Content uploaded by Jean E Stevenson
Author content
All content in this area was uploaded by Jean E Stevenson on Jul 05, 2019
Content may be subject to copyright.
Content uploaded by Jean E Stevenson
Author content
All content in this area was uploaded by Jean E Stevenson on Jun 26, 2019
Content may be subject to copyright.
Accepted Article
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/jocn.14174
This article is protected by copyright. All rights reserved.
DR. JEAN STEVENSON (Orcid ID : 0000-0002-4626-3979)
Article type : Original Article
Factors influencing the quality of vital signs data in electronic health
records: a qualitative study
Jean E Stevenson, (Corresponding author)
PhD, Registered Nurse (RN)
Information School, Sheffield University, 211 Portobello Road, Sheffield, South
Yorkshire, S1 4DP, UK/eHealth Institute, Linnaeus University, SE-39182 Kalmar,
Sweden
email: jean.stevenson.agren@lnu.se
(Tel) +46 480 446473
(Co-authors)
Johan Israelsson,
PhD student, RN
Department of Internal Medicine, Division of Cardiology, Kalmar County Hospital, SE-
39185 Kalmar, Sweden; Kalmar Maritime Academy, Linnaeus University, Kalmar,
Sweden; Department of Medical and Health Sciences, Division of Nursing Science,
Linköping University, Linköping, Sweden
email: johani@ltkalmar.se
Goran Petersson
Professor
eHealth Institute, Linnaeus University, SE-39182 Kalmar, Sweden
email: goran.petersson@lnu.se
Peter A Bath
Professor
Information School, Sheffield University, 211 Portobello Road, Sheffield, South
Yorkshire, S1 4DP, UK
email: p.a.bath@sheffield.ac.uk
FUNDING
This research received no specific grant from any funding agency in the public,
commercial or not-for-profit sectors.
Accepted Article
This article is protected by copyright. All rights reserved.
ABSTRACT
Aims and objectives: The purpose of this study was to investigate reasons for inadequate
documentation of vital signs in an electronic health record.
Background: Monitoring vital signs is crucial to detecting and responding to patient
deterioration. The ways in which vital signs are documented in electronic health records have
received limited attention in the research literature. A previous study revealed that vital signs in
an electronic health record were incomplete and inconsistent.
Design: Qualitative study.
Methods: Qualitative study. Data were collected by observing (68 hours) and interviewing nurses
(n=11) and doctors (n=3), and analysed by thematic analysis to examine processes for measuring,
documenting and retrieving vital signs in four clinical settings in a 353-bed hospital.
Results: We identified two central reasons for inadequate vital sign documentation. First, there
was an absence of firm guidelines for observing patients' vital signs, resulting in inconsistencies
in the ways vital signs were recorded. Second, there was a lack of adequate facilities in the
electronic health record for recording vital signs. This led to poor presentation of vital signs in
the electronic health record and to staff creating paper 'workarounds'.
Conclusions: This study demonstrated inadequate routines and poor facilities for vital sign
documentation in an electronic health record, and makes an important contribution to
knowledge by identifying problems and barriers that may occur. Further, it has demonstrated the
need for improved facilities for electronic documentation of vital signs.
Relevance to clinical practice: patient safety may have been compromised because of poor
presentation of vital signs. Thus, our results emphasised the need for standardised routines for
monitoring patients. In addition, designers should consult the clinical end-users in order to
optimise facilities for electronic documentation of vital signs. This could have a positive impact
on clinical practice and thus improve patient safety.
Keywords: vital signs, electronic health records, patient safety, qualitative study
'What does this paper contribute to the wider global clinical community?'
Given the importance of recording vital signs, it is essential to understand
problems and barriers to monitoring patient deterioration. Inadequate
routines for measuring vital signs and poor facilities in electronic health
records can have a negative impact on the quality of vital signs presentation.
National guidelines on patient surveillance are warranted.
The findings of this study can contribute to knowledge necessary for
improving the design of electronic health records and thus improve patient
safety.
Keywords: vital signs, electronic health records, patient safety, qualitative study
INTRODUCTION
This article presents the second phase of a mixed methods investigation into documentation of
vital signs in electronic health records (EHRs). In the first phase, a quantitative study was
performed and found that vital signs were incomplete and inconsistently scattered throughout
three sections of the EHR (Stevenson, Israelsson, Nilsson, Petersson, & Bath, 2016). The
Accepted Article
This article is protected by copyright. All rights reserved.
current paper presents a follow-up study, using a qualitative approach, the aim of which was to
investigate in greater detail the reasons for inadequate documentation of vital signs in an EHR.
BACKGROUND
Abnormal vital signs are associated with adverse events such as cardiac arrest and unplanned
admission to intensive care units (Goldhill, Worthington, Mulcahy, Tarling, & Sumner, 1999;
Hillman et al., 2002; Kause et al., 2004). Monitoring vital signs is, therefore, crucial to
recognising and responding to patient deterioration (DeVita et al., 2006; National Institute for
Health and Clinical Experience, 2007; Smith, 2010; Smith et al., 2006). Several systems for
detecting patient deterioration have been developed, e.g., early warning score (EWS) systems. A
number of studies have shown that the way in which vital signs can be viewed in a clinical record
can have a major effect on the ability of clinicians to detect deterioration in a patient's
physiological state (Chatterjee, Moon, Murphy, & McCrea, 2005; Elliott, 2015; Horswill,
Preece, Hill, & Watson, 2010; Preece, Hill, Horswill, & Watson, 2012). However, the way in
which vital signs are presented in EHR systems has received minimal research attention. Since
EHRs are increasingly used for all aspects of clinical documentation (Ayatollahi, Bath, &
Goodacre, 2010), including vital signs, it is crucial that we understand the possible impact that
this transition from paper to electronic systems might have. At the time of this research, the
study hospital had been using an EHR for approximately eight years. Our previous work
demonstrated inconsistencies in vital sign documentation indicating the need for further
investigation into all aspects of vital signs monitoring: routines for measuring vital signs; and
recording and viewing vital signs in the EHR.
