Article Text

Download PDFPDF

Living and dying with severe chronic obstructive pulmonary disease: multi-perspective longitudinal qualitative study
  1. Hilary Pinnock1,
  2. Marilyn Kendall2,
  3. Scott A Murray2,
  4. Allison Worth1,
  5. Pamela Levack3,
  6. Mike Porter4,
  7. William MacNee5 and
  8. Aziz Sheikh1
  1. 1Allergy and Respiratory Research Group, Centre for Population Health Sciences: GP Section, University of Edinburgh, Edinburgh, UK
  2. 2Primary Palliative Care Research Group, Centre for Population Health Sciences: GP Section, University of Edinburgh, Edinburgh, UK
  3. 3NHS Tayside, Ninewells Hospital, Dundee, UK
  4. 4Centre for Population Health Sciences: GP Section, University of Edinburgh, Edinburgh, UK
  5. 5MRC Centre for Inflammation Research, University of Edinburgh, The Queen's Medical Research Institute, Edinburgh, UK
  1. Correspondence to H Pinnock, Allergy and Respiratory Research Group, Centre for Population Health Sciences: GP Section, University of Edinburgh, Doorway 3, Medical School, Teviot Place, Edinburgh, EH8 9AG, UK hilary.pinnock{at}


Objectives To understand the perspectives of people with severe chronic obstructive pulmonary disease (COPD) as their illness progresses, and of their informal and professional carers, to inform provision of care for people living and dying with COPD.

Design Up to four serial qualitative interviews were conducted with each patient and nominated carer over 18 months. Interviews were transcribed and analysed both thematically and as narratives.

Participants 21 patients, and 13 informal carers (a family member, friend, or neighbour) and 18 professional carers (a key health or social care professional) nominated by the patients.

Setting Primary and secondary care in Lothian, Tayside, and Forth Valley, Scotland, during 2007-9.

Results Eleven patients died during the study period. Our final dataset comprised 92 interviews (23 conducted with patient and informal carer together). Severe symptoms that caused major disruption to normal life were described, often in terms implying acceptance of the situation as a “way of life” rather than an “illness.” Patients and their informal carers adapted to and accepted the debilitating symptoms of a lifelong condition. Professional carers' familiarity with the patients' condition, typically over many years, and prognostic uncertainty contributed to the difficulty of recognising and actively managing end stage disease. Overall, patients told a “chaos narrative” of their illness that was indistinguishable from their life story, with no clear beginning and an unanticipated end described in terms comparable with attitudes to death in a normal elderly population.

Conclusions Our findings challenge current assumptions underpinning provision of end of life care for people with COPD. The policy focus on identifying a time point for transition to palliative care has little resonance for people with COPD or their clinicians and is counter productive if it distracts from early phased introduction of supportive care. Careful assessment of possible supportive and palliative care needs should be triggered at key disease milestones along a lifetime journey with COPD, in particular after hospital admission for an exacerbation.

Statistics from


“Well, end stage is from the beginning, isn't it, to a certain extent?” [F07.2 Nurse]

Globally, long term conditions such as chronic obstructive pulmonary disease (COPD) are responsible for an increasing proportion of deaths.1 Cancer based palliative care services, predicated on an ability to predict a terminal phase of disease,2 3 are being extended to encompass people dying with non-malignant disease.4,,6 Prognostic indicators have been developed to aid identification of people “at risk of dying,” whose physical, psychological, social, and spiritual needs can be assessed and their care planned.2 7 There is concern, however, that the slow physical decline of patients with COPD, which is punctuated by potentially serious but unpredictable disease exacerbations, may lead to prevarication rather than provision of anticipatory care.8

About half of patients discharged after a hospital admission for COPD will die within two years.9 Markers such as severity of disease, poor nutritional status, comorbid heart disease, depression, impaired quality of life, and older age have all been shown to be associated with an overall poor prognosis.9 10 Accurate predictions of life expectancy for individual patients with COPD, however, remain extremely difficult.7 11 This difficulty with prognosis is compounded by a tendency for doctors who are familiar with patients to overestimate survival.12 The only condition where prognosis is less accurate is dementia.11

People with very severe COPD have a well recognised burden of disabling physical symptoms (especially breathlessness), compounded by comorbidity, psychological distress, and social isolation.13,,17 Despite these issues, the needs of these patients are typically poorly addressed, and many patients have limited access to specialist palliative care services.13 14 The consultation on a strategy for services for COPD in England18 and the standards of care for COPD in Scotland19—which advocate adopting a lifelong approach to preventing, diagnosing, and providing care for people with COPD—acknowledge this deficiency and prioritise access to improved end of life care for those “sick enough to die.”

