Vital signs monitoring and nurse–patient interaction: A qualitative observational study of hospital practice

International Journal of Nursing Studies 56 (2016) 9–16

Contents lists available at ScienceDirect

International Journal of Nursing Studies journal homepage: www.elsevier.com/ijns

Vital signs monitoring and nurse–patient interaction: A qualitative observational study of hospital practice M. Cardona-Morrell a,*, M. Prgomet b, R. Lake b, M. Nicholson c, R. Harrison d, J. Long b, J. Westbrook b, J. Braithwaite b, K. Hillman a,c a

South Western Sydney Clinical School, The University of New South Wales, Australia Australian Institute of Health Innovation, Macquarie University, Australia Intensive Care Unit, Liverpool Hospital, Australia d School of Public Health, Sydney Medical School, The University of Sydney, Australia b c

A R T I C L E I N F O

A B S T R A C T

Article history: Received 1 October 2015 Received in revised form 6 December 2015 Accepted 18 December 2015

Background: High profile safety failures have demonstrated that recognising early warning signs of clinical and physiological deterioration can prevent or reduce harm resulting from serious adverse events. Early warning scoring systems are now routinely used in many places to detect and escalate deteriorating patients. Timely and accurate vital signs monitoring are critical for ensuring patient safety through providing data for early warning scoring systems, but little is known about current monitoring practices. Objective: To establish a profile of nurses’ vital signs monitoring practices, related dialogue, and adherence to health service protocol in New South Wales, Australia. Methods: Direct observations of nurses’ working practices were conducted in two wards. The observations focused on times of the day when vital signs were generally measured. Patient interactions were recorded if occurring any time during the observation periods. Participants (n = 42) included nursing staff on one chronic disease medical and one acute surgical ward in a large urban teaching hospital in New South Wales. Results: We observed 441 patient interactions. Measurement of vital signs occurred in 52% of interactions. The minimum five vital signs measures required by New South Wales Health policy were taken in only 6–21% of instances of vital signs monitoring. Vital signs were documented immediately on 93% of vitals-taking occasions and documented according to the policy in the patient’s chart on 89% of these occasions. Nurse–patient interactions were initiated for the purpose of taking vital signs in 49% of interactions, with nurse–patient discourse observed during 88% of all interactions. Nurse–patient dialogue led to additional care being provided to patients in 12% of interactions. Conclusion: The selection of appropriate vital signs measured and responses to these appears to rely on nurses’ clinical judgement or time availability rather than on policymandated frequency. The prevalence of incomplete sets of vital signs may limit identification of deteriorating patients. The findings from this study present an important baseline profile against which to evaluate the impact of introducing continuous monitoring approaches on current hospital practice. ß 2015 Elsevier Ltd. All rights reserved.

Keywords: Vital signs Patient monitoring Nurse–patient relations Hospital Physiological deterioration qualitative study

* Corresponding author at: The Simpson Centre for Health Services Research, South Western Sydney Clinical School, The University of New South Wales, PO Box 6087 UNSW, Sydney, NSW 1466, Australia. Tel.: +61 2 8738 9373. E-mail address: [email protected] (M. Cardona-Morrell). http://dx.doi.org/10.1016/j.ijnurstu.2015.12.007 0020-7489/ß 2015 Elsevier Ltd. All rights reserved.

10

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

What is already known about the topic  Identifying and responding to deteriorating patients is an essential component of high quality safe care.  Vital signs monitoring at the bedside offers an opportunity for nurse–patient interaction in which patients can contribute to their care by identifying additional care needs.  Early warning systems are routinely used in developed countries to identify and trigger escalation and clinical response for patients whose clinical or physiological condition is deteriorating. What this paper adds  Vital signs monitoring practices in the study wards rarely adhered to the policy-mandated full set of five vital sign measures.  The infrequency with which full sets of vital signs measures were taken suggest that nurses are confident about their clinical judgement, or lack time to comply with state policy to prevent deterioration.  Incomplete monitoring could limit nurses’ ability to recognise deteriorating patients and further exploration of the reasons behind this behaviour would be useful.  The findings from this study present an important baseline profile against which to evaluate the impact of introducing continuous monitoring approaches on current hospital practice. 1. Introduction The most common method for monitoring and documenting a patients’ well-being or deterioration in hospitals is often through the assessment of five vital signs: blood pressure (BP), oxygen saturation of the blood (SpO2), pulse rate, respiratory rate, and body temperature (Ahrens, 2008). Hospital policies in many countries internationally mandate that these measures be taken in all wards, for all patients, at various frequencies per day and there is substantial research literature about the individual measures taken in the monitoring process (Ahrens, 2008). Yet since the introduction of vital signs monitoring and documentation, the practice of this critical aspect of nursing has received limited research attention (Cretikos et al., 2008; Hands et al., 2013). There is a dearth of literature around vital signs monitoring and documentation practices and a lack of data and recommendations about optimal practices (Evans et al., 2001; Lockwood et al., 2004). Patient safety failures, such as system gaps or individual health care provider errors that put patients at risk of harm (Carayon and Wood, 2010; van Beuzekom et al., 2012), have demonstrated that recognising early warning signs of clinical and physiological deterioration can prevent or reduce harm resulting from serious adverse events (Beaumont et al., 2008; Kyriacos et al., 2011; Ludikhuize et al., 2012; van Beuzekom et al., 2012). There is now international recognition of the need to identify and respond to clinical and physiological deterioration in the interest of providing high quality and safe care (ACSQHC,

