An audit of nursing observations on ward patients
Smith, S. et al (2008) Nursing observations on ward patients - results of a five-year audit. This is an extended version of the article published in Nursing Times; 104: 30, 28-29.
Recording vital signs is an important part of the care of patients on hospital wards but problems have been identified with the way these signs are recorded. This article describes the results of five years of annual audits carried out by Kent and Medway critical care network outreach nurses. The results demonstrate an improvement in vital sign recording.
Sally Smith, MSc, DipHE, RN, is consultant nurse, critical care outreach; Jayne Fraser, BSc, RN, is critical care outreach sister; both at Kent and Sussex Hospital, Tunbridge Wells, Kent; Catherine Plowright, MSc, BSc, is consultant nurse critical care; Louise Dennington, BSc, DipHe, RN, is outreach sister; both at Medway NHS Trust Gillingham, Kent; Paul Seymour, BSc, RN, is senior nurse, intensive care, Princess Royal University Hospital, Farnbourgh, Kent; Gemma Oliver, RN, is outreach senior nurse, East Kent Hospitals Trust, William Harvey Hospital, Ashford; Claire MacLellan, RN, is outreach sister, Darent Valley Hospital, Dartford, Kent.
Recording physiological vital signs is one of the core roles of nurses and healthcare assistants in acute care wards. Recent guidance from NICE (2007) recommended that they should be monitored at least every 12 hours in all adult inpatients.
There is evidence that ward staff fail to recognise the significance of changes in vital signs and this can result in a late referral of patients to critical care (NICE, 2007; National Confidential Enquiry into Patient Outcome and Death, 2005). However, guidance is available to enable clinicians to manage this risk (NICE, 2007; National Patient Safety Agency, 2007a).
Critical care outreach teams have been instrumental in developing and implementing track-and-trigger physiological scoring systems for staff to use in order to alert them when a patient's condition changes (Department of Health, 2003). These continue to be recommended although there is little robust evidence to support their efficacy (NICE, 2007; DH, 2005; NCEPOD, 2005). Many teams have set up training and education to help nurses and HCAs to complete timely, vital sign recording and to act on these observations (Smith, 2000).
The Outreach Nurses in Kent (ONIK) group is a Kent and Medway-wide network that aims to develop critical care outreach. It represents eight hospitals across five trusts, and two critical care networks. Seven of the hospitals are in one critical care network.
In 2001 ONIK undertook an audit of vital signs carried out in the acute wards in the Kent and Medway Critical Care Network (Chellel et al, 2002). This categorised inpatients according to DH definitions of levels of care that were first defined as a way of quantifying patient acuity (Intensive Care Society, 2002; DH, 2000). Table 1 lists these definitions; patients classified as level 1 and above were included in the 2001 audit.
Table 1. The new critical care dependency levels (ICS, 2002; DH, 2000)
|Level 0||Patients whose needs can be met through normal ward care in an acute hospital|
|Level 1||Patients at risk of their condition deteriorating, or those recently relocated from higher levels of care, whose needs can be met on an acute ward with additional advice and support from the critical care team|
|Level 2||Patients requiring more detailed observations or intervention including support for a single failing organ system or post-operative care and those 'stepping down' from higher levels of care|
|Level 3||Patients requiring advanced respiratory support alone or basic respiratory support together with support of at least two organ systems. This level includes all complex patients requiring support for multi-organ failure|
Records were checked to ascertain whether the following respiratory signs were documented in the previous eight hours:
Percentage of oxygen delivered (FiO2) - as opposed to litres of oxygen being delivered (unless a non-re-breath bag or nasal speculum was used).
Examination of fluid balance charts included the number of patients who:
Had a fluid balance chart;
Should have had their fluid input and output monitored but did not have a fluid balance chart, for example, patients who had intravenous fluids or drugs;
Had an incomplete fluid balance chart.
Records were also checked for evidence that the patients were being fed in the previous 48 hours.
Outreach nurses conducting the audits worked in pairs to ensure accuracy.
This audit highlighted deficiencies including, failure to record respiratory rate, oxygen saturations, inspired oxygen and fluid intake and output. It also identified that some patients who required a fluid balance chart did not have one.
Following the 2001 audit the ONIK group has worked to improve vital sign monitoring and re-audited the wards annually using the same method each time.
In the 2006 audit patients were categorised according to their level of acuity and the observation charts of all level 1 and level 2 patients were included in the audit.
Results of the audit
There are no results for 2005 as the audit was not conducted across the whole of the network for that year.
Levels of care
The numbers of patients categorised into different levels of care in 2006 are shown in Table 1. Table 2 shows the percentages of patients categorised into different levels in each audit since 2001.
