The All Babies Count Initiative: Impact of a Health System Improvement Approach on Neonatal Care and Outcomes in Rwanda

A health system improvement program combining facility readiness support, clinical training/mentoring, and improvement collaboratives increased quality improvement capacity, improved maternal and newborn quality of care, and reduced neonatal mortality. These results can be used to inform system improvement approach design to transform quality of care and outcomes for newborns. Key Findings A district-wide health system improvement program combining facility readiness support, clinical training/mentoring, and district-wide improvement collaboratives increased quality improvement capacity, improved maternal and newborn quality of care, and reduced neonatal mortality by approximately 35% overall and 49% among high-risk preterm/low birth weight infants. This improvement in mortality was not seen during the same time period in the rest of rural Rwanda. Key Implications Policy makers should consider adopting components of the All Babies Count program into the design of system improvement approaches to transform quality of care and outcomes for newborns. Embedding the design into existing health system structures could help structure improvement in other clinical domains.  ABSTRACT Introduction: Poor-quality care contributes to a significant portion of neonatal deaths globally. The All Babies Count (ABC) initiative was an 18-month district-wide approach designed to improve clinical and system performance across 2 rural Rwandan districts. Methods: This pre-post intervention study measured change in maternal and newborn health (MNH) quality of care and neonatal mortality. Data from the facility and community health management information system and newly introduced indicators were extracted from facility registers. Medians and interquartile ranges were calculated for the health facility to assess changes over time, and a mixed-effects logistic regression model was created for neonatal mortality. A difference-in-differences analysis was conducted to compare the change in district neonatal mortality with the rest of rural Rwanda. Results: Improvements were seen in multiple measures of facility readiness and MNH quality of care, including antenatal care coverage, preterm labor management, and postnatal care quality. District hospital case fatality decreased, with a statistically significant reduction in district neonatal mortality (odds ratio [OR]=0.54; 95% confidence interval [CI]=0.36, 0.83) and among preterm/low birth weight neonates (OR=0.47; 95% CI=0.25, 0.90). Neonatal mortality was reduced from 30.1 to 19.6 deaths/1,000 live births in the intervention districts and remained relatively stable in the rest of rural Rwanda (difference in differences −12.9). Conclusion: The ABC initiative contributed to improved MNH quality of care and outcomes in rural Rwanda. A combined clinical and health system improvement approach could be an effective strategy to improve quality and reduce neonatal mortality.