Evaluating pediatric early warning score (PEWS) as a predictor of PICU admission for children visiting emergency department in a tertiary care health centre

Abstract

Objectives: To obtain cut-off values of PEWS scores that can predict need for PICU transfer at admission. To obtain cut-off values of PEWS scores that can predict unanticipated worsening (PICU transfer from ward after admission) Materials and methods: This was a prospective 12 month observational study of patients aged 1month to12 years admitted in our Emergency department. Approximately 738 patients were included in our study and all were subjected to PEWS score individually by the nurses and the pediatric trainee. Interrater reliability between nurses and pediatric trainee was evaluated. PEWS score was measured at initial assessment (S0) by pediatric trainee and staff nurse in emergency and after one hour (S1) by pediatric trainee alone. Patients were stratified into 3 outcome groups: those admitted to PICU either from ED, transferred from the ward, or those directly admitted to ward from ED. PEWS score was calculated and compared among the groups. Results: Among 1525 patients, 738(48%) were admitted, 252(34.12%) to PICU and 486(65.85%) to ward. AUC for predicting overall admission was 0.69 (95%CI: 0.68–0.77), for PICU 0.86 (95%CI:0.96–1) and for ward 0.75 (95%CI: 0.66-0.75). The sensitivity and specificity in predicting overall admission with cut off of PEWS≥ 1.5 was 76% and 64%, for PICU admission with cut off PEWS ≥ 3 was 100% and 90%, for ward admission with cut off PEWS ≥1.5 was 79% and 61% respectively. Interrater reliability was excellent . Conclusions: The study found elevated PEWS serve as excellent screening tool for prediction to PICU admission and transfer from ward to PICU