OBJECTIVE: To determine whether primary midwife care (caseload midwifery) decreases the caesarean section rate compared with standard maternity care. DESIGN: Randomised controlled trial. SETTING: Tertiary-care women's hospital in Melbourne, Australia. POPULATION: A total of 2314 low-risk pregnant women. METHODS: Women randomised to caseload received antenatal, intrapartum and postpartum care from a primary midwife with some care by 'back-up' midwives. Women randomised to standard care received either midwifery or obstetric-trainee care with varying levels of continuity, or community-based general practitioner care. MAIN OUTCOME MEASURES: PRIMARY OUTCOME: caesarean birth. Secondary outcomes included instrumental vaginal births, analgesia, perineal trauma, induction of labour, infant admission to special/neonatal intensive care, gestational age, Apgar scores and birthweight. RESULTS: In total 2314 women were randomised-1156 to caseload and 1158 to standard care. Women allocated to caseload were less likely to have a caesarean section (19.4% versus 24.9%; risk ratio [RR] 0.78; 95% CI 0.67-0.91; P = 0.001); more likely to have a spontaneous vaginal birth (63.0% versus 55.7%; RR 1.13; 95% CI 1.06-1.21; P < 0.001); less likely to have epidural analgesia (30.5% versus 34.6%; RR 0.88; 95% CI 0.79-0.996; P = 0.04) and less likely to have an episiotomy (23.1% versus 29.4%; RR 0.79; 95% CI 0.67-0.92; P = 0.003). Infants of women allocated to caseload were less likely to be admitted to special or neonatal intensive care (4.0% versus 6.4%; RR 0.63; 95% CI 0.44-0.90; P = 0.01). No infant outcomes favoured standard care. CONCLUSION: In settings with a relatively high baseline caesarean section rate, caseload midwifery for women at low obstetric risk in early pregnancy shows promise for reducing caesarean births.