Design

An explorative generic qualitative study with healthcare professionals (HCPs) working in a Dutch hospital was performed from 2021 to 2022. This study was set-up from an interpretivist/constructivist paradigm, assuming that individual construct their own social realities through interaction and experience [18]. Given our research focus, this paradigm allowed us to approach the topic in a way that prioritizes understanding the participants’ perspectives within their specific context. For this report, standards for reporting qualitative research- guidelines were used [19].

Characteristics and reflexivity

Interviews were conducted by two separate interviewers (JH and LT), focus groups (FG) were led by LT and supported by JH. JH is an experienced nurse in pulmonology and was trained on the job for interviewing. LT is an experienced clinical academic nurse in pulmonology, trained in collecting qualitative data through semi-structured individual interviews, and FGs. Parallel to this study, LT was trained in palliative care. Interviewers were familiar with the aim of this study and had a working relationship with all HCPs who were interviewed. Given the existing relation between interviewer and interviewees, there was a risk of “going native”, in other words, potentially blurring the line between researcher and participant [20]. Interviewers were aware of this and memos were written by them to reflect on the influence they might have had on the results. Additionally, they reflected on this during research meetings. These measures were undertaken to manage potential biases and ensure that findings reflected participants’ perspectives rather than those of the interviewers.

While analyzing data and discussing themes, the entire, multidisciplinary research team was involved. Coding was done by LT and LD, with the latter having no experience in providing care. A senior qualitative researcher EdG supervised the data analysis process. All analyses, including the identification of categories and themes, were subsequently discussed within the entire research team.

Context

This study is performed in a referral center for interstitial lung diseases. The HCPs are specialized in interstitial lung diseases. The care in this center is delivered by an interdisciplinary team with specialized pulmonologists and nurses. A specialist palliative care team is available, who can support palliative care in inpatient and outpatient clinics.

Study population

All HCPs working in active inpatient and outpatient care in this department were eligible for this study. Starting June 1^st, 2021, HCPs were informed about the overall aim of this study and asked to participate via e-mail, and face-to-face in informal meetings. Participants were selected through convenience sampling, based on voluntary participation, with efforts to include a range of roles and experience levels within the multi-professional team.

After fifteen interviews, no new themes or insights were identified. An additional interview with a HCP of differing characteristics and perspective also revealed no new themes. Recruitment ended at March 10^th 2022.

Ethics

This study followed the Declaration of Helsinki [21], Good Clinical Practice Guidelines [22] and the General Data Protection Regulation [23]. This study is not subject to the ‘Medical Research Involving Human Subject Act (WMO) as it does not concern medical scientific research, and participants are not subject to procedures or required to follow rules of behavior. The study protocol was approved by the Medical research Ethics Committees United (MEC-U W21.018). All participants gave verbal informed consent for the use of their data.

Participants’ consent and interview records were documented and securely stored in REDCAP (www.project-redcap.org) and pseudonymized by giving them a code number in transcription procedures. Only LT and JH had access to recordings, transcripts, and the pseudonymization document on two different hard disks.

Data collection

All interviews were conducted between June 2021 and March 2022. The combination of FGs and individual interviews was chosen to first get insight into relevant topics as discussed between professionals. These topics were then explored in greater depth during subsequent individual semi-structured interviews. This sequential approach allowed us to identify shared themes through group interaction, and then examine personal experiences and nuances that may not surface in a group setting. The use of both methods provided complementary perspectives and enriched the overall understanding of the research topic.

HCPs only participated once, either in an interview or a FG, based on personal preference. In FGs, all disciplines were mixed together. During interviews, only the interviewer and the participant were present. Due to COVID-19 regulations one FG was performed online, using Pexip (www.pexip.com). Other FGs and interviews were performed face-to-face in a conference room in the hospital.

Data collection and analysis were alternated, with new questions that raised during the analysis being included in the interview guide for subsequent interviews.

