Mann, S., Bowler, M. (2008) Using an early warning score tool in community nursing. This is an extended version of the article published in Nursing Times; 104: 20, 30-31.
This article describes the development of an early warning score chart. The aim of this tool was to support advanced-level nurse practitioners’ decision-making when managing complex patients with long-term conditions at home.
The tool was developed following the launch of the urgent care team, which was established to prevent hospital admissions and allow patients with both acute and long-term conditions to be managed in the community. The article describes an audit carried out to help construct the tool, and outlines an evaluation of the tool’s effectiveness.
Sandra Mann, BSc, DipN, RGN, is nurse practitioner, Sunderland Teaching PCT; Mandy Bowler, MSc, BSc, OND, DMS, RGN, is senior nurse and business manager, NHS South of Tyne and Wear.
The NHS Plan: A Plan for Investment, A Plan for Reform(Department of Health, 2000a) brought about a new system of earned autonomy, which devolved power from the government to the local health service as modernisation took hold. This plan envisaged that the new NHS would be designed around patients’ needs, where for the first time patients would have new powers and more influence over the way the system works. Within these changes came new ways of working; one of these was the development of a nurse-led urgent care team in Sunderland.
This team began in 2004, with its main aim being to prevent hospital admissions by managing acute and long-term conditions at home. It was the first of its kind nationally and brought great challenges to tradition and professional practice.
The urgent care team (UCT) is primarily concerned with complex patients who require improvements to their disease management to reduce exacerbations of long-term illness. These patients are often described as high-risk, requiring specialist interventions to achieve optimum health, reduce risk of complications and deterioration. They also need assistance and support to prevent further ill-health (DH, 2005).
Although the decision on whether to admit a patient or keep them at home with support from the UCT was initially based on the nurse practitioner’s subjective, individual patient assessment, it soon became evident to the team that this was an area of risk which regularly exposed practitioners to the difficulties of decision-making pathways. This led to the team’s development of an early warning tool. It was designed specifically for patients in the community, to influence decision-making regarding treatment and management plans and to act as a guide for practitioners to select the most appropriate patient pathway.
The development of the nurse-led UCT embraces all the aspects and expectations of the nurse practitioner role, especially in relation to advanced nursing practice (Furlong and Smith, 2005), the ability to act autonomously (Hicks and Hennessy, 1999), to make differential diagnoses using decision-making tools and use problem-solving skills.
However, while it was readily acknowledged across the team that all practitioners are expected to make decisions for which they are accountable, difficulties could be experienced in ensuring consistency in the decision-making process as this is often open to subjectivity. This could lead to differing opinions over which pathway may be best to treat or manage a patient. It could also give rise to conflicting views or advice which seriously compromise patient confidence in the team or safety, depending on the chosen pathway.
It was therefore agreed to introduce a tool to support this decision-making process for the benefit of both patients and staff. The literature on identification of sick patients and early warning scores is well established in the critical care and outreach setting (Naeem and Montenegro, 2005). However, to date there is no evidence available on the application of such tools in primary care.
Developments in Sunderland
The early warning score (EWS) is essentially a tool that quantifies a patient’s condition by assigning a score to each set of physiological parameters. Typically these are: respiratory rate; pulse rate; systolic blood pressure; temperature; urine rate; and a measurement of the patient’s responsiveness. The tool provides a method of assessing patients’ care through calculating a score for a set of vital observations and a series of triggers that warn of actual or potential deterioration (Goldhill et al, 1999).
In the early planning stages, simply adapting the chart and trigger levels used in the local acute trust was considered. This would allow the chart and the tool to stay with the patient throughout her or his journey from home to A&E and on to the ward. However, once this was explored further, it was found that patients seen by the UCT were too sensitive to the parameters used in the acute trust, and that the tool may not be as useful in decision-making in the community setting.
Two physiological triggers causing concern were the respiratory rate and oxygen saturations. Since many patients seen by the UCT were at mid- or end-stage of COPD, they often had vital signs that were outside the normal range values. Although these patients may have restricted lifestyles they are often able to manage some activities of daily living despite these abnormal vital signs.
In order to test this hypothesis the acute trust tool was used over two days on 10 UCT patients. Vital signs were recorded against the tool. This mapping highlighted that long-term disease and ill-health triggered an outcome score on the tool that would, if used in the acute trust, indicate the initiation of a medical intervention or the step-up to a higher dependency of care.
