Patients and family members´ perceptions of interprofessional teamwork in palliative care: A qualitative descriptive study

Abstract Aims To describe patients and family members’ perceptions of interprofessional teamwork in specialised palliative care. Background Interprofessional teamwork is essential when delivering high‐quality palliative care. Little attention has been paid to patients and family members’ perceptions. Design A qualitative descriptive design. Methods Semi‐structured individual interviews were conducted with 20 palliative patients and family members (n = 19) in four palliative wards, which were collected from May 2019 to November 2019. Data were analysed using inductive content analysis. COREQ guidelines were followed. Results Patients’ perceptions of interprofessional teamwork were described as the nature of interprofessional teamwork, a sense of community and patient participation. Family members’ perceptions of interprofessional teamwork were described as the nature of interprofessional teamwork, the diverse expertise and the sense of community. Patients and family members’ perceptions of interprofessional teamwork were nearly identical and were based on observed social situations or their assumptions. They trust that professionals are working interprofessionally, even if the teamwork cannot be observed. In palliative care, the nature of interprofessional care changes together with patients’ condition and family members progressively need more professional support. Conclusions Conducting interprofessional care more openly could benefit the availability of different professionals’ competence to patients and family members. In palliative care, the nature of interprofessional teamwork changes together with the patients’ health condition. More information is needed about what constitutes an interprofessional framework and the required interprofessional competencies in palliative care. Relevance to clinical practice. The findings show the importance of considering the patient's health status when interprofessional care is planned. However, professionals should recognise that a patient's weakening condition changes the focus more to the needs of the family members. It is acknowledged that IP teamwork requires time, but in PC settings, spending time on collaborative practices is not always possible.


| INTRODUC TI ON
Annually, over 56.8 million people all over the world are in need of palliative care (PC). This is because of the ageing population and complex health issues. (World Palliative Care Alliance World Health Organization (WPCA & WHO) 2020) PC can be delivered in different clinical settings from home to hospitals. Specialised PC services are needed when patients' have complex care needs, and the expertise of different professionals is essential to deliver holistic care (Gamondi et al., 2013). In PC, the aim is to ensure the quality of life of patients whose death is imminent and their families (WPCA & WHO, 2020). An IP approach should combine different professionals, patients and their families together to deliver quality care (WHO, 2010).
With the collaboration of different professionals, it is possible strengthen healthcare delivery and also achieve better health outcomes (World Health Organization (WHO) 2010). At the core, interprofessional (IP) teamwork is the activity of the professionals.
However, the goal of the team is to provide quality PC for the patient and their families. An IP team can consist of several different professionals, and being a competent team member means acquiring competencies that are both discipline-specific and relevant to all those joining the IP team. Elements such as quality of life, the relationship between patients and professionals and a multiprofessional approach are described as essential in PC settings. In addition, ethical aspects such as the patient's situation as regard life and death, autonomy and dignity are always present. To achieve better care outcomes, an extensive IP team is probably not necessary in PC settings. It is more important that the team shares the same philosophy and has a joint understanding of the care goals. In specialised PC, more qualified professionals work exclusively with complex issues relating to PC. (Gamondi et al., 2013.) It is suggested that patients and families are involved in both the IP teamwork (WHO, 2010) and the PC (WPCA & WHO, 2020).
Despite the recommendations, there have been very few studies that combine IP teamwork and patient and families' perceptions. The role of the patient and family in PC has been discussed but as regard teamwork, the patients and their family's position are unclear, and their roles are poorly documented (McDonald & McCallin, 2010). In this study, IP care is defined as a process where the roles of different professionals in the care process are specialised, but they are expected to interact (Klarare et al., 2013). The focus of this study is on specialised PC wards, where the main activity is to produce quality end-of-life care for the patients with more complex care needs by utilising the expertise of multiple professionals (Gamondi et al., 2013;Hui et al., 2018).

