Read Detecting and managing deterioration in children text version


Detecting and managing deterioration in children


Alan Monaghan describes how the introduction of a critical care outreach service and a Paediatric Early Warning Score improved management of acutely ill children

Alan Monaghan RN (Child), RGN, PGCHSCE, Lecturer Practitioner, Royal Alexandra Children's Hospital, Brighton

In the past, paediatric intensive care units (PICUs)

were perceived by many as ivory towers. They often had poor relationships with other departments probably caused by them working in isolation, by poor communication and lack of admission and discharge planning. Similar issues in adult critical care resulted in the Department of Health recommending a hospitalwide approach to the identification and referral of critically ill adults (DoH 2000). This led to the development of critical care outreach teams in adult services and an improvement in interdepartmental working. The function of these outreach teams is to improve the provision of critical care prior to admission to

Paediatric Nursing

intensive care and following discharge. Across the UK, adult critical care services have been setting up teams under various titles such as emergency medical team or emergency outreach team. Despite the various titles the teams have three shared aims: early detection of patients at risk of deterioration support and education of ward staff in caring for patients at risk of deterioration improved discharge and follow up after intensive care. This article summarises the process of setting up a paediatric critical care outreach team at Brighton and describes the development of a paediatric early warning

vol 17 no 1 February 2005


Intensive care: Paediatric Patient assessment Outreach



score (PEWS) to assist in detecting children at risk of deterioration. The process of implementing change and the difficulties encountered when implementing new working practices are briefly considered. Data from a pilot implementation are presented to provide evidence for use of a paediatric early warning score in children.

care responsibilities had been taken away by the outreach staff. After consideration, a more educational-based model was devised. As the project lead I felt that if education and support were provided by a team experienced in caring for acutely ill children (instead of them `taking over'), staff satisfaction would improve. Staff would become more skilled in caring for these patients and would ultimately provide good continuity of care to an appropriate standard.

Early detection

Early detection and optimal care in the critically ill adult is associated with improved outcomes (DoH 2000), an association that may apply in children. In one review, 61 per cent of paediatric cardiac arrests were caused by respiratory failure and 29 per cent by shock, both of which are potentially reversible causes (Reis et al 2002). Early recognition and appropriate intervention are therefore equally important in children and may prevent the need for admission to intensive care. There are a number of scoring systems being used to identify adults at risk of deterioration but these have yet to be verified for their transferability to other subjects. Although there is a strong indication that early warning scores may have a place in the paediatric setting, a literature search did not reveal any publications on this topic. There were scoring systems in use for croup and meningococcal septicaemia but little evidence on their use and sensitivity. The use of a paediatric early warning score seems to be a relatively new concept.

An educational approach to outreach

There is evidence that outreach teams can have the adverse effect of deskilling staff resulting in a greater demand for high dependency facilities as staff become reluctant to care for these patients (Mercer et al 1999). With this in mind a series of study days were set up along with short in-house sessions in ward areas on aspects of caring for the acutely ill child. Education of nursing and medical staff was found to be of real importance in the recognition of critically ill patients by Franklin and Mathews (1994). Difficulties in recruiting qualified nurses resulted in a dilution of the skill mix in many areas in their study. Other factors affecting staff responses to deterioration were the increasing use of temporary staff and a reduction of paid study leave to gain these skills (Franklin and Mathews 1994). In addition to staff education, we needed to address approaches to identifying the child at risk of deterioration, particularly in areas that mainly admit adult patients such as accident and emergency departments in district general hospitals with limited paediatric facilities.