Table 2. Levels of care 2006
|Level of care||Number of patients||%|
Table 3. Levels of care comparison 2001-2006
In 2006 over 20% of patients were assigned to level 1 compared with 11% in 2001. The percentage of level 1 patients had almost doubled since 2001 (Fig 1). Between 2004 and 2006 there was an increase of nearly 3.5% in level 1 patients, which coincided with a reduction in levels 2 and 0 patients. The low number of level 2 patients in 2006 may be because the wards were unusually quiet when the audit was undertaken. Local outreach data collected and kept by each team within their own trust actually showed an increase of level 2 ward patients in some hospitals.
The overall drop in level 2 patients may also be related to additional high dependency provision being developed across the network throughout 2005.
Respiratory rate is considered to be one of the most sensitive indicators of critical illness, yet it is a vital sign that is often omitted (NCEPOD, 2005). Between 2004 and 2006 there was a slight improvement in the recording of respiratory rates, although over a five-year period the performance of this observation has been variable with a poor result in 2002 (Fig 2). An improvement from 74% of patients not having respiratory rate recordings in 2002, to 27% in 2006 is probably due to the concerted effort of teams following these poor results and the introduction of track-and-trigger scoring systems in some hospitals. However, this improvement was not sustained in the following years with around a third of acutely unwell patients still not having respiratory rates recorded.
The number of patients who did not have their oxygen saturations recorded in the previous eight hours fell from 56% in 2001 to 7% in 2006 (Fig 3).
The recording of the percentage of oxygen a patient received became more accurate, with less than 10% of the charts showing either a missing or inaccurate recording in 2006 compared with 25% in 2001 (Fig 4). This is encouraging as the numbers of patients at level 1 who are generally more unwell and require closer monitoring increased across the network.
Overall there has been an improvement since 2002 in the recording of respiratory observations.
Fifty-three per cent of patients had a fluid balance chart in 2001 compared with 59% in 2006. Fig 5 shows a comparison across the years of the percentage of patients whose fluid intake and output were recorded on a fluid chart.
The number of patients who should have had fluid intake and output recorded on a fluid chart but did not fell from 41% in 2002, when this was initially audited, to 20% in 2006.
Fluid charts were examined to check they were completed accurately (Fig 7). The criteria for an incomplete chart were:
Fluid input or output was missing;
Totals of input and output were not calculated.
The range of incomplete fluid charts varied across the network from 9% in one hospital to 51% in another. Overall the numbers of incomplete charts was variable with an increase from 16% in 2002 to 49% in 2003 and falling to 22.5% in 2006.
The number of patients not fed during the audit period decreased from 25% in 2001 to 4% in 2006 (Fig 8).
Discussion of results
Overall the results of this audit compared with the initial audit in 2001 are encouraging. Since the original audit the outreach teams have been actively involved in educating staff about the importance and significance of physiological monitoring and observations. This training and education has been provided to both registered nurses and HCAs.
The Network runs the ALERT (Acute Life-threatening Events Recognition and Treatment) training course (Smith, 2000). Track-and-trigger scoring systems are now in place across the whole network. However, it is important to continue this training to maintain standards and improve performance.
The increasing number of patients categorised at level 1 on hospital wards in our audit is reflected in the literature and also recognised by the government in their guidance and policy development. The case mix in hospitals is changing with hospitals caring mainly for those with acute illness (DH, 2004a).
Many patients are older with a wide variety of co-morbidities, undergoing high-risk procedures that require skilled nursing, close monitoring and observation (DH, 2004b; 2004c).
There is a problem that nursing and medical staff have less exposure to acutely ill patients during their training due to the European Working Time Directive, and nurse and medical training is more educationally based (O'Riordan et al, 2003). It is imperative that staff are aware of the significance of vital sign recording, and have the knowledge and skill to interpret them to ensure patient safety (NPSA, 2007a).
This audit has shown an overall improvement in vital sign recording across the Kent and Medway critical care network. The main concern was the recording of respiratory rates.
Poor vital sign recording, and specifically poor assessment of respiratory observations is well documented in the literature (NPSA, 2007a; NCEPOD, 2005; Buist et al, 1999; Goldhill et al, 1999; McGloin et al, 1999; Goldhill and Sumner, 1998).
A core part of the work of the ONIK has been the introduction of track-and-trigger scoring systems that require a full set of vital signs to be undertaken in order to calculate an accurate score (Critical Care Stakeholder's Forum and National Outreach Forum, 2007; Smith et al, 2006). It is recognised as an important part of minimising risk for patients (NPSA, 2007a).
