The facilitation of a PHD is a rather distinct study phenomenon within the home care nursing setting. The ethnographic objective of unveiling the shared aspects of culture, including values, beliefs, knowledge, skills, and actions, is well-suited to this context [27]. Using a wider range of research methods, specifically observational methods, enables a more detailed investigation of HCNs’ clinical practice [28], and we considered Roper and Shapira’s qualitative framework on focused ethnography design to be suitable [29]. This design enabled us to explore HCNs’ bedside practice with a focus on the facilitation of PHD in these private and sensitive situations [28]. The fieldwork involved short-term and targeted data collection sessions with participant observations and individual interviews with HCNs. Thus the participant observations and individual interviews with carefully selected participants could be structured around the study topic [30]. The first author and co-authors strived for flexibility, adaptability, and rigor in the data collection [27]. We employed a data triangulation of participant observations, semi-structured interviews, and analysis of the registrations forms to gather rich data and thus better understand the complex experiences of PHD from the HCNs` perspective. This resulted in valuable insights not generally accessible through standard research methods such as solely retrospective qualitative accounts.

Setting

The study setting was eight different home care services in two municipalities (50,000–80,000 inhabitants) consisting of both rural and urban areas in northern Norway. Norway is a vast country, with rural areas having sparse and intermittent home care services and long travel distances. The Norwegian healthcare system is based on the principles of public welfare and provides free and universal healthcare to all. Municipalities are required to provide free home nursing services to their inhabitants [31], including basic palliative care integrated into general patient care [32]. National policies and government health initiatives have put an increased emphasis on providing palliative care in primary health care closer to the patient`s home [7]. These changes can be attributed to the introduction of the Norwegian Coordination Reform (2012) [33]. This reform has placed a considerable responsibility on the municipalities as they now play a prominent part in meeting the growing demand for health services [33]. Within the municipalities, HCNs provide home care nursing comparable to the role of ‘community nurses’ in Australia and Canada and ‘district nurses’ in the UK [18]. For clarity and consistency, this study will use the term ‘home care nurse’ (HCN).

Participants

HCNs employed in home care areas and who had experience with facilitating PHD were eligible for inclusion. We used purposive sampling to recruit HCNs from the home care services in order to ensure specific knowledge and experience in facilitating PHDs. In co-operation with the nurse mangers in the home care service, the first author identified eligible HCNs that met the inclusion criteria. They were contacted by the managers inviting them to participate, and those who expressed interest in participating were approached by the first author with information and a formal request to participate in the study. Included HCNs were asked if they knew other HCNs with experience in facilitating PHD, thus acting as referral agents. Potential cases of PHDs to observe were identified by an ambulatory cancer nurse who was a knowledgeable insider and who acted as a gatekeeper when mapping appropriate cases for participatory observations. She informed the patients and next of kin (NOK) about the project and helped to establish whether the researcher`s presence was acceptable.

Data collection

Data collection took place from March 2019 to March 2020. In accordance with the ethnographic approach with the goal to study how HCNs think and act in their natural setting, we used individual semi-structured interviews and participant observations. In addition, we registered the number of PHDs in the whole municipality during the period.

Participant observation

The first author observed HCNs when they facilitated PHDs for dying patients. During the participant observations, we gained deeper insights into what HCNs do when caring for dying persons and their NOK. Each observation session included the time HCNs spent preparing for the visits at HCN offices, time spent traveling to and from patient’s homes, and time spent during visits. Due to the sensitive nature of the topic being investigated, the researcher did not take notes in the patient’s homes. However, to avoid memory bias observations, reflective notes, and verbatim quotes were noted immediately after home visits. These were written in accordance with an observation guide inspired by Spradley (1980) [34]. The guide acted as a reminder of what to pay attention to during the observations, such as the physical place, the people involved, the norms of the group, the activities they carried out, and the emotions they felt and expressed. During the observations in the patient’s home, the researcher sat down with the HCN, patient, and NOK when invited. When HCNs carried out clinical tasks, such as managing analgesics or changing a urinary catheter, the researcher adopted an observer role without being invasive or obtrusive. The purpose was to observe and understand from the view of the participants rather than add personal interpretations and explanations.

