Perspectives of nursing home executives on collaboration with GPs and specialist palliative care teams: A qualitative interview study.

Abstract:Background: Nursing home (NH) staff, general practitioners (GPs) and specialist outpatient palliative care teams are expected to cooperate to ensure adequate palliative care for NH residents in Germany. Aim: The aim of this study was to investigate the perspective of NH executives concerning collaboration with GPs and specialist outpatient palliative care teams. Methods: We conducted semi-structured telephone interviews with executives of NHs in the federal state of Baden-Wuerttemberg, Germany. Interviews were analysed by means of structured content analyses. Results: Executives of 20 NHs participated in the study, eight NHs cooperate with specialist outpatient palliative care teams. Content analysis resulted in two main categories: 'general palliative care by primary carers' and 'collaboration with SAPV in NHs', each with three first-order subcategories. The main barriers to adequate palliative care were reported to be lack of palliative care knowledge in GPs and NH staff, refusal of some GPs to cooperate with specialist outpatient palliative care teams and staff shortage in NHs. Specialist palliative care involvement was described to result in improved palliative care. Conclusion: Solutions seem obvious, e.g., further education in palliative care or round tables to discuss collaboration. However, studies show that even comprehensive educational and management interventions to implement palliative care do not always result in long-term effects and further research is needed.

Zusammenfassung: Hintergrund: Mitarbeitende der Pflegeheime, Hausärzt_innen und Teams der spezialisierten ambulanten Palliativversorgung (SAPV) sollten zusammenarbeiten, um die Palliativversorgung für Pflegeheimbewohner sicherzustellen. Ziel: Ziel dieser Studie war es, die Perspektive von Leitungspersonal in Pflegeheimen auf die Zusammenarbeit mit Hausärzt_innen und SAPV zu untersuchen. Methoden: Es wurden halbstrukturierte Telefoninterviews mit leitendem Personal von Pflegeheimen in Baden-Wuerttemberg durchgeführt und mittels inhaltlich strukturierender qualitativer Inhaltsanalyse ausgewertet. Ergebnisse: Insgesamt haben sich Leitende aus 20 Heimen beteiligt, acht davon kooperieren mit der SAPV. Im Rahmen der Inhaltsanalyse wurden zwei Hauptkategorien gebildet „Allgemeine Palliativversorgung durch Primärversorger" und „Zusammenarbeit mit SAPV in Pflegeheimen" und diesen wurden jeweils drei Subkategorien erster Ordnung zugeordnet. Als größte Hindernisse für eine angemessene allgemeine Palliativversorgung wurden genannt: Mangelndes Wissen der Hausärzt_innen und des Pflegeheimpersonals, Weigerung einiger Hausärzt_innen mit der SAPV zusammenzuarbeiten, und Personalmangel in Pflegeheimen. Aus Pflegeheimen, die mit der SAPV zusammenarbeiten, wird berichtet, dass der Einbezug zu einer verbesserten Palliativversorgung führt. Schlussfolgerungen: Es gibt naheliegende Lösungen, wie zum Beispiel Weiterbildung in Palliative Care, doch zeigen Studien, dass selbst umfassende Maßnahmen oft keine langfristigen Auswirkungen haben und weitere Forschung erforderlich ist.

Keywords: Pflegeheim; Palliativversorgung; Kooperation; nursing home; palliative care; collaboration

Palliative care in German nursing homes should be provided through collaboration with general practitioners and specialist outpatient palliative care teams. In practice this is often a challenge.

The perspective of the nursing homes on these collaborations and the impact on residents.

Palliative care is an everyday task for nursing homes (NHs) and their staff. At the same time, studies show shortcomings in palliative care for NH residents with complex needs ([17]): avoidable and potentially burdensome hospital transitions ([2]; [20]), high rates of hospital deaths ([8]) and unsatisfactory symptom control ([12]; [22]).

In NHs, general palliative care is provided by NH staff, general practitioners (GPs) and medical specialists (e.g., oncologists) treating residents with life-threatening diseases in their last weeks and months. There are two models of GP care in German NHs: the GP that was already treating the patient before entering the facility continues to do so in the NH, or NHs cooperate with selected GPs based on a special contract.