To inform current deliberations on how best to provide care for people living and dying with COPD, we undertook an in-depth inquiry seeking to understand the end of life needs of affected patients and their informal and professional carers.


Our study took place over 18 months during 2007-9. Ethical approval was obtained from the Multicentre Research Ethics Committee for Scotland (B), and governance approval was obtained from NHS Lothian, NHS Tayside, and NHS Forth Valley.

Longitudinal qualitative research using multi-perspective, serial interviews offers advantages over the more usual single “snapshot” qualitative techniques in understanding patients' and family carers' evolving and dynamic experience of illness (box 1).20 21 In this study, we invited patients and their nominated informal and professional carers to take part in up to four interviews at 6-9 month intervals. Multidisciplinary perspectives from the research team, an end of study workshop, and views of a lay advisory group were used to ensure breadth of interpretation of the data.22

Box 1 Features of serial multi-perspective interviews20


  • Give important insights into patients' changing experiences over the course of an illness

  • Enable an understanding of relationships and dynamics between patients, their families, and their professional carers

  • Enable an exploration of similarities and differences in the perceptions of patients and their family and professional carers

  • Enable an understanding of (potentially contrasting) individual needs of patients, informal carers, and professional carers

  • Facilitate integration of suggestions for improving services from patients, informal carers, and professional carers

  • Allow the development of a relationship between the patient and the researcher as a result of increased contact with participants, which facilitates discussion of sensitive issues


  • Increased complexity of recruitment

  • Need to be sensitive to the disadvantages (as well as the advantages) if the patient opts for a joint interview with their informal carer

  • Potential for breaches of confidentiality if information from one informant is (inadvertently) relayed to another

Patient recruitment

Primary and secondary care clinicians from general practices (recruited through the Scottish Primary Care Research Network) or from hospital and community specialist respiratory services in Lothian, Tayside, and Forth Valley, Scotland, identified patients with end stage COPD. We provided information about the known predictors of a poor prognosis to aid identification of such patients,7 but specifically suggested that they should use the “surprise question”—“Would I be surprised if my patient were to die in the next 12 months?”8—to identify potential participants. We purposefully sampled to recruit men and women with different ages, social class, rurality, presence of an informal carer within the home, and current smoking status. Significant comorbidity was expected; the only exclusion criteria were inability to participate (for example, because of dementia) or other imminently life threatening illness (for example, lung cancer). A clinical assessment by a respiratory nurse established eligibility, indicators of severity, and markers known to be associated with poor prognosis.7

Recruitment of an “interview set”

At each time point, patients nominated for interview an informal carer (for example, a family member, friend, or neighbour), if they had one, and a key health or social care professional whom they regarded as important to their care at that time, thereby creating “interview sets.” The person(s) nominated could differ at different time points. Our previous studies suggested a sample of 16-20 interview sets would be sufficient to reach data saturation.20


Informed written consent was obtained from all patients at the beginning of the study and reviewed verbally before each interview. Informal and professional carers identified by the patient were asked independently for their consent to participate. Withdrawal of a patient's consent terminated the interview set. If an informal or professional carer withdrew, interviews were allowed to continue with other members of the interview set.

Data generation

A social scientist (MK) carried out the interviews with patients and informal carers at a location of their choice, and with professional carers by phone. In-depth interviews with the patient and his or her informal carers (jointly, if preferred by the patient) were participant led, allowing people to tell their story in their own terms and at their own pace. Issues covered included the experience of living with COPD, the patient's main concerns (whether physical, psychological, social, or spiritual), views on care and treatment, and carers' needs and concerns (for topic guides see web appendix 1 on Interviews lasted between 40 and 150 minutes and were all audio recorded, with consent. Field notes were recorded after each interview.