METHODS
Design
Using a qualitative approach, we examined work processes of medical and nursing staff for
measuring, documenting and retrieving vital signs. Data were collected through observations and
semi-structured interviews between August 2014 and January 2015. The data were analysed using
thematic analysis (Braun & Clarke, 2006). Data collected during interviews were used to
corroborate the results of the observational data. In addition, an EWS system, Vitalpac Early
Warning Scores (ViEWS), was used as a validated model for monitoring vital signs (Prytherch,
Smith, Schmidt, & Featherstone, 2010). In ViEWS, six vital signs (temperature, pulse, respiratory
rate, blood pressure, oxygen saturation and conscious level) are measured twice daily.
Setting
The setting was a 353-bed district general hospital in Sweden. Four independent clinical settings
were included: an acute medical department (cardiology unit (CU), 40 beds); an acute surgical
ward (SW) (28 beds); an infection ward (IW) (18 beds); and an emergency department (ED). The
ED attended to approximately 35,000 patients annually. These clinical areas were selected as they
provided a wide range of patients receiving acute clinical care who would require monitoring of
vital signs. The hospital had no general guidelines for monitoring vital signs. An EHR had been
implemented in 2007.
Accepted Article
This article is protected by copyright. All rights reserved.
Participants
The sampling method was a purposive, convenience sample. Nursing and medical staff who used
the EHR were invited to take part in the study. Several nurses volunteered immediately and,
during observation sessions, further volunteers were recruited. The observational study included
a total of 68 hours of observations in four clinical areas. In the interview study, interviews were
conducted with eleven registered nurses (RNs) and three medical doctors (MDs).
Data collection and analysis
Within each clinical area, data collection was performed in two sequential stages: first, the
observations were undertaken. These were followed by conducting the semi-structured
interviews, providing the opportunity to probe issues raised during observations.
Observations
Observations aimed to explore all aspects of vital sign measurement, documentation and
retrieval: which vital signs were measured, how often they were measured, why they were
measured; where they were documented, how they were documented; how they were retrieved.
Nurses were observed in their everyday duties, and particular attention was given to actions and
activities that were associated with vital signs. Observed data were recorded in field notes by one
of the researchers (JS).
Analysis of observational data
The observational data were analysed using an inductive approach (Elo & Kyngäs, 2008). The
data were coded. How and why vital signs were documented in the EHR were explored. A
matrix (Graneheim & Lundman, 2004) was used to ensure that meaning units were analysed
systematically. Figure 1 displays an example of the matrix. Thus, the data from the observational
study were organised into themes. Three main themes emerged: measuring vital signs,
documenting vital signs and retrieving vital signs. These themes were then used to inform the
questions for the semi-structured interviews.
Semi-structured interviews
An interview guide with semi-structured questions was designed based on the three main themes
that had emerged from the analysis of the observational study. This allowed us to explore issues
raised from the observations. An informal, flexible approach was taken. Each interview lasted
approximately 19 minutes (+/- 9). Interviews were carried out by JS in each of the clinical
settings until data saturation was reached. Data were recorded using digital audio equipment.
Analysis of interview data
Because there were pre-determined themes, the data analysis was carried out using a deductive
approach (Elo & Kyngäs, 2008). Data were coded and interview excerpts were identified to
illustrate the themes.
Combined data analysis
Data sets from both the observational study and the interview study were combined to make one
corpus and used to corroborate one another. Analysis of the measurement of vital signs was
based on and compared to ViEWS. The accuracy of the results was validated by member
Accepted Article
This article is protected by copyright. All rights reserved.
checking within the same clinical areas, i.e., key participants agreed that the findings reflected
their practice (Creswell & Plano Clark, 2007).
Ethical considerations
The Regional Ethical Review Board, Linköping, Sweden, approved the study (DNR 2014/351-
52). The managers of the clinical areas consented to the study. Participants gave their informed
consent, those being observed as well as those being interviewed.
RESULTS
The results are presented using the main themes that were identified: measuring vital signs;
documenting vital signs; and retrieving vital signs.
Measuring vital signs
In the four clinical areas, there was noticeable variation regarding which, and how often,
individual vital signs were measured. These were compared to ViEWS and are shown in Table 1.
In the ED, the routine for measuring vital signs was well-defined and was similar to routines
demanded in ViEWS with all vital signs recorded on admission and subsequent vital signs
recorded according to the seriousness of the patient's condition. The ED used the rapid
emergency triage and treatment system (RETTS) (Widgren, 2013), which mandates that all vital
signs (temperature, pulse, respiratory rate, blood pressure, oxygen saturation and conscious level)
are taken and recorded in the system so that the triage priority can be estimated.
There were well-defined guidelines in the SW. Five of the ViEWS vital signs were recorded
routinely once daily, thereby varying from ViEWS in frequency, and that conscious level was not
included routinely. These vital signs were also recorded when a patient was admitted, when
transferred from ICU or recovery room, and after surgical procedures. The guidelines also
specified that patients who showed any signs of clinical deterioration needed to have vital signs
checked. These guidelines provided a clear routine that seemed to be adhered to by nursing staff
(Field note. 12).
There were some occasions in which isolated vital signs were taken in both the ED and the SW.