For most residents, general palliative care should be sufficient. However, in a survey of German GPs, 71% agreed that NH residents are treated too often in hospitals at the end of life and 54% rated end-of-life care in NHs as "rather poor" ([1]). In Germany specialist outpatient palliative care is provided by SAPV teams ("Spezialisierte ambulante Palliativversorgung"), which were established in 2007. It is available in most regions and provides palliative care for people at home and in NHs. Experts estimate that 10–20% of the dying residents in German NHs need multi-professional, specialist outpatient palliative care teams due to complex needs ([18]). However, various studies and reports have described problems with the collaboration of NH staff, GPs and SAPV – often collaboration with SAPV has not even been established. NH staff name refusal of prescription of SAPV by GPs as one of the challenges in cooperation ([28]). The opinions of GPs on the benefit of SAPV involvement vary strongly: Some do not see any need, others see more SAPV involvement as a solution to the "rather poor" end-of-life care in NHs ([1]). However, the framework conditions for reimbursement and cooperation between general and specialist care are for some a cause of disputes, and this does not encourage good cooperation ([19]).

In home care, positive effects of specialist outpatient palliative care team consultations, e.g., on dying at home or symptom control, have been shown ([5]; [7]). Research into the effects of SAPV in NHs is still at an early stage, but initial studies show promising results. [15] studied residents dying in 46 NHs: Residents with SAPV consultations required less acute care and had fewer burdensome hospital admissions. In an Australian study, specialist palliative care interventions in NHs resulted in reduced length of in-hospital stays, a reduction in the number of hospitalisations and cost savings ([3]). Furthermore, staff reported better quality of death and dying as well as an increased capability to adopt a palliative care approach, complete advance care plans and obtain medical power of attorney ([14]).

To involve SAPV for NH residents, NH executives and staff must be willing to collaborate and a medical doctor (mostly the GP) needs to prescribe SAPV. The involvement of SAPV can vary widely from just providing advice to a complete transfer of responsibility for medical decisions and the treatment of the resident. However, collaboration is not yet widely established.

Against this background, the present study aimed to explore the perspectives of NH executives on the current implementation of palliative care in NHs and the need(s) for action to improve it. The present publication focuses only on those results related to the following research questions:

What experiences do the executives of NHs describe regarding the provision

• • of general palliative care in collaboration with GPs and
• • of palliative care in collaboration with SAPV?

Within the context of a mixed-methods study, we conducted an online-survey and semi-structured telephone interviews with executive staff of NHs in the federal state of Baden-Wuerttemberg, Germany. This publication focuses on the results of the interviews as the online survey did not request information on the topic of collaboration challenges but focused on other topics (e.g., further palliative care education).

In a stratified purposeful sampling strategy, NHs registered at the Statistical State Office of Baden-Wuerttemberg were stratified into subgroups by their number of beds, regions, and providers. Fifty NHs were randomly chosen from the subgroups and were contacted by telephone by LV requesting an interview with staff in executive positions who were able to report on the topic of palliative care in their NH. There were no further requirements regarding the exact position or profession of the respondent. They were informed that the study was funded by Ministry of Social Affairs and Integration of Baden-Wuerttemberg with the intent of exploring needs for action and that results would be reported strictly anonymously.

All study participants provided written informed consent before participation.

The interview guide (see electronic supplementary material ESM1) was constructed based on the results of the online survey and a literature review. The interview topics were structured into three sections: (a) current status, (b) challenges and barriers, and (c) needs for action regarding palliative care in the NH. Collaboration with external services and institutions was a sub-topic in each section. If the topic of collaboration was not mentioned by the interviewee, it was raised and explored by the interviewer.

The interviews were audiotaped and took between 25 and 45 minutes, and notes were additionally taken by the interviewer. They were conducted by a member of the research team (LV; M.A. Health Education, research assistant, female) who had prior work experience in NHs and was supervised by a researcher (KS; M.A. Social Work, research assistant, female) with expert knowledge in qualitative research.

