Objective: To evaluate the effect of World Health Organization Essential Newborn Care course and the American Academy of Pediatrics Neonatal Resuscitation Program training on perinatal mortality in rural India. Methods: This study was part of a multi-country prospective, community-based cluster randomized controlled trial. Birth, 7-day and 28-day neonatal outcomes for all women with pregnancies greater than 28 weeks in the 26 study communities in Karnataka, India were included. Mortality rates pre- and post-Essential Newborn Care training were collected prospectively and then communities randomized to either receive neonatal resuscitation or refresher newborn care training in the control clusters. Results: Consent was obtained on 99% of the 25,096 births. Perinatal mortality for infants ≥500 g decreased from 52 to 36/1000 after newborn care training (RR 0.7; 95% CI 0.5, 0.9); stillbirth decreased from 23 to 14/1000 (RR 0.62; 95% CI 0.46, 0.83) and early neonatal mortality decreased from 29 to 22/1000 (RR 0.74; 95% CI 0.53, 1.03). Mortality was not reduced further with resuscitation training. Conclusions: Using a prepost design, World Health Organization Essential Newborn Care community birth attendant training resulted in a significant reduction in perinatal mortality. In low-resource settings, the newborn care training package appears to be an effective intervention to decrease perinatal mortality. © 2012 Informa UK, Ltd.