Original articleWhat impact did a Paediatric Early Warning system have on emergency admissions to the paediatric intensive care unit? An observational cohort study
Introduction
The reported incidence of cardiopulmonary arrest in hospitalised children is low: 0.2–2.5/1000 (Berg et al., 2008). However, mortality (52%) and morbidity remain high despite advances in resuscitation training, technology and treatment (Tibballs et al., 2005). The devastating physical consequences of deterioration leading to cardiopulmonary arrest are well documented (Meert et al., 2009), as are the psychological effects that emergency admission to the Paediatric Intensive Care Unit (PICU) has on the family (Balluffi et al., 2004). There are significant additional financial costs to the National Health Service of ‘failing to rescue’ deteriorating children in hospital (Duncan and Frew, 2009). Therefore, there is a pressing social and financial need to improve the early identification and treatment of deterioration in hospitalised children.
Research in adults demonstrated that cardiopulmonary arrest or emergency admissions to Intensive Care were often preceded by a period of physiological instability which, once recognised, provided an opportunity for the healthcare team to intervene to improve outcome (Hodgetts et al., 2002, Kause et al., 2004). A similar window of opportunity may exist for hospitalised children (Haines, 2005, Tasker, 2005, Tume, 2004, Tume, 2006).
Paediatric Early Warning systems (PEWs) have been identified as a mechanism to improve safety for hospitalised children (CEMACH, 2008). To date, published studies evaluating PEWs have focused on the performance of individual tools in single centres and their impact on the incidence of respiratory or cardiopulmonary arrest. However, this does not capture other sick children admitted to the PICU as emergencies following acute deterioration or the impact that PEWs has on PICU service delivery.
Section snippets
Study objectives
The study objectives were
- 1.
To understand how the introduction of a Paediatric Early Warning system at a tertiary children's hospital affects emergency admissions to the PICU.
- 2.
To compare the ‘in-house’ cohort of emergency admissions to PICU with a comparable group; emergency admissions transferred to PICU from wards at District General Hospitals (without PEWs in place).
- 3.
To explore the impact that a PEW system had on PICU service delivery.
Setting
The setting was a tertiary children's hospital in the Northwest of England, with 337 in-patient beds, handling 37,000 annual admissions (excluding day-cases). Tertiary specialties include cardiology/cardiac surgery, neurology/neurosurgery, renal, oncology, burns/plastics and neonatal surgery. The hospital has a 24 hour emergency department, a 22 bed PICU and two separate High Dependency Units (total beds 21). The PICU admits 1100 patients annually. Half of those admissions are elective,
Study design
This is an observational cohort study, conducted at a large tertiary children's hospital in the United Kingdom. A summary of the study design is presented in Table 2
Patient-specific information and source data is routinely collected prospectively for all PICU admissions for the Paediatric Intensive Care Audit Network dataset (PICANet) (Table 2). The data is rigorously checked by random case quality control each month and is considered to be both accurate and reliable. Formal permission was not
Results
A total of 958 unplanned PICU admissions over two years were reviewed, for one year before PEWs introduction and one year afterwards.
The impact that PEWs had on PICU service delivery
PICU service delivery is affected by the number of patients admitted as emergencies, their severity of illness and their length of stay. The effect of mismatched PICU bed availability and demand include cancellation of major elective surgical cases at short notice or refusal of external referrals for PICU. This has clinical and cost implications. If there are no available PICU beds in the region, the critically ill child would have to be transferred out of region, with the added clinical risks
Discussion
In the United Kingdom, the National Patient Safety Association (2009), Confidential Enquiry into Maternal and Child Health deaths (CEMACH, 2008) and NHS litigation authority (NHSLA, 2012–2013) have advocated for the widespread use of PEWs, to assist staff in the early identification of children who may be deteriorating. This advice is made based on consensus opinion and as yet the evidence to support this is limited.
In practice, it is recognised that evaluating a complex intervention like PEWs
Limitations
This is a small before-and-after observational study undertaken at a single centre. We acknowledge the limitations of this study design, as described by Joffe et al. (2011) but have attempted to address these by using a comparison group. No attempt was made to test the performance of the individual PEW criteria during this study, or the local compliance with the PEW process, which may have impacted on its effectiveness. There may have been inaccuracies in the timing of the external referral to
Conclusion
Although there is limited evidence of effectiveness for using PEWs, conceptually it makes sense that if deteriorating children could be identified earlier, they would be easier to rescue. Indeed, there is no published evidence of harm from using PEWs. Despite its limitations, this study showed that the implementation of PEWs at a tertiary paediatric hospital reduced the likelihood of dying during the PICU admission, the requirement for PICU interventions and the PICU length of stay for
Contributions
Concept, study design G.S. and C.M. Data collection and analysis G.S., C.M. and S.L. Manuscript preparation G.S., L.T., S.L., P.L. and E.C.
Conflict of interest
None declared.
Acknowledgement
The authors would like to thank Elaine Scott for helping prepare the PICANet data for transfer to SPSS for further analysis.
This project to introduce a PEWs using existing resources to respond to deteriorating patients, won the BUPA patient safety award for 2007.
References (27)
- et al.
In-hospital pediatric cardiac arrest
Pediatric Clinics of North America
(2008) - et al.
Short-term health system costs of paediatric in-hospital acute life-threatening events including cardiac arrest
Resuscitation
(2009) - et al.
Physiological abnormalities in early warning scores are related to mortality in adult inpatients
British Journal of Anaesthesia
(2004) - et al.
Incidence, location and reasons for avoidable in-hospital cardiac arrest in a district general hospital
Resuscitation
(2002) - et al.
A comparison of antecedents to cardiac arrests, deaths and emergency intensive care admissions in Australia and New Zealand, and the United Kingdom—the ACADEMIA study
Resuscitation
(2004) A structured approach to the seriously ill child
- et al.
Traumatic stress in parents of children admitted to the pediatric intensive care unit
Paediatric Critical Care Medicine
(2004) - et al.
Systematic review of paediatric alert criteria for identifying hospitalised children at risk of critical deterioration
Intensive Care Medicine
(2010) - et al.
Developing and evaluating complex interventions – new guidance. Medical Research Council (MRC)
Critical Care
(2008)