All interviews were transcribed verbatim (software: f4/f5transkript).

The transcripts were analysed by structured content analysis according to [11] using MAXQDA, a computer-assisted qualitative data analysis software. The team of two investigators (LV and EN, M.A. International Public Health, male), one of whom had prior qualitative research experience, participated in the coding process, and were supervised by EM. As part of the initiating text work, the two researchers first carefully read through all transcripts, marked notable text passages, and wrote down initial analysis ideas. Against the background of the two main research questions, which are based on the findings of the literature review, the framework conditions, and structures of palliative care in NHs in Germany, two main categories were deductively developed. Using these main categories, each of the two researchers coded 12 of the 20 interviews. In a second step, both researchers developed first- and second-order subcategories for each of the two main categories and coded the twelve interviews again. To meet the criteria of consensual coding, both researchers finally reviewed the consistency of their main categories and subcategories, discussed discrepancies and clarified category definitions. This resulted in a joint codebook that was then used for the final coding process. Again, both coded the remaining eight interviews and, where necessary, recoded the twelve already coded transcripts.

Of the 50 NHs contacted, 23 agreed to an interview, but three interviews could not take place due to organisational reasons. In total, 20 interviews were conducted with 21 interviewees (one interview took place with two interviewees) in 2017. Table 1 provides an overview of the NHs and interviewee characteristics. Four of the 50 NHs could not be reached by phone for an interview request. Of the 23 NHs reached that did not wish to participate, five cited 'no time', four 'no interest', and two 'individual reasons' (e.g., illness) as reasons, while twelve did not give a reason.

Graph

Table 1 Characteristics of nursing homes and interviewees

We report the findings of the two deductively developed main categories and their subcategories.

• • General palliative care by primary carers in NHs. Subcategories were: (1) framework conditions, (2) Factors that contribute to good quality of palliative care and (3) barriers to the adaption of care and medical treatment to the palliative needs of residents by nurses and GPs.
• • Collaboration with SAPV in NHs. Subcategories were: (1) collaboration with SAPV, (2) barriers to access to SAPV for patients and (3) experiences with (the involvement of) SAPV.

For some subcategories there were further second-order subcategories. The full coding system is available in ESM2.

Several different framework conditions imply challenges for palliative care. The most frequently mentioned issue was staff shortage (due to skill shortages and limited financial resources), which was described by many NH executives. Even the basic information and communication needs of residents, relatives and GPs in palliative situations cannot always be met due to lack of time. Furthermore, in some rural areas, a shortage of physicians impairs the quality of medical attention, as one of the interviewees pointed out:

"It is like that with health care in the rural areas. Disastrous. And it's the same with palliative care."

(Interviewee 18)

Additionally, a growing number of residents move into NHs only during the very last phase of their lives or with cognitive impairments, already unable to communicate their end-of-life preferences.

"Since the residents also come to us later and later, it can also often happen that the residents move in when already in palliative status and they have a very short length of stay, so that everything has to happen very quickly."

(Interviewee 07)

The interviewees describe very different ways of addressing the challenges of care in the last months of life and dying. Some NH executives depicted structures, processes and collaborations that reflected a well-functioning everyday approach to palliative care in their institutions. They described factors that contribute to good quality general palliative care: knowledgeable and dedicated NH leadership, (continuous) staff training in palliative care and good collaboration with competent GPs dedicated to the care of the residents, including e.g. out-of-hours accessibility.

Other interviewees described barriers to adapting care and medical treatment to the palliative care needs of residents by nurses and GPs that resulted in situations in which they witnessed shortcomings of general palliative care.

Regarding the situation in the NHs, some executives described limited time and staff resources for palliative care as well es limited palliative care knowledge and awareness of palliative care needs among NH nurses. For example, nurses do not always recognise palliative care needs. Some nurses fear being accused of neglect if they adapt, e.g., nutrition or hygiene protocols in end-of life situations.