This could be an extra blood pressure recording, for example, if a patient seemed unwell. These
additional checks were considered 'extra' recordings to find out if a patient was alright.
Nevertheless, universal guidelines recommend recording complete sets of vital signs (DeVita et
al., 2010).
Findings in the CU and the IW were rather more complex. Neither RETTS nor written
guidelines were used in these clinical areas. Within cardiology:
"The routine is to measure three times a day, morning, afternoon and evening". (Interview 2. Nurse CU)
Decisions on which vital signs should be taken at these routine times were made individually by
nurses:
Accepted Article
This article is protected by copyright. All rights reserved.
"It's blood pressure, pulse, saturation. It's always blood pressure and pulse as a routine, you could say." (Interview 2.
Nurse CU)
Blood pressure and pulse were considered the most important vital signs in the CU. Clinical
judgement drove decisions on whether additional vital signs were required:
"We would take vital signs if a patient became poorly. Then we measure temperature, pulse and blood pressure, and
oxygen saturation - those four. Sometimes we measure respiratory rate if a patient is breathless." (Field note 1 CU)
In the IW, the standard process was to record vital signs on admission and at least twice daily.
Only temperature was routinely included in these twice-daily recordings as it was considered the
most important. The remaining vital signs were measured 'as required' (Interview 12. Nurse IW).
"With us, all patients have their temperature checked - that's what we're interested in." (Interview 12. Nurse IW)
This meant that nurses used their clinical judgement to decide if they would measure additional
vital signs:
"If we notice that a patient's condition worsens, then we decide ourselves as nurses if we should take it more often."
(Interview 11. Nurse IW)
These results show that there was clear variation within the clinical areas for which vital signs
were measured and how often this occurred.
Documenting vital signs
The findings showed that documentation of vital signs was not only carried out in the EHR but
also on paper. A summary of the findings is shown in Figure 2.
Electronic health records
In the EHR, vital signs were documented in three sections known as: the table; the journal; and
the report sheet. Figure 3 shows a screen shot of the table, in which all vital signs were entered as
numbers with no available graph. In all four clinical areas, the most frequently used section of
the EHR was the 'table' (Figure 3).
Tables
Tables had to be generated in the EHR for each new patient:
". . . every time you get a new patient you have to create a new one [table]. (Interview 1. Nurse CU)
Nurses found the table awkward to use, and time-consuming:
". . . there are very many clicks before you can get anywhere. Masses of clicks . . . if there were less clicks it would be
easier." (Interview 12. Nurse IW)
They also found that the table could not accommodate many vital signs, for example, if a patient
was very sick and needed frequent recordings, because then, the table became too 'long':
Accepted Article
This article is protected by copyright. All rights reserved.
"You don't write down every time you check . . . then we enter the first BP and the final BP [in the EHR] instead of
filling in each pulse and blood pressure in the patient table." (Interview 1. Nurse CU)
These findings suggest that the table in the EHR did not support the recording of recurrent vital
signs.
Journal
The journal section of the EHR was used in some instances. In the ED, the first set of vital signs
was automatically transferred from RETTS to the journal. However, as this quotation illustrates .
. . :
"It is only the first vital signs that we write in the RETTS triage module. After that we write vital signs in the table
for vital signs in the EHR, as is done in the rest of the hospital." (Interview 9. Nurse ED)
Not having baseline recordings on the same page as subsequent recordings could make it
difficult to follow a patient's progress or any changes from the time of admission.
A nurse explained that she sometimes wrote vital signs in the journal notes so that she could give
a better account of the patient.
"They [vital signs] are always written in the table but it can be that a summary is written in the journal notes."
(Interview 12. Nurse IW)
Report sheet
Nurses also said that they sometimes preferred using the report sheet to the table. One reason
was that it was quicker:
"I use the report sheet, because it is faster, if I don't have time to create a column [in the table], then I quickly write
it in the report sheet." (Interview 3. Nurse CU)
A further reason for writing in the report sheet was to alert nurses to any abnormalities or
potential problems:
"In the report sheet . . . I write if a patient is deteriorating." (Interview 1. Nurse CU)
"I can tell the whole story there . . . if the patient becomes poorly, I think the next nurse needs to be able to read
what has happened." (Interview 3. Nurse CU)
Nurses felt the report sheet was suitable if they wanted to write additional details about the
patient's care.
Paper charts
Paper charts were used extensively alongside the EHR, so-called 'workarounds', which have been
reported previously (Tucker, 2009). One reason was if a patient required frequent vital signs to
be measured, e.g., if a patient was very ill, unstable, had had surgery or commenced a new
treatment. In cardiology:
Accepted Article
This article is protected by copyright. All rights reserved.
"You have an acutely sick patient and you don't have time to document all this [in the EHR]. Then it is paper charts
that are suitable. That's because we might be taking the vital signs every 5 minutes, so we don't put all of those in the
EHR." (Interview 4. Nurse CU)
"On the paper, it is certainly easier to see. Then you don't need to click in the EHR, you don't always have the page
with the table open in the EHR, so if you have it there on paper, then you can see it quickly." (Interview 3. Nurse
CU)
In the ED:
"We write the vital signs on a paper first when we admit a patient . . .This is just so that we can have it for ourselves,
to simplify, to be able to quickly hand over to the doctor so they don't have to go to the EHR." (Interview 10.
Nurse ED)
The doctor collected this paper when going to see a new patient (Field note ED).
In the infection ward:
"If we have a sepsis patient who is unstable in the ward, then there is a paper chart which is left in the room with the
patient. You don't need to run back and forth between the computer and the patient . . ., instead you have a paper
that is there all the time." (Interview 13. Doctor IW)
When patients were discharged, these paper charts were electronically scanned and stored in a
separate part of the EHR.
