CANHENT NEONATAL CARE
Neonatal mortality remains one of the most persistent public health challenges in low- and middle-income countries, where health systems are often constrained by limited resources, high patient loads, and shortages of skilled health personnel. Although progress has been made in child survival globally, deaths in the first 28 days of life continue to account for a significant proportion of under-five mortality. Many of these deaths are linked to preventable causes such as neonatal sepsis, perinatal asphyxia, and complications of prematurity. In Nigeria, where neonatal mortality remains among the highest globally, strengthening early identification and management of critically ill newborns remains essential to reducing avoidable deaths.
As part of its commitment to supporting evidence-informed action in neonatal and child health, the Collaborative Action for Neonatal and Child Health and Nutrition (CANHENT)spotlights findings from a study conducted at the University College Hospital (UCH), Ibadan. The study examined the usefulness of an admission early warning scoring system in predicting neonatal outcomes in a resource-limited setting. Specifically, it assessed whether a significant modified early warning score (sMEWS) could help identify newborns at increased risk of deterioration and in-hospital mortality, both at admission and after 72 hours of care.
The research was a retrospective analysis of 371 neonatal admissions recorded between January and December 2019. Data were extracted from routine nursing observation charts, and modified early warning scores were calculated using key physiological parameters including temperature, respiratory rate, heart rate, oxygen saturation, and neurological signs. A score of four or more was classified as significant (sMEWS), indicating a higher likelihood of clinical instability. The study assessed newborn status at two time points on admission and at 72 hours post-admission providing insight into early risk detection as well as response to treatment during hospitalization.
The findings showed that nearly half of the newborns admitted required resuscitation at presentation, reflecting the severity of cases managed in neonatal emergency units. Most babies were late preterm to term, and a significant proportion were outborn referrals, emphasizing the ongoing challenge of delayed presentation and inter-facility transfer for critically ill newborns. The overall mortality rate recorded in the study was 13.5 percent, highlighting the continued burden of preventable neonatal deaths even within tertiary-level care settings.
At admission, 36.1 percent of newborns had significant early warning scores, although this proportion declined to 5.9 percent after 72 hours of care. This reduction suggests that many newborns responded positively to stabilization and early clinical interventions. However, the persistence of sMEWS after 72 hours was strongly associated with mortality risk, indicating that newborns who fail to stabilize early may require closer monitoring, timely escalation of care, and improved clinical decision support.
The study also identified key factors associated with significant early warning scores at admission, including being outborn, requiring resuscitation, and diagnoses such as sepsis and perinatal asphyxia. Even after controlling for confounding factors, outborn status and the need for resuscitation remained strong predictors of high sMEWS. Importantly, the study found that sMEWS at both admission and 72 hours were significantly associated with in-hospital mortality, reinforcing the potential value of structured early warning scoring systems as a tool for predicting outcomes in neonatal care.
To further assess predictive performance, the researchers applied receiver operating characteristic (ROC) analysis. The results showed that sMEWS had moderate accuracy in predicting neonatal mortality at both admission and after 72 hours. While not perfect, this suggests that early warning scores can provide meaningful clinical support in settings where neonatal monitoring resources may be limited and where rapid identification of high-risk newborns is critical.
For CANHENT, the study offers useful evidence for strengthening newborn survival strategies in low-resource contexts. It highlights that neonatal deterioration can be subtle, and that early physiological changes may precede severe complications. Standardized early warning systems may improve clinical communication, support timely escalation of care, and enhance triage processes, especially in facilities where staffing and monitoring equipment are limited.
The study also draws attention to the vulnerability of outborn newborns, many of whom are referred late from lower-level facilities. This reinforces the importance of strengthening referral systems, improving emergency newborn transport, and enhancing capacity at primary and secondary healthcare levels. In addition, the strong association between resuscitation at admission and mortality underscores the need for sustained investment in neonatal resuscitation training, essential equipment availability, and emergency response readiness.
In conclusion, this research from UCH Ibadan provides evidence that significant modified early warning scores may serve as a practical tool for identifying newborns at increased risk of mortality in resource-limited settings. While further research is needed to validate and strengthen predictive performance across multiple facilities, the findings support the integration of structured monitoring tools and continuous capacity-building efforts as part of broader strategies to reduce preventable neonatal deaths. CANHENT remains committed to promoting the use of such evidence to inform policy dialogue, improve service delivery, and strengthen neonatal health outcomes across vulnerable populations.