"That is always the biggest fear of the staff, that they might not do something, or that they could be accused later of not having done everything to keep this person alive. So, this fear is still there somehow."

(Interviewee 06)

Similarly, executives reported that some GPs only have limited palliative care knowledge and awareness and often limited information about the resident's individual situation and preferences. That leads to various problems: Some are generally reluctant to withdraw treatment or emergency measures at the end of life. Some have reservations about prescribing opioids for fear of respiratory depression or drug dependence.

Interviewee 20:

"We just had that case two weeks ago. [...] He said he will not prescribe it."

Interviewer:

"May I ask what that was about?"

Interviewee 20:

"Morphine. Sure, morphine. The woman was obviously in pain. Couldn't say it, but she was obviously in pain. [...]"

Not all GPs support pre-planning for emergency situations by prescribing medication and coordinating measures for foreseeable complications and exacerbations.

"We have some doctors who collaborate very well, also with the emergency plan, but we also have doctors who somehow don't want to take the responsibility."

(Interviewee 16)

The executives depicted various collaboration problems between GPs and NH staff: Both the assessment of residents' needs and their treatment led to disputes as they mutually question the competence of the other as well as the assessment of residents' needs. Also, decisions on therapy limitation at the end of life lead to conflicts, if the resident's will is unknown. Another problem is emergency accessibility of the GPs and the resulting situations when they cannot come:

"So, the accessibility [...] the doctors have consultation hours and if it's Wednesday afternoon and the GP is not available, then the doctor on call comes. He doesn't know the resident and then a decision has to be made. And I think that's the point where things don't work out the way we want them to."

(Interviewee 05)

As a consequence of the barriers to adaption of care, interviewees described situations in which residents suffered due to delayed or unsatisfactory symptom control or a lack of adaption of care and treatment in the last hours of life. In cases in which no advance care plans, no advance directives or emergency plans are made – or the physician on site does not know about them – interviewees describe that residents are admitted to hospital against their wish, sometimes treated in intensive care during the last phase of life and die in hospital.

Assessments on whether SAPV is needed in NHs varied widely among the interviewed executives: Some valued the collaboration with SAPV, some had never asked themselves the question, and some stated that there is no need for SAPV if the quality of general palliative care is high:

"The general practitioner has often known the person for many years. And if he has the know-how and the nursing staff has the know-how and attitude, then I would say that is enough. And volunteers from the outpatient hospice service. [ ...] And pastoral care."

(Interviewee 20)

The interviews revealed barriers to access to SAPV for patients. Different requirements on various levels – health care system, NH and GPs – must be met. First, an SAPV team that has sufficient capacity to become engaged with NHs must be available, which is not always the case. Furthermore, one interviewee mentioned that collaboration with SAPV had been rejected as it was delivered by a competing provider. Some of the problems discussed for general palliative care also have an effect here, in particular lack of palliative care knowledge and awareness both in NH staff and GPs – including knowledge of existing services and possibilities of involving SAPV: one interviewee had never heard of SAPV and needed an explanation of what it was; other interviewees reflected an understanding of palliative care as spiritual care, pain management or solely as end-of-life care:

"We only call them, if the general practitioner says 'now we need a palliative care doctor'. Normally it doesn't take long then, until the pain medication is scheduled. It is carried out by the nursing staff. So that there are not weeks of discussions. But the whole thing is usually taken care of within 72 hours. Thus, the person has left us."

(Interviewee 02)

The fact that access to SAPV requires a doctor's prescription (usually GP) was also reported as a barrier by NH executives. While many GPs are prepared to cooperate with SAPV, some feel threatened and refuse, as described in the following:

"We only have that problem with the general practitioners. They sometimes feel threatened, if we add the SAPV team. [...]"

(Interviewee 08)

Some of the interviewees reported workarounds to secure residents' access to SAPV by getting prescriptions from doctors in hospitals or doctors on weekend duty. Others pressure GPs themselves or ask the resident's relatives to do so.

Despite all these barriers, eight NH executives reported cooperating with SAPV; one other executive described ending their collaboration "after the first experience" – without elaborating on the reasons or detailing the experience.

