labor are associated with operative delivery? Mariette Ruckert, Myrthe Peelen, Michelle van Vliet, Mieke ten Eikelder, Nikki van Desel, Katrien Oude Rengerink, Ben Willem Mol, Irene de Graaf Diakonessenhuis, Obstetrics and Gynecology, Utrecht, Netherlands, Academic Medical Center, Obstetrics and Gynecology, Amsterdam, Netherlands, Groene Hart Hospital, Obstetrics and Gynecology, Gouda, Netherlands, Kennemer Gasthuis, Obstetrics and Gynecology, Haarlem, Netherlands OBJECTIVE: At present, there is little scientific fundament behind the interpretation of physical examination during labor and delivery. We aimed to indentify which findings at the start of the second phase of labor were associated with instrumental vaginal delivery (IVD) and emergency caesarean section (ECS). STUDY DESIGN: We performed a prospective cohort study in five hospitals in the Netherlands. Consecutive laboring women, with a viable singleton pregnancy in cephalic presentation and a gestational age of at least 34 weeks, were included. We assessed the predictive value of antepartum and intrapartum characteristics and findings at digital examination prior to the second phase of labor (start pushing) for operative delivery (OD) using logistic regression analysis. We selected variables based on literature and clinical experience. Backward selection was used to develop a prediction model, predictors with a p 0.157 remained in the final model. The discriminative ability of the model was assessed using ROC analyses (perfect when AUC 1). The correspondence between the predicted probability for OD and the observed number of operative deliveries over groups was assessed with a calibration plot (perfect when all points are on the diagonal). RESULTS: Between December 2010 and October 2011, we included 473 women, of whom 74 (16%) had a VID and 22 (5%) an ECS. For 425 women, data on all relevant variables were available. Predictors for OD at start of the second phase of labor were 10 cm dilatation, Caucasian etnicity, nulliparity, epidural analgesia, spontaneous start, molding, engagement and position of the fetal head (Table). The model including these variables showed good discriminative ability: AUC ROC 0.82 (95% CI 0.77-0.87). The model calibrated well (Figure). CONCLUSION: At start of the second phase of labor, antepartum and intrapartum characteristics can predict the need for OD. The prediction model may be implied as a useful tool for clinical decision making and assembling logistics at the delivery ward. 776 The validity of the variable “NICU admission” as an outcome measure for neonatal morbidity Nora Danhof, Melanie Wiegerinck, Anton van Kaam, Ben Willem Mol Academic Medical Center, Obstetrics & Gynecology, Amsterdam, Netherlands OBJECTIVE: The variable “neonatal intensive care (NICU) admission” is a frequently used outcome measure in clinical studies, as can be illustrated by the fact that it is reported as such in nearly one hundred Cochrane reviews from the Pregnancy and Childbirth Group. We addressed the question whether NICU admission is a valid surrogate outcome measure for neonatal morbidity, specifically when comparing different birth settings with different access to the NICU. STUDY DESIGN: We performed a retrospective cohort study among neonates admitted at the NICU of a tertiary hospital in the Netherlands during a ten-year period. The inclusion was restricted to singletons born at term, who were admitted at the NICU for delivery-related morbidity within 24 hours of birth. Patient characteristics and admission data were compared for four groups based on the line of care during delivery, i.e. home birth (group Ia), midwive lead hospital delivery (Ib), secondary care (II) and tertiary care (III). We specifically looked at the mortality rate, defined as the percentage of children that died at any time during NICU admission. RESULTS: We studied 871 newborns of which 53 were born at home, 25 under supervision of a midwife in the hospital, 166 in secondary Calibration plot, showing the average observed operative delivery rate (Y-axis) against the average calculated operative delivery rate (X-axis)