International Journal of Nursing Studies 49 (2012) 519527

Contents lists available at SciVerse ScienceDirect

International Journal of Nursing Studies

journal homepage: www.elsevier.com/ijns

Challenges in transition from intervention to end of life care in intensive

care: A qualitative study
Maureen A. Coombs a,b,*, Julia Addington-Hall b, Tracy Long-Sutehall b
a
b

Cardiac Intensive Care Unit, Southampton University Hospitals Trust, Tremona Road, Southampton, Hampshire SO16 6YD, United Kingdom
Faculty of Health Sciences, University of Southampton, Higheld, Southampton, Hampshire SO17 IBJ, United Kingdom

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 1 March 2011
Received in revised form 11 October 2011
Accepted 19 October 2011

Background: Providing quality end of life care is a challenging area in intensive care
practice. The most demanding aspect for doctors and nurses in this setting is not the
management of care at end of life per se, but facilitating the transition from active
intervention to palliation and nally, end of life care. Whilst there is understanding about
some aspects of this transition, recognition of the complex and inter-related processes that
work to shift the patients trajectory from cure to end of life care is required. This is
important in order to work towards solutions for issues that continue to pose problems for
health care professionals.
Objectives: To identify the challenges for health care professionals when moving from a
recovery trajectory to an end of life trajectory in intensive care.
Design: Qualitative methods of enquiry.
Methods and setting: Single semi-structured interviews with 13 medical staff and 13
nurses associated with 17 decedents who underwent treatment withdrawal in intensive
care were carried out. Participants were drawn from two Intensive Care Units in a large
university-afliated hospital in England.
Findings: Patients who died in intensive care appeared to follow a three-stage end of life
trajectory: admission with hope of recovery; transition from intervention to end of life
care; a controlled death. The transition from intervention to end of life care was reported
as being the most problematic and ambiguous stage in the end of life trajectory, with
potential for conict between medical teams, as well as between doctors and nurses.
Conclusions: End of life care policy emphasises the importance of end of life care for all
patients regardless of setting. These ndings demonstrate that in intensive care, there is
need to focus on transition from curative intervention to end of life care, rather than end of
life care itself so that effective and timely decision making underpins the care of the 20% of
intensive care patients who die in this setting each year.
2011 Elsevier Ltd. All rights reserved.

Keywords:
Intensive care
End of life
Transition
Team working

What is already know about the topic

* Corresponding author at: Faculty of Health Sciences, University of

Southampton, Higheld, Southampton, Hampshire SO17 IBJ, United
Kingdom. Tel.: +44 2380796121; fax: +44 2380796118.
E-mail addresses: Maureen.coombs@suht.swest.nhs.uk,
mc9@soton.ac.uk (M.A. Coombs).
0020-7489/$ see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijnurstu.2011.10.019

 Palliative and end of life care is receiving increasing

policy focus.
 Whilst a signicant evidence base exists to guide end of
life practices in within cancer care, less is known about
end of life care in non-cancer settings.
 For intensive care health professionals, work to date
focuses not on symptom management at end of life but

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

520

discrete aspects concerned with managing the transition

from active care to treatment withdrawal.
What this paper adds
 An end of life trajectory in intensive care provides a
framework that enhances understanding of role and
decision making challenges for health care professionals
in this setting.
 Managing the transition from curative intervention to
end of life care is most challenging and complex stage in
the end of life trajectory with intra- and inter-disciplinary tensions experienced by clinical teams.
 A focus on the transition from curative intervention to end
of life care is required so that timely and effective decision
making about withdraw of treatment is facilitated.
1. Background
Palliative and end of life care is receiving continued
academic and health policy focus. There is an increasing
awareness of differences in clinical practices in end of life
across Europe (Menaca et al., 2011) and internationally
(Ball et al., 2010; Gysels et al., 2011). Furthermore, the need
for end of life care research strategies and evaluation of
health policy for end of life care (Mackenzie et al., 2011) at
a global level has been recognised. Within the United
Kingdom (UK), the main health policy impetus was the
launch in 2008 of the rst end of life care strategy in
England and Wales (Department of Health (DH), 2008), and
the subsequent national end of life care programme
(http://www.endoifecareforadults.nhs.uk). This resulted
in health and social care providers across the country
working to implement strategy recommendations that
emphasise the importance of improving end of life care for
all patients, regardless of diagnosis or setting.
Such policy initiatives do not only highlight the need to
improve quality care during the nal weeks and months of
life. The need to identify people approaching end of life and
the need to identify individual preferences with regard to
place and process of dying is also outlined (DH, 2008).
However whilst evidence exists to guide and support health
care professionals as they initiate such discussions with
patients (Schoeld et al., 2010), the majority of such
knowledge remains cancer care based (Gott et al., 2010)
and may not always be applicable in other care settings. For
example, identifying cancer patients who are nearing end of
life is reported to be less challenging for health care
professionals than managing the transition to palliative and
end of life care (Boyd and Murray, 2010). However, it has
been recognised that both identifying when a patient is
dying and managing the transition from active to palliative
care are challenges for health care staff when patients are
dying from conditions other than cancer (Gott et al., 2007)
and particularly for those patients within intensive care
settings1 (Ravenscroft and Bell, 2000; Coombs and Long,
2008).