For routine vital signs, paper lists with room numbers were used to document at the bedside:
"We have the [paper] list so that we have it when we go round [the patients]. You don't always have time to stand at
the computer. We write it in later. . . . We have a routine that it is written in the EHR before we go home."
(Interview 12. Nurse IW)
The SW had one pre-printed paper chart for the 'once daily' routine recordings, with the room
numbers in the left vertical column and the vital signs in a horizontal line along the top.
"We take one of those lists so that we see that it is done on everyone [vital signs]. Then we see if one is not done
then we go in and take them. The paper is like a checklist. We write it into the EHR later." (Interview 14. Nurse SW)
Thus, paper charts were adopted because the EHR did not fit in with the workflow. Staff felt
that they could obtain a more complete overview of vital signs on a paper chart. Moreover, it
was simpler and quicker and paper charts could be kept at the bedside, allowing documentation
at the point of care.
Paper notes
In addition to pre-printed paper charts, many paper notes were used as and when they were
required on whatever paper happened to be available, e.g., note pads, 'post-its' and, occasionally,
paper towels, were used for taking quick notes of vital signs. In Cardiology:
Accepted Article
This article is protected by copyright. All rights reserved.
"Most often you have a paper in your pocket or a paper towel, (laughs) whatever, so that you have something to
write on and make it easy. That's the way it is . . . so that I remember it until I get to a computer." (Interview 4.
Nurse CU)
"And sometimes that may be towards the end of my shift, as it can be very busy." (Interview 3. Nurse CU)
In the ED, it was observed that when vital signs were checked in locations that were some
distance from the nearest computer, paper notes were used and afterwards the information was
transcribed to the EHR. These excerpts indicate the need for paper notes to ensure 'point of
care' documentation when nurses were busy or were not close to a computer.
Retrieval of vital signs
The methods by which vital signs were retrieved varied. In cardiology, most nurses looked in the
journal first and then the report sheet. For example:
"I usually read the journal and a little in the report sheet. I do not go in so often on the table for vital signs."
(Interview 1. Nurse CU).
Doctors vented most concern about viewing vital signs. A doctor from CU indicated that it was
problematic to see a patient's vital signs at a glance and also said that an excessive number of
clicks was required before you could view whatever you were looking for in the EHR:
"It is very hard to get to this information in [the computer]. I must make many clicks to see those vital signs . . .
They are not easily accessible." (Interview 8. Consultant CU)
This cardiology consultant emphasised how important it is to know the vital signs of acutely sick
patients immediately.
"It's quicker to ask the nurses what the vital signs are than it is to find them in the EHR." (Interview 7. Doctor CU)
". . . I get the report from the night nurse. Most often they tell me the blood pressure, if it has been fine, or if it is
high or low." (Interview 6. Nurse CU).
In the ED, a nurse reported that:
"Our triage vital signs can be seen in the journal part of the EHR . . . you must go in to the table to see what
subsequent vital signs the patient has had". (Interview 9. Nurse ED)
Thus, comparing admission vital signs to subsequent values required users to view two
independent locations in the EHR.
In the IW, nurses often used the paper 'list' to view vital signs. This list was kept on the trolley at
all times so that nurses could easily access the most recent vital signs:
"We have the list so that we have it when we go round. You don't always have time to stand at the computer."
(Interview 12. Nurse IW)
Accepted Article
This article is protected by copyright. All rights reserved.
These paper 'lists' ensured that vital information was easily available. A doctor reported that:
"First I ask the nurse for the latest vital signs. They can often be given verbally." (Interview 13. Doctor IW).
In the SW, it was routine to retrieve vital signs from the table:
"We are so used to it. You have a row with all the vital signs and you can look in the history. And if there is
anything, then they are there." (Interview 14. Nurse SW)
Clearly, various practices for documenting and retrieving vital signs among the four clinical areas
existed.
Summary of results
There were two central factors influencing the quality of vital signs data in the EHR. The first
was associated with practices and routines for measuring vital signs and the second concerned facilities
and functions in the EHR (Table 2).
DISCUSSION
The aim of our study was to develop a better understanding of why documentation of vital signs
in an EHR may be incomplete and inconsistent (Stevenson et al., 2016). We identified two
central reasons that may help to explain this. The first is that the absence of firm guidelines,
practices and routines for observing patient's vital signs can result in many vital signs not being
measured routinely. The second is that the lack of adequate facilities and functions for
documenting vital signs in an EHR, can make it difficult both to enter and to locate and retrieve
the data. Given the importance of recording vital signs (DeVita et al., 2010; Smith, 2010), it is
essential to understand problems and barriers to recording and monitoring patient deterioration.
This is particularly important with the development of EHR systems, and the transition from
traditional paper-based systems. These results can help developers to design systems that fit into
the work practices of health care professionals.
Practices and routines for measuring vital signs
The benefits of early warning score (EWS) systems are widely accepted and have been adopted
internationally for early detection of deteriorating patients (Hammond et al., 2013; Ludikhuize,
Smorenburg, de Rooij, & de Jonge, 2012; Monsieurs et al., 2015). They are associated with the
reduction of cardiac arrest, unplanned admissions to ICU, and death (DeVita et al., 2010; Smith,
2010). Where a hospital, or health care provider does not have a policy for vital sign recordings,
patient surveillance and monitoring can vary extensively depending on the clinical area. Three of
the four areas in our study showed considerable disparity to ViEWS.
