Article Text

Download PDFPDF

Exploring the therapeutic power of narrative at the end of life: a qualitative analysis of narratives emerging in dignity therapy
  1. Glendon R Tait1,2,
  2. Catherine Schryer3,
  3. Allan McDougall4 and
  4. Lorelei Lingard4
  1. 1Department of Psychiatry, Dalhousie University, Halifax, Nova Scotia, Canada
  2. 2Department of Psychiatry, University of Toronto, Toronto, Ontario, Canada
  3. 3School of Professional Communication, Ryerson University, Toronto, Ontario, Canada
  4. 4Centre for Education Research and Innovation, University of Western Ontario, London, Ontario, Canada
  1. Correspondence to Dr Glendon R Tait, Department of Psychiatry, Dalhousie University, Suite 7113A, Abbey J Lane Building, 5909 Veterans Memorial Lane, Halifax, NS B3H 2E2, Canada; glendon.tait{at}


Objective To understand the therapeutic effect of a narrative intervention, specifically dignity therapy, in patients at the end-of-life. To examine the thematic dimensions and shared narrative features of the stories that emerge in dignity therapy and theorise their relationship to the intervention's clinical impact.

Design Resident physicians, as part of an educational intervention, co-administered the dignity therapy protocol with the principal investigator. Interviews were transcribed, edited, and then, within a week, read back to the patient and provided as a document for the patient to keep. A constant comparative approach was taken to identify narratives and thematic patterns.

Participants 12 Patients at the end-of-life were administered dignity interviews by 12 resident physicians, accompanied by the principal investigator.

Setting Palliative care settings in two University of Toronto academic hospitals.

Results Three narrative types emerged, each containing several themes. Evaluation narratives create a life lived before illness, with an overarching theme of overcoming adversity. Transition narratives describe a changing health situation and its meanings, including impact on family and on one's world view. Legacy narratives discuss the future without the patient and contain the parables and messages to be left for loved ones.

Conclusions While the interview protocol guides patients' responses, the commonality of narrative structures across interviews suggests that patients draw on experiences with two familiar genres: the eulogy and the medical interview, to create a narrative order during the chaos of dying. The dignity interview's resonance with these genres appears to facilitate a powerful, and perhaps unexpected sense of agency.

Statistics from

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.


Dignity therapy is a narrative intervention that emerged in response to evidence that the end-of-life experience extends beyond physical discomfort to psychological and existential factors such as hopelessness, burden to others and dignity.1 Dignity therapy uses an interview guide based empirically on patients' concept of dignity at the end of life. Its goal is to foster dignity by helping a patient make meaning of the events in his or her life.2

Research has documented the impact of dignity therapy on indices such as sense of purpose, will to live, sense of suffering and depressed mood.2 These findings prompted a multicentre, international clinical trial to further evaluate its effects. However, the question of how this narrative intervention produces these effects is less well explored. While there has been increasing interest in the narrative dimensions of medicine,3,,9 no research has explored the shared narrative features of the stories that emerge from dignity therapy, or how these features might be implicated in the intervention's clinical impact. This information is crucial for understanding how such an intervention addresses the needs of patients. Similarly, it would be prudent to understand whether there are essential narrative elements of this experience and how these might be integrated routinely into the encounters between dying patients and the healthcare professionals who care for them.

We sought to systematically examine the thematic dimensions and recurrent narrative features that emerge in narratives told in response to the dignity therapy protocol, and to theorise the relationship between these narrative dimensions and the intervention's therapeutic effect.


Theoretical framework

Reflecting our interest in what the dignity interview does through language, we position this work within the theoretical framework of rhetorical genre studies, which investigates language in its social contexts, with special attention to the social action of these events—that is, how specific text types can enable and constrain the actions of participants.10,,12

Study setting

This study occurred in palliative care settings in two academic hospitals associated with one faculty of medicine. This study was part of a broader project where dignity therapy was used as an educational intervention for first-year psychiatry and family medicine residents completing a rotation in palliative care. Residents were accompanied by a senior psychiatry resident trained in dignity therapy at the Memorial Sloan Kettering Cancer Center, one of the dignity trial sites. This study was approved by the Research Ethics boards of both hospitals.


