“Whatever happens, happens” challenges of end-of-life communication from the perspective of older adults and family caregivers: a Qualitative study

Background Effective communication is integral to the delivery of goal-concordant care for older adults and their family caregivers, and yet, it is uncommon in people with serious illness. This study explores the challenges of integrating end-of-life communication into heart failure management from the perspectives of older adults and family caregivers. Methods In a qualitative study of older adults with heart failure and their family caregivers, fourteen semi-structured interviews were conducted with 19 participants in Ontario, Canada. The interviews were transcribed verbatim and thematic analysis was applied to analyze the data. Results Four themes were identified in the context of participants’ understanding of illness: 1) trivializing illness-related challenges, 2) positivity in late life, 3) discomfort in having end-of-life conversations, and 4) reluctant to engage despite need. These challenges often intertwine with one another. Most participants had not engaged in end-of-life discussions with their clinicians or family members. Conclusion The findings provide insights that can inform approaches to integrate end-of-life communication for older adults with serious illness and caregivers. The identified challenges highlight a need for end-of-life communication to occur earlier in illness to be able to support individuals throughout the period of decline. In addition, end-of-life communication should be introduced iteratively for those who may not be ready to engage. Alternative approaches to communication are needed to elicit the challenges that patients and caregivers experience throughout the progression of illness to improve care for people nearing the end of life.


Design, setting, and participants
This study was guided by interpretive description methodology, which acknowledges the constructed and context-dependent nature in which health-related experiences form [26,27]. Using a purposeful sampling technique, older adults and their family caregivers were recruited from a specialized outpatient heart failure clinic in a medium-sized urban teaching hospital in Ontario, Canada (see Table 1 for inclusion criteria). An interdisciplinary team of cardiologists, a nurse practitioner, and a pharmacist, all of whom have specialized training in heart failure, care for patients in the clinic. Over the course of recruitment, 36 patient candidates were identified. Active recruitment concluded after 22 patients were contacted to participate in the study. Of the 22 study candidates that were contacted for recruitment, two patients declined to be approached by the researcher and one candidate could not be approached due to cognitive decline. Recruitment of study participants came to an end when thematic saturation was reached which was determined based on no new data on relevant phenomena of interest being generated. No participants withdrew from the study.

Data collection
In-person, semi-structured interviews were conducted with older adults and their family caregivers as individual or dyad interviews. Interview guides were developed by the lead author (JI) with input from the multidisciplinary research team (i.e. clinicians and researchers in cardiology, palliative care, family medicine and social work). The interview guides included open-ended questions about participants' understanding of illness, their goals of care, and prior engagement in end-of-life discussions. Probes were used to generate discussion on specific aspects of end-of-life communication such as hypothetical scenarios and concerns about the future (see Additional file 1 for interview questions). Adaptations were made to the interview guides as data collection and analysis progressed to refine questions and exhaust emerging concepts. Interviews were conducted by a trained qualitative researcher (JI) who was mentored by the senior author (KK), an experienced qualitative researcher who leads a program of research on the experience of vulnerable populations and their caregivers. Participants were introduced to the study and provided an opportunity to ask questions. The data collection process began by collecting participant characteristics followed by semi-structured interviews. The interviews were audio-recorded and transcribed verbatim by a medical transcriptionist and participants were reminded that their participation was voluntary. To ensure data accuracy and confidentiality, the transcripts were checked against the original recordings and identifiable information was redacted. The majority of the participant interviews were conducted at the clinic in a private room and the remainder of the interviews were conducted in participants' homes based on their preference. For patients and caregivers who were recruited as dyads, the decision to be interviewed individually or together was left to the participants given the sensitivity of the topic under inquiry. No prior relationship between the interviewer and any participants existed. The interviews ranged from 25 to 90 min, although most interviews were completed in 45-60 min. All interviews were completed in English in one session. The data collection period lasted from August 2017 to January 2018.

Data analysis
The analytical process began with reviewing the transcripts at a high-level alongside data collection. To become familiar with the data, transcripts were reviewed and data were coded using a qualitative data management software program, NVivo11 (QSR International). Upon becoming familiar with the data, the first author reviewed the transcripts in depth by applying the specific research question under inquiry. Thematic analysis was used to code the data in order to identify patterns of related constructs to generate codes and themes [28]. Descriptions and interpretations of different excerpts were juxtaposed throughout the transcripts to answer the research question. While reviewing each transcript, relevant data that related to barriers of end-of-life communication were extracted and coded.
To ensure rigour, the process of developing themes and interpreting the findings consisted of reviewing the transcripts multiple times. A qualitative memo was created to document the coded excerpts and their accompanying interpretation, which was reviewed by the senior author (KK). Regular meetings were held with the team as well as knowledge users (e.g. clinicians in cardiology and palliative care) to discuss findings as analysis progressed and as themes became identified. Engaging with the research team and the knowledge users helped to increase the trustworthiness and credibility of the findings as multiple perspectives from various disciplines were accounted for.

