CDC-backed surveillance finds 4 in 5 child deaths preventable across 7 high-mortality countries

A nearly decade-long surveillance effort spanning 7 countries in Africa and South Asia has produced one of the most detailed pictures yet of why stillbirths and young children die in the world's highest-mortality settings — and the findings, published September 8 in the CDC's Morbidity and Mortality Weekly Report, suggest that roughly 80% of those deaths could have been prevented with care that already exists.

The report draws on the Child Health and Mortality Prevention Surveillance (CHAMPS) network, which since 2016 has combined minimally invasive tissue sampling, laboratory testing, verbal autopsy, and clinical record review to determine causes of death with a level of precision that conventional mortality surveillance in low-resource settings typically cannot achieve. Between December 2016 and December 2024, the network identified 18,784 eligible deaths across Bangladesh, Ethiopia, Kenya, Mali, Mozambique, Sierra Leone, and South Africa; families consented to enrollment in 83% of cases, and 8,500 deaths ultimately underwent full postmortem investigation and multidisciplinary cause-of-death review.

Causes shift sharply with age

The report found that cause of death varied predictably as children aged. Among the 3,199 stillbirths analyzed, perinatal asphyxia or hypoxia was implicated in nearly 80% of cases, frequently alongside maternal hypertensive disorders, placental abnormalities, or chorioamnionitis. Among the 3,230 neonatal deaths, the leading contributors were preterm birth complications (about 40%), asphyxia or hypoxia (about 38%), and sepsis (about 37%) — with sepsis often driven by the gram-negative bacteria Klebsiella pneumoniae and Acinetobacter baumannii.

Among the 2,071 deaths of infants and children up to age 5, the picture broadened to include lower respiratory infections, sepsis, malnutrition, malaria, and diarrheal disease, each present in roughly one-fifth to more than one-third of cases. Notably, multiple contributing conditions were the norm rather than the exception: 44% of neonatal deaths and 68% of infant and child deaths involved 2 or more causes acting together, a level of complexity that verbal-autopsy-based surveillance — which typically assigns a single cause — cannot capture.

Postmortem measurements also revealed a heavy undernutrition burden among surviving-age children who died: about 61% were underweight, 61% wasted, and 43% stunted by World Health Organization growth standards.

Infection patterns point to gram-negative bacteria

Among deaths where at least 1 pathogen was implicated, gram-negative bacteria dominated, particularly among neonates, where they accounted for roughly 86% of pathogen-associated deaths. K. pneumoniae alone was tied to nearly half of those cases. Among older infants and children, malaria (chiefly Plasmodium falciparum), pneumococcus, and K. pneumoniae were most common, and nearly 45% of pathogen-linked deaths in that age group involved more than 1 organism. The consistency of gram-negative bacteria across sites, researchers noted, strengthens the case for investment in infection prevention infrastructure and in vaccines targeting these pathogens.

Most deaths traced to fixable gaps in care

Of the deaths with a formal preventability assessment, 80.7% were judged preventable or possibly preventable given locally available resources. Reviewers most often pointed to gaps in antenatal care, obstetric management, and health care–seeking behavior, followed by shortfalls in pediatric clinical management, infection control, and nutritional support. For stillbirths specifically, nearly 69% of recommendations centered on antenatal care and 56% on obstetric management — underscoring how much of the mortality burden traces back to pregnancy and delivery rather than later childhood illness.

The authors cite concrete examples of the surveillance data already shaping policy. In Kenya's Siaya County, CHAMPS findings prompted an expansion of Kangaroo Mother Care and emergency obstetric training, coinciding with a decline in newborn-unit mortality. In South Africa, the data supported wider antenatal ultrasound access and the addition of aspirin therapy for hypertensive pregnancy disorders to the national formulary. In Sierra Leone, evidence that malnutrition contributed to nearly half of under-5 deaths drove the country's first domestic purchase of therapeutic food, and separate HIV findings helped secure an additional $8 million in PEPFAR funding.

Limitations acknowledged

The authors caution that CHAMPS catchment areas are not nationally representative, and because tissue sampling required rapid notification, deaths in health facilities were more likely to be captured than community deaths or those with delayed reporting. Data completeness also varied across sites, and while the network's protocols aimed to distinguish causal pathogens from incidental findings, some misclassification is possible. Preventability judgments, the authors note, reflect expert assessment of local context rather than a formal estimate of how many deaths a specific intervention would avert.

Even so, the researchers argue the network demonstrates that high-precision postmortem investigation is feasible even in resource-constrained settings, and that its core methods — standardized tissue sampling, targeted diagnostics, and structured multidisciplinary review — could be adapted more broadly to strengthen routine mortality surveillance as countries work toward the United Nations' 2030 goal of ending preventable deaths among newborns and young children.

Reference

Madewell ZJ, Dangor Z, Mutevedzi PC, et al. Causes of death among stillbirths and children aged <5 years in Africa and South Asia — Child Health and Mortality Prevention Surveillance, seven countries, 2016–2024. MMWR Surveill Summ. 2026;75(6):1-24. doi:10.15585/mmwr.ss7506a1