2014; National Institute for Clinical Health Excellence, 2007; Agency for Healthcare Research Quality, 2011). In 2005, the United Kingdom (UK) National Patient Safety Agency conducted an analysis of serious incidents resulting in death; of these, 425 occurred in acute/general hospitals and 64 of those deaths resulted from a lack of recognition or response to patient deterioration (Beaumont et al., 2008). The Australian National Safety and Quality in Health Service standards now require all hospitals and day procedure services to support health care providers recognise and respond to a patient’s deteriorating condition (Australian Commission on Safety Quality in Health Care, 2012). Early warning scoring systems are used to protect patients by highlighting signs of deterioration and triggering an escalation of care from health care staff (Osborne et al., 2015). These systems are used routinely in both the UK and Australasia (Green and Williams, 2006; Hogan, 2006; Chaboyer et al., 2008). For example, in New South Wales the ‘Between the Flags’ system was introduced in 2010 to improve recognition of and response to deteriorating patients in public hospitals across the state (Clinical Excellence Commission, 2013; Hughes et al., 2014). This system is designed to identify patients who show early signs of deterioration using intermittent vital signs monitoring and documentation to identify when there is a need for on-site clinical review by a ‘home team’ or rapid response by the medical emergency team. Effective early warning systems rely on vigilant staff taking vital signs measures in accordance with health system policy to provide the necessary data from accurate and timely observations (Bellomo et al., 2012). The UK National Institute for Health and Clinical Excellence (NICE) recommend that physiological observations should be taken at least every 12 h and more frequently if abnormal physiology is detected (National Institute for Clinical Health Excellence, 2007). In New South Wales (NSW) Australia, a minimum set of vital signs must be monitored and documented at least three times per day at eight hourly intervals, unless more frequent monitoring is ordered by a clinician such as in the case of patients requiring rapid response calls (New South Wales Health, 2014). A minimum set of observations in our hospital includes five vital signs (respiratory rate, heart rate, blood pressure, temperature and SpO2) and two observations (level of consciousness, pain score) (New South Wales Health, 2014). To date, vital signs monitoring has operated by nurses going to the patient’s bedside at given intervals to collect clinical and physiological data and make an assessment of the patient’s condition. This model is open to wide variation in practice between wards and hospitals due to the knowledge and experience of staff (Lockwood et al., 2004). Patients may experience substantial deterioration between monitoring intervals and the system is vulnerable to the delay or omission of critical data capture during busy periods or times when there are fewer staff available such as weekends and evenings (Buist and Stevens, 2013; Hands et al., 2013). This study explores current vital signs monitoring practices in an Australian health care setting, adherence to the health service policy in New South Wales,

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

and the nature of patient–nurse interactions that arise in the monitoring process, including the resulting additional or altered care received. 1.1. Study objectives  To develop a baseline profile of nurses’ vital monitoring practices in two wards at a large Australian urban teaching hospital: the frequency and type of vital signs data collected; the frequency and type of patient observations; and the frequency of chart review and documentation of collected data in patients’ charts.  To assess current practice against the New South Wales Health policy.  To characterise the nurse–patient interactions arising during times of the day when vital signs monitoring typically occurs.  To identify additional care resulting from nurse–patient interactions during vital signs monitoring. 2. Methods The ‘Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studies’ was used to guide the reporting of this work (von Elm et al., 2007). 2.1. Ethical approval This study was granted ethical approval by the SouthWestern Sydney Local Health District Human Research Ethics Committee. 2.2. Study design An observational, cross-sectional design was used. Researchers undertook unobtrusive observations and recorded nurses monitoring practices on a standardised data collection sheet developed for this project in order to assess the frequency with which monitoring and documentation of the five vital signs, other patient observations (level of consciousness, pain, urinary output, bowel motion), and patient–nurse interactions occurred. Consent was obtained from each participating nurse prior to inclusion in the study, but during the observation times no researcher–nurse interactions took place. 2.3. Setting The study was undertaken between 26 June and 15 August 2014 in one respiratory ward and one neurosurgery ward of a large urban teaching hospital in New South Wales, Australia. This 800+ bedded hospital provides diagnostic and imaging services, emergency and trauma care, maternity, paediatric, cancer care, mental health, ambulatory care, allied health, and medical and surgical services. It serves a multicultural population of over 180,000 people speaking 140 different languages and with a median age of 33 years. The study wards were identified through purposive sampling to provide data about vital signs monitoring practices from different types