Two interview guides were developed (S1 Appendix), for FGs and semi-structured interviews, based on literature and the research team’s experiences. All questions were open ended. The interview guide for FGs was designed to initiate discussions among HCPs, encouraging them to identify and discuss important themes. FGs were conducted prior to individual interviews and were used to optimize the interview guide for individual interviews. See interview guides for more details,

Data processing

FGs and all interviews were digitally audio recorded, transcribed verbatim, and then anonymized. Transcription and anonymization were performed by working students. Subsequently, transcripts were uploaded into the Atlas.ti 9 application together with memos. From that point onward all data was processed in Atlas.ti 9.

Data analysis

A thematic analysis was performed following the six phases described by Braun and Clarke [24], here shown in Italics. Two researchers (LD, LT) first familiarized themselves with the data by reading and re-reading the transcripts. Initial coding was performed independently and inductively on two transcripts in Atlas.ti (LT and LD), using open and detailed coding while staying close to participants’ wording.

Based on this initial coding, a preliminary coding framework was developed and applied to the subsequent transcripts (LT, LD). Each interview was discussed to reach shared understanding and refine interpretation.

Codes were then organized into meaningful clusters, first manually and later using Atlas.ti functions. These clusters formed the basis for developing potential themes and subthemes, which were reviewed, refined, and named through discussion with the multi-professional team.

Theme development was approached as an active and interpretative process, consistent with reflexive thematic analysis, rather than as a passive emergence from the data.

Descriptive statistics were used to present characteristics of the participants

Trustworthiness

To enhance credibility, memos were written after each FG and individual interview, during transcription, and throughout the data analysis process. Member checking during interviews helped ensure that participants’ perspectives were accurately captured.

To ensure dependability, an audit trail was maintained throughout the whole process of data collection and analysis, documenting all decisions made during the study. Coding and theme development were discussed and reviewed by multiple researchers to strengthen consistency.

Transferability was supported by providing detailed descriptions of participants’ roles, experiences, and the context in which the study took place, allowing us to assess the applicability of findings to other settings.

Confirmability was promoted through the use of reflexive memos and regular team discussions, which helped minimize interviewers’ and researchers’ biases and ensure that interpretations were grounded in the data.

Lack of a shared vision on palliative care

Unlike disease-specific treatment, palliative care lacks a unified approach. Consequently, each healthcare professional (HCP) may adopt their own interpretation of palliative care, leading to variability in services and care provided. This diversity is reflected in three subthemes: use of language, collaboration within the team, and collaborations outside the team.

Use of language

Participants defined the onset of the palliative phase in diverse terms, including “no longer curable,” “when progression is noted”, “final phase of life,” “upon a terminal diagnosis,” and “a hybrid event”. This variability extends to discussions on personalized and patient centered care. While some aim to align care with patients’ wishes and values, others may center care around the patient, without actively involving them in decision making. HCPs admit to not verifying their assumptions with the patient and their partner, tending instead to project their own feelings and values.

‘Then we often notice that people haven’t heard the term “palliative care” very often, or that it has never really been brought up.’ (participant 2)

Collaboration within the team

The lack of a unified understanding, and clear definitions regarding palliative care, results in confusion about responsibilities and task allocation within the team. Roles of nurses and physicians often remain ambiguous, leading to inconsistent patient care. While some HCPs proactively explore and manage their patients’ palliative care needs, others overlook necessary discussions on palliative care. This inconsistency is particularly pronounced when patients transition between specialists, a situation frequently encountered with specialists on fixed-term contracts. Although one participant has a positive experience and saw potential benefits in this approach, the prevailing lack of clarity undermines consistent care delivery.

‘It’s important to do it as a team: but that also means that the team needs to be on the same page... That’s why I find “palliative [treatment]” a difficult term to convey to patients. That’s why I prefer using symptom-focused to differentiate it from other treatments.’ (participant 1)

Few participants suggested that established protocols and guidelines could enhance inter-professional collaboration in palliative care. However, they also explained that current palliative guidelines are not always applied by all HCPs.