| BACKG ROU N D
IP care has been highlighted when delivering PC (European Association for Palliative Care (EAPC) 2020; Hui et al., 2018;McDonald & McCallin, 2010) and professionals from different disciplines are required to work together for the good of patients receiving PC and their families (Connolly et al., 2016;McDonald & McCallin, 2010;WHO, 2010).
Because of the special nature of PC, different professionals' participation in the care goal discussions are essential and all professionals play an important role in patient care (You et al., 2015). Although patient-centred care is designated as one of the key features in IP teamwork (WHO, 2010), it might be challenging, even unrealistic in life-limiting situations (Sanderson et al., 2017). It is known that it might be challenging for patients and family members to accept bad news or difficult to understand the limitations of available treatments (You et al., 2015). Moreover, the possibility for patients to participate in care meetings might be limited because of their current health status.
In some cases, like communication, the role of family members might be even greater than the patients. (Sanderson et al., 2017.) Whether or not a patient and the family want to be involved, care should be conducted interprofessionally and holistically.
The findings show the importance of considering the patient's health status when interprofessional care is planned. However, professionals should recognise that a patient's weakening condition changes the focus more to the needs of the family members. It is acknowledged that IP teamwork requires time, but in PC settings, spending time on collaborative practices is not always possible.

K E Y W O R D S
family members, interprofessional care, palliative care, patient, qualitative research, teamwork What does this paper contribute to the wider global clinical community?
• Patients and family members' perceptions are needed for interprofessional teamwork when the goal is to provide high quality and individual patient-and familycentred PC.
• PC, where the care relationship has a specific ending and family members increasingly need support, the IP teamwork has special characteristics compared with other fields of care.
• Interprofessional education is needed to receive all the potential of the team to provide holistic PC without causing discomfort to the patient and family members.
To deliver PC effectively in IP teams, professionals should be competent in their own profession but also interprofessionally (Witt Sherman et al., 2017). IP competencies are needed when professionals are working with other professionals, patients, families and organisations (Canadian Interprofessional Health Collaborative (CIHC) 2010; Interprofessional Education Collaborative (IPEC) 2016). IP teamwork consists of certain types of elements such as the clarification of the professional members' role in the team, patient-and familycentred care, communication (CIHC, 2010;Hepp et al., 2015;IPEC, 2016;WHO, 2010;Wood et al., 2009) and shared documentation (Wilhelmsson et al., 2012;Wood et al., 2009). Because of the nature of PC as regard the elements mentioned previously certain types of professional attributes are preferred, such as respect, empathy (Ciemins et al., 2015), sensitivity and communication skills (Ciemins et al., 2015). IP teamwork correlates patient-staff communication and patients may experience the execution of communication in a different way compared with the professionals (von Knorring et al., 2020).
Patients and their family members as part of the team and in team meetings have been studied in PC (Sanderson et al., 2017), but to the best of our knowledge, the literature reveals gap concerning how patients and family members experience IP teamworking during PC in hospital settings. Although the goal of PC is to provide quality care for the patient and their families with the help of different professionals working in IP teams (Gamondi et al., 2013), little is known about how patients and family members experience IP teamworking in PC; therefore, this study aims to describe patients and family members' perceptions of IP teamwork in a specialised PC ward.
The ultimate goal was to add knowledge of IP teamwork in specialised PC delivery from the point of view of patients and their family members in order to promote care delivery and patient safety and also to encourage the systematic involvement of patients and their family members in IP teamworking. This study is a part of a larger study considering evidence-based practices and the competence of professionals in palliative and end-of-life care; the aim of the larger study is to improve quality of care and patient safety.

| Aim
Aim of this study was to describe patients and family members' perceptions of IP teamwork in a specialised PC ward.

