{
  "abstract": "Paediatric early warning scores (PEWS) with a system of response have been introduced into a number of inpatient units, emergency departments and indeed primary care facilities. Their aim is laudable: to identify seriously ill children and recognise potential clinical deterioration, with the aim to intervene and prevent morbidity and mortality. There are many scores and systems out there with variable efforts to validate them. Many of these have poor predictive values, and so there are many false positives; some do not show acceptable sensitivity and specificity, and none have been shown to reduce mortality. Despite these limitations, they are now almost universally used in inpatient units in the UK and around the world. Most attempts at validation have been from high-income countries. Some have been designed for specific use, such as on oncology units or cardiac units, where some ‘abnormal’ parameters may be acceptable in that patient group. So, it is important to use a tool that has been validated in the patient group you are looking after, and even more reason why an inpatient-validated scorers should not be used in the emergency department where the patient population and acuity of patients differ. These scores are starting to be used in resource-poor settings where patient population and acuity, again, differ. So, it is with real interest that the study by Andersen et al (Pediatrics 2026; 158:e2025072357.) has looked at a modified PEWS specifically for a paediatric population with severe acute malnutrition (SAM) and examined its longitudinal performance while in the hospital. This work was conducted with Médecins Sans Frontières (MSF) on children hospitalised with SAM in Madarounfa, Niger. This novel MSF-PEWS uses scores for nine clinical indicators, summarised into colour categories (green, orange, yellow, red) and assessed regularly throughout hospitalisation. It included 4125 inpatient children. They estimated the association of colour category at admission with subsequent clinical deterioration and mortality, as well as the association between time-varying colour category and mortality within 48 hours and time to discharge or death. Children in the red MSF-PEWS colour category at admission had a 2.58 (95% CI 1.87 to 3.55) times greater risk of death during treatment compared with children in the green category at admission. In time series analyses, children in the red category at any given time were 11.9 (95% CI 11.3 to 12.6) times more likely to die in the following 48 hours than children in the green category. More severe colour categories were associated with longer lengths of hospital stay and shorter times until death. This simple score was effective at identifying children at higher risk of adverse clinical outcomes, which included mortality, during inpatient treatment for complicated SAM in Niger. This is a simple score and approach and is one of the first validation studies in this setting. However, there are still many problems as the editorial by Engmann (Pediatrics. 2026;158(1):e2025075563) points out, and when one looks at this paper in more detail, there are ‘more questions than answers’.",
  "authors": [
    {
      "affiliations": [],
      "name": "BMJ Publishing Group Ltd and Royal College of Paediatrics and Child Health"
    }
  ],
  "title": "More on paediatric early warning systems",
  "uid": "7f47bf74-889c-5565-9ecb-710d53976772"
}