In order to develop a tool more suited to use in the community, an audit was conducted to help determine more meaningful vital sign parameters.
One hundred patients’ notes were chosen randomly. These were patients who had either been seen regularly or as a single intervention over the previous two years. The vital signs recorded were: systolic blood pressure; pulse rate; respiratory rate; oxygen saturations; urine output; and temperature.
Patient outcome was also recorded to identify whether the patient had stayed at home, been admitted to hospital or referred to a community matron or other specialist nurse.
The results of the audit identified that not all 100 patients had complete sets of vital signs recorded. While 88% had oxygen saturations recorded, only 64% had respiratory rate and a mere 4% had urine output recorded.
Average values of each of the vital signs were recorded and compared against normal values. The average respiratory rate was 28 breaths per minute and oxygen saturations were 90%.
Using the averages, a variable grid was devised to reflect the findings. The same principles of other scoring grids were applied, that is, the further the deviation, the greater the score (see Fig 1 for the scoring chart).
As well as physiological variables, the team felt there were other factors which needed to be taken into consideration as areas that influenced decision-making. Assessing the clinically unwell patient in their own home requires a holistic approach. Furthermore, assessing an individual’s ability to manage activities of daily living during the period of illness was essential.
The decision to include ‘gut feeling’ based on nurse intuition is a debatable one. While the tool’s aim was to support consistency in patient management and reduce risks to patient safety or team credibility, it was acknowledged that experience and intuition play an important part in nursing acute patients at home. The team, like Miller (2000) and Benner (1984), believed that intuitive judgement distinguishes the expert from the novice, with the expert no longer relying on the analytic principles to connect their understanding of the situation to the appropriate action.
Since the tool was to act as a guide rather than a definitive decision-making tool, the team felt it important to include gut feeling as a valuable score. However, this forms part of the overall score and does not stand alone. The vital signs are the trigger.
Other additional triggers included failure to improve despite treatment. This is especially useful when reassessing a patient after initial first-line treatment, for example, after a single dose nebuliser for breathing difficulty. If there is no improvement then this would trigger a warning.
The final trigger is for patients with multiple conditions or co-morbidities. This category is at higher risk of repeated admissions to hospital due to some forms of degenerative disease coupled with long-term conditions (DH, 2006).
An action plan was devised based on the findings of patient outcome from the audit. This action plan is available on the opposite side of the chart (Fig 2) and the chart is placed in the patient’s notes. The score is reassessed at each visit by the team.
Within the lower score ranges nurse practitioners are guided to refer the patient to either a community matron, district nurse or specialist nurse depending on patient needs. Bowler (2006) identified how the UCT and community matron role work together to reduce hospital admissions for patients with long-term conditions by supporting them through acute exacerbations at home. No patient who scored less than five subsequently required admission to hospital in the next 12 hours.
24/7 is another community-based nursing team which provides step-down support to patients and assists them in maintaining independence in their own home. Farnborough Court is a social services intermediate care centre that offers short-stay rehabilitation for patients until they can return home.
If no further intervention is required then the patient can be discharged back to the care of their GP.
Higher scores indicating clinical deterioration will direct practitioners to either review the patient again within eight hours or directly admit them to the clinical decisions unit in the acute trust.
Supporting the tool’s use
Throughout the planning stages, both formal and informal teaching sessions took place to encourage staff participation to influence and shape the tool as well as to share learning on how to apply and use it. The new chart would guide practitioners to specific patient outcomes. However, it was stressed that it should be used in conjunction with clinical assessments or findings and should not replace them.
All staff, particularly healthcare assistants who regularly took and recorded vital signs, were given further education and training to ensure that improvements in recordings were noted in all patients in future.
The tool provides complex assessment which is shared with other services involved in patients’ longer-term care and management, especially those with long-term disease supported by community matrons or specialist nurses.
Patients’ individual norm can be monitored and the tool applied by specialists such as heart failure nurses or community matrons when monitoring for signs of deterioration. It can be equally as valuable during communication between teams.