| Design
A descriptive qualitative design with semi-structured individual interviews was used to describe the interprofessional teamwork comprehensively from patients' and family members' perspectives. Data were inductively analysed at the surface and manifest level on the experiences of patients and families in their own words. (Sandelowski, 2010.) The Consolidated Criteria for Reporting Qualitative Research (COREQ checklist) was followed when preparing the manuscript (Tong et al., 2007)
The suitability of the patient was assessed individually by the PC ward staff. The inclusion criteria for patients were as follows: 18 years old or older, Finnish-speaking, inpatient, patient has received palliative hospital care for at least one week, are incurably ill, receiving only symptomatic not delaying treatments and good fundamental care, and is capable of willingly sharing her/his perceptions. The patient and the family members participated separately, consequently their family relationship was not determined. The patient named the family member who could be asked to join the study. Participation of the patient was not a prerequisite for the participation of the family member and vice versa.

| Data collection
Before starting the data collection, researchers met the ward staff and provided them with information about the whole research project in general and recruitment process (e.g. the criteria for suitable participants). The ward staff gave oral and written information about the study to the possible participants and asked them if they would take part in the study; those who participated gave their written informed consent and returned it in a sealed envelope. The contact person, who was a registered nurse on the ward, informed the interviewer when it was suitable to conduct the interviews.
Data were collected between May and November 2019 by two interviewers interviewing participants individually. All participants were interviewed face-to-face except for one relative who was interviewed over the telephone. Interviews were undertaken in clinical settings. A calm place on the ward was requested in which to conduct the interviews to minimise distractions and to obey the duty of confidentiality; the place was chosen by the interviewee or the ward staff. Only the interviewee and interviewer were present.
The semi-structured interviews were based on previous framework describing the goal of the IP team and competencies of professionals working in PC to provide quality care for the patient and their families (Gamondi et al., 2013). All interviews were digitally audiorecorded with the consent of the interviewee and were transcribed verbatim. The participants were asked to describe their perceptions of the following: (1) which professionals have participated in their care in this current ward, (2) have these professionals worked as a team/collaboratively in care and (3) how can the teamwork be observed and described. Supportive questions were asked, and concrete examples were requested along with the initial questions. The interview guide was pilot tested, and these data were also included in the final data as there were no need for changes.
The interviews were conducted between May and November 2019. Patients' interviews lasted approximately 24 minutes (ranging from 9 minutes to 50 minutes) and family members' interviews lasted approximately 33 minutes (ranging from 13 minutes to 1 hour 13 minutes). Out of respect for the patient's health condition and family members' well-being, the interviews were highly focused and relatively short in duration. In three cases, the interviews were interrupted due to treatment being needed for the patient and the patient's health condition. In some cases, the patient's health condition challenged the data collection (e.g. their mouth was so dry, it was hard to understand their speech or incoherence caused by a brain tumour). The participants' background information (age, gender, illness, the length of the illness and the length of the treatment) was requested orally at the beginning of the interview.

| Data analysis
The interviews were analysed using inductive content analysis because it is applicable when studying perceptions (Graneheim et al., 2017). The analysed data only consisted of the interview material and the data from the patients and the family members' interviews were analysed separately based on the aim of the study (Kyngäs et al., 2020). The data analysis process was conducted individually by one researcher (PK) and checked by two other independent researchers (EH and LS); it was also discussed with the research group. The analysis process was conducted through the following phases (Table 1): At first, the interview material was read through several times to become familiar with the content. The data were examined to find units for analysis, and then, relevant expressions corresponding to the aim of the study were identified from the transcripts. The selected sentences were simplified by removing unnecessary words. Expressions were coded and compared to identify similarities and differences and based on this comparison, grouped into subcategories by inductive category forming. The classification was continued by summarising the subcategories into the categories (Lindgren et al., 2020.). Data saturation was achieved with 20 patients and 19 family members (Elo et al., 2014).

