Effect of an integrated neonatal resuscitation package on adverse neonatal outcomes in public health facilities, Ethiopia: a non-randomized quasi-experimental study

Each year, an estimated 900,000 neonatal deaths are attributed to intrapartum hypoxic events (birth asphyxia) worldwide. Survivors often suffer neurological impairments, with disproportionately high rates in Sub-Saharan Africa, including Ethiopia. Effective neonatal resuscitation (NR) reduces this risk, but its impact is limited by delayed or interrupted ventilation. Innovative strategies promoting early and continuous ventilation are needed. Therefore, this study aimed to evaluate the effect of an integrated NR package on adverse neonatal outcomes in public health facilities in Debre Birhan, Ethiopia. A non-randomized controlled quasi-experimental study with baseline and endline measurements was conducted in Debre Birhan city public health facilities from March 1, 2024, to January 30, 2026. A total of 1260 asphyxiated babies were enrolled. Baseline assessments were conducted in both control and intervention sites. Providers in the control group received standard NR simulation training, while those in the intervention group received NR training enhanced with an integrated device with a real-time feedback system. Intervention facilities also used NeoBeat and an upright bag-mask with PEEP during clinical resuscitation. Data collected via Kobo Toolbox were analyzed using Stata 17. Baseline comparability was assessed using chi-square or Fisher’s exact and independent t-tests. A Difference-in-Differences (DID) analysis was used to evaluate the intervention’s effect on composite adverse neonatal outcomes. After accounting for six participants lost to follow-up, the final analysis included 1,254 newborns. Adjusting for maternal demographics, clinical conditions at admission, and delivery-specific variables, the intervention package was associated with an 18.7%-point lower probability of the adverse composite neonatal outcome (average treatment effect (ATE) = -18.7; p = 0.011; 95% CI [-30.9, -6.5]). An integrated neonatal resuscitation package was associated with a lower probability of adverse neonatal outcomes. Because of the non-randomized design, the small number of facilities and the bundled intervention, causal inference is limited and the effect of individual components cannot be determined. Larger, preferably cluster-randomized, and economic evaluations studies are needed before scale-up. Registered in Pan African Clinical Trial Registry with the unique identification number of PACTR202605670069858. Registered on 04 May 2026, retrospectively registered. Link: https://pactr.samrc.ac.za/TrialDisplay.aspx? TrialID=41112 .