What is deterioration?

Deterioration is the progressive worsening of the physiological condition of a patient. Failure to detect or act on preventable deterioration is a source of harm in clinical settings.

It is known that some children who die or deteriorate unexpectedly in hospitals will have observable features before the seriousness of their condition is recognised. It has been estimated that approximately one in five children who die in hospital have avoidable factors contributing to their death, with up to half having other potentially avoidable factors.9 A previous confidential enquiry into childhood death concluded ‘there should be ways of telling if something is wrong with a child as early as possible, for example, an early warning scoring system’.9

Unique challenges of children

Children are a very different population group when compared with adults. Therefore, we cannot directly replicate NEWS2, but there are similarities and learning that can be shared. Children and young people (CYP) are a complex population because they:

1. Exhibit a vast change in physiology between birth and adulthood.

2. Tend to compensate physiologically, making it difficult to spot the acutely ill child early.

3. Often decompensate very rapidly.

4. May not communicate their symptoms as effectively as adults.

5. Are usually accompanied by parents/carers who possess unique insight into their child’s well-being.

6. Have lower morbidity and mortality rates compared with adults.

Several issues emerge when considering the impact of an intervention designed to improve the response to deterioration. First, all types of deterioration need to be considered. Recent national policy focused on sepsis,10 but deaths from asthma and other diseases continue to occur.11 Second, clinical judgement and parent/carer concerns must be able to override any score and prompt escalation where appropriate. Evidence and case reviews highlight that inability to respond to concerns are often involved in missed deterioration. Third, nearly half of hospitals in England operate paper systems.12 While there is a clear national direction towards implementation of digital care records, paper systems still need to be in place for organisations who are still on this journey, or when digital processes fail.

What are PEWScores and PEWS (systems)?

Paediatric Early Warning Scores (PEWScores) describe any tool collating vital signs and other observable or measurable characteristics producing thresholds for response. This may be a numerical score based on the individual components summated into an overall number (with the score determining the level of escalation required), or trigger based with any one component exceeding predefined thresholds. Score and trigger-based tools are sometimes brought together under the collective title of ‘track and trigger’ tools.

Responding to a track and trigger alert can be an action for an individual (via the child’s normal clinical team) or by a hospital wide team such as a rapid response team (RRT). The RRT comprises individuals, who may not be involved in the normal care of the deteriorating patient, but have specialist skills in these situations.

PEWS (Paediatric Early Warning Systems) describe the entirety of the process including PEWScores if present and the response (single clinician or RRT) and organisational polices surrounding escalation and communication.

Standardised capture and record of the child’s observations and evaluating them to inform the wider PEWS process is critically important. This could fall between score and system, but is considered here because it is often undertaken by junior staff, who will often move between different centres. There are examples of Human Factors research demonstrating that minor variations in the dataset, layout (on paper or digital systems) or methodology in such circumstances could lead to failure.13

What is the evidence for PEWS?

Chapman et al 14 examined paediatric alert criteria, advising that studies were needed to determine which physiological parameters or combinations best predicted serious adverse events and a subsequent review of Track and Trigger systems by Chapman15 found considerable variation in their performances. A more recent systematic review16 found limited evidence that PEWS incorporating a paediatric track and trigger tool reduced deterioration or death in practice. It did highlight the potential for track and trigger systems benefit communication, teamwork and empowerment of junior staff. Standardising PEWS, with uniform data collected across hospitals and subjected to machine learning, will enable exploration of these possible benefits. Child mortality is rare and is not a useful surrogate outcome for the success of a PEWS. The EPOCH study,17 the first multicentre study of a PEWS concluded ‘implementation of the Bedside Paediatric Early Warning System compared with usual care did not significantly decrease all-cause mortality among hospitalized pediatric patients’ but there were trends in earlier intensive care admission. It is necessary to ensure the national SPOT programme develops a constellation of factors to support detection, planning and action in response to deterioration across all settings; these three components derived from a systematic review of factors enabling the effectiveness of systems to detect deterioration in children and adults.18

