End‐of‐life priorities of older adults with terminal illness and caregivers: A qualitative consultation

Abstract Background As older adults approach the end‐of‐life (EOL), many are faced with complex decisions including whether to use medical advances to prolong life. Limited information exists on the priorities of older adults at the EOL. Objective This study aimed to explore patient and family experiences and identify factors deemed important to quality EOL care. Method A descriptive qualitative study involving three focus group discussions (n = 18) and six in‐depth interviews with older adults suffering from either a terminal condition and/or caregivers were conducted in NSW, Australia. Data were analysed thematically. Results Seven major themes were identified as follows: quality as a priority, sense of control, life on hold, need for health system support, being at home, talking about death and competent and caring health professionals. An underpinning priority throughout the seven themes was knowing and adhering to patient's wishes. Conclusion Our study highlights that to better adhere to EOL patient's wishes a reorganization of care needs is required. The readiness of the health system to cater for this expectation is questionable as real choices may not be available in acute hospital settings. With an ageing population, a reorganization of care which influences the way we manage terminal patients is required.


| INTRODUC TI ON
Increased use of emergency services and hospitalizations among older people who are dying 1 often includes intensive procedures 2 that can prolong suffering and are too late to be of benefit. 3,4 Evidence that patients or families has been consulted regarding their preferences for future care and how this consultation has occurred is limited but critical to the provision of appropriate end-of-life (EOL) care. 5 Older patients and their families are usually provided with information about hospital-based treatment options 6,7 regardless of whether they wish to spend their last days in an acute care hospital. 8,9 You et al 10 report that patients and families lack understanding of the implications of life-sustaining treatments, with Wilmott and colleagues 11 finding that a patient's substitute decision makers do not always act in the patient's best interest. As a result, a patient's wishes may not be known or honoured. Additional factors that can further complicate EOL decision making are as follows: the low public awareness 12 ; cultural values affecting care preferences at the EOL; 13,14 family denial of the patient's prognosis; 15 potential cognitive impairment in old age; 16 conflicting family pressures; 17,18 and the level of a doctor's professional expertise in communicating a terminal prognosis sensitively. 19 Health professionals providing quality EOL care across all health services must have an understanding of the family and patient's perception of what is appropriate and contributes to high-quality care, and what constitutes a "good death." [20][21][22] This information is necessary to fully inform clinical and other support staff providing EOL health services. Existing data that demonstrate the use of medically inappropriate treatments at the end-of-life and the importance of engaging in advance care planning may assist to inform more honest end-of-life discussion. 23,24 Research to date on EOL care has been predominantly conducted in the cancer realm. 25 With more people dying from diseases of ageing, 26 this research, although informative, does not take into consideration the EOL trajectory of other terminal conditions. The Australian government recognizes the importance of providing high-quality EOL care 27 and developing guiding principles and essential elements for the provision of safe and highquality EOL care. 28 Despite the Australian government's support for EOL care, a recent Australian-based study found that only fourteen per cent of non-cancer patients in the last year of life, with irreversible conditions which were considered amenable for palliative care, received specialist palliative care compared to more than two-thirds of cancer patients. 29 These non-cancer patient conditions included the following: heart, renal and liver failure; COPD; HIV/AIDS; dementia; and Motor neurone, Parkinson's and Huntington's disease.
This study aimed to determine older terminally ill patients and caregivers' priorities, perceptions and appropriateness in EOL care.

| Sample and data collection
All of the consumer EOL advisory group members were invited to participate in Focus Group Discussions (FGD) or In-depth Interviews (IDI). A total of 24 (65%) agreed to participate. Ninety-minute FGDs were conducted on the same day in April 2016 and 90-minute (IDI) were undertaken during the month of June 2016. All FGDs comprised both terminally ill patients and caregivers in each group. One member of the study team (EL) conducted all IDI with patients being at home, which were necessary to capture perspectives of those who were unable to participate in FGD. This was due to geographical difficulty, as 30% of the Australian population reside outside major cities 30 or due to the participant's poor physical health from their terminal condition which impacted on their mobility and transport capacity. Members of the study team developed the FGD guide which included four main topics EOL care, quality of life factors, family impact and health-care provision (Appendix S1). The IDI guide that reflected the FGD topic was also generated for those who lived away from the city where the study was conducted or for those volunteers who were too ill to attend the focus groups. Written consent was obtained from each participant. Three study team members with either a psychology (RH) or a nursing background (EL, LH), who were trained in qualitative methods, facilitated the three 90-minute FGDs in a private meeting room on a University campus. The facilitator guided participants through each of the topics. The FGDs were audio-recorded and written notes were also taken by the team members. One team member (EL) was present for all FGDs and listened to the transcripts from all FGDs to ensure consistency. EL conducted all IDI using the agreed guide.