1
Of 89,682 patients admitted to 180 NHS adult, general critical care
units between April 1st 2008 and March 31st 2009, 15,358 (17.1%) died
(ICNARC, 2010).

Most deaths within critical care follow after a decision to

withdraw or withhold life sustaining treatments has been
made (Lautrette et al., 2006; Keenan et al., 1998). The
decision to withdraw treatment occurs once there is
realisation within the intensive care clinical team that
patient recovery is not an outcome. Discussions with
patient, family and extended care teams then focuses on
palliation and end-of-life management (Sprung et al., 2003).
Once agreement on end of life care has been reached
amongst all involved, end of life care is often achieved
through the withdrawal of life sustaining treatments
including ventilation and support drugs (Bewley, 2000)
together with a focus on maximising patient and family
comfort.
However, identication of patients who will not survive
is a signicant challenge in the intensive care unit (ICU) as
the patients condition is often complex and death may be
an unexpected development during the care trajectory.
Identication is further compromised by lack of agreement
between different disciplines regarding the goals of end of
life care (Cook et al., 2006; Robichaux and Clark, 2006;
Bach et al., 2009; Long-Sutehall et al., 2011; Festic et al.,
2011) and a lack of fora and motivation to discuss end of
life care (Yaguchi et al., 2005). This paper reports on
ndings that illustrate the complexities involved in
facilitating the transition from active treatment to end
of life care within intensive care.
2. Study aim and design
The paper is drawn from a larger study exploring end of
life care and decision making in an intensive care setting
(Coombs et al., 2009). The study took place over a 12 month
period from 2008 to 2009. The aim of the study was to gain
an in-depth understanding of end of life care in intensive
care units from multiple perspectives. This was undertaken through qualitative interviews with doctors and
nurses involved in patient management during withdrawal of treatment.
3. Research setting
The study was undertaken in two intensive care units
(ICU) located in a large university teaching hospital in
England. One unit was a large 20 bedded mixed adult
intensive/high dependency care unit managed by consultant intensivist staff. The second site was a tertiary
specialised unit with 15 cardiac intensive care beds,
managed by consultant cardiac anaesthetic staff.
4. Participants
There were 67 potential cases during the recruitment
timeframe of which 35 met the inclusion criteria. The
reasons for exclusion from the study were evidence in the
patient notes of: extreme grief (n = 12); complex family
issues (n = 10); and police/coroner involvement (n = 10).
Staff involved in care of the 35 cases was sent to study
recruitment packs inviting them to participate in the
study. Relatives were also interviewed in the study, and
data from these reported elsewhere (Coombs et al., 2009).

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

521

Table 1
Health care interviews linked to deceased patient.
Case no.

Patient medical details

Number of interviews
Nurse

Doctor

01

84 year old male: specialist management post subarachnoid haemorrhage and hydrocephalus.
56 year old female: hepatorenal syndrome, decompensated alcoholic liver disease and
respiratory failure.
78 year old female: collapse, chest infection, diarrhoea, respiratory failure and hypotension.
72 year old female: acute abdomen (perforated duodenal ulcer) and renal failure. Developed
septic shock and respiratory failure.
71 year old female: respiratory failure, prolonged and complex ventilation support.
62 year old male: pulmonary brosis and chronic pulmonary emboli.
76 year old female: hepatorenal failure, decompensated alcoholic liver disease, septic and
respiratory failure.
84 year old female: ischemic bowel, respiratory and renal failure. Admitted following
emergency laparotomy.
77 year old male: readmission for respiratory failure. Re-do aortic valve replacement 3
month previous.
78 year old male: post operative re-admission for respiratory failure, pneumonia, renal
failure post cardiac surgery.
69 year old male: type II respiratory failure following cardiac failure. In situ implantable
cardioverter debrillator.
75 year old female: mitral valve replacement. Difcult surgery. Pus around heart. Open chest
with high inotropic support.
55 year old male: Salvage cardiac surgery undertaken
75 year old female: tricuspid valve replacement. Renal failure with increasing inotropic
therapy.
79 year old gentleman: post operative emergency valve replacement, partial pericardectomy.
Treatment withdrawn after pathology results conrmed malignancy.
73 year old male: respiratory management, infective endocarditis, work up for surgery
previous Fallots repair. Multisystem failure. Increasing therapies.
53 year old male: out of hospital arrest. No resuscitation for 15 min. Multi-system failure.