Health and social care professionals were asked about their perception of the patients' and informal carers' needs, available services, and barriers to the provision of care. Bereavement interviews were, where possible, conducted with both informal and professional carers. As we came to appreciate the effect of the very long disease trajectory of COPD, we used subsequent interviews specifically to explore the patients' “story” of their condition.

Data analysis

Interviews were transcribed and managed with a qualitative data software programme (QSR NVivo version 7; QRS International, Doncaster, Australia). Analysis was iterative throughout the study, which allowed emerging themes to be explored in later interviews; deviant cases were actively sought. The interviews with individual patients, informal carers, and professional carers were coded separately and then analysed: (a) as serial interviews with individual patients and carers to identify how needs evolved over time; (b) as integrated “patient/informal carer/professional carer” interview sets; and (c) as integrated “patient sets,” “informal carer sets,” and “professional carer sets.” Comparing and contrasting across and within these data sets highlighted emerging themes and, importantly, also divergence of perspectives.20

All transcripts were coded by MK (assisted by the study administrator), using a thematic narrative approach,23 reflecting the research questions and themes raised by the participants. We used categories informed by sociological theory on health and illness, such as Bury's work on biographical disruption (see web appendix 2 on Frank's typology of illness narratives was used to categorise patients' narratives into three types: restitution, quest, or chaos (box 2).25 Restitution narratives are satisfactory stories of illness and recovery. Quest narratives tell stories of determined action and significant events that led to understanding of an illness, a search for cure, adaptation to disability, or campaigning to raise awareness or better care for the condition. By contrast, chaos narratives appear to be a disjointed series of events within which neither the narrator who is living within the story, nor the listener, can discern a clear purpose or direction.

Box 2 Frank's typology of illness narratives

Arthur Frank's classic text The wounded storyteller: body, illness and ethics (1995) suggests three main types of illness narrative25

1) Restitution narrative

“Yesterday I was healthy, today I am sick, but tomorrow I will be healthy again”

This narrative recounts a belief in a restoration to health and portrays illness as transitory. It is the easiest narrative to tell and to hear.

2) Quest narrative

“I have a health problem but I’m using the experience to improve the situation (for myself or others)”

In this narrative the patient uses the illness experience as an opportunity to embark on a personal quest; for example, to adjust to life with illness or to improve care and support for people with a particular condition. This narrative portrays illness as an opportunity to learn and improve, and is often found in celebrity accounts.

3) Chaos narrative

“This happened. Later on this happened. And also this happened”

This is the least heard narrative because “lived chaos” cannot be told. The story can only begin to emerge once there is an ability, however tenuous, to stand outside the chaos. Telling and hearing a chaos narrative is frustrating because it appears disjointed and without causal sequence or purpose (an anti-narrative).

The researcher and principal investigator (MK and HP) met frequently to discuss the emergent findings, to aid data synthesis and interpretation. In addition, both the lay advisory group and the multi-disciplinary steering group regularly reviewed the evolving themes and thoughts on the illness narratives in the light of their areas of expertise, knowledge of the literature, and experience.22

End of study workshop

To bridge the (typically wide) gulf between research and practice, we convened a national multi-disciplinary end of study workshop to which we invited 27 academics, policy makers, and government representatives. This allowed us not only to share and receive critical feedback on our preliminary conclusions, but also to raise awareness of our emerging findings among those responsible for service planning and provision (see web appendix 3 on Extensive notes were made from recordings of the four break-out groups, in which participants were invited to reflect critically on the findings and discuss and debate the implications for policy and practice.

Lay advisory group

We recruited 10 people with COPD from local “Breathe Easy” groups, a support group network run by the British Lung Foundation for people living with a lung condition and those who look after them. This lay advisory group, which was facilitated by MK, met regularly throughout the study to discuss the interview topics, the emerging findings, and the methods of disseminating the findings (see web appendix 4 on The comments, suggestions, and perceptions of the group were used throughout the iterative analysis process to aid interpretation. Four members of the group attended the end of study workshop.


