Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Nursing observations on ward patients at risk of critical illness

  • Comment

VOL: 98, ISSUE: 46, PAGE NO: 36

Annie Chellel, MA, PGDE, BSc, BA, RNT, RGN, outreach at the Kent and Sussex Hospital, Maidstone and Tunbridge Wells NHS Trust

Jayne Fraser, BSc, RGN andVeronica Fender, BSc, RGN, outreach at the Kent and Sussex Hospital, Maidstone and Tunbridge Wells NHS Trust;Debbie Higgs, BSc, RGN,Stefa Buras-Rees, RGN,Lorraine Hook, RGN, RSCN, DipN andLucy Mummery, RGN, outreach at East Kent Hospitals;Claire Cook, RGN andSara Parsons, RGN, outreach at Darent Valley Hospital, Dartford;Claire Thomas, RGN, outreach at Medway Maritime Hospital, Chatham

Research indicates that ward care of critically ill patients is suboptimal (Garrard and Young, 1998; McQuillan et al, 1998; McGloin et al, 1999) and suggests that the signs of impending critical illness or cardiac arrest are being missed (Franklin and Mathew, 1994; Schein et al, 1990).

Respiratory rate is a key predictor of both cardiac arrest and admission to intensive care (Fieselmann et al, 1993; Goldhill et al, 1999), yet outreach teams have become aware, through anecdotal evidence, that ward nurses often fail to record this vital sign.

Translocation of chemical mediators from the gut is also recognised as a contributing factor in the development of sepsis (Thelan et al, 1998; Kompan et al, 1999) and so hyponutrition in critical illness may be considered as another factor in suboptimal care (Moore et al, 1992; Heyland et al, 1993; Griffiths, 1997).

Outreach teams were set up across the country in response to Comprehensive Critical Care: A Review of Adult Critical Care Services (Department of Health, 2000). The aim was to extend the principles of critical care treatment beyond intensive care units (ICUs) and high dependency units (HDUs) and to ensure that critically ill patients receive an appropriate level of care, regardless of their location.

Outreach teams, which should consist of nurses, doctors and physiotherapists with critical care knowledge and experience, were given three specific objectives by the DoH:

- ’To avert admissions by identifying patients who are deteriorating and either helping to prevent admission or ensuring that admission to a critical care bed happens in a timely manner to ensure best outcome;

- ’To enable discharges by supporting the continuing recovery of patients on wards postdischarge from hospital and their relatives and friends;

- ‘To share critical care skills with staff in wards and the community to enhance training opportunities and skills practice and to use information gathered from wards and the community to improve critical care services for patients and relatives.’

The document classifies patient care into four levels (Table 1), which reflect the severity of illness and patient dependency instead of the ‘geographical’ description of patients that was used previously. For example, ward patients are now described as being levels 0 and 1, HDU patients are level 2 and ICU patients are now level 3.

Premature discharge from ICU results in poor outcomes for level 3 patients (Daly et al, 2001) and this has resulted in a cautious approach to ‘stepping down’ from level 3 to level 2 to ward care. HDU and ICU beds in Kent are located in the same unit which ensures that they are used flexibly. However, because ICU bed occupancy in Kent is around 95 per cent (according to an unpublished audit by Medical Associated Software House Ltd), this means HDU bed occupation is dominated by recovering ICU patients.

It is not known what proportion of ward patients are also critically ill, at levels 1, 2 or even 3, but cannot gain access to a critical care bed. The level of this unmet demand for critical care remains unquantified.

Kent has outreach teams practising in all four trusts although not in all seven hospitals. Outreach nurses in the south east had been meeting regularly for the purposes of sharing good practice, mutual support, debate and communication.

The groups have now become part of the critical care network created by the NHS Modernisation Agency. The outreach nurses in the Kent group have established specific aims, including research and audit activity to inform the future development of critical care provision.

Through discussion it became apparent that the problems faced by the critically ill ward patient are shared across the country. Intensive care nurses visiting patients on the wards were discovering that level 2 and even level 3 patients were being nursed on the wards and that the detailed observation and interpretation of vital signs carried out in intensive care was not evident in the wards, even for patients who were very sick.

Aims of the survey

The survey of outreach nurses in the Kent group of trusts was carried out to provide a snapshot of the situation on the wards for those patients who are, or who are at risk of becoming, critically ill. The survey aimed to establish:

- The number of patients at each level of care as a crude indicator of the demand for: critical care at level 2 and outreach services;

- The level of observation of basic indicators of critical illness being carried out on ward patients above level 0 who are at risk of critical illness or whose condition is deteriorating;

- The nature of outreach services being offered in Kent.