0
1

1
1

0
1
0

1
1
1

0
1

1
0

13

13

02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
Total

Staff associated with 20 cases returned reply slips

indicating a willingness to be interviewed. Interviews
were successfully arranged with staff associated with 17 of
the cases. Due to rota and leave commitments, interviews
were not achieved with three members of staff. Patient
cases recruited to the study were representative of the
demographics and case mix of the sites. All patients were
receiving ventilator support and were therefore sedated.
Family members were proxy decision makers. Length of
stay for patients in this study ranged from 1 to 22 days, the
mean was 5.4 days, the mode was 2 days.
Twenty-six health care staff linked to the care given to
seventeen deceased patients were recruited to the study and
interviewed (Table 1). Interviews with three health care
staff occurred for two patient cases, interviews with two
staff for ve cases and one member of staff for ten cases. A
range of intensive care experience was demonstrated across
all medical and nursing staff groups. Nurses interviewed
ranged from senior experienced staff (greater than 20 years
in intensive care and holding a senior sister/charge nurse
post) to more junior staff nurses with less than three months
experience in speciality. Doctors interviewed ranged from
consultant grade with twenty years experience in intensive
care to registrar grade with four years experience in
medicine. The interview duration was 2560 min. Cases
recruited to the study were representative of the demographics and case mix of the sites. No patients in this study
had advance decisions or Do Not Attempt Cardio-Pulmonary
Resuscitation orders in place.

5. Research methods
Health professionals were recruited to the study
through retrospective case note review of all intensive
care patients who died and where end of life discussions
were evidenced in the patients medical notes. Inclusion
criteria for the case note review included: death not
resultant from cardiac arrest; documented end of life care;
and no risk of complicated bereavement issues.2
Case note review was undertaken by dedicated medical
and nursing staff on each unit. A sampling template was
used to record key events (date of admission, date of death,
summary of patient progress, decisions and end of life
discussions held) and the health care professionals involved.
Within a fortnight of a patient death that fullled the
inclusion criteria, three members of the clinical intensive
care team (doctors and nurses) with the most signicant
documented involvement in the patients care received
study information and recruitment letter inviting them to
a single face to face interview. Replies from potential
participants were sent direct to the researcher.
The location and time of interview was chosen by the
participant and written consent was obtained prior to the

2
Complicated bereavement issues were dened as: police/coroner
involvement, care subject of internal investigation/family complaint,
complex family issues (physical or mental health) or extreme grief
observed and documented in patient notes.

522

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

start of the interview. All interviews were audio recorded

and commenced with the open question: Can you tell me
about what happened around the time of (patient name)
death? A series of probes were used to elicit additional
information about the end of life decisions made and the
process of care withdrawal.
Data analysis was an iterative approach utilising
constant comparison identifying themes as coding progressed. Data collection and analysis occurred concurrently enabling reection on previously collected data.
Field notes and memos were recorded to capture analytical
and methodological issues, and theoretical insights. These
were later used to inform development of the study
ndings in the later stages of data analysis (phases 2 and 3
of the analysis).
Interviews were transcribed verbatim after digital
audio-recording. Expressions, such as laughter and expletives were included in the text. All the transcripts were
checked by MC against the recorded data for accuracy.
A three phased approach was used to interpret the data.
In the rst phase, the transcribed interviews were read
several times by MC to get insight into the broad content of
the individual accounts and perspectives. From this phase
of data analysis, an overview summary of each interview
was then documented on a summary sheet. In the second
phase, data derived from transcripts were reviewed lineby-line. Individual codes were developed with any
similarities or differences across interviews noted at this
stage. In the third phase of data analysis, the developed
codes were grouped into related themes and compared
with one other.
A sample of transcripts were independently analysed
by TLS to validate emerging codes and themes. The coding
reliability between the two researchers (MC and TLS) was
greater than 95% of agreement. These developed themes

were then linked to form a conceptual framework (the end

of life trajectory). The nal analysis was circulated to the
study research team and was further examined and
validated during the meeting of the studys Advisory Board.
Credibility and study rigour were maintained throughout the undertaking of this study. To increase the potential
of producing credible ndings, the researcher invested
time to gain rapport with staff in the study sites through
joining staff for handover, coffee and meetings. Through
this, the researcher and the study became more familiar to
staff enabling trust and facilitating disclosure.
Reective eld notes and memos were written
throughout the research with the aim of detailing
important analytical and methodological issues, ideas
and theoretical insight. These initiatives added explanatory rigour to researcher judgements and decision making,
providing an audit trial of the study. Furthermore,
independent analysis of the data by another member of
the research team (TLS) assisted in establishing credibility
and ensuring value-free and conrmable analysis.
Ethical committee approval was given for this study by
the Isle of Wight, Portsmouth and South East Hampshire
Research Ethics Committee (REC 08/H0501/65).
6. Results
6.1. End of life trajectory in intensive care
During the patients stay three key stages were
identied that described the end of life trajectory in
intensive care: admission with hope of recovery; transition
from intervention to end of life care; and nally, a
controlled death (Fig. 1). In each stage of the trajectory
health care staff held a particular role with some stages
generating specic challenges and tensions related to

Fig. 1. End of life trajectory in intensive care: key stages and themes.