Interviews

Interviews are important in focused ethnography due to the constricted time frame of fieldwork [35]. We employed semi-structured interviews with a set agenda, yet with the flexibility to pursue the expressed ideas of the participants [36]. The interviews were used to collect data on what we could not or had not observed, such as attitudes and emotions. The interviews helped validate the observations and thus provided directions for future observations, as well as observations giving directions for subjects requiring further explanation in the interviews. The interview guide (additional file A) was developed for this study by the first author in collaboration with the co-authors based on the literature review and the study aims. The guide included open questions such as “Can you describe how you facilitate a planned home death” and “What challenges do you face when facilitating planned home deaths”. When appropriate, follow-up questions were asked, moving from the general to the more specific by probing (e.g., “could you tell me more about that” or “what do you mean when you say that”. The interviews took place at a suitable location at the HCNs’ places of work. The interviews took an average of 50 min and were audio recorded and transcribed verbatim. The interview guide was fine-tuned over the course of the study as observations revealed issues we wanted to explore further. The interviews as a whole sought to encourage participants to articulate their experiences, beliefs, practices and attitudes in order to identify practice patterns that are characteristic of this sub-culture through active listening and attention to non-verbal signs and by showing empathy [34]. Although the participants were offered the opportunity to read the transcribed material, they all declined.

Registrations of the number of home deaths

The registration form was reviewed and accepted for use by a knowledgeable insider. The documentation was completed by an ambulatory cancer nurse employed by the home care service, responsible for monitoring all PHDs within the municipality throughout the data collection year. The information gathered from these records was utilized to determine the count of PHDs that were successfully provided by the home care services, along with the tally of requested PHDs that were not fulfilled, along with an explanation for the unmet requests. If the patient did not die at home, even though it was an expressed wish, the alternatives for boxes to check why included insufficient personnel, lack of competence, lack of equipment, changes in the patient’s wishes, the NOK`s situation, the need for advanced symptom management, and others. No identifying patient information was registered.

Data analysis

Field notes and transcribed interviews were read and reread to get an overview of the data. The transcribed interviews were analyzed first, followed by a separate analysis of the field notes. Then all the data, including data from the registration forms, were analyzed together to identify similarities or discrepancies. We used NVivo 11 (QRS International, 2017) to organize the data. The initial analysis was conducted by the first author and discussed and developed in collaboration with the co-authors. Roper and Shapira`s framework (2000) was employed using the following five steps: coding for descriptive labels, sorting to identify patterns, identifying outliers, generalizing constructs and theories, and taking notes of reflections and insights made by the researcher [29].

This is a circular process, not linear or chronological. However, to describe the process we present it here as continuous steps. Answering the study`s aim, related and similar segments of text were grouped and coded to form meaningful categories we called descriptive labels. The length of the data segments in the descriptive labels were reduced in size without losing content in order to form manageable units. We used in vivo coding when applicable. We started with broad codes, such as ‘next of kin’, that were later broken into smaller concepts, i.e., ‘feeling of concern for next of kin’. We clustered these labels in order to identify patterns and themes, visually separating them into general and abstract themes that could explain relationships in the cultural setting of the home care services. We then rearranged the labels in order to identify patterns, such as similarities or differences in how HCNs performed or reflected on PHD, in their cultural setting. These similar patterns formed categories, which were of manageable sizes. From the sorted categories, we identified themes and drew connections between the emic perspectives of the participants and the etic interpretations of the researchers in order to construct theoretical understandings that included both perspectives. We formulated and defined overall themes that included relevant categories. Themes are defined as larger sociological categories, group behaviors, shared opinions, etc. [37]. These were defined again and compared with compatible existing constructs, theories, literature, and research in line with the analytical framework [29].

In every step of the analysis we were aware of the potential for outliers, namely statements or actions that do not ‘fit’ with the rest of the data [29]. We identified two statements as outliers from the interviews. These were followed up and examined to see if they were extraordinary responses, i.e., true outliers, or if they were common traits shared with other HCNs.

Note-taking was done continuously throughout all phases of the data collection and the analysis to record first impressions, identify preconceived notions, and support the audit trail. The credibility of the coding, categories, and identification of the outliers was checked for accuracy by the team of authors.

We translated the findings and quotations from Norwegian into English. We strived to correctly translate the participants’ expressed views and descriptions. Verbatim quotations are used to exemplify the themes and represent the participants` understanding as accurately as possible in order to give a better understanding than the researcher’s paraphrasing could.

Ethical considerations

Because the researcher could inadvertently gain information regarding other patients not relevant to the study when observing HCNs’ work, an exemption from the duty of confidentiality was obtained from Regional Committee for Medical and Health Research Ethics in Northern Norway (REC) (2019/605). According to Norwegian law, no further ethical approval was required. All collected data were anonymized and kept secure and confidential at all times. This study was performed according to Norwegian Centre for Research Data (NSD) (77,356) regulations (now known as the Norwegian Agency for Shared Services in Education and Research). The HCNs were informed verbally and were provided with a participant information sheet and gave their written informed consent to participate in the interviews and observations. All patients and NOKs who were approached agreed to participate in the study. All patients and NOKs were informed about the study, and written consent was collected that permitted the researcher to observe the work of HCNs in their homes.

