Background: Child mortality is high in Ethiopia, but there are no reliable data on the causes of these deaths.
Methods: We established a death notification system in health facilities and three communities in eastern Ethiopia, at demographic surveillance areas in Kersa/Haramaya (rural) and Harar (urban). We collected ante-mortem data and post-mortem samples from stillbirths and deaths in children aged <5 years. Cause of death (CoD) was determined by an expert panel integrating clinical data, verbal autopsies, molecular and conventional microbiology, and histopathology of post-mortem tissue biopsies. Causes were classified as underlying, comorbid, and immediate and analysed separately for stillbirths, neonatal deaths and child deaths (1–59 months). Findings: We screened 2237 death notifications over two years; among 307 eligible deaths where the families were approached, 195 (64%) consented for post-mortem sampling. CoD was determined in 193 (99%) including 114 stillbirths, 59 neonates and 20 child deaths. Common underlying causes for stillbirth were perinatal asphyxia/hypoxia (53%) and birth defects (21%). Among 59 neonatal deaths, the underlying cause was perinatal asphyxia in 29%, and the immediate cause was neonatal sepsis in 60%. Malnutrition was the leading underlying cause for child deaths (75%) and infections were common immediate and comorbid causes. Pathogens were identified in 19 (95%) child deaths, mostly Klebsiella pneumoniae and Streptococcus pneumoniae. Interpretation: Perinatal asphyxia/hypoxia, infections, and birth defects accounted for 68% of stillbirths and child deaths. Most deaths were preventable with feasible interventions such as improved maternity services or folate and vaccines uptake. Funding Information: This work is supported by the CHAMPS grant to Emory University. The parent grant is from The Bill and Melinda Gates Foundation (INV-009109). Declaration of Interests: No competing interests were disclosed. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the CDC. Ethics Approval Statement: Every post-mortem investigation was preceded by an informed written consent by the child’s caregiver. The study was approved and renewed annually by the London School of Hygiene and Tropical Medicine Ethics Board (ref 14394) and by the Haramaya University Ethical committee and National Ethics Committee in Ethiopia (last ref MoSHE 04/246/661/21).
Keywords: Child mortality, CHAMPS, cause of death, surveillance, stillbirths, Ethiopia, minimally invasive tissue sampling