| E THIC AL CONS IDER ATIONS
This study was conducted according to the Declaration of Helsinki (World Medical Association (WMA) 2013). Ethical approval was requested to undertake the study and was received from the Ethics committee for Human Sciences at the University of Turku (15/2019).
In addition, permissions were requested and obtained from the organisations from which the data were collected. All participants received oral and written information concerning the ongoing study.
Participation was voluntary, anonymity was guaranteed during the whole research process and participants provided informed written consent for their participation in the study. The participants were informed of their right to withdraw their participation at any time, and confidentiality was observed at all times during the study. A written information letter including the researchers' contact information was provided, and the participants had an opportunity to ask questions when receiving the oral information and also during the interviews. The participants were informed that participating in the study would have no effect on the care they were receiving. Extra attention was paid to the vulnerability of the participants. The terms used were chosen to be as neutral as possible, and the interviewers endeavoured to keep the length of the interviews short. The

Original expression Simplification Subcategory Category
'It is really hard for comment because I have not needed the help of any other professional'.
It is hard to comment if the help of certain professionals has not been needed. Health statusdependent teamwork Pain effects negatively on patients thinking.
'The library personnel and the singer came on the same day, and there were those volunteers too. There were too many people, so I said nicely to the volunteer, that "thank you kindly, but my things are arranged well this moment'.
Having too many contacts have its effect on patients' ability to participate.
'Actually, this whole process has proceeded so quickly that just a few professionals have been participating my care'.
The disease has proceeded so quickly that only a few professionals have participated in the care.
well-being of the participants was monitored during the interviews, and interview was suspended with a low threshold.

| Characteristics of participants
A total of 20 patients and 19 family members from four different specialised PC wards in Finland participated in this study ( Table 2).
Nine of the family members were spouses of the patient, five were children, two were friends, and in one case, each a sister, a mother and a father. Seventeen of the participating patients were suffering from an incurable cancer (different types), and the diagnosis of three patients was unknown or not mentioned. The relatives of the family members who participated in the interviews were all suffering from some form of cancer, except one patient who had ALS.

| Patients' perceptions of interprofessional teamwork in specialised palliative care
Patients' perceptions of interprofessional teamwork in specialised palliative care were described as follows: Nature of interprofessional teamwork, Sense of community and Patient involvement ( Table 3). As regard health status-dependent teamwork, the patients often described that they were too tired, sick or in severe pain to participate in teamwork. Health-status dependent teamwork was present in patients' perceptions. End-of-life care gives special input to IP teamworking when the diagnosis has been given with a very short notice or the prognosis is very bad or the disease itself has rapidly deteriorated. A few patients described the IP in PC as demanding, because of the many contact people and tight schedules, which often meant that all the contacts were on the same day. In these cases, patient felt that some of the professionals needed to be excluded.

| Family members' perceptions of interprofessional teamwork in specialised palliative care
Family members' perceptions of interprofessional teamwork in specialised PC can be described as Nature of interprofessional teamwork in specialised palliative care, Diverse expertise and Sense of community (Table 4).
The first category describes the nature of interprofessional teamwork in specialised PC consisting of two subcategories: visible interprofessional teamwork and invisible interprofessional teamwork.
Family members described professionals as working as a team and every professional making an effort to get things done. Family members saw doctors and nurses at the centre of teamwork, but they also described teamwork among just one group of professionals, such as the nurses, as helping each other in care situations. Invisible teamwork can occur in two ways; when the teamwork cannot be noticed, or there is a lack of IP teamwork. A few family members could not comment on the teamwork question because the teamwork in the care unit could not be seen. Teamwork was also hard to describe because they did not know which professionals were participating and which one was not involved. Two family members described that there was no teamwork between different professionals. They said it appeared that professionals were only doing the work appointed to them, but nothing else. Family members described the team spirit among IP team. They felt that the atmosphere was warm, and everyone was getting along well. When conflicts or arguments were not noticeable, it gave an impression of a good team spirit. In addition, they felt that the number of years the professional had worked together helped 'the team feeling', as one family member described: 'Probably they have worked together for many years. The encounter was warm. And even if the patient is present, the atmosphere was really warm'.
The patients and family members' possibilities to participate in the teamwork were also the way that they sensed interest and enthusiasm from the professionals on the ward. They compared their previous experiences and felt that in the current palliative ward, the professionals, in general, were participating more actively. Considering their own participation in IP teamwork, family members had both experiences; they have had the possibility and not had the possibility to communicate with the professionals. Family members felt that the nature of IP teamwork changes when the patients' health status changes. At first, patients are more interested in participating in teamwork, but when the disease progresses and the patient's well-being is deteriorating, family members are the people who need more support from the professionals.
'When you recognise that the condition of the relative is getting worse and we are not able to manage with the situation anymore or don´t know what should do. Then some kind of contact at least is needed, how to start the evaluation process to find out if more support is required'.