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

decision-making. The most problematic stage was transition from the active stage of intervention to end of life care.
Whilst the complete trajectory is presented in Fig. 1 to
illustrate the end of life trajectory in intensive care, this
paper focuses on the transitioning stage from intervention
to end of life care. Findings are supported by exemplar
quotes.
6.2. Transition from intervention to end of life care
For critically ill patients in intensive care, the stage from
intervention to end of life care was initiated as result of an
increasing awareness amongst the clinical team of
continued deterioration in the condition of the patient,
or a lack of patient response to the treatments and
interventions administered. Whilst maximal therapies
continued, there was focussed communication between
nursing and medical staff about whether treatment should
continue or whether a review of treatment direction was
required. Family members were also involved in these
discussions, although this usually occurred once consensus
from the health care team had been reached. Decision
making within transition from intervention to end of life
care was characterised by three stages: making a diagnosis
of dying; managing end of life consensus; and pushing the
door open to facilitate family grieving.
6.3. Making a diagnosis of dying
A key phrase often used by health care staff at this stage
was that treatment was futile and that the patient was
dying. In some situations this was clear to all and very
evident whilst at other times patient outcome was less
clear and difcult to predict.
Eventually she declared herself and all of her numbers
went off and it was obvious that inside (she) was not
doing.
(Case 01, nurse interview).

You need to take each case on its own merits. Most of our
end of life decisions are related to futility of treatment and
prospect of sensible recovery and the balance of those two
are different in different patients. . .bearing in mind the
potential quality of life as perceived by the family and the
medical and nursing staff.
(Case 13, consultant interview).
Whilst doctors and nurses all held opinions on whether
the patient was dying or not, it was medical staff on
intensive care who held authority to make a diagnosis of
dying. This diagnosis was an important point in the
patients trajectory as it heralded a re-focussing of patient
treatment objectives and instigated a different focus for
discussions with the patients family.
However, making a diagnosis of dying was not only
informed by objective clinical data and decision making
variables. Individual viewpoints, beliefs and experiences of
medical staff were also factors that made identication of
the transition point from intervention to end of life care,
problematic.

523

I could have made the decision in another day or two, but I

didnt think it was in any ones best interest to do that when
he obviously had no chance of surviving and we were just
articially keeping him alive on a ventilator. I felt it was
morally wrong to do that.
(Case 06, consultant interview)
It is never black and white, but it shouldnt be grey versus
grey. If its black and white youve probably left it a bit too
late. All of us have a different tipping point, based on our
personality, its based on our experience, and its based on
our moral and religious framework
(Case 03, consultant interview).
Data demonstrated how patients and family were fully
supported by medical and nursing staff about the lack of
recovery whilst the diagnosis of dying was made. Medical
staff gave formal intermittent updates through planned
family conferences, whilst nursing staff offered on-going
bedside care. This support focussed on gaining knowledge
on family membership, family understanding of events to
date, and family coping in the face of the critical illness of
their family member. It also included explaining and
emphasising salient points of the medical update.
Making a diagnosis of dying was an important catalyst
that instigated movement towards a trajectory of end of
life intensive care. All intensive care medical staff
identied that this was a demanding decision due to its
ultimate outcome: the patients death. Indeed it was seen
to be an intrinsic part of what an intensive care consultant
job entailed. However, different perceptions were held
about diagnosing dying in other areas.
the diagnosis of dying is no longer made on the wards,
when you go up to review a patient and write in the notes,
this patient is dying, it is such an eye opener and
completely refocuses peoples minds. Doctors are not
taught how to make a diagnosis of dying any more . . .in
intensive care we have learned how to make a diagnosis
and deal with it. Unfortunately we are an acute palliative
care service sometimes.
(Case 07, consultant interview)
Although intensive care patients were managed primarily by intensive care teams, the referring ward based
teams visited the units and made contribution to patient
management decisions. Therefore achieving agreement
from all team members that recovery was no longer a
patient option was the next challenge in transitioning from
cure to treatment withdrawal.
6.3.1. Managing end of life consensus
Achieving consensus about the transition from intervention to end of life with referring clinical teams was seen
to be problematic at times by doctors and nurses in
intensive care.
The medical decision is very hard for them (doctors). I can
always pass the buck to them. But it is hard when I have to
side step awkward questions from the family when you
know agreements havent been reached amongst the
medical team. End of life can be a nasty area it can get