Collected data would be pooled so that individual hospitals, wards, outreach teams and trusts would not be identifiable.

The survey

Three sets of data were identified:

- Data about the number of patients at each level of care during the specified week in November 2001 when the survey was carried out;

- Data about the basic nursing observations and nutritional care being carried out on those patients who were defined by the outreach team as level 1, 2 or 3 (in other words, those requiring support above that normally offered on the ward). This data concerned the recording of the patient’s respiratory rate, oxygen saturation (SpO2) and inspired oxygen percentage (FiO2), the feeding of patients, and the recording of fluid intake and output. In the absence of fluid balance charts, the outreach nurse collecting the data was required to make a professional judgement on whether one was needed;

- Data about the nature and variation of the outreach service offered by each trust including the number and grade level of nurses employed, the number of hospitals covered by outreach and the hours worked by outreach nurses.

The first and second sets of data were gathered using a standard form during a randomly selected five-day period in November 2001. Each outreach team in Kent collected the data as part of their daily trawl in which they visit each ward and ask the senior nurses to identify the patients whose condition is a cause for concern, or who require outreach support. The nature and purpose of the survey was explained to the senior ward nurse and permission obtained for collecting data from the ward observation charts.

Following discussion with the senior nurse, all patients at level 0 were identified, counted and excluded. Patients at levels 1, 2 or 3, were assessed by the outreach nurse and their level of care confirmed.

The outreach nurse examined ward documentation of observations on all these patients and details were recorded on the standard form. This noted whether the patient’s respiratory rate, oxygen saturation and inspired oxygen had been recorded in the previous eight hours.

If a fluid chart was being used, the outreach nurse checked whether it was complete and whether the patient had been fed within the previous 48 hours. If there was no fluid balance chart, the outreach nurse, who was collecting the data and assessing the patient, decided whether the patient needed one.

Any observations carried out by outreach nurses as part of their assessment process were not included in the survey and outreach support was carried out if necessary. If an urgent intervention was required it was agreed that the survey would be abandoned and continued at a subsequent visit to the ward. In the event, none of the data collectors had to abandon the survey for this reason. ICUs, HDUs, A&Es, paediatric wards and obstetric units were not included in the survey. The third set of data about outreach provision was gathered at the meeting.

Survey findings

A total of 1,873 ward patients (from 82 wards in four trusts) were included and their level of care assessed. Of these, about 88 per cent (1644) were level 0 patients; 10 per cent (191) were level 1; less than 2 per cent (35) were level 2; and less than 1 per cent (only three patients) were considered to be at level 3 (Table 2). There were 229 patients (12 per cent) above level 0 whose care needs were in excess of those normally managed on the ward.

Of the 229 patients above level 0 (those who require close observation for signs of critical illness), 127 (55 per cent) had not had their respiratory rate recorded in the previous eight hours and 83 (36 per cent) had not had their oxygen saturation recorded in the previous eight hours (Table 3). For 62 patients (27 per cent) there were no records of the inspired oxygen concentration they were being given.

Some 123 patients (54 per cent) were on a fluid balance chart, but 51 of these had charts that were incomplete. Thirty-nine patients who needed a fluid chart, as assessed by a critical care nurse, did not have one. The survey also found that 60 patients (26 per cent) had not been fed in the last 48 hours (Fig. 1).

Five nurse-led outreach teams in Kent cover six hospitals. One team is supervised by a nurse consultant in critical care, one by an H-grade clinical nurse specialist, and the others are supervised by F-grade or G-grade sisters. All nurses have a background and/or specialist qualifications in intensive care. The outreach teams in Kent visit the wards every day and all but one promotes the use of an early warning or patient at risk scoring system as a tool to assist nurses in assessing the level of critical illness. Each scoring system is slightly different having been adapted from published scores to meet local needs.

All teams have established good multidisciplinary relationships and subsequent referral is made following the initial assessment by outreach nurses. Only one outreach team offered 24-hour cover seven days a week at the time of the survey. The remaining teams offer a Monday-to-Friday service, with one team offering a service on Saturday. The hours covered are from 9am to 6/7pm. Outreach teams hand over to night practitioners who have traditionally supported ward nurses in caring for the acutely sick.


Of the 1,873 adult ward patients in acute hospitals included in this survey, 12 per cent required more care and support than would normally be available at ward level. Around 2 per cent of these patients were at levels 2 and 3 (the criteria for a critical care bed), suggesting that the unmet demand for critical beds may be low.

The limitations of a snapshot survey must be considered. It is impossible to say whether the sample used is representative of the adult ward population, or whether the week in which the survey took place was representative of the hospitals’ level of activity.