Ethics
Prior to data collection, the ethics application was approved from the Research Ethics Boards at Sinai Health System (July 12, 2017). All study participants provided written informed consent to participate in the study, which included consent for audio recording and transcription of the interview, and for the findings to be used in the dissemination of knowledge upon removing identifiable data.
Four main challenges of engaging in end-of-life discussions were identified: trivializing illness, positivity in late life, discomfort in having end-of-life conversations, and reluctant to engage despite need (additional quotes can be found in Table 5). Although four challenges are discussed independently, these challenges are intertwined with one another. The challenges are identified in the context of participants' understanding of illness, which is reported in a separate article [29]. Briefly, participants had detailed knowledge of heart failure management and its self-care behaviors, and yet, participants appeared limited in their understanding of the consequences of illness and its progressive nature. Participants did not connect declines in health as part of the progressive illness trajectory; rather declines in health were perceived as temporary health states and hospitalizations were thought to be a routine part of illness management. Despite challenges they faced, participants adapted to the challenges of heart failure, which appeared to influence their perception of overall health. Most participants had not engaged in prior end-of-life discussions.
Challenges of end-of-life communication Theme 1: trivializing illness -"Everything is…I'm managing fine" Participants described the manifestations of heart failure as interrupting their day-to-day living commonly due to symptoms including fatigue and difficulty walking. Symptoms of illness impeded patients' ability to perform day-to-day tasks such as cooking and shopping. At the same time, participants described the ways in which the recommended self-care behaviours of heart failure (e.g. dietary restrictions and taking medications) interfered with aspects of life they once enjoyed such as eating their favourite foods. However, even as patients described a loss of independence and certain pleasures in life, they emphasized their ability to continue managing under their circumstances. For example, a 90-year old patient with advanced heart failure, and other chronic conditions described her difficulty with walking up the stairs, which further affected her ability to perform other tasks such as preparing meals for herself. Throughout the interview, the participant reiterated her ability to manage "fine" (PT07). Hence, it appears that despite the difficulties experienced participants have a tendency to maintain their circumstances. As one caregiver put it: CG03: it's not enough to say to a patient, "How do you feel?" It's useless… if his family physician were to phone him or email him and say, "How are you doing?" he'd say, "Fine." He just wouldn't be around, that's all.