Project development

In February 2001, a working group was established at Brighton and Sussex University Hospitals NHS Trust to develop a score modified for children. We also investigated the possibility of applying the principle of outreach within the women and children's directorate. This was felt to be a logical way to approach the increasing problem of caring for highly dependent children in the ward area. In October 2001 a pilot of outreach began with a team comprising a clinical nurse educator lead and anaesthetic staff grades on PICU attachment after hours. At first, ward staff referred patients about whom they had concerns and intensive care staff would attend, assess and care for these children and arrange transfer to intensive care. At this time there was no formal assessment tool available. Unfortunately this approach appeared to result in an element of deskilling among ward-based staff. Following debriefing sessions, staff stated that they felt undermined and undervalued as

vol 17 no 1 February 2005

Developing the early warning score

A multidisciplinary planning group was set up with representation for all who would be involved in caring for these patients. The group members gave their ideas on why staff generally had concerns regarding a patient's condition and the markers used to judge severity. It was decided to adopt an early warning score as an aid to help nurses assess patients objectively using vital signs. As no paediatric scoring system could be found at the time, we decided to develop a paediatric score based on the available adult systems. One concern was that some adult systems use blood pressure as one of the main predictors of deterioration. In children, hypotension is associated with late signs of shock and is a

Paediatric Nursing 33


Fig 1. Royal Alexandra Hospital For Sick Children, Brighton ­ Paediatric Early Warning Score


Behaviour Playing/ appropriate Cardiovascular Pink or capillary refill 1-2 seconds






Lethargic/confused Reduced response to pain


Pale or capillary refill 3 seconds

Grey or capillary refill 4 seconds. Tachycardia of 20 above normal rate

Grey and mottled or capillary refill 5 seconds or above. Tachycardia of 30 above normal rate or bradycardia.


Within normal parameters, no recession or tracheal tug

>10 above normal parameters, using accessory muscles, 30+% Fi02 or 4+ litres/min

>20 above normal parameters recessing, tracheal tug. 40+% Fi02 or 6+ litres/min

5 below normal parameters with sternal recession, tracheal tug or grunting. 50% Fi02 or 8+ litres/min

Score 2 extra for 1/4 hourly nebulisers or persistent vomiting following surgery

potentially a pre morbid sign: according to Hazinski (1992) it is a sign of decompensated shock and multi system failure. We also realised that if we used a system that relied on vital signs parameters alone, three or four different scores would be needed to allow for different age variables. The development proved difficult because of this factor. There is a danger in implementing a very complex scoring system or a number of systems as staff would find it difficult and time consuming to carry out their assessment. Any clinical scoring system should be easy to use and not open to different interpretation by different users. It was hoped that the paediatric early warning system would be used by all grades of staff and that it would not generate too much extra work.

signs are scored exactly as observed so the child who is uninterested in his or her surroundings would score three ­ lethargic. We initially considered including more detail in the behavioural assessment but this would have made the score too complex and open to misinterpretation. Colour and capillary refill were chosen to assess cardiovascular signs rather than mean arterial blood pressure. Both signs are used as not all staff are skilled in assessing capillary refill. Respiratory rate was included along with oxygen demand. Goldhill et al (1999) found that respiratory rate and adequacy of oxygenation were important physiological indicators of a critically ill ward patient and could be assessed without special equipment. This removes any reliance on equipment such as saturation monitors being available. Mean respiratory parameters are used in order to increase sensitivity. Having assessed the parameters (see Figure 1) the nurse calculates the child's total score, which dictates one of four actions: informing the nurse in charge increasing the frequency of observations calling for medical review and informing the outreach team calling out the full medical team and outreach team If the child's score was in the red column, or greater than four, the protocol recommends calling out the full

vol 17 no 1 February 2005

The scoring system

Each clinical area was assigned a link person from the planning group to promote the score and give feedback on its use. This helped in cascading the information and aided implementation. The scoring system focused on three components of assessing a child: behaviour colour/cardiovascular status respiratory status. Behaviour was felt to be an important observation criterion as it is often an early sign of a shocked child and something the parents may also recognise. Behavioural

34 Paediatric Nursing


team. This action could be adapted by other units taking into account the local facilities and resources. For example in the absence of on-site paediatric intensive care support, a red score or four score could be adapted to involve consulting a lead centre. Several studies have found that nurses and junior medical staff often fail to notify the senior physician of clinical deterioration leading up to cardiac arrest (for example Franklin and Mathews 1994). It was hoped that the paediatric early warning score (PEWS) would prompt action. It would also provide an objective assessment tool to prevent contributing factors from affecting judgement ­ such as a busy ward. The PEWS also allows the nurse to call medical staff without having to give a lengthy justification over the phone which could waste valuable time.

