Background Neonatal deaths in developing countries make the largest contribution to global mortality in children younger than 5years.90%of deliveries in the poorest quintile of households happen at hom e.We postulated that a community -based participatory in tervention could sig-nificantly reduce neonatal mortali ty rates.Methods We pair -matched 42geopolitical clusters in Makwanpur dis-trict,Nepal,selected 12pairs rand omly,and randomly assigned one of each pair to interven tion or control.In each intervention cluster(average population 7000),a female facilitator convened nine womens g roup meetings every month.The facilitator supported groups through an ac-tion -learning cycle in which they id entified local perinatal problems and formulated strategies to address them.We monitored birth outcomes in a cohort of 28931women,of whom 8%joined the groups.The primary outcome was neonatal mortality rate.Other outc omes included stillbirths and maternal deaths,uptake of anten atal and delivery ser-vices,home care practices,infant m orbidity,and health -care seeking.Analysis was by intent ion to treat.The study is registered as an International St andard Randomised Controlled Trial,number ISRCTN31137309.FindingsFrom 2001to 2003,the neonatal morta lity rate was 26.2per 1000(76deaths per 2899livebirths)in intervention clusters compared with 36.9per 1000(119deaths per3226livebirths)in controls(adjusted odds ratio 0.7095%CI 0.53-0.94).Stillbirth rates were similar in both groups.The maternal mortality ratio was 69per 100000(two deaths per 2899livebirths)in intervention clusters compared with 341per 100000(11deaths per 3226livebirths)in control clusters(0.220.05-0.90).Women in intervention clusters were more likely to have antenatal care,institutional delivery,trained birth atten-dance,and hygienic care than were controls.Interpretation Birth outcomes in a poor rural popula tion improved greatly through a low cost,potentially sustainable and scalable,participatory intervention with wo men’s groups. Background Neonatal deaths in developing countries make the largest contribution to global mortality in children younger than 5years.90% of deliveries in the poorest quintile of households happen at hom e.We postulated that a community-based involvement in tervention could sig-nificantly reduce neonatal mortali ty rates. Methods We pair-matched 42 population of clusters in Makwanpur dis-trict, Nepal, selected 12 pairs rand omly, and randomly assigned one of each pair to interven tion or control. In each intervention cluster (average population 7000), a female facilitator convened nine women’s roup meetings every month.The facilitator supported groups through an ac-tion-learning cycle in which they id entified local perinatal problems and formulated strategies to address them. WeCombined birth outcomes in a cohort of 28931women, whom whom 8 % joined the groups. primary outcome was neonatal mortality rate. Other outc omes included stillbirths and maternal deaths, uptake of anten atal and deli very ser-vices, home care practices, infant m orbidity, and health-care seeking. Analysis was by intent to treat. The study is registered as an International St andard Randomized Controlled Trial, number ISRCTN31137309.FindingsFrom 2001to 2003, the neonatal morta lity rate was 26.2per 1000 (76deaths per 2899livebirths) in intervention clusters compared with 36.9per 1000 (119deaths per3226livebirths) in controls (adjusted odds ratio 0.7095% CI 0.53-0.94) .Stillbirth rates were similar in both groups. The maternal mortality ratio was 69per 100000 (two deaths per 2899livebirths) in intervention clusters compared with 341per 100000 (11deaths per 3226livebirths) in control clusters (0.220.05-0.90) .Women in intervention clusters were likely likely to have antenatal care, institutional delivery, trained birth atten- dance, and hygienic care than were controls. sterpretation Birth outcomes in a poor rural popularization improved greatly through a low cost, potentially sustainable and scalable, participatory interventio nwith wo men’s groups.