11

of care and patient groups. One ward was an acute surgical ward (neurosurgery) which included a younger patient group and the other ward was a chronic disease ward (respiratory) with a higher proportion of elderly patients. The respiratory ward employs 38 nurses and the neurosurgery ward employs 34 nurses in addition to five trainee/ student nurses in each. The nurse to patient ratio on those wards is 1:4 during the day and 1:10 overnight. Vital signs monitoring practices differ for different patient groups, as do deterioration thresholds for calling a rapid response team. For example, additional monitoring of acute patients post-surgery is necessary. Consultation with the ward managers prior to the study established there to be support for the research in both wards, making the baseline observational work feasible before the pilot introduction of continuous monitoring technology. 2.4. Participants All consenting registered nurses in each ward were eligible to be observed; trainee/student nurses were not eligible due to their lack of experience, not reflecting majority practice on the study wards. We sought to observe at least 50% of the nurses in each ward over the three-month study period in an attempt to provide a representative sample from which to create a baseline profile of nursing vital sign monitoring practices in these wards. An opportunistic sample of nurses was included, with a total of 44 nurses approached across the two wards to participate in the study. In accordance with the ethical approval obtained, nurses were briefed about the study by their ward managers prior to the observations to make them aware of the project and that they may be invited to take part. Participation was voluntary and the ward managers did not have a role in either recruiting nurses for the study or observing them. Individual nurses were approached on the ward by a member of the research team, asked if they wished to participate, and if so, gave written consent to being observed as they worked. Nurses were free to decline consent if they did not wish to participate. Only two nurses declined to participate. Observations were only conducted in open ward areas, as opposed to private or isolation rooms, and not when a patient’s curtain was drawn. Patients were advised that observational work was being undertaken by researchers in the ward and they could choose to have the curtain drawn if they did not wish their interactions to be observed. 2.5. Outcome variables and measurement used The research team gathered observational data regarding three aspects of care: (1) the vital signs monitoring practices of the nurses (pulse rate, blood pressure, respiratory rate, temperature, SpO2); (2) other patient observations conducted by nurses beyond the five vital signs (e.g., level of consciousness, urine output, pain) and (3) the duration and content of nurse–patient interactions in care, many of which were during the vital signs monitoring process or the wider patient observations undertaken by nurses. The outcome variables collected and

12

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

Table 1 Information captured during observations. Category

Description

Measure

Vital Signs

The frequency with which vital signs monitoring occurred. The frequency with which each individual vital sign was measured. The equipment used to take vital signs measures. The frequency with which vital signs measures and observations details were recorded immediately. The frequency with which patients’ charts were checked and the results recorded immediately. The frequency with which the standard patient observations occurred. The frequency with which each individual patient observation occurred. Non-standard patient observations that were taken and their frequency. The duration of each nurse–patient interaction. The most common reasons for initiating nurse–patient interactions. The duration of dialogue between patients and nurses during interactions. The types of dialogue occurring between nurses and patients. The frequency with which vital signs were re-measured during the same interaction. Other nurse responses to vital signs measures – such as patient care undertaken.

Numeric frequency Numeric frequency

Documentation

Patient observations

Nurse–patient interactions

Nurse–patient Discourse

Actions in response to vital signs measures

the format in which they were recorded by the researchers are shown for each variable in Table 1. 2.6. Procedure Once written consent was obtained, one researcher followed the nurse through the course of their duties, recording information on the observational data collection forms (Supplementary file 1). A new form was completed for each interaction that the nurse had with the patient to document the nurse’s practice. Observation sessions lasted up to two hours to minimise interference in nurses’ daily activities. However, if the target nurse went to a meeting or went on a break, they were not observed during those periods. Nurses on the selected wards conducted three rounds of vital signs monitoring during the day, one in the morning, one in the early afternoon, and a third in the evening. The exact time of the vital signs rounds varied depending on the activity on the ward and the completion of other required tasks, such as showering patients and changing sheets. In order to capture the main periods in which vital signs monitoring and documentation took place, observations were conducted on weekdays in the mornings and in the early afternoons on each ward. Observations were not conducted in the evening or on weekends. The observations were undertaken by three members of the research team, one who had a nursing background and two who had social and applied science backgrounds. All observers were knowledgeable regarding vital signs monitoring practices and were briefed on local hospital culture by an experienced intensive care nurse from the study site. The observers were also trained in the use of the observational checklist by one of the lead authors prior to undertaking observations. In the participating hospital, the patient observations conducted by nurses could include pain level, blood sugar

