Objective To test the null hypothesis that no correlation exists between transvaginal digital examination and the gold standard technique of intrapartum transabdominal ultrasound determination of fetal head engagement, and secondarily to compare the performance of attending physicians and senior residents in depicting fetal head engagement by transvaginal digital examination. Methods Two hundred and twenty-two consecutive patients in labor >37 weeks' gestation with normal singleton cephalic-presenting fetuses and with either ruptured or intact membranes were included. Of these, 119 were nulliparous and 103 were multiparous. Transvaginal digital examinations were performed by either attending physicians or senior residents and were followed immediately by transverse suprapubic sonographic assessments by a single sonographer. The fetal head was considered engaged on transvaginal digital examination if the leading part of the fetal bead was positioned at least at maternal ischial spine station 0, and on transverse suprapubic ultrasound if the fetal biparietal diameter was below the maternal pelvic inlet. Examiners were blinded to each other's findings and the effect of examiner experience on the rate of agreement between the two techniques was assessed. The effect of several other independent variables upon the rate of agreement between the two modalities was also assessed. Results Overall, transvaginal digital examinations were consistent with ultrasound determinations with a raw percent agreement rate of 85.6% (95% confidence interval (CI), 80.8-90.3); kappa = 69.5% (95% CI, 59.4-73.9; P < 0.001). When stratified according to parity, the raw percent agreement rate for nulliparous patients was 81.5% (9S% CI, 73.4-88.0); κ = 60.7% (95% CI, 45.9-64.1; P < 0.001), and for multiparous patients it was 90.3% (95% CI, 84.1-95.9); kappa = 80.4% (95% CI, 63.0-87.5; P < 0.001). Maternal age, gravidity, maternal body mass index, gestational age, cervical dilatation, effacement, membrane status, ischial spine station of the fetal bead, fetal bead position at ultrasound assessment, birth weight and mode of delivery did not significantly affect rate of agreement. Parity did not affect examination consistency in multiparous patients, but in the whole group increasing parity increased the rate of agreement between modalities. Presence of combined spinal epidural anesthesia significantly increased the rate of agreement in the complete group and among multiparous but not among nulliparous patients. Conclusion These data demonstrate a high rate of agreement (85.6%) between ultrasound determination and transvaginal digital assessment of fetal bead engagement. Examiner experience bad no effect. These data support the use of intrapartum transabdominal assessment of fetal bead engagement. Copyright (C) 2003 ISUOG. Published by John Wiley & Sons, Ltd.