Abstract
OBJECTIVE: The Bishop score is the most commonly used method to assess the readiness of the cervix for induction. However, it was created without modern statistical methods. Our objective was to determine whether a simplified score can predict vaginal delivery equally well. METHODS: Data were analyzed for 5,610 nulliparous women with singleton, uncomplicated pregnancies between 37 0/7 and 41 6/7 weeks of gestation undergoing labor induction. These women had all five components of the Bishop score recorded. Logistic regression was performed and a simplified score created with significant components. Positive and negative predictive values and positive likelihood ratios were calculated. RESULTS: In the regression model, only dilation, station, and effacement were significantly associated with vaginal delivery (P<.01). The simplified Bishop score was then devised using these three components (range 0-9) and compared with the original Bishop score (range 0-13) for prediction of successful induction, resulting in vaginal delivery. Compared with the original Bishop score (greater than 8), the simplified Bishop score (greater than 5) had a similar or better positive predictive value (87.7% compared with 87.0%), negative predictive value (31.3% compared with 29.8%), positive likelihood ratio (2.34 compared with 2.19), and correct classification rate (51.0% compared with 47.3%). Application of the simplified Bishop score in other populations, including indicated induction and spontaneous labor at term and preterm, were associated with similar vaginal delivery rates compared with the original Bishop score. CONCLUSION: The simplified Bishop score comprised of dilation, station, and effacement attains a similarly high predictive ability of successful induction as the original score. LEVEL OF EVIDENCE: II.