In this study, clear guidelines such as the RETTS tool used in the ED and the guidelines used in
the SW had observable benefits; vital signs were recorded routinely and not only when there was
a clinical indication. Precise guidelines did not exist in the CU or the IW. The association
between clear procedures for recording specific vital signs have been demonstrated, with EWS
systems having been shown to enhance the frequency of recordings, particularly of respiratory
rate (Bunkenborg, Samuelsson, Poulsen, Ladelund, & Åkeson, 2013; Cahill et al., 2011; McBride,
Accepted Article
This article is protected by copyright. All rights reserved.
Knight, Piper, & Smith, 2005; McCormick, 2005; Odell et al., 2007; Sharpley & Holden, 2004). A
previous study showed low frequency in the documentation of respiratory rate (Stevenson et al.,
2016). The current study found that decisions for measuring vital signs were made at
departmental or ward level, or left to individual nurse's clinical judgement. A number of studies
have shown that clinical judgement can be subjective and should not replace the systematic
recording of vital signs (Cuthbertson & Smith, 2007; Fullerton, Price, Silvey, Brace, & Perkins,
2012; Preece et al., 2012; Zarabzadeh et al., 2013).
Given the advantages of evidence-based patient surveillance systems such as EWS, this study
underlines the importance of a hospital wide policy for detecting patient deterioration (Smith et
al., 2006). Although many health authorities in Sweden have now adopted EWS systems, there is
not, to date, a national policy. Such a policy could have the potential to improve patient safety
and promote equality in care across Sweden.
Facilities and functions for documenting and retrieving vital signs in the EHR
Recording vital signs in EHRs has received minimal attention in previous research; however, the
results of this study elucidated several problems when considered alongside key requirements of
electronic record systems. In health care settings, interfaces need be streamlined to work
processes and meet specified goals to be safe, efficient and effective (Kushniruk & Patel, 2004;
Peute, Spithoven, Bakker, & Jaspers, 2008), and an EHR should facilitate clinical work and take
account of work routines (Ovretveit, Scott, Rundall, Shortell, & Brommels, 2007). Clearly, the
EHR in this study did not meet these demands for recording vital signs. In another Swedish
study, five out of nine hospitals used paper charting alongside the EHR to support the
documentation of vital signs, indicating that these hospitals also experienced difficulties in vital
signs documentation (Skyttberg et al., 2016).
In our study, there were three main concerns when documenting and retrieving vital signs in the
EHR: consistency, correctness and currency (Russ et al., 2010). First, there was a lack of
consistency as vital signs could be documented in three locations. The variation in the location
of vital signs could lead to difficulties in viewing consecutively recorded vital signs, and thus,
immediate changes in a patient’s condition, or trends, might easily be missed. For example, in the
ED, vital signs from RETTS were automatically charted in the journal with subsequent vital
signs manually recorded in the table section of the EHR. This could hinder appropriate clinical
decision making and could compromise patient safety.
Second, vital signs were not adequately presented. Previous research has shown that the way in
which vital signs are presented visually has a profound impact on the ability of clinicians to
identify patient deterioration and emphasises the need for coloured charts and plotted graphs;
the possibility to easily identify trends enhances the processing of information, enables rapid
detection of changes and facilitates informed decision making (Chatterjee, Moon, Murphy, &
McCrea, 2005; Preece et al., 2012). Viewing vital signs in a table, such as that used by health care
professionals in our study, meant that changes and trends could not be recognised easily or
quickly (Figure 3). Previous studies have shown that vital signs should be presented on plotted
graphs which are colour coded to ensure clinicians are able to identify patient deterioration at a
glance (Chatterjee et al., 2005; Horswill et al., 2010; Preece, Horswill, Hill, & Watson, 2010).
Accepted Article
This article is protected by copyright. All rights reserved.
Third, the vital signs recorded in the EHR were not current. Entering and accessing vital signs
was time consuming and cumbersome resulting in the staff delaying entry of vital signs into the
EHR; the delay could be from a few minutes after measurement, to several hours, or could be
delayed until the end of a shift. If frequent vital signs were required, paper observation charts
were used instead so these would not be added to the EHR until they were scanned in when a
patient was discharged. Thus, vital signs in the EHR were often not current and the latest vital
signs were not available for clinical assessment.
"Workarounds are non-standard methods for accomplishing work blocked by dysfunctional
processes" (Tucker, 2009). The 'workarounds' mentioned above were initiated to fill functional
gaps. The staff in this hospital used paper notes and checklists, because it was quicker and more
convenient; they carried out the 'official' documentation in the EHR at a later time. When
frequent vital sign recordings were necessary, workarounds in the form of paper observation
charts were used; it was quicker and easier and it meant that vital signs could be entered at the
point-of-care and were available for viewing beside the patient. It is important, therefore, that
EHR systems are designed to facilitate easy recording of vital signs, as well as other clinical
measurements, in order to avoid the need for workarounds.
The paper observation charts created by staff were not optimal as they were all in the form of
tables, thereby not meeting the recommendations of evidence-based observation charts that have
coloured bands and plotted graphs, which have the possibility to detect changes quickly and
facilitate informed decision making (Chatterjee et al., 2005; Horswill et al., 2010; Preece et al.,
2012). Therefore, if paper charts are employed, we recommend that a standardised evidence-
based chart is adopted to ensure timely detection of patient deterioration.
EHRs are used in all hospitals in Sweden and there are several providers. There is a dearth of
research into the way in which vital signs are presented in EHRs: this study helps to develop a
better understanding of the problems associated with recording important information in EHRs.
Clearly, the way in which vital signs were recorded in this EHR did not concur with evidence-
based guidelines for paper observation charts, and this suggests that developers and providers of
EHRs need a better understanding of how to accommodate vital signs, and of their importance.
