The safety of planned homebirths remains controversial in Western countries. The aim of the present study was to compare outcomes in women who planned, and were selected to, homebirth at the onset of labour with women who planned for a hospital birth.

Data from 1631 planned homebirths between 1990 and 2007 were compared with a random sample of 16,310 low-risk women with planned hospital births. The primary outcomes were intrapartum intervention rates and complications. Secondary outcomes were perinatal and neonatal death rates.

Primiparas who planned homebirths had reduced risks for assisted vaginal delivery (OR 0.32; 95% CI 0.20-0.48), epidural analgesia (OR 0.21; CI 0.14-0.33) and dystocia (OR 0.40; CI 0.27-0.59). Multiparas who planned homebirths had reduced risks for operative vaginal delivery (OR 0.26; CI 0.12-0.56), epidural analgesia (OR 0.08; CI 0.04-0.16), episiotomy (OR 0.48; CI 0.31-0.75), anal sphincter tears (OR 0.29; CI 0.12-0.70), dystocia (OR 0.10; CI 0.06-0.17) and postpartum hemorrhage (OR 0.27; CI 0.17-0.41). We found no differences in cesarean section rate. Perinatal mortality rate was 0.6/1000 (CI 0-3.4) and neonatal mortality rate 0.6/1000 (CI 0-3.4) in the homebirth cohort. In the hospital birth cohort, the rates were 0.6/1000 (CI 0.3-1.1) and 0.9/1000 (CI 0.5-1.5) respectively.

Planning for homebirths was associated with reduced risk of interventions and complications. The study is too small to make statistical comparisons of perinatal and neonatal mortality.