Objective: To assess differences in adverse maternal and neonatal outcomes before and after closure of a secondary obstetric care unit of a community hospital in an urban district. Study Design: Retrospective cohort study using aggregated data from National Perinatal Registry of the Netherlands (PERINED) in the very urban region of Amsterdam, consisting of data of five secondary and two tertiary hospitals. We assessed maternal and neonatal outcomes in singleton hospital births between 24 +0 weeks of gestational age (GA) up to 42 +6 weeks. Data of 78.613 births were stratified in two groups: before closure (years 2012–2015) and after closure (2016–2019). Results: Perinatal mortality decreased significantly from 0.84 % to 0.63 % (p = 0.0009). The adjusted odds ratio (aOR) of the closure on perinatal mortality was 0.73 (95 % CI 0.62–0.87). Both antepartum death (0.46 % vs 0.36 %, p = 0.02) and early neonatal death (0.38 % vs 0.28 %, p = 0.015) declined after closure of the hospital. The number of preterm births decreased significantly (8.7 % vs 8.1 %, p=<0.007) as well as number of neonates with congenital abnormalities (3.2 % vs2.2 %, p=<0.0001). APGAR < 7 after 5 min increased (2.3 % vs 2.5 %, p = 0.04). There was no significant difference in SGA or NICU admission. Postpartum hemorrhage increased significantly from 7.7 % to 8.2 % (p=<0.003). Perinatal mortality from 32 weeks onwards was not significantly different after closure 0.29 % to 0.27 %. Conclusions: After closure of an obstetric unit in a community hospital in Amsterdam, there was a significant decrease in perinatal, intrapartum and early neonatal mortality in neonates born from 24 +0 onwards. The mortality decrease coincides with a reduction of preterm deliveries. The increasing trend in asphyxia and postpartum hemorrhage is of concern. Centralization of care and increasing birth volume per hospital may lead to improvement of quality of care. A broad integrated, multidisciplinary maternity healthcare system linked with the social domain can ...