6040 Validating the national PEWS and LqSOFA in general practice: a retrospective cohort study

Objectives Children with acute illnesses commonly present in general practice. Most have mild and self-limiting infections but a minority of children have a serious illness. Clinical tools are needed to help identify which children are at high risk of serious illness and require hospital admission and which children can be safely managed at home. A widely used tool, the National Institute for Health and Care Excellence (NICE) Traffic Light system was recently validated in general practice and performed poorly.¹ A new standardised paediatric early warning score (National PEWS) is currently undergoing development and validation in hospitals throughout England, with plans for implementation into general practice.² Another score, the Liverpool Quick Sequential Organ Assessment (LqSOFA), was developed and validated within a paediatric Emergency Department and performed well in this setting.³ The aim of this study was to determine the accuracy of the National PEWS and LqSOFA in general practice, for identifying children who need admitting to hospital. Methods This study uses data from a retrospective cohort of acutely unwell children aged under five years presenting to general practice in England and Wales.⁴ Study data was linked with hospital admission data from NHS Digital within the Secure Anonymised Information Linkage (SAIL) Databank. Variables from each score were mapped onto study data to calculate score totals. Diagnostic accuracy was calculated using our primary outcome: hospital admissions within two days of the index GP consultation. Results A total of 6,703 children were included in the study; few children scored highly on either score (displayed in figure 1). Using a threshold of one point or more, the National PEWS had a sensitivity of 81.0% (95% confidence interval: 71.0% to 88.1%) and specificity of 32.5% (31.2% to 33.8%), with positive and negative predictive values (PPV/NPV) of 1.9 (1.5 to 2.5) and 99.1 (98.4 to 99.4) respectively, and an area under the curve (AUC) of 0.66 (0.59 to 0.72). Using the same threshold, the LqSOFA had a sensitivity of 30.6% (21.8% to 41.0%), specificity of 84.7% (83.7% to 85.6%); PPV 3.0 (0.1 to 1.0); NPV 98.7 (98.4 to 99.0); and an AUC of 0.58 (0.53 to 0.63). Conclusion Overall, the results demonstrate that neither the National PEWS nor the LqSOFA are accurate for identifying acutely unwell children requiring hospital admission from general practice. Both scores demonstrated poor discrimination for predicting hospital admissions and serious illnesses, suggesting that neither score is suitable for implementation in this setting. References Clark, et al. British Journal of General Practice. 2022. RCPCH. RCPCH website. 2019 Romaine ST, et al. Pediatrics. 2020. Hay, et al. Annals of Family Medicine. 2016.