The characteristics of the 21 patients recruited to the study are given in table 1. Figure 1 gives details of the 92 interviews—23 of the patient interviews included the informal carer, and seven of the interviews with professional carers related to more than one patient. Table 2 explains our convention for describing the participants.

Figure 1

Schedule of interviews over the 18 month study

Table 1

Characteristics of the 21 patients with chronic obstructive pulmonary disease interviewed for the study

Table 2

Conventions for describing patients and interviews

Overview of findings

The major physical and social problems faced by the patients and their informal carers, and the considerable difficulties in accessing appropriate health and social care services, were expected findings13,,17 and will therefore not be dwelt on in this paper. Rather, we focus on the reasons underpinning why current strategies for extending palliative care services to people living and dying with COPD is proving such a challenge 8 and suggest alternative approaches to the current policy direction.

Our report is structured under the following themes:

  • I. Acceptance of COPD as “a way of life”

  • II. The story of COPD

I Acceptance of COPD as “a way of life”

Apparent throughout the patients' interviews was a sense of “acceptance” in the face of severe disease and social difficulties. COPD was something that had to be coped with “as best you can” [F07.1]. Distressing breathlessness was accepted as “the way it has worked out” [F02.1]. Patients did not actively seek out information about their condition; indeed several patients indicated that they would “rather not know” [F07.1]. For many patients their situation was made considerably worse by unsuitable housing that they seemed powerless to influence, leaving them to “do the best you can with what you've got” [F02.1].

“Oh I don't ask for anything do I? So, if I asked I think I would get, I am quite sure I would because I do get excellent help from the doctors, so as I say I don't ask … we manage as we are, yes, but for how long I don't know.” [F10.1]

Although this passivity might be construed negatively, or as weary resignation to insurmountable problems, in some contexts it could also be seen as representing an appropriate adaptation to a way of life.

Interviewer: “What has helped you most over these last two years that I’ve been coming to see you?”

Patient F10.4: “Learning to accept it. To stop fighting against it and just accept it in my mind.”

The only exception was a relatively young patient with α1 antitrypsin deficiency. He and his wife had actively searched for information on the internet, joined a national organisation for people with “Alpha 1,” and campaigned to improve their circumstances.

“I've done a lot of shouting at the council for four years to get re-housed and it didn't do much. I had to threaten them with a lawsuit.” [L06.1 wife]

II The story of COPD

An early observation was the contrast between the chaotic stories told by people with COPD and the well constructed and rehearsed narratives of people with, for example, lung cancer.26 Patients with COPD struggled to tell a coherent illness story distinct from their life story. These data are summarised in table 3.

Table 3

The story of COPD: comparing and contrasting perspectives

A story with no beginning

It proved impossible for our participants to identify a “beginning” to their COPD story. Some simply acknowledged that they had “always been bothered” [L03.1] with symptoms that dated back to their “younger years” [T06.1]. Others identified a milestone (for example, point of diagnosis or a severe exacerbation) and used it as a beginning, although it was clear that disabling symptoms had preceded this beginning by many years. A different beginning could be selected in subsequent accounts, demonstrating the fluidity of the stories.

Some patients responded with well structured stories of unrelated health problems (“I had a major op in the infirmary … They thought I had cancer.” [L04.1]). Others answered the question of “when?” by explaining “how” smoking or their job had caused the symptoms.

Interviewer: “So when did it all start?”

Informal carer T05.1 [wife]: “See he was a smoker …”

Patient T05.1: “We used to put that rubber on carpets and vulcanise it … we didn't realise that we were inhaling all that smoke.”

The insidious onset and lack of clear beginning to the disease was echoed by the clinicians, who described a point of “formal diagnosis of the actual COPD problem” [F10.1 GP] that was preceded by years of an “informal” recognition of the developing symptoms.