The definition of the levels of care that a patient required was determined by discussion between a senior ward nurse and an experienced critical care nurse and therefore represents shared understanding of local practice and professional consensus rather than a precise, objective definition. It was agreed within the outreach group that this was a good way of defining patient level given the generic nature of the loose definitions provided by the DoH document concerning critical care (DoH, 2000).

The findings reveal that basic nursing observations on the sickest ward patients are incomplete. Such omissions imply that basic and easily obtainable indicators of impending critical illness and deteriorating respiratory function are going undetected.

This places the patient at risk of further deterioration which might have been prevented with timely medical intervention, and supports evidence from the literature that ward care is suboptimal (McQuillan et al, 1998). The reasons behind these omissions cannot be inferred from this survey and more sophisticated research is required to explain this phenomenon.

The findings of this survey suggest that the greatest risk to the critically ill ward patient is not the lack of critical care skills in ward-based nurses, as implied by the Comprehensive Critical Care document. Instead it is a lack of basic nursing observation and patient assessment being undertaken on the ward.

The dramatic shift in nurse-patient ratios as the patient steps down from level 3 care to HDU and then to the wards is the most striking difference between critical care and ward environments. This is a factor which may explain the findings of this survey.

In the ICU, the nurse-patient ratio is 1:1 and in the HDU it is 1:2. On the wards, the ratio slips to approximately 1:15. Even this is an optimistic estimate, since more than one outreach team involved in the survey reported that one qualified nurse to 30 patients at night was not an exceptional occurrence. If outreach is to offer a service to the critically ill ward patient that reduces the impact of this sudden shift, a ratio of 1:10 would seem to be appropriate, being approximately half-way between the HDU and the ward ratio (Table 4).

Assuming that the survey was representative - with 12 per cent of ward patients requiring outreach support - and that a ratio of one outreach nurse is needed for 10 ward patients at level 1 or above, it is possible to suggest a formula for calculating the number of outreach nurses needed in an acute hospital:

No of acute adult ward beds X 0.12

A hospital of 250 acute beds might require three outreach nurses. This is speculative but because outreach is an educative, advisory and supportive service, adjustment does not need to be made to account for 24-hour cover. The number would relate to the number of nurses needed to run a daytime outreach service.

In Kent the number of outreach nurses in relation to the number of patients above level 0 in the survey gives a ratio of 0.06 (Table 4) indicating that there are not enough outreach nurses in Kent to provide a service which improves the nurse-patient ratios for critically ill ward patients, although this varies between trusts and between hospitals.

Conclusion and recommendations

This survey suggests that the number of ward patients at level 1 or above is 12 per cent. Only 2 per cent of patients were at level 2 and 3. Research on a larger scale is needed to establish if the findings of this survey are representative.

The severity of illness and dependency level of patients on the wards must be investigated and measured if we are to establish the demand for outreach services and the level of unmet need for level 2 critical care beds. The critical care networks could be used to extend this survey to a national level.

Deficiencies in basic nursing observation of respiratory function and fluid balance were revealed in this survey, confirming the anecdotal evidence received from outreach nurses. Inadequate nurse-patient ratios on the wards are most frequently blamed for this.

The findings of this survey offer a basis for calculating the number of outreach nurses needed, based on the percentage of patients who are above level 0.

Suboptimal ward care has been researched and debated from the medical perspective. This survey suggests that nursing observation is also suboptimal, creating a further risk to the critically ill ward patient. This is a serious threat to the safety of ward patients because without basic observation data, assessment by nurses cannot take place, early signs of impending critical illness are missed and the triggers for summoning early medical intervention are not acted upon.

Inadequate nurse-patient ratios may be the underlying cause, but nursing research is needed to find out why ward nurses are not assessing the physical condition of their patients and failing to record basic observations such as respiratory rate. This has implications not only for the quality of care given to patients, but also for nurse education. The findings of this survey suggest that the most significant impact of critical care outreach may be its educative role, making quantitative intensive care measures irrelevant. Outreach may improve the care of the critically ill ward patient, but do so indirectly, through teaching physical assessment skills and promoting the accurate and frequent observation of basic vital signs.

There is an urgent need for multidisciplinary research to investigate the nature, extent and causes of suboptimal ward care. The problems of caring for critically ill patients on the wards must be clearly understood, so that educational institutions, and medical and nursing teams can support the principles of comprehensive critical care. Research needs to be coordinated nationally and commissioned by the DoH to ensure that it provides reliable and valid evidence as a basis for improving services for critically ill patients who are being cared for on wards.

  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.