Adherence to the bedside paediatric early warning system (BedsidePEWS) in a pediatric tertiary care hospital

BACKGROUND: The aim of this study is to describe the adherence to the Bedside Pediatric Early Warning System (BedsidePEWS) escalation protocol in children admitted to hospital wards in a large tertiary care children’s hospital in Italy. METHODS: This is a retrospective observational chart review. Data on the frequency and accuracy of BedsidePEWS score calculations, escalation of patient observations, monitoring and medical reviews were recorded. Two research nurses performed weekly visits to the hospital wards to collect data on BedsidePEWS scores, medical reviews, type of monitoring and vital signs recorded. Data were described through means or medians according to the distribution. Inferences were calculated either with Chi-square, Student’s t test or Wilcoxon-Mann–Whitney test, as appropriate (P < 0.05 considered as significant). RESULTS: A total of 522 Vital Signs (VS) and score calculations [BedsidePEWS documentation events, (DE)] on 177 patient clinical records were observed from 13 hospital inpatient wards. Frequency of BedsidePEWS DE occurred < 3 times per day in 33 % of the observations. Adherence to the BedsidePEWS documentation frequency according to the hospital protocol was observed in 54 % of all patients; in children with chronic health conditions (CHC) it was significantly lower than children admitted for acute medical conditions (47 % vs. 69 %, P = 0.006). The BedsidePEWS score was correctly calculated and documented in 84 % of the BedsidePEWS DE. Patients in a 0–2 BedsidePEWS score range were all reviewed at least once a day by a physician. Only 50 % of the patients in the 5–6 score range were reviewed within 4 h and 42 % of the patients with a score ≥ 7 within 2 h. CONCLUSIONS: Escalation of patient observations, monitoring and medical reviews matching the BedsidePEWS is still suboptimal. Children with CHC are at higher risk of lower compliance. Impact of adherence to predefined response algorithms on patient outcomes should be further explored.