524

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

very political amongst the teams and need a lot of

mediation to get a consensus before going to the family.
(Case 11, nurse interview)
Its easier to box people into a corner and get them to carry
on, than trying to persuade them to stop. Momentum is
difcult to stop.
(Case 15, consultant interview).
Once again, the personal beliefs and values of clinicians;
the nature of clinical specialities and specic knowledge
base of acute and chronic, rather than critical illness; and
the longevity of relationships between speciality referral
teams and patients were all cited as confounding factors
impacting on decision making during the transition from
active intervention to end of life care.
In-house teams see themselves as patient advocates. They
are very attached to certain patients. It also depends on
why the patient is on ICU. If there is a surgeon involved and
there has been a catastrophe it will be very hard for them
to accept (that the patient will not recover). If there is a
medical consultant, they may have known the patient for
7, 8, 9 years and dealt with them in clinic on a monthly
basis. They (Other doctors) may resist (the decision to
withdraw treatment) quite forcibly. If they are resistant I
say why dont we give them (the patient) a couple of days
and see what is happening? Ultimately as ICU clinicians
we take the decision, but it is very unlikely that we will go
against their wishes.
(Case 03, consultant interview)
Whilst discussions with referring teams occurred
behind the scenes, nursing staff continued to support
the family as previously and took opportunity to inuence
decision making through careful and selective presentation of facts during ward rounds. The more experienced
nursing staff also highlighted how they would offer their
opinion on the direction of care to senior medical staff
across the teams.
The doctors saw her as numbers, if we could get her
numbers sorted . . . and the doctors she was admitted
under, they wouldnt stop, wouldnt let us stop. Im all in
favour of throwing everything at the patient for as long as
they need it, and it is working. Trying to get the Consultant
to agree on withdrawing was quite difcult.
(Case 02, nurse interview)
It was clear from the interviews that nurses in intensive
care understood the need for consensus in decision
making. However, nurses expressed frustration when
discussions about end of life care were focussed solely
on mortality and morbidity issues. For nurses, preadmission quality of life, potential for physical and mental
functioning after discharge, and a perception of sustained
patient/family suffering whilst in intensive care were also
important areas that informed their opinion on the need to
transition to end of life care.
And I actually thought, why did they do this operation on
this lady? This is absolutely ridiculous, you know, she
would, she had been living in quite awful conditions at

home. You know smearing her faeces around the house,

not eating properly, you know just generally not caring for
herself because of her dementia and they had gone ahead
and operated on a lady with all those problems.
(Case14, nurse interview).
It was not dignied. . . we shouldnt have carried on for as
long as we did
(Case 12, nurse interview).
Gaining consensus across the health care team was an
essential part of moving from active treatment to end of
life care. However the need to engage all clinical teams in
this process impacted on the timescale of events within
intensive care, and specically on the timing of the nal
stage in transition; that of approaching the family
regarding treatment withdrawal.
Surgeons do not understand the misery that goes onthey
do not spend that much time with the patients. Playing the
surgeon on side having to involve them in the decision
slows things down
(Case 16, consultant interview).
6.3.2. Pushing the door open: facilitating family grieving
Once agreement had been secured across the medical
and nursing teams involved in the patients management
and care, an end of life focus was then introduced in
discussions with the family. Such discussions were usually
staggered and staged with the aim of enabling families to
understand lack of progress or lack of recovery; the lack of
treatment options remaining; and the outcome for the
patient.
I also need to prepare the family I wont just drop it on
them. I plant little seeds. Its a process.
(Case 03, consultant interview)
Part of the process is talking with the family. You set up
their mind that they are not likely to survive and you give
them a little bit of extra time to see what is going on. So
when you come to that sort of discussion, it is not a bolt
from the blue. Having a relative on ICU is an emotional
rollercoaster. You have no power, no control, someone who
is your nearest and dearest is critically unwell, and you
have different messages from doctors and nurses. Your life
becomes entirely revolving around your relative. You are
constantly anxious, you take in little bits of good news,
which you take huge benet from and then by denition
its a rocky road and you get huge dips of despair. The
burden of ICU can be quite onerous.
(Case 04, Consultant interview)
Medical and nursing staff were acutely aware of the
stress on relatives at this time and the need for families to
have time to assimilate information. All staff stated that
families were an intrinsic part of end of life decision making,
but were not there to take responsibility for care decisions.
Time was a valuable commodity to give families opportunity
to move from a position of hope for recovery to understanding the discussions and impact of end of life care.
During this time nurses worked alongside medical staff
to gauge family understanding of the current situation and
to prepare the family for the end of life discussions. Nurses

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

appraised medical staff of the views of the family and

shared with the medical team their assessment of family
coping.
The main nursing role is being an advocate for the patient,
to bring it (end of life care) up for discussion, whether
withdrawal is an option (for the patient), making sure the
family are always kept up to date. Liaising between the
family and the doctors.
(Case 01, nurse interview)
However, it was clear that nurses experienced frustration during slower transitioning to end of life care as a
result of prolonged decision making across medical team.
Strong language (e.g. torturing, ogging the person to
death until rigour mortis is almost set in) was often used
by nurses when talking about such cases. Indeed, this
frustration appeared to impact on the nursing role during
the nal part of the end of life trajectory; that of a
controlled death.
Whilst transitioning to end of life care could take hours/
days/weeks, the controlled death stage in intensive care
was rapid (min/h). In contrast to the other stages in the end
of life care trajectory, nurses took a dominant role here,
impacting on end of life care in several ways.
The nurse at that point is almost like, you see them rack up
again . . . they move into full care mode.
(Case 13, consultant interview).
The doctors step back and the nurses lead. . .. I tried to
take everything away I could, cover the back of the bed
space and give her back to them as they know her. An
emergency hair comb thats my big thing you need to
look good while youre dying. Her hands were out and she
smelt nice. It is important; those are the things that matter,
because you cant save them anymore
(Case 02, nurse interview)
7. Discussion
By adopting a trajectory as the vehicle to illustrate
stages and transitions associated with end of life care in
intensive care, we are following in the footsteps of others.
Whilst this is not a novel approach per se (Glaser and
Strauss, 1968; Boyd and Murray, 2010), it is valuable in
that it encapsulates the patient journey from admission
with hope of recovery, transition from intervention to end
of life care, through to a controlled death. Findings
reported here noticeably identify the challenges that exist
for health care professionals as they seek to move from
curative interventions to end of life care.

