Pediatric Emergency Warning Systems in Niger for Severe Acute Malnutrition: More Questions Than Answers

In this issue of Pediatrics, Anderson et al publish their findings, which assess the clinical utility of an early warning system score among children in Niger with complicated severe acute malnutrition (SAM).1 SAM is defined by the World Health Organization (WHO) as a weight for height/length Z-score under −3 SDs below the median on the WHO Child Growth Standards, clinical signs of bilateral pitting edema, or midarm circumference of less than 115 mm.2 To better appreciate the utility of this work, some contextualization is helpful.Niger has a population of 27 million people, covers a land area of approximately 1.3 million km2 (490 000 square miles), and is the largest landlocked country in West Africa. Approximately 80% of this equatorial country’s land mass lies in the Sahara Desert—the world’s largest hot desert. Niger is nearly the same size as South Africa and Peru, approximately twice the size of France and Texas, and is the world’s 22nd largest country.3 Over 82% of the population is rural, and for the past 2 decades, Niger has ranked in the bottom 5 of all nations on various human development index scores.4 In Niger, 48% of males and 26% of females attend and complete a mandatory 6 years of primary school. The under-5 mortality is one of the highest in the world at 117 deaths per 1000 live births, as is the total fertility rate, where an average of 6 children are born per woman. Average life expectancy is 60 years, and recent estimates are that there are 3 physicians and 30 nurses/midwives per 100 000 people. By contrast, in neighboring Nigeria, there are over 10-fold physicians (39 per 100 000 people) and nearly 6 times the number of nurses/midwives (170 per 100 000), while in Sweden, there are 430 physicians and 1160 nurses/midwives per 100 000 people.5 On the Global Hunger Index, a standardized composite score that measures the multidimensional aspects of hunger and includes domains of undernourishment, child stunting, child wasting, and child mortality among countries, Niger scored 34.1 and was ranked 117 out of 123 countries in 2025. Of note, this score has decreased from over 50 in the mid-2000s to present values.6 In Niger, acute malnutrition afflicts 1.7 million children, and SAM occurs in over 400 000 cases.7It is against this backdrop that Anderson et al assess the clinical utility of a pediatric early warning system (PEWS) score developed by Médecins Sans Frontières (MSF) staff to assess SAM. This score, named MSF-PEWS, has been scaled up to more than 150 centers in over 20 countries. The authors enrolled 4125 children aged 6 to 59 months diagnosed with SAM complicated by either an inability to eat or compounding illnesses severe enough for admission to an inpatient therapeutic feeding center.In this vulnerable population of admitted infants, the authors ascribed a score based on morbidity severity, then summarized these scores into a spectrum of color categories, with red representing the sickest children and green the least sick that warranted admission. The authors used these colors throughout hospitalization to track subsequent clinical deterioration and mortality. Not surprisingly, perhaps, the authors found that the average time to discharge was 46 hours for the sickest (red color) children, compared with 21 hours for the least sick (green color) children, and that there was a 12-fold higher risk of dying within 48 hours among children assigned a red compared with those assigned a green color. The authors also acknowledge that while deploying a PEWS system involves a 2-fold process—a scoring component and a clinical decision-making action-oriented response algorithm, their work only assesses one component—the scoring phase.1The PEWS system has a checkered history. PEWS were initially introduced in high-income hospital settings; however, recent analyses call into question the utility of such systems. The EPOCH trial, published in 2018 and conducted in 7 countries, included 21 hospitals and over 550 000 patient-days across Europe and New Zealand. This trial concluded that compared with normal bedside practice, a PEWS system did not improve all-cause mortality.8 Although there is a paucity of PEWS data from low-income countries, a nonsystematic scoping review of PEWS in resource-limited and humanitarian settings reported improvements in mortality and reduction in clinical deterioration events in one single-site, low-middle income country among 258 pediatric oncology patients.9,10 There are even fewer data assessing the performance of PEWS specifically in SAM in low-income settings. Secondary analysis of data from 843 children with SAM conducted in Kenya and Malawi suggests a correlation between warning score and inpatient mortality,11 while another study conducted in northern India among 5 different PEWS scores concluded that though none showed optimal accuracy, there was a wide range of performance of these PEWS tools examined.12 Of note, these studies, like this one in Pediatrics, examined the PEWS scores, but not the entire PEWS system, which includes a response algorithm.The next steps before recommending the use of MSF-PEWS to triage children across SAM or a range of conditions would be to validate and compare it to other PEWS scores, and, crucially, link a score to an action-oriented response algorithm. This would enable assessment of whether the score plus the action-oriented response algorithm decreases all-cause mortality in children. Until this information is ascertained, equipoise about the intrinsic utility and opportunity cost exists. Despite this, there are practical considerations and possible benefits to this work. Given the paucity of trained health care workers in Niger and worldwide, task-shifting is a pragmatic reality. In such cases, less trained and experienced health care workers can be deployed to determine and use these PEWS scores to admit patients into a facility, for regular clinical monitoring, and to execute on locally-developed therapeutic management protocols. These scores may also help administrators in the annual planning and contribute to the limited data on PEWS and SAM in low-resource settings.1There are some interesting innovations that are starting to be deployed to address SAM prevention and management strategies.13–15 One approach uses artificial intelligence, big data, and machine learning in high-frequency monitoring states, while other approaches leverage the integration of climate change, weather patterns, and agricultural systems to better predict and assess nutritional shocks and mitigate against adverse outcomes.It is a well-established axiom that where one is born, or where one lives, plays an outsized role in determining one’s morbidity and mortality risk. Nowhere is this more evident than in Niger, where the burden of SAM remains unacceptably high among children. Clearly, SAM represents the tip of the spear of a continuum of challenges with multiple complex antecedent factors. In addition to the availability of healthy food, broader sociopolitical, accountability, and national prioritization considerations are necessary to drive solutions. This study raises more questions than it provides answers and highlights the urgent need for an assessment of the utility of the MSF-PEWS System, which addresses, together, scoring and response components. Such an effort needs to have a good cost-benefit component to ascertain whether it is practical, feasible, and scalable for country governments to adopt.