Article Text
Abstract
Background Current ambulance service policy requires paramedics in certain parts of the UK to transport children aged 0–2 years to hospital, regardless of their presenting complaint. A number of Paediatric Early Warning Scores (PEWS) exist to detect deterioration in the hospitalised child. This study aims to evaluate and understand the potential utility of PEWS in an ambulance service setting.
Method This study is a retrospective analysis of patient reports over a 12 month period (June 2013–June 2014). PEWS are calculated using ambulance vital signs and compared against the following ED discharge outcomes: admission to hospital, GP referral and discharge home. Data analysis consisted of the variables of sensitivity; specificity; positive and negative predictive values; positive and negative likelihood ratios; and ROC curve for the PEWS values against the three main outcomes.
Results From a randomised sample of 300 patient records, 131 were excluded based on the exclusion criteria listed above, leaving 169 complete data sets that were included for analysis. Of the 169 record analysed, 100 (59.2%) were discharged to home, 30 (17.8%) referred to their GP, and 18 (10.7%) were admitted following assessment in the ED. PEWS for admission showed low sensitivity (6.8%–10.12%). PEWS for GP referral also demonstrated low sensitivity (15.53%–18.12%). PEWS for discharge to home showed higher sensitivity and specificity than other outcomes.
Discussion Overall, PEWS has some degree of high specificity in all outcome measures, but often with wide confidence intervals. PEWS is weakly sensitive across all outcomes, and this is demonstrated in the AUCs for each outcome measure. As a potential diagnostic test to predict ED outcome, in this study PEWS is shown to perform poorly. Further work is therefore required to determine the utility of PEWS, or other early warning scores, for use in an out-of-hospital setting.