we resuscitated her and sustained her life. Once we had

settled her so that her life was no in immediate danger,
we went into a phase of extracting as much information
as we could about her health, her family situation, and
her wishes. (Case 07, Consultant interview).

525

As reported, one of the challenges for the transition of

intensive care patients from cure to end of life care is the
identication of those who will not survive. In attempting
to address this problem, several approaches have been
used to improve prognostic decision making including use
of organ/functional failure modelling (SUPPORT, 1995),
length of stay and mortality rates (Lilly et al., 2000) and
prognostic modelling using Acute Physiology and Chronic
Health Evaluation scores (Barnato and Angus, 2004).
Whilst such approaches offer guidance to clinicians, they
do not appear to offer the prognostic clarity intended, and
do not appear to assist in gaining consensus between
teams who may hold differing views on whether on-going
treatment is futile. As discussed by Ridley (2002)
recognising futility is important to: allow discontinuation
of distressing interventions when there is no hope of
recovery; enable families to prepare for death; and allow
for re-allocation of scarce resource.
Participants in this study spoke of the futility of
treatments. However of note is how nursing staff appeared
to link futility and dying more closely than medical staff.
Nurses would often describe such patients as actively
dying. This apparent disassociation for doctors between
futility and dying is similarly reected in the medical
literature where the concept of futility appears to be
separate from the concept of dying (Helft et al., 2000;
British Medical Association, 2006). Further exploration is
required to determine whether reframing the concept of
futility to incorporate a diagnosis of dying would facilitate
more timely decision making to transition from curative
intervention to end of life care.
A key point directing transition from curative intervention to end of life care occurred when consensus from
the whole clinical team occurred. It was clear that this was
a complex process that happened over time and was
dependent on team awareness of impending patient death
and successful negotiation of the complex team decisionmaking processes. At times, this required medical staff to
collaborate and also challenge their medical colleagues.
Nursing staff also used strategies to inuence medical
decision making whilst managing their own frustrations
and supporting families at the bedside. Interestingly
contemporary models of palliative care (Lynn and Adamson, 2003; Murray et al., 2005) do not highlight such
complex decision making during the transition stage from
curative intervention to end of life care, and the resultant
impact of this on health care staff.
Whilst transitions towards end of life care are reported
as leading to tension between doctors and nurses (Pugh
et al., 2009; Long-Sutehall et al., 2011), this study also
indicates difculties between intensive care and referring
clinical teams. Cassells ethnographic study (2003)
describing the divergent ethical beliefs and practices of
surgeons and anaesthetists is seminal in building understanding about the source of such intradisciplinary
tensions, explicating the rationalisation of decision-making against professionally encultured belief systems. More
recent work (Pugh et al., 2009) indicates that one in ve of
119 NHS intensive care consultants surveyed, disagreed
with withholding or withdrawing treatment based on their
individual belief system. This suggests that more work is

526

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

needed to explore the impact of personal and professional

beliefs in staff as a potential barrier or catalyst to
withdrawing treatment, and how this may be linked to
denitions of futility and dying.
A further consideration that speaks to individual belief
systems is the need to acknowledge the emotional labour
involved in making the transition from curative interventions to end of life care. Whilst the emotional labour of
intensive care nurses involved in delivering care during the
nal hours of life (Stayt, 2009) has been acknowledged
within the nursing literature, the emotional and cognitive
demands made on all team members when transitioning to
end of life care, needs wider acknowledgement and
exploration.
Whilst this study raises interesting insights into the end
of life trajectory in intensive care, there are study
limitations to be acknowledged. As staff were invited to
participate in this study, this was a self-selecting sample. It
was noted that once cases were selected for the study,
certain staff would always respond to an interview request,
whilst others did not reply. This raises the possibility that
the issue of end of life care was important for staff
interviewed, or that staff participated due to outstanding
issues about bereavement. Whilst motivations for participation were not explored, all interviewees were able to
engage in the interview experience and reect on their
experiences.
A further limitation of this study was that several of the
consultants were interviewed more than once. The
consultant team were key in holding authority to make
end of life decisions and it was anticipated that consultants
could be interviewed several times. This is clearly a
limitation of the study design and its potential for study
bias is acknowledged.
Due to its qualitative design and convenience sampling
methods, this study is not directly transferable to other
populations or contexts. Its credibility will be conrmed if
ndings have meaning for other staff involved in end of life
care in ICU, and in its ability to inform understanding in
this area. However, ndings from this study conrm and
extend theoretical knowledge in the area of end of life care
in intensive care. Further studies that examine end of life
decision making and specically the transition from
intervention to end of life care in other critical care
specialities would be useful.
8. Conclusion
It is clear that the intensive care setting presents unique
challenges when translating end of life care policy into
practice. Adoption of approaches such as the Liverpool
Care Pathway (Liverpool Care Pathway Central Team UK,
2007) into an environment where the greatest challenge is
diagnosing the end of life, as opposed to the delivery of care
in the last few hours, presents obvious difculties. There is
currently limited guidance on how to manage transitioning from cure to end of life discussions amongst teams.
Until such approaches are developed and adopted, intra
and interdisciplinary tensions arising from diverse professional motivations for cure within end of life care decision
making will remain (Pattison, 2006). One of the greatest