The study sample consisted of 12 dignity interviews that were conducted by first-year residents. This was a convenience sample in that both patients and residents were volunteers and could accommodate the interview schedule. One invited patient declined; four additional residents consented but could not accommodate interview schedules. Participants had terminal diagnoses, with fewer than 6 months to live, and did not have cognitive impairment. Patients were identified as potential participants by their palliative care physician; informed consent was obtained by the principal investigator. Interested patients were provided written information, and the principal investigator secured their informed consent.


Resident physicians were provided with a guide that outlined the rationale and nature of the intervention, and they met with the principal investigator before the interview for a brief educational session at which time any questions were addressed. While resident physicians led the interviews, to ensure that the interview approximated Chochinov's protocol,2 the principal investigator made minor contributions to the interview, such as asking questions to help the patient to clarify or elaborate.

The questions of the dignity interview were asked in a 1-hour semistructured interview (see Box). The recorded interview was transcribed, yielding transcripts from 8 to 17 pages in length. The resident physician edited the transcript—for example, if the recorded transcript included a conversation from a nurse who entered the room, this would be deleted. The patient's wording was preserved and no grammatical alterations were made.

Box The dignity interview question guide2

  • Tell me a little about your life history: particularly the parts that you either remember most or think are the most important?

  • Are there specific things that you would want your family to know about you, and are there particular things you would want them to remember?

  • What are the most important roles you have played in life (family roles, vocational roles, community service roles, etc)? Why were they so important to you, and what do you think you accomplished in those roles?

  • What are your most important accomplishments, and what do you feel most proud of?

  • Are there particular things that you feel still need to be said to your loved ones, or things that you would want to take the time to say once again?

  • What are your hopes and dreams for your loved ones?

  • What have you learnt about life that you would want to pass along to others? What advice or words of guidance would you wish to pass along to your (son, daughter, husband, wife, parents, others)?

  • Are there words or perhaps even instructions that you would like to offer your family to help prepare them for the future?

  • In creating this permanent record, are there other things you would like to include?

Within a week, the resident physician and principal investigator returned to read the document to the patient. While rare, patients could then request to add or omit details, which were made prior to the final document being returned to patients.

Data analysis

Four researchers participated in the data analysis: an expert in discourse analysis, a communications researcher, a psychiatrist and a research assistant with discourse analysis training. Our early analytical process used a constant comparative approach,13 which involved individual reading of transcripts and group discussion of emergent thematic patterns, through which we came to a consensus on an overall narrative pattern of thematic elements. The transcripts were coded for grammatical features that identify narratives and for thematic features of the narratives. The group held regular meetings to review emerging patterns, attend to discrepancies, refine thematic categories and describe the emergent narratives.


Patients' stories fell into three overarching narrative types, described below with the themes and subthemes present in each. In this report, names, dates, places and narrative details that would identify individuals, families or communities have been changed to preserve anonymity.

Evaluation narratives

In evaluation narratives, patients reflected on their early lives before their current illness emerged. Throughout the evaluation narratives, one overarching theme recurred: overcoming adversity. Evaluation narratives frequently related patients' experiences of illness, poverty or persecution, and described how these obstacles were overcome as the following examples illustrate: When the Americans and the Japanese went to war, we had to evacuate and we had to stay 100 miles away from the coast. Some were sent to camps, others chose to just come out east. . . at that time, there were several very intelligent young men who graduated and, because of the war situation, they got no job except maybe gardening or mundane type work. But my people were strong. (P4) . . .we wore hand-me-downs, but so did everybody. You never thought twice about that. We've been talking lately about food packages. . . in those days. There. . . would be a cardboard thing in between the biscuits. . . And by golly, we took that cardboard out and we put it in the kitchen drawer, and when your shoes started to get holes in the bottom of them you'd grab about three of those cardboard papers, cut them to fit, and that was how you soled your shoes. Like, who had a nickel to go the shoemaker? … And you knew that every other kid on the block was doing the same thing, so who cared? (P3)