This study adds to the evidence base on vital signs in EHRs. It demonstrates shortfalls in this
EHR, one widely used in Sweden (Nilsson, 2013), when it comes to documenting and retrieving
vital signs. In light of our study, it could be time for a standardised policy for documentation of
vital signs in EHRs to promote patient safety and enhance equality of care. This would mean
that those who procure systems for use in acute care could insist on basic evidence-based
characteristics in EHRs and require developers to involve health care professionals in the design,
development and testing of systems. However, the degree to which managers involve research-
based evidence in decision making is unclear (Walshe & Rundall, 2001). Furthermore, there is a
need of consensus among stakeholders when introducing interventions such as electronic
documentation (Ericson, Hammar, Schönström, & Petersson, 2017).
Accepted Article
This article is protected by copyright. All rights reserved.
Strengths and weaknesses
A limitation of the study was that it just examined the use of a single EHR system. This could
mean that the findings may not be generalisable to all EHRs. However, the general issues that
were raised in relation to this particular EHR system are likely to be transferable to similar
systems in which there are problems with documentation of vital signs. A strength of this study
was that it complemented previous quantitative research on vital sign documentation in an EHR
(Stevenson et al., 2016). It elucidated reasons for incomplete documentation identified in the
quantitative study by examining these issues more deeply, and thus, completed the second phase
of a mixed methods investigation (Creswell, 2013). An additional strength of our study was that
it was carried out in different areas and examined a broad selection of data on vital sign
monitoring within one hospital. A limitation of qualitative methods based on observations is that
staff may behave differently when they know that they are being observed (Kumar, 2011). On
the other hand, two qualitative approaches - observations and interviews - provided triangulation
of data sources and ensured consistency of the results in this qualitative study (Johnson,
Onwuegbuzie, & Turner, 2007).
CONCLUSIONS
Monitoring patients' vital signs is essential to safe patient care. This study makes an important
contribution to knowledge about the documentation of vital signs in an EHR, by identifying
problems and barriers that may occur. It has demonstrated the need for improved facilities for
electronic documentation of vital signs. Moreover, it has confirmed the need for standardised
routines for measuring vital signs. Complete and current documentation may be possible in
EHRs when issues of safety and user-friendliness are addressed and iteratively evaluated. Until
then, parallel paper systems are likely to persist. Previous studies have shown that nursing
documentation in the EHR system can be difficult and inefficient (Stevenson & Nilsson, 2012);
however, this research has demonstrated that inadequate facilities for documenting vital signs
could have an adverse impact on patient safety.
RELEVANCE TO CLINICAL PRACTICE
Inconsistencies in recording vital signs may delay recognition of patient deterioration and lead to
adverse outcomes. Patient safety may have been compromised because of poor presentation of
vital signs. Thus, our results emphasised the need for local and national guidelines on patient
monitoring to improve clinical practice. Moreover, designers of electronic health records need to
consult with clinical end-users in order to optimise facilities for electronic documentation of vital
signs. Applying these suggestions could have a positive impact on patient safety.
REFERENCES
Ayatollahi, H., Bath, P. A., & Goodacre, S. (2010). Factors influencing the use of IT in the
emergency department: a qualitative study. Health Informatics Journal, 16(3), 189-200. doi:
10.1177/1460458210377480
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in
Psychology, 3(2), 77-101. doi: 10.1191/1478088706qp063oa
Bunkenborg, G., Samuelsson, K., Poulsen, I., Ladelund, S., & Åkeson, J. (2013). Lower incidence
of unexpected in-hospital death after interprofessional implementation of a bedside
track-and-trigger system. Resuscitation, 85, 424-430.
Accepted Article
This article is protected by copyright. All rights reserved.
Cahill, H., Jones, A., Herkes, R., Cook, K., Stirling, A., Halbert, T., . . . Gattas, D. J. (2011).
Introduction of a new observation chart and education programme is associated with
higher rates of vital-sign ascertainment in hospital wards. BMJ quality & safety, 20(9), 791-
796. doi: 10.1136/bmjqs.2010.045096
Chatterjee, M., Moon, J., Murphy, R., & McCrea, D. (2005). The “OBS” chart: an evidence based
approach to re-design of the patient observation chart in a district general hospital
setting. Postgraduate Medical Journal, 81(960), 663-666.
Creswell, J. W. (2013). Research design: Qualitative, quantitative and mixed methods approaches. Thousand
Oaks: Sage.
Creswell, J. W., & Plano Clark, V. (2007). Designing and conducting mixed methods research (1st ed.).
Thousand Oaks: Sage.
Cuthbertson, B. H., & Smith, G. B. (2007). A warning on early-warning scores! British Journal
Anaesthesia, 98(6), 704-706. doi: 98/6/704 [pii]10.1093/bja/aem121
DeVita, M. A., Bellomo, R., Hillman, K., Kellum, J., Rotondi, A., Teres, D., . . . Galhotra, S.
(2006). Findings of the First Consensus Conference on Medical Emergency Teams.
Critical Care Medicine, 34(9), 2463-2478. doi: 10.1097/01.CCM.0000235743.38172.6E
DeVita, M. A., Smith, G. B., Adam, S. K., Adams-Pizarro, I., Buist, M., Bellomo, R., . . . Winters,
B. (2010). "Identifying the hospitalised patient in crisis"- a consensus conference on the
afferent limb of rapid response systems. Resuscitation, 81(4), 375-382. doi:
10.1016/j.resuscitation.2009.12.008
Elliott D, A. E., Perry L, et al (2015). Clinical user experiences of observation and response
charts: focus group findings of using a new format chart incorporating a track and trigger
system BMJ Qual Saf, 24, 65-75.