| Data analysis
The audio-recorded qualitative data were transcribed verbatim and managed using NVivo software (version 11 QSR, International Pty Melbourne, Victoria, Australia).
Thematic content analysis was used to elicit themes from participants regarding whether quality or length of life was most valued, attributes of good quality living, the effect of EOL involvement on caregivers and their experiences and expectations of health professionals providing this care. Two team members (RH, EL) undertook the thematic analysis, 31 initially repeatedly reading the transcripts and then labelling the text in the NVivo software. Each team member then independently grouped labels into related themes around consistent or divergent issues arising. These researchers held iterative discussions in which they reflected on the emerging themes 32 and then refined the themes into a final set of agreed categories.
A third researcher (LH) who observed the FGD independently reviewed the categories and themes within these for face validity.

| RE SULTS
The final sample included 24 participants, 17 females and 7 males.   Figure 1 illustrates these themes and the factors within each theme that were assessed by the participants as being a priority for optimal EOL care. Each theme is also described below with supporting quotes.  Carer participants described EOL care as emotionally difficult.

| Putting family life on hold
Feelings of guilt, denial, distress and sadness were noted in addition to the physical and logistical challenges of caring for a loved one.
Although caregiver participants sought to adhere to patients' preferences regarding not prolonging their life, conflicting feelings occurred as participant caregivers also did not want to lose their loved one.
I think I've had both those feelings; wanting them to hang around, but also wishing them a speedy goodbye for their sake.

| Being at home (at EOL)
Participants agreed that it was generally preferable to stay at home for as long as possible. Consultation with health providers and choice regarding the location of their treatment and care was identified as important.
Dad was in palliative care and basically he didn't want to be there at all…one day mum went in and he said take me home….So he was home for a week and then passed away at home, but at least he met his wishes.
(Male, Caregiver) The hospital is an alternative, I would never say the home is an alternative… the hospital has to be the last place on earth in any country that you would need to go simply to die.
(Female, Caregiver) Yet caregiver participants identified cases in which this was not possible, such as when the person or their caregiver did not have the ability to provide the care required. In these cases, carer participants often reported a need for greater support from other social or community-based services to facilitate care at home.
We knew we couldn't deal with it at home ourselves, but had there been other kinds of care, that would have been perfect. He would have still been in his own garden and done his own little pottering around the place as he always did (Female, Caregiver) She couldn't cope at home and she's gone into an old age home and that's much better all-round than trying to cope at home.

| Talking about death
Openness about impending death, honesty and transparency be-

| Competent and caring health professionals
Participants identified that a consultative, patient-centred care approach was critical. They stated that health professionals, who were compassionate, respectful, and ensured patient dignity, played a significant role in providing quality EOL care to patients and caregivers.
…Good relationships with the primary health team is what I think is absolutely essential…. …A good relationship someone who understands you and understands the family and who will work with other professionals….
(Female, Motor Neuron Disease) She came in to find her with an oxygen mask on and we had said no resuscitation. The doctor said something like she won't need that anymore and walked out the door. That was how my sister discovered mum had died.
(Female, Caregiver) to enhance the relevance of health services research. 34 The information collected in our consultation covered recent experiences in the health system and home settings and is of relevance for clinicians and health service planners. IDI supplemented the FGD findings with extensive details from less physically mobile health service consumers.

| D ISCUSS I ON
A possible limitation of our study is that the majority of participants were females and caregivers. However, as females are often the informal caregivers of chronically ill patients 36 , this may in fact be representative of the reality of informal caregivers. The fact that we only conducted three FGD could also be considered as a limitation; however, saturation was rapidly achieved even with three FGD. While our study confirmed that consulting patients and families about this sensitive topic is feasible in Australia, this consultation did not happen at a time of acute medical crisis. It could be argued that our study did not take into account patient and family preferences at those critical times, as studies have shown that preferences can change over time depending on a person's state of health. 59 However, we believe the views our participants are further enriched by the ability for retrospection without the influence of an acute emotion.

| CON CLUS IONS
This consultation identified priorities and preferences by consumers through their experiences of the delivery of EOL care. It con-

I N FO R M E D CO N S E NT
Informed consent was obtained from all individual participants included in the study.

ACK N OWLED G EM ENTS
We thank members of the consumer advisory group for their insightful contributions to this research. Mr Chhorvoin Om provided initial assistance with concurrent thematic interpretation of data from one of the focus groups.

CO N FLI C T O F I NTE R E S T
The authors declare they have no conflict of interest.