We were unable to gain knowledge about HCN’s facilitation of home death without close contact with dying patients and their NOK. Therefore sensitive issues and ethical concerns were carefully considered. We reflected on the potential to intrude on an intimate and private experience and how to address any distress we may inadvertently cause. Guided by Sivell and colleagues (2019), we used sensitive and open questioning, researcher self-disclosure, awareness of power relationships, and adaptation to the individual needs of the patients on a case-by-case basis [38].

Findings

Twenty-one HCNs were approached with requests to participate in the study, and one declined participation without providing a reason. Twenty HCNs were interviewed, 17 of whom were women, and the HCNs were 25–63 (mean 42) years old with 1–24 (mean 12) years of experience. Their experience in facilitating PHDs ranged from two cases to more than ten cases (Table 1). Three of these were also observed when they were facilitating PHDs. They were selected because they were on duty when the patients had scheduled visits. Of the nine home care service areas contacted, one did not have any HCNs that met the inclusion criteria at the time of study recruitment. One interview was interrupted after 30 min due to a work-related emergency.

The researcher followed three different HCNs facilitating PHD for two dying patients and their NOK. This amounted to seven different observations, including four with the HCN, NOK, and the patient, two with the HCN and NOK, and one with just the HCN. These observations included visits to the patients’ homes, bereavement conversations, and one session with an HCN in a car, observing her getting supplies for a PHD and making phone calls about the PHD. From the registration forms, we counted eight patients wanting to die in their own homes. Five of those died at home. The reasons for not dying at home were in all three cases related to the NOK’s situation. In total, participatory observations amounted to 8.5 h. The researcher took on different roles during the observations. At times, like during medical interventions, she assumed the role of a silent observer when the HCNs were focused on clinical tasks. On other occasions, it was natural for her to engage in conversations with the patient and/or their NOK, for example, when they offered coffee and biscuits.

Table 1 Characteristics of the study participants

The analytical process led to the identification of three main themes that characterized HCNs’ experiences of facilitating PHDs, namely going above and beyond, juggling the patient’s and NOK’s needs, and demanding work We present them here separately but recognize that the themes are interrelated and overlapping. HCNs find the work of facilitating PHDs rewarding, and it is important to them to guarantee availability and flexibility for patients and NOK. However, they reported challenges when facilitating PHDs in the forms of staff shortages and high workloads. In addition, they expressed concern for the well-being of NOK as they juggle the needs of patients and the needs of NOKs. The three themes are presented in the following sections.

Going above and beyond

HCNs in this study clearly expressed positive attitudes about facilitating PHDs and emphasized that this was rewarding work. The HCNs commented on the value of facilitating PHDs numerous times, and they used words like rewarding, meaningful, and purposeful and said that they took pride in it. One nurse said:

‘You are alone most of the time in the home care service. I found it really scary the first time. But at the same time it was nice to see…it was her greatest and last wish, to die at home. It was really nice making that happen for her. Even though it was hard for me, it was fulfilling to see that we made her wish come true.’ (HCN 18).

HCNs stressed how they did everything within their power to give patients and NOK the best possible care. Observations also revealed how HCNs repeatedly told patients and NOK to call the HCN if they required assistance without specifying which tasks they referred to. An HCN explained it like this in an interview:

‘We don’t usually give away the on-call number, but in the case of PHD we give it to both the patient and NOK so they can easily contact us, whenever and for whatever reason.’ (HCN11).

The HCNs expressed going above and beyond, and some of the HCNs from the more rural home care areas demonstrated a collective drive or an ethical duty to provide PHD as an option by using phrases like ‘We do everything in our power’ and ‘We want it to happen’. One HCN stated in the car after a house call:

‘Like I said, we do everything we can and more to make this happen.’ (Observation 5).

Several HCNs stated that they would go far to aid patients and NOK, and some of them mentioned how they had rescheduled or cancelled their planned vacations to ensure nurse coverage during a summer when a patient in their home care area wanted to die at home. Another HCN explained how she had put a folding cot in her office so she could spend the night there and thus have a shorter drive in case the dying patient and their NOK needed anything. This was the first of the two outliers identified in the data and was used to confirm the notion of HCNs going out of their way to accommodate a patient’s wish to die at home.