Discussion
In this qualitative study of older adults with heart failure and their family caregivers four challenges of end-of-life discussions were identified: trivializing illness-related challenges, positivity in late life, discomfort in having end-of-life conversations, and reluctant to engage despite need. Uniquely, these challenges were identified within the context of participants' understanding of illness, which is critical, as it sets the foundation for individuals to perceive a need to engage in end-of-life discussions. Our findings corroborate the work of previous research on the experiences and needs of people with serious illness at the end of life. In a qualitative study of older adults with heart failure, Klindtworth and colleagues reported that participants downplayed their symptoms and the challenges of illness as a process of adapting to living with a serious illness [30]. In a longitudinal qualitative study of 828 interviews with patients with serious illness and their caregivers, Kendall and colleagues also found that participants were reluctant to discuss the end of life, preferred to remain positive, and adapted to new norms as they progressed in illness and faced disabling symptoms [31]. These intertwining challenges point to the need for alternative approaches for communicating with older adults and caregivers. The struggles that older adults and caregivers regularly face in their home environments can be indicative of declines in health and one's ability to care for themselves. Our study found that caregivers appear to be in distress over managing the challenges of decline, and yet, may be initially reluctant to engage in end-of-life discussions. This underscores the need for more effective communication to elicit the struggles that remain unknown to clinicians during a period of decline. Effective communication may enable clinicians to coordinate services that older adults and families' can benefit from such as home care, social care services and palliative care. This is important in order to improve support for caregivers of people with advanced heart failure, which has recently been identified as a research priority [32]. Regardless of one's discipline, providing clinicians with training on effective communication in serious illness is one way to integrate end-of-life discussions earlier in the illness trajectory. Communication is a fundamental competency in medicine, and yet, clinicians repeatedly report feeling underprepared to initiate or have end-of-life discussions [33][34][35]. In the United States, changes to the undergraduate curricula have led all medical education programs to include provision of at least some material on death and dying. However, less than a quarter of programs offer separate courses on end-of-life care [36]. With an increasing aging population, it ought to be necessary for medical education programs to include rotations in palliative care training for medical students to gain exposure and clinical opportunities to interact with patients and families in real-world end-of-life situations [34]. In fact, desire for such opportunities has been cited in previous research [34,37]. In addition, interventions to improve communication for both clinicians and patients are being tested. In the Serious Illness Care Program, Lakin and colleagues found that communication skills training for primary care clinicians led to improved documentation and the comprehensiveness of goals-ofcare discussions [38]. A patient activation intervention that empowers patients with heart failure to initiate goals-of-care discussions has also been found to be feasible and beneficial to both patients and clinicians [39]. Dougherty and colleagues found that patients viewed the over-the-phone coaching and the pre-visit patient activation outline to be facilitators of goals-of-care discussions. However, these kinds of interventions are inconsistently available to people with serious illness in most healthcare systems. Hence, there remains a great need to increase capacity to introduce end-of-life communication earlier for people with serious illness.
The patient-and family-related challenges of integrating end-of-life communication are not limited to the four that are identified in this study. Moreover, the identified challenges intertwine with other barriers posed by the healthcare system and present clinicians with a moral dilemma. Specifically, clinicians have a duty to share information and present services that could benefit patients and caregivers such as end-of-life discussions that have been found to improve patient and caregiver reported outcomes [40]. However, clinicians ought to respect individual preferences to wait to engage in end-oflife discussions. Our findings suggest that although patients and caregivers may seem hesitant to discuss endof-life care, they have needs and concerns that are not being expressed. In a qualitative study of patients' perspectives on physician behavior during end-of-life discussions, Abdul-Razzak and colleagues found that assessing a patient's readiness to engage may be important before initiating end-of-life discussions [41]. As assessing readiness can be a challenge for physicians, they found that patients might prefer to be directly asked to have end-of-life discussions. Together, these findings stress the importance of integrating end-of-life communication as an iterative process, particularly in a highly variable illness such as heart failure, to ensure that patients and caregivers are provided with opportunities to discuss concerns and challenges at different points along the illness trajectory.

Limitations
The findings should be interpreted in the context of several limitations. First, the clinical nature of heart failure presents unique challenges for end-of-life communication. However, all patients had multiple chronic conditions and it is uncertain to what extent this study captured how the complexity of multimorbidity affected engaging in end-of-life communication. Nonetheless, many older adults with heart failure have multiple chronic conditions, and therefore, our sample well represents the older adult population. Second, in any qualitative study it can never be determined the extent to which participants were willing to share their thoughts and insights. This is potentially of greater relevance given the sensitivity and discomfort that surrounds endof-life discussions. However, it is also possible that participants felt comfortable confiding in a situation where their identity would remain anonymous which appeared to be the case as participants shared their concerns. It is also possible that the presence of caregivers in dyad interviews could have influenced the data, which was unaccounted for in our study. We found dyad interviews to generate rich discussions between patients and caregivers, however, this study cannot speak to how the data would differ had all interviews been solo interviews. Third, patients and family members were recruited from one outpatient heart failure clinic. This clinic also operates as a specialized heart failure clinic, which in our study setting means that patients are cared for by an interdisciplinary team of clinicians with training in heart failure management, which reflects the in-depth knowledge of heart failure management that participants had. This limits the transferability of our findings as not all patients may reflect the sample recruited in this study and the care setting. Lastly, participants in this study were predominantly Caucasian, and therefore, largely reflect the perceptions of people from one ethnic background.

Conclusion
This study identified four challenges of engaging in endof-life communication from the perspective of older adults with advanced heart failure and family caregivers. These findings provide useful clinical insights that can inform approaches to integrate end-of-life communication into serious illness care. The findings highlight a need for end-of-life communication to begin earlier in the course of illness to better support patients and caregivers during a period of decline as well as for communication to occur iteratively for those who may not be ready to engage. Future research is needed to explore and test alternative communication approaches that effectively unravel the challenges of illness among older adults with serious illness and their caregivers earlier in the illness trajectory.
Additional file 1. Sample interview questions