We investigated the maternal characteristics, labour performance and delivery mode of 497 nulliparas entering labour spontaneously at term to identify the obstetric factors which influenced their choice of analgesia; 51.7% of these women used epidural analgesia. They were shorter (163 versus 165 cm, p = 0.002) and the mean gestation was 3 days greater than those not using epidural analgesia (40.2 versus 39.6 weeks, p = 0.0007). Median birth-weight in the epidural group was greater by 155 g (3,450 versus 3,295 gs, p = 0.0001). Analysis of the labour characteristics showed a lesser cervical dilatation on admission, significantly longer latent and active phases of labour and second stage length in the epidural cohort. The need for oxytocin augmentation was significantly greater in the epidural group, both prior to and after insertion. Oxytocin augmentation was strongly associated with an increased risk of operative intervention regardless of analgesia. Selection of intrapartum analgesia is not a random event and epidural analgesia appears to be an indicator of abnormal labour patterns. To further investigate the impact of analgesic methods on nulliparous labour we are currently conducting a prospective randomized controlled trial.