On a field visit to a dying patient’s home one Monday, the patient and NOK explained to the HCN that the patient had been briefly admitted to an institution during the weekend due to exacerbation of their cancer pain. The HCN asked why they had not called her during the weekend. The spouse of the patient replied that they knew she was at her cabin, given she had the weekend off. The HCN then said: ‘You should have called me anyway. I am always available for you.’ (Observation 1).

Another HCN commented in the interview:

‘We go above and beyond. Every time. We want it to happen as much as they do. After all, it is somebody else`s life, or in fact somebody else’s death. You only get one chance.’ (HCN12).

Juggling the patient’s and NOK’s needs

The dynamics between the NOK, the patient, and the HCNs emerged as an important issue due to conflicting needs. According to HCNs, NOK are often in danger of being exhausted, worn out, and scared. Their needs were highlighted as equally important as the patients’ needs but more challenging to care for. Their needs were described as not homogenous, nor was the way they coped with the situation. When asked ‘What do you consider important when facilitating for planned home deaths’, one HCN replied:

‘We don’t have enough focus on family rotations. They don’t get it. I don’t think it is explained in detail. The family has to work shifts to make it happen.’ (HCN2).

The HCNs in our study pointed out how on several occasions the NOK were an essential factor in facilitating PHDs and how PHDs could not be facilitated without them. There were similar perceptions among the HCNs that PHDs were not possible without NOKs being present and willing to take on the majority of the day-to-day care.

'What I see is the next of kin. A lot of weight is put on them. It’s useless if…Honestly, it is not enough with just a spouse.'

(HCN 13)

The participants expressed that facilitating PHDs for patients who had young children was a challenge. To see their emotional struggles took a toll on the HCNs, and several of them reflected on such cases:

‘It ended up being challenging for the children…the last couple of days. That’s why XX went to the hospital and died there. It can seem ok in the beginning. Like, ok, we’re all together as a family…but, I think it gets… The children cried a lot and didn’t want to go to school, and it just got too complicated at home. I would not do that again.’ (HCN2).

Much of the work the HCNs did was organized around the medical needs of the patients, but they focused a lot of attention on the NOK. Meeting their needs and listening to their worries and comforting them were sometimes the biggest issues. HCNs normally plan their shift according to a worklist; a detailed list of patients to visit, their care needs and estimated time usage. However, visits to PHDs were not timed like other visits, and the HCNs seemed to exhibit more flexibility and discretion in time management when visiting dying patients. The HCNs never looked at their phones or appeared to be busy or rushed when meeting with patients or NOKs. However, after leaving the dying patient and NOK, the HCNs explained that they were in a hurry, and now had other patients waiting for them.

NOK were perceived as essential when facilitating PHD and as making a substantial contribution to the care of the dying. However, barriers to PHDs lie in the workload put on NOK when patients die at home. The NOK could be so exhausted that they do not feel safe with the patients at home, even though it is the expressed wishes of the patient.

‘Quite often it feels like what you do is primarily for the next of kin, but that’s just the way it is.’ (HCN3).

Two different HCNs both described a fear of NOK getting flashbacks after a period of palliative care in the home:

‘I find next of kin the most challenging factor. Because they are so different, some are fine with things and handle it well, but others don’t. They are so important in making this happen, and we must make sure that we don’t get it wrong, and they end up having terrible flashbacks later on.’ (HCN11).

‘I think I would say that next of kin are a barrier for planned home deaths. They are so important, and you must make sure that it does not get overwhelming. That they feel that they have to do this because it is expected and then it backfires later, and they get painful flashbacks.’ (HCN9).

This can imply that HCNs work within a tension that exists in the home, where different needs are not being met. The following excerpt from the field notes made on a visit to a dying patient and NOK further illustrates this point:

‘The nurse (HCN6) asks the patient if more frequent visits from the HCNs could be an option, particularly with regards to assistance when showering. The patient replies (annoyed?) it is not necessary yet. That they don’t want more people coming into their home. The spouse stares down in the table. The nurse looks closely at her and asks directly but gently, “Do you need some assistance, some relief?” The spouse tears up but does not look up.’