The exception again was the patient with α1 antitrypsin deficiency, who told a dramatic and well rehearsed story about how the diagnosis was confirmed the day before his wedding. Although he acknowledged that he had been “slowly deteriorating” for many years before diagnosis, the story was told as a sudden disruption to life.

Informal carer L06.1 wife: “Day before the wedding we found out he was basically a dying man.”

Interviewer: “Good heavens, how awful!”

Patient L06.1: “So my lungs was away and I was 41.

Interviewer: “That must have been a shock.”

Informal carer L06.1 [wife]: “Slightly!”

Patient L06.1: Because I was always active as a worker like, 18 hours a day sometimes, always busy then suddenly it was stop, because I couldn't do it anymore.”

A middle that is a way of life

Unlike the narratives of people with cancer and heart failure, where accounts of the period from diagnosis are characterised by a clear sense of a developing plot with one event leading to another,26 the patients with COPD told “chaos narratives” characterised by directionless stories that were typically indistinguishable from their life story and natural ageing.

“I fear I am getting worse (which is understandable), it's like any other illness. It's like ageing, you are getting older and that really is the illness is getting worse as well.” [F06.1]

The participants told illness narratives about their exacerbations (for example, being increasingly breathless and rushed into hospital) or a referral for assessment for lung transplantation (such as having high hopes that were dashed after complex assessments at a distant hospital). Exacerbations were described as independent episodes interrupting “normal” life; there was little sense of a linkage, or a developing trajectory. The very slow evolution of COPD meant that some participants who had experienced a long interval between exacerbations were almost able to tell a “restitution narrative.” Table 4 shows a restitution narrative told by a 75 year old man [T01] who was able to describe himself as “well” after several months in which his very severe COPD was relatively stable. This is set against a background of a chaos narrative in which little changes over the 18 month study.

Table 4

An example of a “restitution narrative” told against a background of a “chaos narrative” in which little changes over the 18 months study

Although clinicians were aware of the chronic deteriorating nature of COPD, they told a parallel story of a “way of life” and colluded (at least linguistically) with the idea that the patient was “well” between exacerbations.

“People like Mr X who doesn't really bother us that much, we really only see him when he's not well.” [F08.1 GP]

By contrast, the patient with α1 antitrypsin deficiency and his wife told a classic “quest” narrative. They knew and accepted that the story would end in his death, but were determined to use the intervening time fighting to improve care for him and others with the condition.

“They're not interested in people like us, as I say, well, because we've got … the wrong illness, the wrong age, well, they've freely admitted he's too young!” [L06.3 wife]

An unpredictable and unanticipated end

Patients were aware that their symptoms were severe and appreciated that the future was uncertain, but—consistent with the chaos narrative—“the end” was described in terms reminiscent of a normal expectation of death rather than as an anticipated consequence of an unfolding story. In keeping with findings from surveys of attitudes to death among the public in the United Kingdom,27 our participants were aware of their own mortality (for example, an 85 year old patient commented that she “did not know” whether she would still be alive for future interviews[L05.1]). Death, however, was generally not considered as an imminent threat, and end of life wishes were generally not discussed with professional carers, friends, or family.

“I know when the nurse first said, ‘I’m referring you to [a day centre run by the local hospice]' I thought oh God! This isn't terminal! Not me!” [T01.3]

In telling the story of an exacerbation, several patients said “I thought I was going to die” (in the past tense), but in keeping with the restitution narrative of the acute attack, once they were restored to normal the threat receded. Interestingly, the professional carers sometimes used language that echoed the patients' normalisation of the prospect of death.

“… we are all going to die aren't we? But it is a case of picking the time and place [to discuss it].” [L06.1 hospital doctor]

Professional carers were aware that patients were deteriorating, but a transition point for palliative care was elusive. The uncertain prognosis (“Is this it? Another year? Three years?” [F09.1 nurse]) was universally cited as a barrier, but there were other more subtle influences of particular relevance in the context of long standing relationships. Some professional carers struggled with moving from “always being able to do something” to acknowledging “for the rest literally what can we do?” [F07.2 nurse]. In contrast to consultations with people with cancer where “I think it seems to come up more naturally to talk about [death]” [L03.1 GP], a familiar and comfortable pattern of consulting prevented initiating an unlooked for discussion about the future.