Evaluation narratives contain three prevalent subthemes. First, patients recounted narratives, as illustrated below, about their heritage, including the social, cultural or political contexts of their birth, their parents' careers or personalities and, similarly, their upbringing: I was born just after the war when things in London were quite scarce. My mother, my father and I used to go away a lot of weekends which was not possible for many people. We had a car, so we used to go to the seaside or go away for weekends. (P6)

Second, patients discussed maturation, including marriage, raising children, spouses and careers: It was very bad times because it was a war and a new country. It wasn't easy. . . my husband's family was living in Europe. He wasn't in the camp; he was in the army. So, after the war it was very bad for [his family], so we used to send them parcels. . . . We brought the brothers here, we helped them with everything. . . We shared whatever we had. (P7)

Third, patients spoke of their social lives or their travels: [Our cottage] was more like a shack. . . The roof leaked. There was no electricity, it was strictly propane. There was no running water. There was a pail-a-day toilet. It was the basics, but, boy, we loved it. It was a sandy beach and it was a lake. You know those lakes, they're mostly all rock and once in a while, because of the current or whatever, you might get a sandy beach. Well, that's what we had. It was beautiful. (P3)

Transition narratives

Transition narratives discuss the impact of past events on the context of dying. Throughout all transition narratives, the overarching theme is a changing health condition. The previous statement of ‘living a good life’ inherently indicates a transitional state: the transition from ‘living one’s life' to a life that ‘has been lived’. My life's been quite happy, no very happy really. I’ve been very lucky apart from this, this is a bit of a stinker and that's mild. (P6)

Naturally, the major topics addressed during transition narratives involved patients' illness-related concerns. For example, some patients discussed how they have experienced societal conceptions of illness since their diagnoses. Yet, patients more often related how their terminal diagnosis had changed their lives and the lives of their loved ones: My mother lived until she was 88; she had smoked for like 50 of those years and never had a sickness in her life. Because I was her stature and whatnot, I just figured I was going to live until I was 88 as well. So it's been a bit of a shock. I feel cheated out of 20 years because I’ve always led a very healthy life. I’ve never smoked, never been a drinker so to speak; am a golfer, curler and all those kind of Canadian things. . . (P2)

Aside from illness-related concerns, transition narratives also recounted how patients' past experiences had influenced their current worldview. Subthemes within patients' transition narratives spoke of personal growth, changing character and a sense of altruism—karma, the ‘golden rule’ and forgiveness—themes not mentioned in evaluation narratives: I've really softened; I feel I’ve softened and been more understanding; just more understanding where maybe I wasn't before. . . It's been really a gift to have this part of my life because I swear having this kind of life ending experience is a good thing. (P2)

Along with discussing new insights as a coping mechanism for loved ones, several patients' transition narratives described writing autobiographical narratives in the form of letters for their loved ones— another way that past events have an effect on the present context: I spent hours sitting at the computer sobbing and writing the letters and I think what I’ve written is sufficient. It was written with a clear mind. If I’d have left it even until now I think I wouldn't have been able to write as clearly as I did and offer them as much comfort as I think there would be in the letters. (P6)

Legacy narratives

Legacy narratives are the parables and messages patients wished to leave for their loved ones: I really hope my legacy will embrace them and influence them to living, not necessarily my life but just to live a good life, a charitable life, that kind of thing. That would be my wish because charitable means a lot of things. I would mean just be gracious, open and just generous in that way, not necessarily with money but generous with her spirit and stuff. (P2)

Patients' discussions of legacy involved three primary subthemes. The first situated the patient's newfound perspective when facing death. Patients remarked that facing death has left them with few regrets, often feeling ‘more alive’ than ever before: I actually feel pretty alive now, ironically, but I think that's just wisdom. . . Your perspective changes when you're facing death, so I think you do see things differently and time is of essence and you appreciate things more. . . It's almost like you step outside of the group and you look into the group and a lot of the little chatter goes away. In my head I find I can find more peace in my thinking. Things aren't important that used to be important. (P8)