involved the outreach co-ordinator. Eighty three per cent of the patients improved following intervention; 17 per cent deteriorated and required admission to the PICU. Staff identified three children who were unwell and required medical intervention during the pilot and who they thought should have scored higher. Analysis of these children's audit forms and observations showed that they had all developed a tachycardia prior to their deterioration. It was also noted that certain patients, such as children who have had a tonsillectomy, may not display classic signs during the early phases of deterioration. Prolonged vomiting was felt to be a more prominent sign of bleeding in these cases and was added as a factor to score. A subsequent audit found a direct relationship between the PEWS and the recording of respiratory rate: 80 per cent of children without a recorded PEWS also had no respiratory rate recorded. Goldhill (2000) states that sensitivity, specificity and usefulness for this type of score in adults has yet to be demonstrated; it is important to evaluate the validity and reliability of the PEWS. We plan to test inter-rater reliability of the PEWS among all grades of staff and a study is underway to evaluate the sensitivity of the PEWS in reflecting the child's severity of illness. Staff experience of both the PEWS system and the outreach service was very positive: a survey of staff experience found that out of 33 staff on acute medical and surgical wards, 88 per cent felt that the outreach service increased their confidence in caring for the child at risk of deterioration. Eighty per cent reported that the PEWS system improved their confidence in recognising the child at risk of deterioration.


Department of Health (2000) Comprehensive Critical Care ­ A Review of Adult Critical Care Services. London, DoH. Franklin C, Mathews J (1994) Developing strategies to prevent in hospital cardiac arrest: analysing responses of physicians and nurses in hours before the event. Critical Care Medicine. 22, 244-247. Goldhill DR (2000) Medical emergency teams. Care of the Critically Ill. 16, 209-212. Goldhill DR et al (1999) Physiological values and procedures in the 24 hours before ICU admission from the ward. Anaesthesia. 54, 529-534. Hazinski FM (1992) Nursing Care of the Critically Ill Child. 2nd edition, pg 181. St Louis, Mosby. McQuillan P et al (1998) Confidential inquiry into the quality of care before admission to intensive care. British Medical Journal. 316, 1853-1858. Mercer M et al (1999) Medical emergency teams improve care. British Medical Journal. 318, 54. Reis A et al (2002) A prospective investigation into the epidemiology of in-hospital paediatric cardiopulmonary resuscitation using the international ustein reporting style. Paediatrics. 109, 200209.

Implementation and audit

The initial introduction of the scoring system was met with a variable response. Some staff could not see why we needed a score as they felt they were quite capable of recognising patients at risk. Staff were also concerned about the assessment being time consuming and adding extra paperwork. We timed how long it would take to score a patient and found it to be about 30 seconds on top of a standard set of observations. This time was found to decrease as the nurse became more familiar with the system. The standard four-hourly observation charts did not allow for different timings of observations and recording of PEWS. Incorporating the score into the standard observation chart prevented the need for additional paperwork. This proved to be popular with most staff as it provided more flexibility for documentation.


The PEWS was piloted for three months and patients were reviewed using an audit tool which captured the patient's observations, PEWS score, who was called, what actions were taken and the outcome. We also audited children who should have scored highly but did not (such as children who required intensive care admission with PEWS scores below four). During the pilot there were 30 patients who scored four, warranting a call for medical staff to review the patient's condition. Ninety six per cent were seen within 15 minutes and the same percentage required medical intervention. In 54 per cent of these cases, care

vol 17 no 1 February 2005


The use of the Brighton PEWS system is now standard practice in our trust and we have used it for several years to assess acutely ill children. Along with the development of the outreach team, this has meant that children showing signs of deterioration are assessed by appropriate medical and nursing staff and receive optimum care during their acute episode. As our audits show, staff in ward areas have an increased level of confidence in caring for the acutely ill child. Further work is required to show the benefits of the outreach team in order to secure funding for what staff see as a valuable service PN

Paediatric Nursing 35


Detecting and managing deterioration in children

4 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

Detecting and managing deterioration in children