Current use of PEWS

A survey of organisations providing services to children in ED, assessment, or inpatient settings in 201119 showed use of PEWS, in the Great Britain had increased from 21% in 2005 to 85%.20 In 2018/2019, a national survey21 generated a snapshot of current PEWS use and RRTs. In this survey, 100% organisations in England reported using a PEWS chart (paper or electronic). The survey found wide variation across the country of the chosen chart model, scoring system, make-up of the RRT and age segmentation. However, there were similarities in some parameters scored—heart rate and respiratory rate contributed to the score in all the PEWScores. Respiratory effort, oxygen therapy, oxygen saturation and abnormal consciousness were in approximately 75% of PEWScores.

The survey found a plethora of other practices, tools and initiatives to detect and escalate management of deterioration in children. These included simulation training, safety huddles, learning from incidents and successful episodes of care. Since the 2011 survey at least 25% of organisations in England have introduced electronic PEWScores (e-PEWS) with a further c. 30% planning to implement e-PEWS within a year. With over half of the organisations using e-PEWS shortly, it is critical to develop an electronic format alongside a paper-based system.

PEWS charts variation

Using a purposive sampling technique, a selection of PEWS charts from around the UK were brought together. The critical findings from this review are framed in table 1.

Of note NEWS2 for adults starts at 16 years of age and is used on adult wards.6 Children transition into adult services between 16 and 25 years old, depending of specialty, service and organisation. To reduce human error, we propose a pragmatic approach, to solely use PEWS on paediatric wards (even if the young person is over 16 years of age) and NEWS2 on adult wards. It was also noted that many neonatal departments use the NEWTT tool22 to cover the first month of life. An overview of the acute deterioration response tools used at different stages of life is summarised in figure 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

A summary of early warning systems used at different ages throughout the life course. CQUIN, Commissioning for Quality and Innovation; NHSE/I, NHS England and NHS Improvement; RCN, Royal College of Nursing; RCPCH, Royal College of Paediatrics and Child Health.

The case for standardisation

There are a number of arguments in favour of a standardised systems. First Scotland, Ireland and Northern Ireland have implemented systems without evidence of harm. The failure of England to do so would put children in an iniquitous position compared with children in devolved nations and adults. Second, the majority of parameters used have greater similarity than dissimilarity, making the move to a single system potentially less complex. However, it is acknowledged there will be significant local challenges in deimplementing current systems and these need both local and national support. Third, while child mortality is lower than that in adults, it is expected that there will be long-term gains in reduced morbidity and mortality from a standardised system with a national dataset calibrated by continuous analysis and adjusted in real-time through algorithmic developments.

Development of an e-PEWS model ties in with the NHS Long Term plan23 of embedding digital services. With standardisation, system-wide data collection and system interoperability, individual and collective physiological and other results will be visible across the system will and follow the patient from one care setting to the next

To aid strategy and deployment, three stages of PEWS maturity have been identified (outlined in table 2): the basic levels (1+2) should be achievable by all providers of acute paediatric services, if not already in place. Level 3 will require a system-wide approach across sustainability and transformation partnerships or integrated care systems, including more systematic action to identify acute serious illness in the community.

A standardised PEWS in a wider context

Given the complexity of the implementation task, the SPOT Board agreed it would first focus first on developing a system that worked in an acute inpatient hospital setting. This pragmatic approach mirrors how NEWS2 was developed. However, cross-system work will progress in parallel to ensure that emergency departments as well as community, ambulance and primary care services engage in future (contextual) development and their approaches will be aligned with the standardised PEWS approach, although not necessarily containing all components. The objective is to develop a system of scoring and escalation which delivers a flexible, but synergistic, approach to acute deterioration in CYP. In particular, it is acknowledged that mental health deterioration can exhibit in various ways, which may not necessarily be captured by a traditional PEWS approach based on physiological variables. Comprehensive sets of observations may not be suitable for other community settings but it will be important to align key components to aid communication in referral and transfer.