Interviewer: “I was going to ask you whether you have talked to him at all about what might happen in the future and how things might progress?”

Professional carer L04.1 [GP]: “No, not really. He usually has got his own [agenda in the consultation]. It's more reassurance about how he is and chatting generally and he just likes a bit of social discourse I think.”

Again, an exception was the patient with α1 antitrypsin deficiency, who was one of only two participants in our study to have discussed the end with his doctors. He and his wife spoke openly from the very beginning of the first interview about his death and their plans for it. When faced with a decision about the use of morphine in a potentially life threatening exacerbation, his wife referred to their plans for future care.

“It wasn't a difficult decision for me actually because having spoken about it at length before, you know, when he had bad episodes, about, you know, what we wanted to happen etc.” [L06.1 wife]

End of study workshop: comments on the data

Participants in the end of study workshop offered additional explanations for our observation of acceptance of the way of life among patients with COPD and debated the practicalities of providing supportive end of life care in the context of a condition in which it is not possible to identify a time point for transition to the palliative care approach (box 3).

Box 3 Extracts from the notes for the four multidisciplinary discussion groups at the end of project workshops

A way of life

“‘Living with COPD’ is a better term. The beginning of the dying process is the first cigarette in the bicycle shed at the age of 12.” [Group IV]

“No beginning for the healthcare professional, because it is already advanced before the professional realises there is a story.” [Group II]

“The trajectory—slowly going down. Every time they come out of hospital they go down a step—it's not dramatic. Cancer is dramatic; the start to COPD is ‘boring’.” [Group II]

“Earlier diagnosis is making it even harder—very long trajectory.” [Group III]

Passive acceptance

“Are we talking about the professionals or the patients …???” [Group II]

“Could be adaptation—a way of dealing with the problem, getting on with their lives around this problem. Not necessarily a bad thing.” [Group III]

“Key is acknowledgement that one has a health condition that threatens ones life. Patients with COPD only accepted that they had an illness when they were in hospital or had an infection—otherwise life was ‘normal.’” [Group I]

Transition to palliative care

“It's not about transition, it's about integration.” [Group II]

“The difficult prognosis is a ‘cop out’ because the need is there. Can still support people.” [Group I]

“At least if they are on the register, someone in the practice is talking to them, whereas they are forgotten otherwise.” [Group II]

“Might be useful to think of milestones. Hospital admission (or second hospital admission) is the obvious one—should that trigger a change of gear?” [Group III]

Comparison with cancer care

“Applying the cancer model across the board is inappropriate.” [Group I]

“We have picked up the term ‘palliative care’ from cancer, maybe we need a different language?” [Group IV]


Summary of findings

The participants in our study described severe symptoms that caused major disruption to normal life, but often in terms implying acceptance of the situation as a way of life rather than an illness. The chaos narratives of their disease stories were impossible to distinguish from their life stories, lacked a clear beginning, and had an unpredictable and unanticipated end. Professional carers' perspectives echoed this chaos narrative, contributing to the difficulty in defining a point when palliative care might be appropriate. The policy focus on identifying a time point for transition to palliative care has little resonance with these accounts from patients, informal carers, and professional carers, and may be counterproductive if it distracts professional carers from timely consideration of providing much needed supportive care.

Passive acceptance, weary resignation, or comfortable adaptation?

The social and clinical burden described by our study participants reflects accounts in the current literature.13,,17 Some researchers have commented on a sense of “sad resignation”28 and a tendency for people with COPD to “marginalise” their condition as the result of “old age.”29 Delegates at the end of study workshop suggested a range of explanations for this “passive acceptance,” including “stoicism,” “weary resignation” after years of futile attempts to improve their circumstances, “recalibration” analogous to the response shift recognised in quality of life research,30 or an adaptive coping strategy. Habraken et al attributed the “silence” of people with end stage COPD to them considering their limitations as normal and regarding themselves as ill only during acute exacerbations.31 Our data support this interpretation and enable us to offer a theoretically based interpretation.

