OBJECTIVE: Higher risks of uterine rupture have been reported among women attempting vaginal birth after caesarean (VBAC) particularly following induction with prostaglandins, compared with women who do not labour. This study aimed to estimate these risks as well as that associated with oxytocin use. DESIGN: Population-based retrospective cohort study involving all women who had their first births by caesarean. In their second birth, risks of uterine rupture among women without labour and women who had labour augmented or induced were compared with women who gave birth after spontaneous labour. SETTING: Four Australian states in 1998-2000. POPULATION: Women on pregnancy outcome databases with a second birth after a prior caesarean for their first birth. METHODS: From 29, 008 women identified from the databases, those with uterine rupture were identified and validated using hospital case records. MAIN OUTCOME MEASURE: Uterine rupture. RESULTS: The risk of complete uterine rupture among women without labour was 0.01%. The risk in spontaneous labour without augmentation was 0.15%, considerably higher when there was augmentation with oxytocin (1.91%). The risk with induction of labour was 0.54% for oxytocin alone, 0.68% for prostaglandin alone, 0.63% without either and 0.88% when they were combined. Compared with spontaneous labour, risks were increased three- to five-fold for any induction, six-fold for prostaglandin combined with oxytocin and 14-fold for augmentation with oxytocin. CONCLUSIONS: Careful consideration should be given to the use of oxytocin for augmentation of labour or induction by any method for women with a previous caesarean in view of increased risks of uterine rupture.