The second subtheme involved how patients believed their loved ones should conceive of time. These narratives espoused a ‘seize the day’ message for loved ones and endorsed appreciating life before it is too late. The body disintegrates, so you shouldn't spend too much effort on that part of it. The fact is that everyone will go through it at some point. I think that's what we all have in common. So, don't get caught up in things that are not important. Stay focused on what's important. (P8)

Last, legacy narratives involved the subtheme of condolence in the form of advice on how loved ones should move on with their lives. These narratives discussed forgiveness and faith as paths to general truths or better living; patients encouraged loved ones to remember the ‘good times’ and to go on and lead ‘good lives’. [My family is going to] miss me and I know they will. I know my children and [my husband], particularly but life goes on. . . you have the funeral; it's all cleaned up; people walk away and life goes on. That's how I see it. They have to go home. They have to deal with things. . . Get on with life and just embrace whatever I’ve left in memory. (P2)


Patients' narratives across these 12 interviews follow similar thematic patterns. The dignity interview guide seemed to provide a time-oriented template that assisted patients in generating these narratives: questions relating to a patient's life history often generated evaluation narratives, while questions relating to the patient's legacy generated legacy narratives. However, while the dignity interview sets up a timeline that, similar to the eulogy, invites patients to shape their narratives into past, present and future orientations, it does not completely control or predict the narratives that patients create. In fact, the transcripts provide some striking examples of how patients take up the interview questions in surprising, but similar ways. For example, nothing in the dignity interview suggests that patients outline narratives of overcoming adversity, and yet they respond with evaluation narratives that highlight this theme. Similarly, nothing within the guide directs patients to discuss how their worldview has changed since facing death, yet patients assert repeatedly that they ‘have never felt more alive’ and often that they can conceptualise the dying experience as positive.

Our data suggest that patients' responses draw on two familiar and closely related text types or genres: the eulogy and the medical interview. A mourning community uses the eulogy to construct a coherent narrative from the chaotic and troubling event of death.14 Similarly, patients in our study use the dignity interview to construct a coherent narrative from the chaotic and troubling experience of dying. The three narrative types described in our results are clearly reflective of the conventional structure of the modern eulogy.15 Evaluation narratives contain elements of the Epainos, which praises the deceased and recounts vivid past narratives (such as overcoming adversity); transition narratives resemble the Paramythia in expressing appreciation of the deceased's good life; legacy narratives share with the Epilogue its testament to continuing bonds with the deceased. Remarkable in the dignity interview narratives, however, is a radical change from the eulogy's traditional locus of agency. In the dignity interview, patients are not only the subject of the eulogy, they are the agent of its creation as well, co-constructing it with the physician interviewers, through the interview protocol. In fact, the daughter of one patient (P1) approached the principal investigator over a year after her mother participated to share that her mother's legacy document was read by her rabbi as a eulogy at her mother's funeral.

The second relevant genre echoing through this narrative experience is the conventional medical interview. The traditional medical interview is predicated on the physician's solving of a ‘puzzle’.16 This objectification supports particular communication patterns, such as the tendency for physicians in general medical practice to repeatedly interrupt patients.17 In the past 20 years, in response to calls for more whole person care, new interview methods such as ‘patient-centred clinical methods’ have emerged.18 Despite this, most patients with chronic and terminal disease will have had experience with the more conventional method.19 And, while medical interviewing practices have evolved, are physicians investing in the intimate engagement required to witness a patient's story, or, as suggested by evidence of declining empathy over the course of medical training, is the medical interview still practiced in a manner that maintains distance rather than increasing intimacy?20 21 The dignity interview is predicated on facilitating patients' existential meaning-making and has as its first function providing patients with the opportunity to create a legacy document which captures their unique story. And, just as it may be a new experience for patients to express their memories, opinions, hopes and dreams in an interview with a physician, so too it may be a unique experience for physicians to solicit and witness them.22 Toolan23 suggests the role of ‘audience’ is necessary for the production of narratives. That the intervention requires the healthcare provider to switch from being an interrogative gatherer of information to being an ‘audience’ for the patient's story may facilitate the transformation from ‘patient’ back to ‘person’. This may foster a powerful sense of agency.