Implementing the tool
The tool was implemented during 2007 and since then has been used on every patient referred to the UCT. Following some early minor modifications to make the chart design more user-friendly, its implementation was successful and has consistently supported decision-making processes throughout the team.
An evaluation of its effectiveness after the first 12 months in use suggested changes in the parameters for oxygen saturation levels based on patient outcomes. Although these parameters are not consistent with other EWS tools in use in acute care, the modification has reduced over-sensitivity in community-supported exacerbations of extremely complex patients with long-term conditions.
Implications for practice
The NHS faces the challenge of responding to patients’ needs and expectations by encouraging and supporting new ways of working, and this includes patient choice. One of the biggest changes is to challenge traditional roles and views of how to deliver care by appropriately skilled practitioners.
Patients are very much at the centre of this change, and the recognition of those whose condition is deteriorating or is at risk of doing so is key to avoiding unnecessary hospital admissions and keeping care closer to home. Critical care without walls (outside specialist units) (DH, 2000b) ensures that appropriately trained, highly skilled, safe practitioners deliver the best possible care in the community.
The use of an early warning score tool in the community came with its own set of challenges. However, once overcome, these developments have provided both staff and patients with an extremely useful and efficient trigger tool. This also provides objectivity and a standard approach to decision-making regarding individual patients’ needs.
The innovative development of the urgent care team has provided new and exciting opportunities for nurses willing to embrace change and enhance their role. It has established urgent and emergency care services that are responsive and more local than the traditional hospital setting. This model has since been introduced across Gateshead PCT following the integrated management arrangements of NHS South of Tyne and Wear.
Supporting urgent care in the community means nurses have to deal with complex patients with a wide range of healthcare needs and illnesses. This requires standards of assessment and an ability to recognise and act on actual and potential deterioration in patients’ conditions.
Nurses working at an advanced level in the UCT need to demonstrate accountability and competence in decision-making regarding differential diagnosis and when to refer patients on to others. This is essential for patient safety and management.
Findings before the tool’s development suggested variance on patient pathways depending on which practitioner had carried out the assessment. This may have been attributable to the practitioner’s level of competence, expertise or perhaps a degree of risk-taking. Using this tool as a guide should support practitioners to use reflective practice, clinical supervision and constructive decision-making processes to discuss, compare and develop competencies and ensure consistencies in practice.
The introduction of an assessment standard certainly reflects a sound evidence base to influence improvements in practice. However, it is essential that practice development is a continuous process and is seen as a participative process for all involved.
This requires an affinity with transformational leadership skills, in particular the skill of helping colleagues to develop ideas and articulate and think these through.
The success of the UCT and the implementation of the tool would not have been possible without the support of all members of the team at every level including: Bev Atkinson, director of provider development; and Corrine Layton and Julie Jordan, both senior nurse practitioners, who helped to develop, test and evaluate the tool.
Benner, P. (1984) From Novice to Expert. New York, NY: Addison Wesley.
Bowler, M. (2006) Use of community matrons for care of long-term conditions. Nursing Times; 102: 33, 31-33.
Department of Health (2006) Our Health, Our Care, Our Say: A New Direction for Community Services. www.dh.gov.uk
Department of Health (2005) Supporting People with Long-term Conditions. Liberating the Talents of Nurses who Care for People with Long-term Conditions. www.dh.gov.uk
Department of Health (2000a) The NHS Plan: A Plan for Investment, A plan for Reform. www.dh.gov.uk
Department of Health (2000b) Comprehensive Critical Care: A Review of Adult Critical Care Services. www.dh.gov.uk
Furlong, E., Smith, R. (2005) Advanced nursing practice, policy, education and role development. Journal of Clinical Nursing; 14: 9, 1059-1066.
Goldhill, D.R. et al (1999) The patient at risk team: identifying and managing seriously ill ward patients. Anaesthesia; 54: 9, 853-860.
Hicks, C., Hennessy, D.S. (1999) A task-based approach to defining the role of the nurse practitioner. Journal of Advanced Nursing; 29: 666-673.
Miller, D. (2000) Leadership choices: self renewal or self neglect. Creative Health Care News; 4: 1, 1-3.
Naeem, N., Montenegro, H. (2005) Beyond the intensive care unit: A review of interventions aimed at anticipating and preventing in hospital cardiac arrests. Resuscitation; 67: 1, 13-23.