Not so much an illness, more a way of life

In her seminal book Hard earned lives: accounts of health and illness from East London (1984), Cornwell identifies three categories of health problems: normal illnesses, real illnesses, and health problems that are not illnesses (table 5).32 Although our study participants all recognised the current “reality” of their illness, the lack of story, the causal link with lifestyle, and the lifelong trajectory of the disease suggest that, despite the severity of their symptoms, they classified COPD as a health problem that is not an illness. Symptoms were described as stemming from a lifetime of exposure to fumes, smoking, or both, and breathlessness was considered “part of getting older.” By contrast, exacerbations were often classified as real illnesses, but once the acute episode was over patients felt they were “back to normal.” The only exception was the man with α1 antitrypsin deficiency, in whom a genetic cause, a clear medical diagnosis, a young age, and severe symptoms signified a real illness.

Table 5

Categories of health problems according to Cornwell's book Hard earned lives: accounts of health and illness from East London (1984)32

The lack of biographical disruption

The term “biographical disruption” describes the major disruptive experience of developing a chronic illness and the consequent rethinking of a person's biography and self concept.24 33 34 More recently, the idea of a “biographically anticipated” event35—exemplified by older peoples' experience of osteoarthritis36 and stroke37—considers that for some people in some circumstances, certain illnesses can be a “normal crisis.” Our findings extend these concepts by showing that in a very slowly progressive condition such as COPD, patients may have no sense of biographical disruption at all. In such individuals there is no illness narrative separate from their life narrative—rather, our study shows that people gradually adjust their sense of self over the years to fit within the limitations imposed by the condition. This lack of disruption may be at the root of the patients' and carers' acceptance and passivity, such that they neither demand nor use services. Clinicians, especially those who have a long term relationship with the patient, may share this passive acceptance of the patient's way of life, contributing to the difficulties in identifying a transition point to palliative care.

The lack of a public story

Cancer has a well publicised “public story” in which death plays a prominent role.38 People with cancer are aware that they have a potentially fatal condition and tell a dual narrative of despair (“I'm going to die”) balanced by hope (“but I might get better”).23 By contrast, COPD is a relatively unknown condition—even the name causes confusion—with no public story to which patients can relate. As such patients often have no expectation of death and no despair, and equally no hope of cure.

Comparison with other illness trajectories

Our findings raise interesting comparisons with other disease trajectories. COPD has a lifelong trajectory (effectively from the first exposure to cigarette smoke, which may be in utero),39 and the onset of symptoms is so insidious as to be imperceptible. In contrast, despite their chronic nature, most other types of long term organ failure typically have a relatively acute presentation (for example, myocardial infarction may present as heart failure).

Another comparator might be patients who have congenital conditions. The concept of “continual biographical revision” has been used to describe how young people with cystic fibrosis attempt to normalise their lives,40 although the genetic cause of this disease and its impact during childhood distinguishes it from COPD. Morbid obesity, which is caused by lifestyle, may be a parallel health problem that is not an illness. The poorly understood disease trajectory of the frail elderly echoes recent understanding of accelerated lung ageing in smokers,41 and the biomedical model of decline in lung function,39 and may resonate best with the disease trajectory of COPD.

Strengths and limitations of study

Our 21 participants may not fully represent the diversity of people with very severe COPD. In particular, none was from an ethnic minority background, although the study cohort encompassed a broad range of demographic, social, clinical, and healthcare backgrounds. All our participants were smokers or ex-smokers, reflecting the profile of COPD in the UK, and it cannot be assumed that their attitudes can be extrapolated to non-smokers. However, the causal link of COPD with lifestyle factors described by our participants encompassed environmental exposure to a broad range of pollutants as well as smoking. Interviewing patients and informal carers together may have resulted in perspectives being modified or withheld; our experienced researcher was aware of this potential problem and could adopt strategies to enable independent voices to be heard.20 We were aware that researchers' attitudes influence design, data collection, and analysis of qualitative studies, and used our multidisciplinary professional team and lay advisers to ensure balanced interpretation of the data.22 In addition, the comments of the delegates at the end of the study workshop added to our understanding of the emerging themes and their practical and policy implications.

