Objective To determine whether a national quality improvement programme implementing continuous positive airway pressure (CPAP) at government hospitals in Malawi improved outcomes for neonates prioritised by an algorithm recommending early CPAP for infants weighing 1.0–1.3 kg (the 50th percentile weight at 30 weeks’ gestation). Design The analysis includes neonates admitted with respiratory illness for 5.5 months before CPAP was introduced (baseline period) and for 15 months immediately after CPAP was implemented (implementation period). A follow-up data analysis was completed for neonates treated with CPAP for a further 11 months. Setting and patients Neonates with admission weights of 1.0–1.3 kg before (106 neonates treated with nasal oxygen) and after implementation of CPAP (153 neonates treated with nasal oxygen, 103 neonates treated with CPAP) in the newborn wards at Malawi government district hospitals. Follow-up analysis included 87 neonates treated with CPAP. Intervention Neonatal CPAP. Main outcome measure We assessed survival to discharge at 23 government district hospitals with no significant differences in transfer rates before and after implementation of CPAP. Results Survival improved for neonates with admission weights from 1.0 to 1.3 kg treated with CPAP (30.1%) as compared with neonates of the same weight band treated with oxygen during the baseline (17.9%) and implementation (18.3%) periods. There was no significant difference in survival for neonates treated with CPAP during the implementation and follow-up periods (30.1% vs 28.7%). Conclusions Survival for neonates weighing 1.0–1.3 kg significantly increased with a nurse-led CPAP service in a low-resource setting and improvements were sustained during follow-up.