Selection of options Numeric frequency Numeric frequency Numeric frequency Numeric frequency Text description and numeric frequency Time in minutes and seconds. Selection of options and text Time in minutes and seconds. Selection of options and text Numeric frequency Text

levels (BSL), bowel movements, level of consciousness (LOC), pupil dilation and/or strength. LOC was measured visually and/or by speaking with the patient. Pain level and bowel movements were assessed by asking the patient directly and pupil dilation was measured by visual assessment. BSL was measured with an electronic device and strength by asking the patient to perform an action such as pushing against the nurse’s hand. Further patient observations included urine output, food and/or fluid intake, numbness or pins and needles, feeling or movement, dizziness, nausea, clamminess of the skin, and breathing. The appropriate observation measures are determined by the hospital staff and all measures would not always be taken. During the observation periods, data about the variables shown in Table 1 were collected in addition to details of nurse–patient interactions: the amount of time spent in each interaction, the nature of any dialogue that took place and the resulting care provided to the patient. An ‘interaction’ was defined as an instance where the nurse spoke to, assisted or provided any care for a patient. Interactions included instances when the patient was asleep or otherwise unresponsive. Where nurses left the patient briefly during an interaction to collect equipment or speak to another staff member and then immediately returned to the patient, this was considered a single interaction. 3. Analytic strategy Text data were independently assessed by the three members of the research team who conducted the observations using content analysis (Krippendorff, 2013). Labels were generated to reflect the types of equipment and interactions recorded. Through a process of discussion between the three research team members, labels were then merged into categories of equipment for recording

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

13

vital signs and the nature of nurse–patient interactions. Upon completion of the observations, the data were entered into the Statistical Package for the Social Sciences (IBM SPSS v22) for analysis. Descriptive statistics were used to calculate frequencies for each of the outcome variables.

handover sheet 2% of the time and elsewhere 3% of the time. Vital signs measures were recorded at the bedside 86% of the time and directly outside the room 14% of the time.

4. Results

The term ‘patient observations’ is used to describe a broader range of assessments beyond the five vital signs measures. We observed nurses taking patient observations a total of 140 times during the 68 h and 41 min of observation. Patient observations were taken 59 times on the respiratory ward and 81 times on the neurosurgery ward. On the respiratory ward the most frequent patient observation was bowel movement queries, observed in 33 instances. The strength test was observed most frequently on the neurosurgery ward, in 48 instances. On 21 occasions patient observations were conducted without vital signs measures being taken. Several further patient observations were observed which are shown in Table 4.

4.2. Patient observations

Across the two study wards, 42 nurses consented to be observed; 22 from respiratory and 20 from neurosurgery. There were 79 observation sessions conducted with these staff members, of which 46 were in the respiratory ward and 33 in the neurosurgery ward. A total of 68 h and 41 min of observational data were collected. In the respiratory ward, 34 h and 30 min of data was collected and in the neurosurgery ward, 34 h and 11 min of data was collected. A total of 441 nurse–patient interactions were observed. 4.1. Vital signs measurement Vital signs were assessed in just over half of the observed nurse–patient interactions (229/441; 52%). Of the 229 vital sign related interactions, 21% (n = 48) involved a full set of vital signs measures (BP, SpO2, pulse rate, respiratory rate, and temperature). In the respiratory ward a full set of measures was taken on 43 of 140 vital signs interactions (31%) and in the neurosurgery ward a full set of measures was taken on 5 of 89 vital signs interactions (6%). BP, pulse rate and SpO2 were the most frequently measured (Table 2). The automated monitor was used most frequently (95% of vital sign related interactions), followed by a digital thermometer which was used in 87% of vital sign related interactions. The sphygmomanometer was used in only 1% of interactions (Table 3). Vital sign measurements were observed to be immediately documented on 93% of occasions (214 of 229 vital sign interactions). Of the 214 instances when they were documented, vital sign measures were documented on the patient’s paper Chart 95 Chart 95% of the time (89% of occasions overall of the 229 vital signs interactions), on a

4.3. Nurse–patient interactions Around half of the interactions we observed were initiated for the purpose of taking vital signs measures (49%), but often interactions were initiated for other reasons and vital signs measures taken during the interaction. The most common reasons for initiating interaction–besides taking vital signs measures – were to administer medication (9%), to check on a patient (8%), and in response to a verbal request from another nurse, the patient, or a friend or relative of the patient (5%). The average duration of each nurse–patient interaction was four minutes on both wards. Dialogue between the nurse and patient was observed in 88% of the 441 interactions (n = 387). Dialogue was more common in the respiratory ward, occurring in 216 interactions compared to 171 interactions in the neurosurgery ward. The majority of dialogue was classed as health related, but not vital sign related (n = 286). Examples of this

Table 2 Frequency of vital signs measures. Vital signs measure

Respiratory

Neurosurgery (N)

Total (N)