The improvement in the other respiratory parameters, including recording for oxygen delivery, is encouraging and may be due to increased awareness among ward staff following teaching and training. Outreach teams are also more aware of these patients and ensure adequate monitoring is in place when they review them and often work alongside ward nurses in the clinical area. Oxygen saturations are easily obtained from the automated observation machines now used on many wards.
Fluid balance charts and nutrition
The number of patients requiring fluid balance charts over the past five years has increased. This may be because of a rise in acuity of patients and it corresponds with an increase in numbers of level 1 patients in the audit. The ALERT course emphasises the importance of urine output measurements, and the administration of fluid to treat hypotension and sepsis (Smith, 2000
The track-and-trigger scoring systems may have encouraged staff to monitor fluid balance more closely. These systems have been shown to help nurses communicate with medical teams and articulate concern about a patient's condition in a way medical teams understand (Andrews and Waterman, 2005).
The number of patients who required a fluid chart and did not have one is still unacceptable, although overall the percentage has decreased since 2003.
The accuracy of the fluid charts varied considerably across the network. When asked, ward nurses often complained of not having time to calculate the totals (personal communication) and they often felt that the wards were too short-staffed. One trust has developed a fluid balance standard that will set the minimum level of documentation required to care for patients.
Feeding patients has improved and it is rare to find acutely unwell patients without dietitian input, and a nutrition plan in place. This is an encouraging result given the importance of adequate nutrition for all hospital patients.
There is evidence in the literature of patients being allowed to become malnourished, and not having their nutritional requirements assessed or met (NPSA, 2007b). All the trusts in our network use the Malnutrition Universal Screening Tool (MUST) (NPSA, 2007b) to assess patients' nutritional needs and this addresses the risk of malnutrition in vulnerable patients.
A lot of work has been undertaken to meet Essence of Care and Patient Environment Action Team (PEAT) standards in the hospitals. Changes in the way wards organise their care, such as protected mealtimes for patients, have helped improve this aspect for care, which was borne out in our audit results.
Vital sign recording
The NICE (2007) guidance recommends that all inpatients must have their vital signs recorded as a minimum every 12 hours. It also recommends the use of track-and-trigger systems to alert staff to patients whose condition is deteriorating.
The 12-hour standard is based on professional consensus among the NICE working group. There is no evidence for optimal frequency of observations, although many studies have cited changes in vital signs, particularly respiratory rates, as a key clinical indicator of deterioration (Butler-Williams et al, 2005; NCEPOD, 2005; Goldhill and McNarry, 2002; Hillman et al, 2001; McGloin et al, 1999; McQuillan et al, 1998; Hillman et al, 1996; Schein et al, 1990).
None of these studies define the parameters for vital signs, thus making specificity and sensitivity difficult to assess. This is understandable as baseline vital signs can vary between patients. For example, a normal respiratory rate for an older person with COPD is different to that of a younger fitter adult. Without baseline data such as respiratory rate for individual patients, signs of impending deterioration can be missed (Sherman, 2002).
The majority of published studies do not investigate in detail the complex extraneous variables involved in recording vital signs. These include ineffective multidisciplinary team working and inadequate care planning for the patient. Delegation of vital sign recording to HCAs who may not fully understand the relevance of the observations and undertaking vital sign recordings at set times in the day rather than according to patient need can also lead to failure to recognise a deterioration.
The influence of the skill mix in ward teams, leadership styles of ward managers and management structures within wards have not been critiqued.
Our audit has shown that overall there has been an improvement in vital sign recording, fluid balance assessment and feeding of patients against a backdrop of increasing numbers of acutely ill patients in the wards. Some areas remain problematic.
The ONIK nurses plan to continue their teaching, training and audit of vital sign recording. Future audits will be altered to reflect NICE (2007) guidance, and will be supported by written standards and protocols. They may include additional parameters such as consciousness level, mean arterial blood pressure and temperatures. We plan to extend the audit to all patients, not just those who are classified as level 1 and 2.
The track-and-trigger scoring systems are also being reviewed across the network to reflect an aggregated level of response to increasing scores. This means that certain actions will be required by ward staff according to the score. These actions or responses prompt the nurse taking the observations to consider whether to increase frequency of observations, whether to use a cardiac monitor and which level of doctor to inform. This may include calling the critical care team to review the patient. These will also be audited in future work to ensure compliance.
It is hoped that improvements will continue as the critical care outreach teams implement the NICE guidance through education and training programmes. We have been fortunate in receiving strong managerial support for the work.