The executives of NHs that cooperate described their positive experiences with the SAPV. Table 2 describes the positive effects of collaboration with SAPV as reported by NH executives. In their experience, SAPV provides a solution to various problems that they do not always find easy to solve within the primary care system.

Graph

Table 2 Positive experiences with (the involvement of) SAPV

On the other hand, executives also revealed issues concerning the collaboration between SAPV and GPs. SAPV is usually only involved in an advisory role – the realisation of measures is the responsibility of the GPs and NH staff. If they do not agree – as described in decision-making situations in general palliative care – the situation is often not resolved collectively:

"What we notice over and over is that the doctor is often not very open to suggestions from palliative care specialists. Be it ... the topic is mostly drug therapy. [...] the doctor doesn't want interference in his own territory."

(Interviewee 02)

Consequences of non-collaboration and conflicts between GP, SAPV and NH staff include unclear communication with residents and relatives, insufficient symptom control, and avoidable and unwanted (by patients, relatives and sometimes even NH staff) hospitalisations, as in the following example:

"We had a resident with cancer, he could have stayed in his room [at our NH], a single room. Everything was arranged [with SAPV]. The daughter with two small children lives a few hundred meters away. I had to go out for a moment. When I came back, the general practitioner had referred him to the hospital palliative care unit [...]. She [the daughter] had been able to leave her kids at home. After that she always had to look for a place to take them to, when visiting her father. I don't know why the doctors do that."

(Interviewee 01)

In general, all eight executives of NHs that frequently collaborated with SAPV described this as a positive experience.

Implementation and organisation of palliative care in NHs in Baden-Wuerttemberg varies widely – both regarding general palliative care and collaboration with SAPV. Some executives described knowledgeable staff and cooperations with competent GPs, others described a range of problems due to, e.g., staff shortage and lack of knowledge and awareness both among GPs and NH staff. Also, assessment of the need and collaboration with SAPV varies. Executives of NHs that collaborated with SAPV reported improvements in palliative care. Whether NH residents receive adequate palliative care depends on (a) regional availability and access to GPs and SAPV, but also on (b) sufficient knowledge and awareness of both NH staff and GPs to identify the need for specialist palliative care and then initiate it in collaboration with SAPV; similar findings were reported by [25].

From the perspective of NH executives, GPs are key in palliative situations, which reflects the results of other studies ([4]; [28]). GPs' qualifications in palliative care, their awareness of residents' palliative care needs and their willingness to visit and cooperate greatly influences the quality of palliative care ([6]; [26]). GPs also decide whether SAPV is involved. NH executives viewed palliative care training of GPs as an important prerequisite for improvement.

The same is true vice versa: GPs reported training for NH staff as an essential measure for change in surveys on the same topic ([1]; [28]; [6]), which is confirmed by some of the NH executives we interviewed. In line with other studies, executives also stated that, within NHs, staff shortages, staff turnover and a lack of time are underlying factors that influence care of residents with palliative care needs ([28]; [16]).

Collaboration problems between GPs and NH staff include discrepancies in the assessment of residents' needs and disputes over prescription of medication and SAPV. GPs do not always acknowledge and respect the perceptions and competence of NH staff ([10]; [21]), which can leave NH staff feeling powerless and morally distressed ([27]).

The descriptions of the interviewees about consequences of shortcomings in collaboration between NH staff and GPs in palliative care are similar to those from various national and international studies. They also describe a need for improvement in advance care planning and symptom control, and the avoidance of unnecessary hospital admissions ([23]; [1]; [17]; [22]).

NH executives' opinions on the need to collaborate with SAPV varied widely in our study. While it has become an irreplaceable support for some NH executives, others see no need for it as they collaborate with knowledgeable and dedicated GPs. Few seemed to lack knowledge of what SAPV is and when to include it in care. In cases of collaboration with SAPV, their experiences reflect the benefits reported in the literature, with positive effects on symptom management, hospitalisation, and place of death of residents, but also an increasing expertise of NH staff through experiential learning ([3]; [14]; [15]).