% of 229 vital signs related interactions

5 83 80 80 73 6

48 221 216 217 204 51

21 97 94 95 89 22

(N) All BP SpO2 Pulse rate Temperature Respiratory rate

43 138 136 137 131 45

Table 3 Equipment used to measure vital signs. Equipment

Respiratory (N)

Neurosurgery (N)

Total (N)

% of vitals taken

Automated monitor Digital thermometer Manual resp. rate Sphygmomanometer

136 128 45 2

82 72 6 1

218 200 51 3

95 87 22 1

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

14

Table 4 Frequency of other patient observations. Observation type

Respiratory (N) Neurosurgery (N) Total (N)

33 Bowel movements Strength test 6 Urine output 2 Breathing related 2 Numbness 0 Touched hands (clammy) 1 Food/fluid intake 0 Feeling and movement 1 Pins and needles 1 1 Nausea Dizziness 1 Stick out tongue 0 More than one 1 10 Total

21 48 0 1 2 0 3 1 6 0 0 1 2 16

54 54 2 3 2 1 3 2 7 1 1 1 3 26

included discussion of treatments or medical tests. Dialogue related to vital signs was also common (n = 144). In the remainder of cases dialogue was general discussion (n = 46). In the majority of cases nurses recorded the vital signs measures they made and took no further action; however in 48 of the 387 instances nurses responded to the vital signs measures with further actions. The most frequent further action related to vital signs measurement was a remeasure, undertaken in 25 instances in the respiratory ward and in nine instances on the neurosurgery ward. Blood pressure was the most frequently re-measured vital sign, with 16 re-measures. Temperature was re-measured seven times, SpO2 and pulse rate were both re-measured five times and respiratory rate was remeasured once. On the respiratory ward, re-measurement during the same interaction was undertaken using the same equipment on 18 of the 25 instances. Staff on the neurosurgery ward used a different piece of equipment to take re-measurements on eight of the nine instances, including an automated monitor (n = 1), a digital thermometer (n 2), a watch for manual respiratory rate (n = 1) or a sphygmomanometer (n = 4). Fourteen other actions were observed in response to the nurse–patient interaction. The most common of these were consultation with a doctor (n = 3) or with another nurse (n = 3) to determine the appropriate course of action or whether to escalate the patient. Doctors were asked to review the patient, provide advice or change something in the patient’s chart. The following adjustments to patient care were immediately made in response to the vital signs interaction: adjusting oxygen levels (n = 2), increasing monitoring frequency (n = 2), administering water (n = 1), connecting a saline drip (n = 1), putting an oxygen mask on the patient (n = 1), and checking for tenderness postsurgery (n = 1). 5. Discussion This observational study provides novel data presenting a profile of nursing vital signs monitoring practices in Australia. The findings highlight the importance of clinical

judgement in vital signs monitoring and documentation, and the opportunity for nurse–patient interaction the process presents. This work, therefore, has important implications for ensuring nurses have the necessary clinical knowledge around vital signs monitoring. Vital signs data were well-documented in the study, details were immediately recorded in patients’ notes in accordance with NSW Health policy in 89% of the instances observed (New South Wales Health, 2014). Despite NSW Health policy requiring all five measures to be taken at least three times per day on general adult patients, similar to other studies, we found that full sets of five vital signs were rarely measured (6% to 21%) (Ludikhuize et al., 2012). This study suggests that the selection of appropriate vital signs measures on a day-to-day basis relies on nurses’ clinical judgement, but other factors such as time constraints and work-distractions may have also influenced practice. With the low nurse:patient ratios at night, it would not be unusual to find lower compliance with the monitoring policy, as has been observed in other studies (Buist and Stevens, 2013). The assessment of vital signs is often considered a basic, skill-based task that falls to more junior and less experienced nurses (Hogan, 2006). Yet appropriate vital signs monitoring and documentation, and the resulting changes in care rely on the clinical knowledge of nursing staff in addition to their ability to utilise a range of technology. Patients admitted to acute hospitals today have increasingly complex health problems and a higher chance of becoming seriously ill during their admission (Ryan et al., 2004). As demand for beds grows, those previously cared for in high dependency units (HDU) or intensive care units (ICU) are now being cared for in general wards. For example, patients in half of 3160 acute hospital admissions in a large study had at least one early sign of critical illness while on the wards (Harrison et al., 2006). It is, therefore, more important than ever that ward staff can identify early signs of critical illness. In the present study, all vital signs measures were only taken in up to 21% of instances of vital signs measurement. The low frequency with which full sets of measures are taken reflects the findings of other studies (Ludikhuize et al., 2012), and may reflect nurses’ confidence in the decision not to take certain measures based on their knowledge of the patient, or lack of time to take a full set due to competing demands. This infrequency is not aligned with NSW Health Policy, and suggests that on some occasions nurses may have failed to take and document important vital signs measures, limiting the ability to recognise deteriorating patients. Further work exploring why particular vital signs are not taken on each occasion would be useful to explain this low rate. Health care services must ensure that nurses at all levels are knowledgeable in vital signs management and recognise the importance of this as a clinical process that ensures safe patient care (Boulanger and Toghill, 2009). Senior nurses have a key role in providing oversight of monitoring process to reduce avoidable harm (Harrison et al., 2005). Beyond ensuring that deteriorating patients are identified, vital signs monitoring and documentation offer an opportunity for patients to engage with nursing staff and participate in their care. Research evidence indicates

