Reliability of the Bishop score before labour induction at term

Objective: To evaluate the agreement within three pairs of observers regarding the Bishop score and an informal global evaluation of the cervix (favourable/unfavourable). Study design: We conducted a reliability study of the Bishop score. Three pairs of examiners (A–B, A–C and D–E) performed independently a cervical examination in 156 term pregnant women admitted for labour induction. We calculated the proportion of agreement and the Kappa coefficient. Results: Perfect agreement between two observers for the Bishop score was found in 44 women (28%). Accepting a difference of one point between the observers, agreement increased to 66%. Weighted Kappa coefficients for the Bishop score were 69, 54 and 35% for each pair of observers. Kappa coefficients for the informal evaluation of the cervix were 64, 45 and 46, respectively. Conclusion: Agreement between two observers evaluating the cervix is fair to substantial. An informal evaluation of the cervix is as reliable as the Bishop score.