A network-wide course on caring for the highly dependent ward patient has been introduced. The critical care teams run this over five days across five consecutive weeks, teaching the importance of vital sign monitoring, assessment and interpretation skills. There are also competencies for staff to work towards after they have completed the taught element. The efficacy of the programme is being evaluated. The network continues to be committed to delivering the ALERT course, and a variety of in-house training days for ward nurses and doctors. There is also a close working relationship with university partners who provide graduate and postgraduate programmes on acute care. A new MSc module on acute care crisis aimed at ward staff working has been developed. Evaluation of this will be undertaken once students have completed the first module.
Ward teams have received both the audit reports and presentations locally of their improvements. This has been encouraging for them, and they continue to work closely with the outreach nurses to strive for a high standard of vital sign recording and assessment skills.
Andrews, T., Waterman, H. (2005) Packaging: a grounded theory of how to report physiological deterioration effectively. Journal of Advanced Nursing; 52: 5, 473-481.
Buist, M.D. et al (1999) Recognising clinical instability in hospital patients before cardiac arrest or unplanned admission to intensive care. A pilot study in a tertiary-care hospital. The Medical Journal of Australia; 171: 1, 8-9.
Butler-Williams, C. et al (2005) Increasing staff awareness of respiratory rate significance. Nursing Times; 101: 27, 35-37.
Chellel, A. et al (2002) Nursing observations on ward patients at risk of critical illness. Nursing Times; 98: 46, 36-38.
Critical Care Stakeholder's Forum and National Outreach Forum (2007) Clinical Indicators for Critical Care Outreach Services. www.dh.gov.uk
Department of Health (2005) Quality Critical Care - Beyond 'Comprehensive Critical Care'. www.dh.gov.uk
Department of Health (2004a) National Standards, Local Action: Health and Social Care Standards Planning Framework 2005/6-2007/8).www.dh.gov.uk
Department of Health (2004b) Standards for Better Health. www.dh.gov.uk
Department of Health (2004c) The NHS Improvement Plan: Putting People at the Heart of Public Services. www.dh.gov.uk
Department of Health (2003) Critical Care Outreach: Progress in Developing Services. www.dh.gov.uk
Department of Health (2000) Comprehensive Critical Care: A Review of Adult Critical Care Services.www.ics.ac.uk
Goldhill, D.R., Sumner, A. (1998) Outcome of intensive care patients in a group of British intensive care units. Critical Care Medicine; 26: 8, 1337-1345.
Goldhill, D.R. et al (1999) Physiological values and procedures in the 24 hours before ICU admission from the ward. Anaesthesia; 54: 6, 529-534.
Goldhill, D.R., McNarry, A. (2002) Intensive care outreach services. Current Anaesthesia and Critical Care; 13: 6, 356-361.
Hillman, K.M. et al (2001) Antecedents to hospital deaths. Internal Medicine Journal; 31:6, 343-348.
Hillman, K.M. et al (1996) Identifying the general ward patient at risk of cardiac arrest. Clinical Intensive Care; 7: 242-243.
Intensive Care Society (2002) Guidelines for the Introduction of Outreach Services.www.ics.ac.uk
McGloin, H. et al (1999) Unexpected deaths and referrals to intensive care of patients on general wards. Are some cases potentially avoidable? Journal of The Royal College of Physicians of London; 33: 3, 255-259.
McQuillan, P. et al (1998) Confidential inquiry into quality of care before admission to intensive care. British Medical Journal; 316: 7148, 1853-1858.
National Confidential Enquiry into Patient Outcome and Death (2005) An Acute Problem? www.ncepod.org.uk
NICE (2007) Acutely Ill Patients in Hospital: Recognition and Response to Acute Illness in Adults in Hospital. www.nice.org.uk
National Patient Safety Agency (2007a) Safer Care for the Acutely Ill Patient: Learning from Serious Incidents. 5th Report from the Patient Safety Observatory. www.npsa.nhs.uk
National Patient Safety Agency (2007b) Nutritional Screening: Structured Investigation Project. www.npsa.nhs.uk
O'Riordan, B. et al (2003) Implementing a critical care course for ward nurses. Nursing Standard; 17: 20; 41-44.
Schein, R. et al (1990) Clinical antecedents to in-hospital cardiopulmonary arrest. Chest; 98: 6, 1388-1392.
Sherman, F.T. (2002) Vital signs are still- well, vital. Geriatrics; 57: 3, 9.
Smith, G. et al (2006) Hospital-wide physiological surveillance - a new approach to the early identification and management of the sick patient. Resuscitation; 71: 1, 19-28.
Smith, G. (2000) Alert Life-threatening Events. Recognition and Treatment: A Multi-professional Course in Care of the Acutely Ill Patient: Course Manual. Portsmouth: Portsmouth Institute of Medicine, Health and Social Care Faculty of Science.