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

patient participation in care offers benefits to patient safety (Tobiano et al., 2015). Sahlsten et al. (2008) provide a conceptual analysis of patient participation in nursing, in which participation is characterised as a strong existing nurse–patient relationship, with nurses relinquishing their power, engaging in mutual exchange of information and knowledge with patients, and collaborating with patients in intellectual and/or physical activities in their care. Vital signs monitoring and documentation at the bedside provides an opportunity for patients to be more readily involved in the exchange of information about their condition and to contribute to decision-making about their care. This was evident in the present study in which nurse–patient dialogue was observed in over half of the vital signs monitoring interactions. The dialogue led to additional care processes on 12% of occasions. Interacting with and involving patients in their vital signs assessments may provide patients with a feeling of being cared for and an opportunity to raise other concerns. Evidence to date indicates that whilst patients are empowered when they are able to participate in their care, engaging with health professionals can be challenging (Peat et al., 2010). Barriers include the power dynamic between patients and health professionals in terms of medical knowledge, patients discomfort challenging and asking questions, and health professionals willingness to invite patients to contribute and act upon their wishes (Peat et al., 2010). Our findings also have implications for the introduction of technology for the continuous monitoring of vital signs, which was under consideration by our hospital management at the time of this observational study. The development of electronic, automated, and continuous vital signs monitoring systems has been identified as a strategy to potentially address challenges associated with intermittent vital signs monitoring practices (Bellomo et al., 2012; Brown et al., 2014). Continuous monitoring systems allow nursing staff access to vital signs data around the clock, which can often be viewed without the need to go to the patients’ bedside. These systems enable a number of patients to be monitored by a nurse from a single location (Welch et al., 2012; Zimlichman et al., 2011). Introduction of continuous monitoring offers great potential for enhancing the safety and quality of patient care, such as leading to the early detection of deterioration and reducing the number of adverse events and possibly providing nurses with more time to engage with their patients. Continuous monitoring also facilitates the collection of evaluative data on the monitoring process (Smith et al., 2009). But, as with many changes in practice, innovation may also lead to unintended outcomes. For example, with no need to attend to the patient at their bedside, an opportunity for nurse–patient dialogue may be reduced Research evidence is needed to fully understand the positive and negative implications of the use of continuous monitoring in general wards. 5.1. Limitations This study has several limitations and its results should be considered in the context of these. Observations

15

wereonly conducted during the daytime and during certain time periods. Different nursing practices in the evening, weekend, or overnight may have been overlooked. The study was undertaken at a single site and only in two wards. The data may therefore not be representative of vital signs monitoring practices in other wards within the study hospital or in other hospitals. While direct observation is useful to capture information about processes occurring in their natural context, this methods suffers some limitations (Kawulich, 2005). Using unobtrusive observations did not facilitate insights into the reasons behind the practices observed. For example, we did not capture data about the reasons why certain vital signs readings were not taken, which may have been due to the nurses’ understanding of the patient’s condition or knowledge of earlier vital signs measures. Likewise we did not capture data about the reasons behind the re-taking of vital signs measures. Observational data are shaped by the stance of the researcher, which can be a source of bias (Kawulich, 2005). The research team members observing the study were academic researchers and independent from the hospital but one researcher had a nursing background which may have influenced their perspective on the observations undertaken. The use of two non-nursing observers and the fact that all observers were not practicing clinically in the study hospital or any hospital reduced the potential for bias. Furthermore, observational data may be subject to the Hawthorne effect, with those being observed adapting their behaviour as a result (McCambridge et al., 2014). 6. Conclusion Our findings suggest that whilst vital signs assessments are well-documented in the majority of cases, the full range of measures are rarely obtained. The selection of appropriate vital signs measures and response to these appears to be influenced by nurses’ clinical judgement or competing demands rather than adhering to mandated policy. The implication of incomplete sets of vital signs is the potential to miss the timely recognition of deteriorating patients. The findings from this study present an important baseline profile against which to evaluate the impact of introducing continuous monitoring technology in these two wards on nurse–patient interactions and early detection of patient deterioration. Acknowledgements We thank the ward nurses for their participation in this study. Our gratitude goes also to Dr. Julie Johnson and Dr. Jacqueline Milne, both formerly associated with the Australian Institute of Health Innovation, for their involvement in the early phase of this project. Conflict of interest: Until 2012 KH was a consultant clinician for a company developing a mobile monitoring system to identify deterioration early. His role as consultant has ceased but he retains stock options. RH and JL received fees for their expert involvement in interpretation of qualitative data and