Chochinov's group has recently published a study wherein they sought to examine the content of dignity transcripts.24 They saw the experience as being one that allows patients to express their values, including ‘family’, ‘pleasure’ and ‘caring’, ‘sense of accomplishment’ and ‘rich experience’. Our study builds on this work in three ways. First, while we found similar values in our transcripts, such as the importance of family and the primacy of caring, we identified values not described by Hack et al,24 one of the most important being ‘overcoming adversity’. Second, our analysis takes another step, mapping the interviews' dominant narrative types or storylines in which these values play out; this step was crucial to being able to understand the social context, and action of the intervention, revealing more about the ‘how’ of the intervention. Third, we have described the agency afforded patients as they construct their narrative responses. This is a powerful—and perhaps unexpected—sense of agency that emerges from the dignity interview's resonance with two familiar genres, the eulogy and the medical interview. Patients are cast as the agent (rather than just the object) of their own eulogy. And they can tell a physician—cast in the role of listener—their human story rather than just their illness story.


This study used a small, convenient sample of patients who had volunteered for the dignity interview in the context of an educational intervention for resident physicians. As with any qualitative study, the narrative patterns found in our study cannot be said to be exhaustive for all patients. Future research would need to explore narratives from a more institutionally diverse sample of patients to explore the transferability of our findings. It would also be important, in a larger study, to probe for discrepant cases to enrich our understanding. As well, the impact of the educational context of the dignity interviews in our study is unknown; are these dignity narratives different in some way due to the resident as an audience that the patient is helping to ‘teach’?

Conclusions and future directions

Acting as the story tellers of their own lives, patients create a narrative sense of order and meaning from the chaotic experience of dying. They achieve this narrative order through evaluation, transition and legacy narratives. Our findings suggest that the ability to achieve a sense of narrative agency is a potent aspect of dignity therapy; continued research is required to better understand the mechanisms by which narrative agency is produced and its relationship to the clinical efficacy of this intervention. We have argued that the agency here seems related to the resonance with, and dissonance from, adjacent genres, such as the eulogy and the traditional medical interview. If a key part of fostering agency is not only the interview event itself but also having the physician in the role of listener—an ‘audience’ to authentically witness, co-construct and document a patient's life story—then can the same sense of agency be achieved with non-medical interviewers? Our results, which emerged from interviews administered by resident physicians, albeit co-administered by the principal investigator trained in dignity therapy, suggest that expertise might not be the only relevant quality. Also important may be the context (medical setting), the interviewer's identity (health professional) and the frame provided for the medical personnel to truly act as an ‘audience’. Key ingredients may be the interview's departure from traditional medical interviews and the subsequent surprise of patient agency in this narrative event. There are undoubtedly many ways in which health professionals bear witness to patients' stories and foster agency. Since interventions such as dignity therapy are practically difficult to make widely available, future research must continue to examine how best to educate professionals to bring such ingredients to not only a formalised intervention, but to each and every encounter with dying patients who are eager to share not only a ‘history’, but a story.


The authors would like to acknowledge Dr Brian Hodges, who supervised the fellowship research of GRT and provided inputs to the design and execution of the study, and supported the funding applications. The study could not have been carried out without the excellent training provided to GRT by Dr Shannon Poppito at the Memorial Sloan Kettering Cancer Centre. The authors thank Drs Dori Secceracia and Jeff Myers for their assistance in identifying participants. The authors also thank Dr Susan Abbey for encouraging the pursuit of the idea that led to this study.



  • Funding This study was supported by the Royal College of Physicians and Surgeons of Canada Medical Education Research Grant and the Sociobehavioral Cancer Research Network, with funds from the National Cancer Institute of Canada.

  • Competing interests None.

  • Ethics approval Ethics approval was obtained from the Health Sciences Research Ethics boards of University Health Network and Sunnybrook Health Sciences Centre, the two University of Toronto hospitals where the study took place.

  • Provenance and peer review Not commissioned; externally peer reviewed.