Article Text

Download PDFPDF
Charted observations and the use of a paediatric early warning tool did not predict the majority of cardio-respiratory arrest calls in a paediatric hospital
  1. D Wright,
  2. G Sefton,
  3. M Horan
  1. Paediatric Intensive Care, Alder Hey Children's Hospital NHS Foundation Trust, Liverpool, UK


Aim To asses whether the routine use of observation charts incorporating a Paediatric Early Warning Tool (PEWT) are effective for screening inpatients, predicting those at risk of acute deterioration, in a paediatric hospital.

Method A retrospective case note audit of all cardiac arrests calls for inpatients between May and December 2009 inclusive. Case notes, observation charts and computerised patient information records were studied to analyse the 24 h period preceding the arrest call to assess whether the deterioration could have been detected earlier. We define the PEWT as ‘triggered’ when documented on the observation chart, a medical review is requested within the PEW time targets and the PEWT logged onto the hospital computerised patient information system.

Results There were 72 arrest calls during the 8 month period of which 56 met the inclusion criteria to the study. One set of notes was unavailable, 55 sets of data were analysed. The reason for arrest calls were: apnoeas (n=15), sudden desaturation (n=15), airway obstruction (n=6), increased work of breathing (n=8), cardiac arrest (n=6), sudden bradycardia (n=3) and seizures with desaturation (n=2). Of 55 arrest calls analysed, 10 (18%) patients triggered the PEWT in the preceding 24 h. Six of these had an arrest call within 30 min of triggering PEWT. Of the 45 patients who did not trigger the PEWT, 17 (38%) patients should have triggered the tool on documented observations. Although the PEWT was not triggered, staff were concerned enough to request a review by the medical team for 10 of these 17 (59%) patients. In total only 27 (49%) of cardiac arrest calls could have been predicted by either observation charts or the PEWT. Of the cardiac arrest calls 28 patients (51%) remained on the ward after intervention. Two (4%) resuscitations were unsuccessful. 25 (45%) were transferred to PICU and 3 later died.

Conclusions The use of observation charts incorporating a PEWT was only 49% sensitive at predicting cardiac arrest calls inpatients in a tertiary paediatric hospital. This sensitivity is being reduced further by incorrect implementation of the PEW process when the tool is ‘triggered’.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.