16

M. Cardona-Morrell et al. / International Journal of Nursing Studies 56 (2016) 9–16

refinement of this manuscript. None of the other authors has a conflict of interest to declare. Funding: This research was funded by a grant from the National Health and Medical Research Council (NHMRC grant reference 1054146). The funding body had no role in the conduct of the research or interpretation of findings, the choice of journal or the decision to publish. Ethical approval: The study was approved by the South Western Sydney Local Health District Human Research Ethics Committee (project # 13/220). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. ijnurstu.2015.12.007. References Australian Commission on Safety and Quality in Health Care, 2014. Vital Signs 2014: The State of Safety and Quality in Australian Health Care Sydney. ACSQHC. (cited 2014 November). Available from: http:// www.safetyandquality.gov.au/publications/vital-signs2014-the-state-of-safety-and-quality-in-australian-health-care/ Agency for Healthcare Research Quality, 2011. 5 Million Lives Campaign: Institute for Healthcare Improvement, AHRQ. (cited 2015 October). Available from: https://psnet.ahrq.gov/resources/resource/4688. Ahrens, T., 2008. The most important vital signs are not being measured. Aust. Crit. Care 21 (1), 3–5. Australian Commission on Safety Quality in Health Care, 2012. National Safety and Quality Health Service Standards Sydney. (cited 2015 November). Available from: http://www.safetyandquality.gov.au/ publications/national-safety-and-quality-health-service-standards/ Beaumont, K., Luettel, D., Thomson, R., 2008. Deterioration in hospital patients: early signs and appropriate actions. Nurs. Stand (R. Coll. Nurs. (Great Britain): 1987) 23 (1), 43–48, PubMed PMID: 18814476. Epub 2008/09/26.eng. Bellomo, R., Ackerman, M., Bailey, M., Beale, R., Clancy, G., Danesh, V., et al., 2012. A controlled trial of electronic automated advisory vital signs monitoring in general hospital wards. Crit. Care Med. 40 (8), 2349–2361. Boulanger, C., Toghill, M., 2009. How to measure and record vital signs to ensure detection of deteriorating patients. Nurs. Times 105 (47), 10– 12, PubMed PMID: 20063615. Epub 2010/01/13.eng. Brown, H., Terrence, J., Vasquez, P., Bates, D.W., Zimlichman, E., 2014. Continuous monitoring in an inpatient medical-surgical unit: a controlled clinical trial. Am. J. Med. 127 (3), 226–232. Buist, M., Stevens, S., 2013. Patient bedside observations: what could be simpler? BMJ Qual. Saf. 1–3. Carayon, P., Wood, K.E., 2010. Patient safety: the role of human factors and systems engineering. Stud. Health Technol. Inf. 153, 23–46, PubMed PMID: PMC3057365. Chaboyer, W., Thalib, L., Foster, M., Ball, C., Richards, B., 2008. Predictors of adverse events in patients after discharge from the intensive care unit. Am. J. Crit. Care: Off. Publ. Am. Assoc. Crit. Care Nurs. 17 (3), 255– 263, PubMed PMID: 18450682. Epub 2008/05/03.eng. Clinical Excellence Commission, 2013. Between the Flags Program. Keeping patients safe. Standard Adult General Observation Chart. (cited 2013 September). Available from: http://www.cec.health.nsw.gov. au/__documents/programs/between-the-flags/2013/sago-12.2013. pdf. Cretikos, M.A., Bellomo, R., Hillman, K., Chen, J., Finfer, S., Flabouris, A., 2008. Respiratory rate: the neglected vital sign. Med. J. Aust. 188 (11), 657–659. Evans, D., Hodgkinson, B., Berry, J., 2001. Vital signs in hospital patients: a systematic review. Int. J. Nurs. Stud. 38 (6), 643–650. Green, A.L., Williams, A., 2006. An evaluation of an early warning clinical marker referral tool. Intensiv. Crit. Care Nurs. Off. J. Brit. Assoc. Crit. Care Nurs. 22 (5), 274–282. Hands, C., Reid, E., Meredith, P., Smith, G.B., Prytherch, D.R., Schmidt, P.E., et al., 2013. Patterns in the recording of vital signs and early warning