Conclusions and policy implications

Our findings challenge current assumptions about models of care and have two important implications for clinicians and policy makers. Firstly, our findings suggest that, in contrast to other conditions, COPD is perceived as “a way of life” rather than an illness that disrupts life. This may underpin acceptance of and adaptation to increasing disability and major health and social needs, in a lifestyle that has become familiar over many years to patients, informal carers, and professional carers. Recognition by healthcare professionals of the risk of “passive acceptance” may enable services to be developed that actively identify and seek to address needs.

What is already known on this topic

  • The needs of people with severe chronic obstructive pulmonary disease (COPD) are poorly addressed, despite the recognised burden of disabling physical symptoms, psychological distress, and social isolation

  • Policy makers are extrapolating to patients with COPD models of palliative care designed for people dying from cancer, in whom a transition point to a terminal phase can be relatively easily identified

What this study adds

  • In contrast with many other long term conditions, COPD is perceived by patients and their informal carers as a “way of life”

  • Patients with COPD tell a “story” of a health problem with no clear beginning, that is indistinguishable from their life story, and with an unpredictable and unanticipated end

  • Clinicians may need to use a proactive management strategy in patients with COPD to minimise the risk of suboptimal management associated with patient and carer inertia.

  • Rather than looking for a transition point to switch to a palliative care approach, holistic assessments of needs could be linked with milestones throughout the patient's journey (such as a hospital admission or retirement on medical grounds), which would progressively integrate supportive care into the good care for people with COPD

Secondly, current models of palliative care for non-malignant disease, adapted from the traditional cancer model, are predicated on an ability to identify “palliative” patients whose end of life care needs should then be assessed and addressed (and who in some systems should eschew curative care3).2 Our data suggest that a point of transition to palliative care is meaningless and impractical in COPD, a condition with no coherent story and an unanticipated end. Linking palliation of symptoms and supportive care to identifying a point of transition thus risks “prognostic paralysis.”8

We therefore propose linking holistic assessments of supportive and palliative care needs with milestones throughout the patient's journey. Suitable milestones might be diagnosis, retirement on medical grounds, starting long term oxygen therapy, hospital admission for an exacerbation of COPD, or (from a clinicians' perspective) a positive answer to the “surprise” question.8 The palliative care approach thus becomes progressively integrated with good care of a lifelong progressive disabling condition, with palliative care specialists available to advise on management of intractable symptoms (especially breathlessness). The historically low profile of COPD and its lack of public story is a further barrier to effective provision of supportive care. Voluntary organisations may choose to champion the disease and communicate a story to aid public understanding of this silent, lifelong, progressively disabling condition.



  • This is a reprint of a paper that first appeared in BMJ 2011;342:doi:10.1136/bmj.d142

  • We are grateful to Janet Winter, Colette Lamb, and Anne Aitken, who undertook the initial clinical assessments, and to Marie Pitkethly, Colette Fulton, and June McGill, from Scottish Primary Care Research Network, who helped with recruitment. Susan Buckingham provided administrative support, and David Chinn assisted with coordination of the project within the work of the Primary Palliative Care Research Group. We thank the members of the lay advisory group, recruited by Wendy Halley from the local British Lung Foundation Breathe Easy groups, who worked with us throughout the study, and Patrick White, Declan Cawley, and Tony Davison for their perceptive comments on our draft manuscript.

  • Funding This study was supported by the Chief Scientist's Office of the Scottish Government. HP is supported by a primary care research career award from the Chief Scientist's Office of the Scottish Government.

  • Competing interests All authors have completed the Unified Competing Interest form at (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Ethical approval Ethical approval was provided by the Multicentre Research Ethics Committee for Scotland (B), and governance approval was obtained from NHS Lothian, NHS Tayside and NHS Forth Valley. Informed written consent was obtained from all patients at the beginning of the study and reviewed verbally before each interview. Informal and professional carers identified by the patient were asked independently for their consent to participate.

  • Data sharing We do not have consent to share data.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.