Elo, S., & Kyngäs, H. (2008). The qualitative content analysis process. Journal of Advanced Nursing,
62(1), 107-115. doi: 10.1111/j.1365-2648.2007.04569.x
Ericson, J., Hammar, T., Schönström, N., & Petersson, G. (2017). Stakeholder concensus on the
purpose of clinical evaluation of electronic health records is required. Health Policy and
Technology, 6, 152-160.
Fullerton, J., Price, C., Silvey, N., Brace, S., & Perkins, G. (2012). Is the Modified Early Warning
Score (MEWS) superior to clinician judgement in detecting critical illness in the pre-
hospital environment? Resuscitation, 83(5), 557-562.
Goldhill, D. R., Worthington, L., Mulcahy, A., Tarling, M., & Sumner, A. (1999). The patient-at-
risk team: identifying and managing seriously ill ward patients. Anaesthesia, 54(9), 853-860.
doi: ana996 [pii]
Graneheim, U. H., & Lundman, B. (2004). Qualitative content analysis in nursing research:
concepts, procedures and measures to achieve trustworthiness. Nurse Education Today, 24,
105-112.
Hammond, N., Spooner, A., Barnet, A., Corley, A., Brown, P., & Fraser, J. (2013). The effect of
implementing a modified early warning scoring (MEWS) system on the adequacy of vital
sign documentation. Australian Critical Care, 26, 18-22.
Hillman, K. M., Bristow, P. J., Chey, T., Daffurn, K., Jacques, T., Norman, S. L., . . . Simmons,
G. (2002). Duration of life-threatening antecedents prior to intensive care admission.
Intensive Care Medicine, 28(11), 1629-1634. doi: 10.1007/s00134-002-1496-y
Horswill, M., Preece, M., Hill, A., & Watson, M. (2010). Detecting abnormal vital signs on six
observation charts: An experimental comparison (S. o. Psychology, Trans.) (pp. 1-25).
Sydney, New South Wales, Australia: Australian Commission on Safety and Quality in
Health Care.
Johnson, R., Onwuegbuzie, A., & Turner, L. (2007). Toward a Definition of Mixed Methods
Research. Journal of Mixed Methods Research, 1(2), 112-133.
Kause, J., Smith, G., Prytherch, D., Parr, M., Flabouris, A., & Hillman, K. (2004). A comparison
of antecedents to cardiac arrests, deaths and emergency intensive care admissions in
Accepted Article
This article is protected by copyright. All rights reserved.
Australia and New Zealand, and the United Kingdom-the ACADEMIA study.
Resuscitation, 62(3), 275-282. doi: 10.1016/j.resuscitation.2004.05.016
Kumar, R. (2011). Research methodology (3rd ed.). New Delhi: Sage.
Kushniruk, A., & Patel, V. (2004). Cognitive and usability engineering methods for the
evaluation of clinical information systems. Journal of Biomedical Informatics, 37, 56-76.
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. Journal of Critical Care, 27(4),
424.e427-424.e413. doi: 10.1016/j.jcrc.2012.01.003
McBride, J., Knight, D., Piper, J., & Smith, G. B. (2005). Long-term effect of introducing an
early warning score on respiratory rate charting on general wards. Resuscitation, 65(1), 41-
44. doi: 10.1016/j.resuscitation.2004.10.015
McCormick, J. (2005). Design and implementation of an observation chart with an integrated
early warning score. Practice Development in Health Care, 4(2), 69-76.
Monsieurs, K. G., Nolan, J. P., Bossaert, L. L., Greif, R., Maconochie, I. K., Nikolaou, N. I.,
Zideman, D. A. (2015). European Resuscitation Council Guidelines for Resuscitation
2015: Section 1. Executive summary. Resuscitation, 95, 1-80. doi:
http://dx.doi.org/10.1016/j.resuscitation.2015.07.038
National Institute for Health and Clinical Experience, c. g. (2007, July 2007). Acutely ill patients
in hospital: recognition of and response to acute illness in adults in hospital. Retrieved 6
December, 2009, from http://www.nice.org.uk/nicemedia/pdf/CG50FullGuidance.pdf
Nilsson, C. (2013, 26 September 2013). De är störst på journalsystem. IT i Vården. Retrieved
from http://itivarden.idg.se/2.2898/1.525235/de-ar-storst-pa-journalsystem
Odell, M., Rechner, I. J., Kapila, A., Even, T., Oliver, D., Davies, C. W., . . . Rudman, K. (2007).
The effect of a critical care outreach service and an early warning scoring system on
respiratory rate recording on the general wards. Resuscitation, 74(3), 470-475. doi: S0300-
9572(07)00079-2 [pii]10.1016/j.resuscitation.2007.01.035
Ovretveit, J., Scott, T., Rundall, T. G., Shortell, S. M., & Brommels, M. (2007). Improving quality
through effective implementation of information technology in healthcare. International
Journal for Quality in Health Care, 19(5), 259 - 266.
Peute, L. W., Spithoven, R., Bakker, P. J., & Jaspers, M. W. (2008). Usability studies on
interactive health information systems; where do we stand? Studies in Health Technology and
Informatics, 136, 327-332.
Preece, M., Hill, A., Horswill, M., & Watson, M. (2012). Supporting the detection of patient
deterioration: Observation chart design affects the recognition of abnormal vital signs.
Resuscitation, 83(9), 1111-1118. doi: 10.1016/j.resuscitation.2012.02.009
Preece, M., Horswill, M., Hill, A., & Watson, M. (2010). The development of the Adult
Deterioration Detection System (ADDS) chart. Sydney. New South Wales, Australia:
Australian Commission on Safety and Quality in Health Care.