Overall, the situation is complex, with underlying issues such as skill shortages, which need to be addressed at the health policy level, staff turnover, which is also caused by facility-related factors, and everyday communication and collaboration problems. While on the surface solutions seem obvious – e.g., improving working conditions and wages for NH staff, further education in palliative care, round tables to discuss collaboration – it is complex in practice. Studies have shown that even comprehensive educational and management interventions to implement palliative care in NHs do not always result in permanent changes ([13]; [24]).

There are several limitations of the study: Even though we provided information on definitions and principles of palliative care, we found a different understanding of the term among some interviewees. For example, some NH executives defined palliative care as end-of-life care or spiritual care ("including priests"). Although repeated attempts were made to ensure mutual understanding in the interviews, we cannot always be sure that we have interpreted the interviewees' statements correctly.

Furthermore, analysing in a multi-professional team, we aimed at intersubjective comprehensibility by communicative validation and documentation of the coding process, involving a member who had experience of working in NHs. Nevertheless, misunderstandings and consequent biased interpretations cannot be ruled out. Similarly, even though we tried to implement the quality criteria of openness in interview conduction and question wording, and assured anonymity in reporting, socially desirable answers cannot be ruled out. Social desirability, strategic motives, or variations in understanding of palliative care might also have influenced the self-selective choice of interviewee by the NHs. We might not always have spoken to the most knowledgeable person.

The sample was heterogeneous as it included rural and urban NHs, different providers and NHs with and without comprehensive experience of collaboration with SAPV. However, only a few NH executives reported experience of collaboration with SAPV; with respect to that, our results provide important exploratory evidence, but might not be comprehensive in terms of content.

The interviews only reflected the perspective of the NH executives – for a more comprehensive picture of collaboration, the views of GPs, SAPV teams, residents and their relatives would add valuable information.

Despite new laws, as well as many initiatives and projects, there is still a great need to improve palliative care in German NHs.

Research on solutions for everyday practice needs to address various questions concerning collaboration between SAPV, GPs and NH staff:

• • How can the overlying problem of staff shortages, which poses a barrier to solving the problems in many NHs, be addressed?
• • How can basic knowledge about palliative care and the services offered by SAPV be disseminated within NHs?
• • Where is the line between what general palliative care in NHs should achieve and when specialists should be called? Are there easily applicable criteria to decide which residents need the involvement of SAPV?
• • How can mutual understanding and communication between professionals be fostered to ensure that residents' preferences are respected, their needs are recognised, and adequate care is adapted and coordinated?

Implementation and organisation of palliative care in German NHs varies widely. Whether NH residents receive adequate palliative care depends on regional availability and access to GPs and SAPV, but also on the knowledge and awareness of both NH staff and GPs when identifying residents' needs. Collaboration problems are reported to hinder high quality palliative care in some cases. They are caused by many factors, ranging from a lack of trust in the competence of the collaboration partners to a direct refusal to involve the specialists of SAPV.

The electronic supplementary material is available with the online version of the article at https://doi.org/10.1024/1012-5302/a000952.

• ESM1. Interview guide.
• ESM2. Overview on main categories, first- and second-order subcategories, and contents.

We thank all study participants. We also express our appreciation to Prof. Annette Riedel, Dr. Margrit Ott and Bettina Couné for providing valuable advice throughout the study process. Finally, we would like to thank the Ministry of Social Affairs and Integration of Baden-Wuerttemberg (Ministerium für Soziales, Gesundheit und Integration Baden-Württemberg) for funding the project.

EM, KS and GB were involved in the conception and design of the study as well as acquisition of funding. Data were collected by LV and EM, which was supervised by KS. LV, EN and EM contributed to the analysis and interpretation of data. GB supervised the study. EM wrote the first draft of the manuscript. All authors read, revised and approved of the final manuscript and agree to be accountable for all aspects of the manuscript.

The study was approved by the Ethics Committee of Albert-Ludwigs-University Freiburg, Germany (Approval No. 264/16). All procedures performed in the study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments.