Objective To determine whether home cervical ripening is associated with acceptable safety margins compared to in-hospital cervical ripening during induction of labour. Design A prospective multicentre observational cohort study using routinely collected data. Setting Twenty-six UK maternity units; 18 offering only in-hospital cervical ripening and 8 offering both in-hospital and home cervical ripening. Population Women with singleton pregnancies, no previous caesarean section, at or beyond 37 weeks gestation having induction with details of cervical ripening method and location recorded. Methods Home cervical ripening using a balloon catheter was compared to in-hospital cervical ripening using prostaglandin. Multivariable logistic regression was performed for the primary outcome. Exposure: Cervical ripening at home using balloon catheter. Main outcome measures Primary outcome: neonatal unit admission within 48 hours of birth for 48 hours or more. Additional neonatal, maternal, labour progress and process outcomes were reported. Results Of 17,530 eligible women, 515 had balloon cervical ripening at home and 4332 had prostaglandin cervical ripening in a hospital that did not offer home cervical ripening. The primary outcome following home cervical ripening with balloon was not higher {16/515 (3.1%) vs 208/4332 (4.8%)}, but with substantial uncertainty on adjusted analysis consistent with a 64% lower risk through to an 81% higher risk. Conclusions Home cervical ripening using balloon catheter may be as safe for babies as using prostaglandin in hospital in low and moderate-risk groups, but further safety data are required.