Preece, M. H., Hill, A., Horswill, M. S., Karamatic, R., & Watson, M. O. (2012). Designing
observation charts to optimize the detection of patient deterioriation: Reliance on the
subjective preferences of healthcare professionals is not enough. Australian Critical Care.
doi:10.1016/j.aucc.2012.01.003
Prytherch, D. R., Smith, G. B., Schmidt, P. E., & Featherstone, P. I. (2010). ViEWS--Towards a
national early warning score for detecting adult inpatient deterioration. Resuscitation, 81(8),
932-937. doi: S0300-9572(10)00242-X [pii]10.1016/j.resuscitation.2010.04.014
Russ, A., Saleem, J., Justice, C., Woodward-Hagg, H., Woodbridge, P., & Doebbeling, B. (2010).
Electronic health information in use: Characteristics that support employee workflow
and patient care. Health Informatics Journal, 16(4), 287-305.
Sharpley, J. T., & Holden, J. C. (2004). Introducing an early warning scoring system in a district
general hospital. Nursing in Critical Care, 9(3), 98-103.
Accepted Article
This article is protected by copyright. All rights reserved.
Skyttberg, N., Vicente, J., Rong, C., Blomqvist, H., Koch, S., & Chen, R. (2016). How to
improve vital sign data quality for use in clinical decision support systems? A qualitative
study in nine Swedish emergency departments. BMC Medical Informatics and Decision
Making, 16, 1-12. doi: 10.1186/s12911-016-0305-4
Smith, G. B. (2010). In-hospital cardiac arrest: is it time for an in-hospital 'chain of prevention'?
Resuscitation, 81(9), 1209-1211. doi: 10.1016/j.resuscitation.2010.04.017
Smith, G. B., Prytherch, D. R., Schmidt, P., Featherstone, P. I., Knight, D., Clements, G., &
Mohammed, M. A. (2006). Hospital-wide physiological surveillance-a new approach to
the early identification and management of the sick patient. Resuscitation, 71(1), 19-28. doi:
10.1016/j.resuscitation.2006.03.008
Stevenson, J. E., Israelsson, J., Nilsson, G. C., Petersson, G. I., & Bath, P. A. (2016). Recording
signs of deterioration in acute patients: The documentation of vital signs within
electronic health records in patients who suffered in-hospital cardiac arrest. Health
Informatics J, 22(1), 21-23. doi: 10.1177/1460458214530136
Stevenson, J. E., & Nilsson, G. C. (2012). Nurses' perceptions of an electronic patient record
from a patient safety perspective: a qualitative study. Journal of Advanced Nursing, 68(3),
667-676. doi: 10.1111/j.1365-2648.2011.05786.x
Tucker, A. L. (2009). Workarounds and Resiliency on the Frontlines of Health Care. from
http://www.webmm.ahrq.gov/perspective.aspx?perspectiveID=78 - ref8back
Walshe, K., & Rundall, T. G. (2001). Evidence-Based Management: From Theory to Practice in
Health Care, 429.
Widgren, B. (2013). RETTS (rapid emergency triage and treatment system). A little practical
guide., 2014, from http://vardgivarwebb.lio.se/pages/206450/RETTS. Handledning.pdf
Zarabzadeh, A., O'Donahue, J., O'Connor, Y., O'Kane, T., Woodworth, S., Gallagher, J., &
O'Connor, S. (2013). Variation in health care providers' perceptions: decision making
based on patient vital signs. Journal of Decision Systems, 22(3), 168-189.
Meaning unit
Condensed
meaning unit and
description close
to the text
Condensed
meaning unit and
interpretation of
underlying
meaning
Sub-theme
Theme
Nurse takes the VS:
T,P, BP and SaO2
of the second
patient in this
room
Takes vital signs-
T, P, BP and SaO2
Selecting which
vital signs taken
according to
individual patient
needs
Clinical judgement
Measurement
Nurse writes VS in
her notebook after
checking each
patient
Writes VS in
notebook
Using paper
instead of the EHR
Documentation on
paper in notebook
Documentation
Note: VS=vital signs; T=temperature; P=pulse; BP=blood pressure; SaO2=oxygen saturation
Figure 1. Examples of meaning units, condensed meaning units and themes
Accepted Article
This article is protected by copyright. All rights reserved.
Table 1. Frequency of vital signs recorded routinely
ViEWS
Emergency
Department
Surgical ward
Cardiology unit
Infection ward
Times per day
2
Admission + as
per protocol
1
3
2
Temperature
Yes
Yes
Yes
As required
Yes
Pulse
Yes
Yes
Yes
Yes
As required
Respiratory rate
Yes
Yes
Yes
As required
As required
Blood pressure
Yes
Yes
Yes
Yes
As required
SaO2
Yes
Yes
Yes
As required
As required
Conscious level
Yes
Yes
As required
As required
As required
Total
6
6
5
2
1
Note: ViEWS=Vitalpac Early Warning Scores, SaO2=saturation of oxygen
Figure 2. Documentation of vital signs within the EHR and the paper-based system.
Accepted Article
This article is protected by copyright. All rights reserved.
Figure 3. Screen shot of the table used for vital signs in the EHR. N.B. this is an example for illustration
purposes, not an authentic copy.
Table 2. Summary of results
Practice and routines for measuring vital signs
Facilities and functions in EHR
No clear policy on which vital signs to measure
Varied practice and routines from ward to ward
Variability in which vital signs were measured
Decisions on routines for measurement of vital signs
made at ward level or left to clinical decisions of
individual nurses
Usability issues - excessive clicking
Not suitable for frequent vital signs
Quicker and easier to document on paper
(workarounds)
Quicker and easier to write in one section than
another
In RETTS - first set of vital signs in journal. Subsequent
vital signs in table
