of BS and BRS, following removal of the "foetal station" parameter due to its low discriminative capacity: BSM and BRSM. Nulliparity was given a value of 0 points, and multiparity a value of 3 points for BSM, and 2 for BRSM. Modified indices showed a higher predictive ability (AUC) for vaginal delivery than the original indices, for both BS (0.70 vs. 0.62) and for BRS (0.69 vs. 0.62). Replacing the "foetal station" parameter with parity in BS and BRS, improves predictive capacity with regard to the original indices in order to determine the outcome of labour. Key Words. Bishop Score. Induction of labour. Outcome of labour. Parity. Predictive model.