scores: compliance with a clinical escalation protocol. BMJ Qual. Saf. 22 (9), 719–726. Harrison, G.A., Jacques, T.C., Kilborn, G., McLaws, M.-L., 2005. The prevalence of recordings of the signs of critical conditions and emergency responses in hospital wards – the SOCCER study. Resuscitation 65 (2), 149–157. Harrison, G.A., Jacques, T., McLaws, M.L., Kilborn, G., 2006. Combinations of early signs of critical illness predict in-hospital death-the SOCCER study (signs of critical conditions and emergency responses). Resuscitation 71 (3), 327–334. Hogan, J., 2006. Why don’t nurses monitor the respiratory rates of patients? Brit. J. Nurs. 15 (9), 489–492 (Mark Allen Publishing) PubMed PMID: 16723921. Epub2006/05/26.eng. Hughes, C., Pain, C., Braithwaite, J., Hillman, K., 2014. ‘Between the flags’: implementing a rapid response system at scale. BMJ Qual. Saf. 23 (9), 714–717. Kawulich, B.B., 2005. Participant observation as a data collection method. Forum: Qual. Soc. Res. 26 (2), 43. Krippendorff, K., 2013. In: Third ed. Inc. SP (Eds.), Content Analysis. An Introduction to its Methodology, Third Edition. The Annenbrrg School for Communication, University of Pennsylvania, Thousand Oaks, CA, 456 p. Kyriacos, U., Jelsma, J., Jordan, S., 2011. Monitoring vital signs using early warning scoring systems: a review of the literature. J. Nurs. Manag. 19 (3), 311–330, PubMed PMID: 21507102. Epub 2011/04/22.eng. Lockwood, C., Conroy-Hiller, T., Page, T. Vital Signs, 2004. Systematic review. JBI Rep. 2, 207–230. Ludikhuize, J., Smorenburg, S.M., de Rooij, S.E., de Jonge, E., 2012. Identification of deteriorating patients on general wards; measurement of vital parameters and potential effectiveness of the Modified Early Warning Score. J. Crit. Care 27 (4), 424.e7–424.e13. McCambridge, J., Witton, J., Elbourne, D.R., 2014. Systematic review of the Hawthorne effect: new concepts are needed to study research participation effects. J. Clin. Epidemiol. 67 (3), 267–277, 08/13/accepted, PubMed PMID: PMC3969247. National Institute for Clinical Health Excellence, 2007. Acutely ill patients in hospital Recognition of and response to acute illness in adults in hospital. NICE Clinical guideline 50, pp. 5824. (cited 2015 September). Available from: http://www.nice.org.uk/nicemedia/live/11810/ 58247/7.pdf. New South Wales Health, 2014. Liverpool Hospital Policy and Guideline Manual, Observations of Adult Patients, LH_PD2014_P01.55. Osborne, S., Douglas, C., Reid, C., Jones, L., Gardner, G., 2015. The primacy of vital signs – acute care nurses’ and midwives’ use of physical assessment skills: a cross sectional study. Int. J. Nurs. Stud. 52 (5), 951–962. Peat, M., Entwistle, V., Hall, J., Birks, Y., Golder, S., 2010. Scoping review and approach to appraisal of interventions intended to involve patients in patient safety. J. Health Serv. Res. Policy 15 (Suppl 1), 17–25, PubMed PMID: 20075123. Epub 2010/01/28.eng. Ryan, H., Cadman, C., Hann, L., 2004. Setting standards for assessment of ward patients at risk of deterioration. Brit. J. Nurs. 13 (20), 1186– 1190. Sahlsten, M.J., Larsson, I.E., Sjostrom, B., Plos, K.A., 2008. An analysis of the concept of patient participation. Nurs. Forum 43 (1), 2–11, PubMed PMID: 18269439. Epub 2008/02/14.eng. Smith, L.B., Banner, L., Lozano, D., Olney, C.M., Friedman, B., 2009. Connected care: reducing errors through automated vital signs data upload. Comput Inf. Nurs.: CIN 27 (5), 318–323. Tobiano, G., Marshall, A., Bucknall, T., Chaboyer, W., 2015. Patient participation in nursing care on medical wards: an integrative review. Int. J. Nurs. Stud. 52 (6), 1107–1120. van Beuzekom, M., Boer, F., Akerboom, S., Hudson, P., 2012. Patient safety in the operating room: an intervention study on latent risk factors. BMC Surg. 12, 10, PubMed PMID: 22726757. Pubmed Central PMCID: PMC3407486. Epub 2012/06/26.eng. von Elm, E., Altman, D.G., Egger, M., Pocock, S.J., Gøtzsche, P.C., Vandenbroucke, J.P., et al., 2007. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Bull. World Health Org. 85 (11), 867– 872, 06/20/received. 08/13/received. 08/20/accepted. PubMed PMID: PMC2636253. Welch, J., Moon, J., McCombie, S., 2012. Early detection of the deteriorating patient: the case for a multi-parameter patient-worn monitor. Biomed. Instrum. Technol./Assoc. Adv. Med. Instrum. (Suppl(Fall)), 57–64. Zimlichman, E., Shinar, Z., Rozenblum, R., Levkovich, S., Skiano, S., SzyperKravitz, M., et al., 2011. Using continuous motion monitoring technology to determine patient’s risk for development of pressure ulcers. J. Patient Saf. 7 (4), 181–184.