(Observation 2)

These excerpts demonstrate some of the challenges faced by HCNs when trying to balance aiding the NOK and respecting the patient’s wishes, all while clinically assessing the patient’s needs. None of the HCNs performed a systematic assessment of the NOK’s needs or wellbeing, but all HCNs expressed a concern for their unmet needs. The HCNs were attentive and supportive by listening to their grievances. HCNs were observed trying to provide reassurance and to console the NOK before, during, and after the death of their loved ones. As seen in the field notes, some NOK were unable to reconcile with reneging on the promise of caring at home until death, even though the situation demanded a hospital admission, as the following demonstrates:

‘During a bereavement follow up, the HCN (HCN4) asks the widow if she is troubled by a bad conscience for the patient dying in xxx (an institution). The widow whispers: “It tears me apart”. The nurse replies: “But you were so tired and worried in the end. You couldn’t have done it!”…. “I should have stuck it out”, the widow replies.’ (Observation 7).

The HCNs explained how the patient’s spouse or NOK often needed a space to talk freely without the patient being present. These talks often took place on the porch when they escorted the HCNs out of the home after visits. As described in the fieldnotes:

‘The spouse leads us out in the hallway while the patient remains in the living room as we say goodbye and get our coats on. She follows us outside on the steps in the cold weather and tells us quietly, almost whispers that she does not get any breaks and that she is very, very tired. She starts crying silently, clearly not wanting the patient to hear or see her like this. The nurse pats her on the back, asking if there are any other relatives that can be with the patient while she gets a break.’ (Observation 3).

Demanding work

All HCNs stated that even though the work was rewarding, it took its toll on them, both emotionally and physically. The reasons for this were blurred work boundaries, and they all repeated numerous times that they would go a long way to support patients and their NOK. They described their work as being a game of solitaire, where the challenge was to make it all add up because even though they had an ongoing PHD they still had 300 other patients requiring visits and care. They explained how PHDs require extensive coordination as well as correspondence with other professions and institutions such as the general physician, the palliative team at the local hospital, and the pharmacy. Most of these calls and follow ups were performed in the car on their way to and from patients’ homes and between other visits. Some HCNs reported personal health challenges related to their work, and four HCNs stated in interviews that they have had periods of sick leave as a direct consequence of facilitating PHDs that were especially strenuous and time-consuming. They talked about other HCNs making an advent calendar for dying patients’ children and questioning where the limits to their jobs were supposed to be drawn, and who was responsible for drawing those lines. HCNs reported that the patients and NOK are satisfied with, and even grateful for, the care provided. However, four of the HCNs in this study asked themselves, ‘At what cost?’ (HCN3, HCN7, HCN15, HCN17).

‘It is a horrible feeling. Doing everything you can, but it is never enough.’ (HCN7).

All HCNs in this study stated that there were not enough nurses in the workforce, and they called for more qualified and knowledgeable staff. The shortage hindered them from doing a good job and kept them from suggesting to eligible patients that dying at home might be an option for them. They slot palliative patients in where they can, while at the same time making sure the dying patient and NOK experience prioritization, availability, and flexibility by, for example, not registering time usage during visits. Many of them spoke of the constant stressor of being understaffed and having a huge workload, and this became worse when there were PHDs to facilitate. One HCN stated:

‘We are stretched pretty thin as is.’ (HCN4).

They talked at length about the frustrations of having to manage the care of patients on their worklist while being flexible and accessible to dying patients and their NOK’s individual needs. To fulfill these requirements within a given timeframe was reported as challenging, and the understaffing in combination with the high workload created a void where they felt they made compromises on the quality of care. They implied that continuity of care could only be achieved when every patient received the right kind of care at the right time.

HCN8: ‘There have been times where we advised against PHDs because we didn’t have the required competence and we didn’t have enough nurses to the extent that we had to be honest and say, we want to but it is not possible…’.

Interviewer: ‘Is that hard? Saying no to patients?’

HCN8: “Yes, very hard. Very hard. We took it personally, and it felt like a failure in that good palliative notion, that good palliative environment and attitude, the hospice philosophy we are trained in, and our core values. But we had to restrict ourselves and analyze the situation, and if you disagree and have objections, take it up with the person in charge.”

Another HCN put it like this:

‘We have become extremely mentally worn out.’ (HCN9).

One HCN in an urban area was skeptical to the notion of facilitating PHDs and stated in the interview that they did not have any PHDs in her area in recent years. She explained how it would not be justifiable due to the lack of nurses. This was identified as the second outlier in that it did not fit with the patterns identified in the rest of the data.

Evidently, HCNs’ feelings for PHD were both positive and negative. Most HCNs expressed a solid commitment to the patient’s and NOK`s needs. However, many of the difficulties experienced by the HCNs were described as organizational challenges. They explained how municipalities have failed to prioritize home care to the extent necessary to ensure adequate nursing coverage and how additional qualified staff and a reduction in workload could provide coordinated, continuous, and appropriate care.