challenges across health care teams continues to be

consensus acknowledgement and acceptance of the
concept of futility of treatment and the concept of dying.
Acknowledgements
We are grateful to the staff from the two Intensive Care
Units who participated in this study during 20082009,
sharing their experiences to increase understanding in this
area of practice.
The authors also acknowledge the valuable contribution of Angela Rogers (AR), Senior Research Fellow on the
original research project.
Contributors: MC had the original idea for the overall
protocol and led the study. All authors and AR contributed
to the development of the research protocol. AR conducted
the interviews with participants. MC and TLS undertook
data analysis. All authors contributed to the nal version of
the paper. MC is guarantor.
Conict of interests: None declared.
Funding: The study on which this paper is based was
funded by a competitive Research for Patient Benet bid,
supported by the National Institute for Health Research
programme.
Ethical approval: Ethics committee approval was given
for this study by the Isle of Wight, Portsmouth and South
East Hampshire Research Ethics Committee (REC 08/
H0501/65). All participants gave informed consent before
taking part and have given written consent to their
interview data being included in publications.
References
Bach, V., Ploeg, J., Black, M., 2009. Nursing roles in end-of-life decision
making in critical care settings. Western Journal of Nursing Research
31, 496512.
Ball, C.G., Navsaria, P., Kirkpatrick, A.W., Vercler, C., et al., 2010. The
impact of country and culture on end of life care for injured patients:
results from an international survey. Journal of Trauma 69 (6), 1323
1333.
Barnato, A.E., Angus, D.C., 2004. Values and role of intensive care unit
outcome prediction models in end of life decision making. Critical
Care Clinics 20 (345362), viiviii.
Bewley, J.S., 2000. Treatment withdrawal in intensive care: the decision
making process. <http://www.avomn.nhs.uk/bristolitutrainees/dissertations/jeremy_bewley_dissertation.pdf> (accessed 07.08.11).
Boyd, K., Murray, S., 2010. Recognising and managing key transitions in
end of life care. British Medical Journal 341, c4863.
British Medical Association, 2006. End of Life Decisions: Views of the
BMA. British Medical Association, London.
Cassell, J., Buchman, T.G., Streat, S., 2003. Surgeons, intensivists, and the
covenant of care: administrative models and values affecting care at
the end of life. Critical Care Medicine 31, 15511559.
Cook, D., Rocker, G., Giacomini, M., Sinuff, T., Heyland, D., 2006. Understanding and changing attitudes toward withdrawal and withholding
of life support in the intensive care unit. Critical Care Medicine 34 (11
Suppl.), S317S323.
Coombs, M., Long, T., 2008. Managing a good death in critical care: can
health policy help? Nursing in Critical Care 13 (4), 208214.
Coombs, M., Long-Sutehall, T., Addington-Hall, J., 2009. End of Life Care in
intensive care settings: a case study approach to explore decision
making and processes. Final report for the NIHR Research for Patient
Benet Research Programme, October, Available at <eprints@soton.ac.uk>.
Department of Health, 2008. End of Life Care Strategy. Promoting a High
Quality Care for All Adults at End of Life. Department of Health,
London.
Festic, E., Wilson, M.E., Gajic, O., Divertie, G.D., Rabatin, J.T., 2011. Perspectives of physicians and nurses regarding end-of-life care in the

