Validation of pediatric early warning score in pediatric emergency department.

Abstract

One of the most important functions of the emergency department (ED) is to assess patient status. Only one, the pediatric early warning score (PEWS), has been designed for ED with acceptable validity, but it has never been validated in Thailand. The objective of this study was to validate PEWS in predicting hospitalization in children visiting the ED. During the initial phase, two triage nurses performed blind scoring (in order to determine inter-rater reliability using kappa statistics) for the first 30 patients who presented to the ED at Ramathibodi Hospital between March and May 2014 and who were aged < 15 years. The second phase then consisted of validation and involved 1136 patients. Patients who presented with trauma, psychiatric, dental and surgical concerns were excluded. Validity of the scoring system in predicting admission was assessed using area under the receiver operating characteristics (ROC) curve (AUC), sensitivity, and specificity, positive predictive value (PPV) and negative predictive value (NPV). Phase I demonstrated good inter-rater reliability (kappa = 0.75). In phase II, of the total group of 1136 patients, 168 patients (14.8%) were admitted: 162 to the general ward and six to the intensive care unit (ICU) during the study period. AUC for predicting overall, ICU, and general ward admission were 0.73 (95%CI: 0.68-0.77), 0.98 (95%CI: 0.96-1) and 0.71 (95%CI: 0.66-0.75), respectively. The sensitivity and specificity in predicting overall admission with a cut-off of PEWS ≥ 1 was 78% and 60%, respectively (PPV, 28%; NPV, 95%). Sensitivity and specificity in predicting ICU admission with the cut-off PEWS ≥ 3 was 100% and 91%, respectively (PPV, 5%; NPV, 100%). Using the cut-off PEWS ≥ 1, sensitivity and specificity in predicting ward admission were 77% and 59%, respectively (PPV, 24%; NPV, 94%). PEWS can be helpful in assessing patient status in pediatric ED with acceptable validity and can serve as a potentially excellent screening tool for prediction of ICU admission.