| DISCUSS ION
The aim of this study was to describe patients and family members' perceptions of IP teamwork in specialised PC. Overall, the perceptions of both groups had similar elements but also varied to some extent. Both described the nature of IP teamwork in similar ways, whether it is visible or invisible and somewhere in the background.
They also described more feelings or assumptions rather than concrete examples of the ways in which the teamwork was con- family members. Communication has been described as one of the key skills in IP teamwork (Hepp et al., 2015;Wilhelmsson et al., 2012;Wood et al., 2009) yet it is known that patients can experience communication differently when compared to the experience of professionals (von Knorring et al., 2020). Studying this aspect by comparing the perceptions of patients, family members and professionals would give valuable information about IP teamwork.
Although the perceptions were nearly alike, only family members connected patient and family member participation to the IP teamwork as being important, and to be recommended (Gamondi et al., 2013;WHO, 2010). Another difference between patients' perceptions and family members was that the family members focused more on the clarity of the professional's role (CIHC, 2010;Hepp et al., 2015;IPEC, 2016;Wood et al., 2009), probably because in many cases, family members have been actively participating in care and taking care of their close one for a long time. Family members also focused on how knowledge in the IP team was organised and if the correct professionals were working with right care-related tasks.
The most remarkable difference between the perceptions were that the patients focused on the quality of the care provided (Gamondi et al., 2013;Hepp et al., 2015;Wood et al., 2009) has an effect on the IP teamwork (Sanderson et al., 2017). When the patient is incurably ill and a prognosis for the future health status cannot be given, naturally it influences how the IP teamwork was seen or experienced. For family members, there might be difficulties to let go of active and rehabilitative care and focus more on the quality of symptomatic care. In general, the IP care frameworks highlight active participation from patients and family members (CIHC, 2010;IPEC, 2016). The IP approach is also seen as positive and self- process. The analysis process was checked by two independent researchers, then the final output was discussed and verified by the research group to increase the trustworthiness of the analysis (Willis et al., 2016). Nonetheless, qualitative descriptive study is interpretative to some extent (Sandelowski, 2010). The credibility of the research was increased by using direct quotations when reporting the study findings.
The transferability of the study findings was enabled by describing participants´ demographic information. Geographically, the interviews were held in the southern parts of Finland where the population density is higher compared to other parts of the country. The study context and inclusion criteria for participation were described as precisely as possible to enable the transferability of the results. Furthermore, some of the descriptions began to recur which were considered a sign of a saturation. (Elo et al., 2014.) All these factors allow readers to assess the possibility to transfer the results to another context after careful consideration.
Dependability was confirmed by giving detailed information about the aim of the study, describing the participant recruitment and data collection process. Moreover, the data analysis process was described phase by phase and supported with tables to provide transparency. Hence, during the recruitment phase, providing only the main goals of the ongoing study minimised any possible changes in professionals' behaviour during data collection.
Confirmability was increased by asking additional questions or for clarification during interview if necessary and also by using the expertise of the research group in the data analysis process. Repeat interviews were not carried out nor were transcripts returned for comments/feedback, because of weakening condition of the patients. (Schwandt et al., 2007.) In addition, few agreed interviews, both with patients and family members, were cancelled because of the changes in the patients' health status or because they were died.

| CON CLUS ION
The perceptions of patients and family members had similar elements but also varied to some extent. However, the care goals differ to some extent when compared to other fields of care.