M.A. Coombs et al. / International Journal of Nursing Studies 49 (2012) 519527

intensive care unit. Journal of Intensive Care Medicine, doi:10.1177/
0885066610393465 (accessed 18.05.11).
Glaser, B.G., Strauss, A.L., 1968. Time for Dying. Aldine, Chicago.
Gott, M., Barnes, S., Parker, C., Payne, S., Seamark, D., Gariballa, S., Small,
N., 2007. Dying trajectories in hear failure. Palliative Medicine 21, 95
99.
Gott, M., Ingleton, C., Bennett, M.J., Gardiner, C., 2010. Transition to
palliative care in acute hospitals in England: qualitative study. BMJ
Online, doi:10.1136/bmj.d1773 (accessed 01.06.11).
Gysels, M., Pell, C., Straus, L., Pool, R., 2011. End of life care in sub-Saharan
Africa: a systematic review of the qualitative literature. BMC Palliative Care 10 (6), doi:10.1186/1472-684X-10-6 (accessed 10.10.11).
Helft, P., Siegler, M., Lantos, J., 2000. The rise and fall of the futility
movement. The New England Journal of Medicine 343, 293296.
Intensive Care National Audit and Research Centre, 2010. CMP Case Mix
and Outcome. Summary Statistics. <https://www.icnarc.org/documents/summary%20statistics%202008-9.pdf> (accessed 20.05.11).
Keenan, S.P., Busche, K.D., Chen, L.M., Esmail, R., Inman, K.J., Sibbald, W.J.,
1998. Withdrawal and withholding of life support in the intensive
care unit: a comparison of teaching and community hospitals. Critical
Care Medicine 26 (2), 245251.
Lautrette, A., Ciroldi, M., Ksibi, H., Azoulay, E., 2006. End-of-life family
conferences: rooted in the evidence. Critical Care Medicine 34 (11
Suppl.), S364S372.
Lilly, C.M., De Meo, D.L., Sonna, L.A., 2000. An intensive communication
intervention for the critically ill. American Journal of Medicine 109,
469475.
Liverpool Care Pathway Central Team UK, 2007. Non-cancer brieng
paper. Brieng paper. Liverpool: Marie Curie Palliative Care Institute.
<http://www.mcpcil.org.uk/> (accessed 01.04.10).
Long-Sutehall, T., Willis, H., Ugboma, D., Palmer, R., Addington-Hall, J.,
Coombs, M., 2011. Negotiated dyinghow nurses shape withdrawal
of treatment in hospital critical care units. International Journal of
Nursing Studies, doi:10.1016/j.ijnurstu.2011.06.003 (accessed
01.08.11).
Lynn, J., Adamson, D.M., RAND Health, 2003. White Paper. Living Well at
the End of Life. Adapting Health Care to Serious Chronic Illness in Old
Age, 2003. RAND Health, Santa Monica, USA.
Mackenzie, t., Innes, J., Boyd, M., Keane, B., Boxall, J., Allan, S., 2011.
Evaluating the role and value of a national ofce to coordinate

527

Liverpool Care pathway implementation in New Zealand. International Journal of Evidence Based Healthcare 9 (3), 252260.
Menaca, A., Evans, N., Andrew, E.V., Finetti, F., Gomez-Batiste, X., Higginson,
I.J., Harding, R., Pool, R., Gysels, M., 2011. End of life care across Southern
Europe: a critical review of cultural similarities and differences
between Italy, Spain and Portugal. Critical Reviews in Oncology Haematology, doi:10.1016/j.critrevonc.2011.06.002 (accessed 10.10.11).
Murray, S.A., Kendall, M., Boyd, K., Sheikh, A., 2005. Illness trajectories and
palliative care. British Medical Journal 330, 10071011.
Pattison, N., 2006. A critical discourse analysis of provision of end of life
care in key UK critical care documents. Nursing in Critical Care 1, 198
207.
Pugh, E.J., Song, R., Whittaker, V., Blenkinsopp, J., 2009. A prole of the
belief system and attitudes to end of life decisions of senior clinicians
working in a National Health Service Hospital in the United Kingdom.
Palliative Medicine 23, 158164.
Ravenscroft, J., Bell, M.D., 2000. End of life decision making within
intensive careobjective, consistent, defensible? Journal of Medical
Ethics 26, 435440.
Ridley, S., 2002. Outcomes in Critical Care. Butterworth Heinemann,
Oxford.
Robichaux, C.M., Clark, A.P., 2006. Practice of expert critical care nurses in
situations of prognostic conict at the end of life. American Journal of
Critical Care 15, 480491.
Schoeld, P., Carey, M., Love, A., Nehill, C., Wein, S., 2010. Would you like
to talk about your future treatment options?
BMJ Online,
doi:10.1136/bmj.d1773 (accessed 19.07.11).
Sprung, C.L., Cohen, S.L., Sjokvist, P., Baras, M., For the Ethicus Study
Group, 2003. End-of-life practices in European Intensive Care Units.
JAMA 290, 790797.
Stayt, L.C., 2009. Death, empathy and self preservation: the emotional
labour of caring for families of the critically ill in adult intensive care.
Journal of Clinical Nursing 18 (19), 12671275.
SUPPORT, Principal Investigators, 1995. A controlled trial to improve care
for seriously ill hospitalised patients: The study to understand prognoses and preferences for outcome and risk of treatments. Journal of
American Medical Association 274, 15911598.
Yaguchi, A., Truog, R.D., Curtis, J.R., 2005. International differences in end
of life attitudes in the intensive care unit. Archives of Internal Medicine 165, 19701975.