Collaboration beyond the team

One FG emphasized the importance of systematic documentation strategies to enhance collaboration among HCPs within the hospital. Effective communication between inpatient and outpatient clinics is crucial, sharing information on outcomes and underlying values and rationales, to ensure proper continuity and follow up. However, other participants raised concerns about the appropriate level of detail for documentation and how to discern what is critical to record.

The role of the Palliative care Advice Team (PAT), a specialized multi-professional palliative care team, was discussed in one FG and two interviews. Participants suggested that the PAT’s task should be to provide advice and coaching to ILD HCPs, rather than assuming direct care responsibilities or communicating directly with patients. This recommendation stems from the PAT’s limited knowledge of ILD and the belief that palliative care is integrated into the ILD care. However, the same interviewees also mentioned that the PAT is consulted excessively due to relative inexperience of trainee doctors and novice nurses in the inpatient ward. Additionally, during FG, some participants expressed unawareness of the PAT’s consulting role for the outpatient clinic and advocated for its increased involvement.

‘I think that when having such a conversation about approaching end-of-life, you need to understand what the disease is, what the impact of the disease is, psychologically, socially, spiritually. We were also just talking about continuity [of HCPs]; I don’t think they [the Palliative advisory team] are the right people. I think they might know which elements should be addressed in the conversation… They can be coaching, but shouldn’t lead the conversation. So really advisory.’ (participant 13)

Participants mentioned a wide range of perspectives on transmural collaboration, leading to varied approaches, and different levels of engagement with general practitioners (GP). Communication methods differ between HCPs, some prefer direct telephone communication for each patient, while others opt for letters. Nurses often maintain communication with primary care providers. Most pulmonologists continue to communicate with patients and/or their families, even after coordination of care has transitioned to the GP. Participants frequently attributed this ongoing contact to the perceived lack of ILD knowledge among primary care provider and to a personal desire to maintain a humane, personal connection.

‘I think we only reach out [to contact the General practitioner] in the final stage [when a patient is dying]. Then we briefly mention it in the letter. Sometimes we call. But I think we’re a bit late with that as well. But the question is, if you [nurses] do it earlier, is that sufficient or not?’ (participant 9)

Learning-driven environment

Participants state that there exists a shared learning environment for acquiring knowledge and skills related to disease specific care. Within this environment, participants expressed confidence in the abundance of physical and medical knowledge, as well as in the ongoing development and dissemination of new insights. However, this shared learning-driven environment appears to be lacking when it involves palliative care and non-physical concerns and needs. In these areas, it depends on individual HCPs and his or her learning approach. Participants highlighted that currently, professional development relies heavily on individual initiative. While some participants reflect on their conversations and seek to learn from their mistakes, few mention peer to peer coaching, or group supervision as potential interventions.

’I do reflect on that myself. I don’t really do it with the patient… If they mention something about the conversation, I will respond to this.’ (participant 13)

Findings indicate there appears to be a noticeable gap in the development of relevant knowledge and skills when it comes to palliative care and end-of-life conversations. This gap is compounded by the absence of a structured approach to facilitate the exchange of this knowledge and skills among HCPs. Trainings, workshops and a culture of feedback are lacking, as evidenced by the knowledge and skills mentioned by participants.

‘I lack knowledge and experience to effectively guide individuals from other cultures in the palliative phase.’ (participant from FG 3- from the wall: ‘not satisfied with..’)

Participants explicitly mentioned the pivotal role of knowledge in initiating and conducting end-of-life conversations. They noted that the required knowledge varies depending on the HCPs profession and their responsibility. While physical symptoms are explicitly discussed, there is a notable absence of consideration for psychological, social and spiritual concerns such as coping mechanisms, loneliness, and grief. Despite expressing a desire to see patients as whole individuals, participants predominantly focused on the physical dimension. A multidimensional approach is only mentioned by HCPs with specialized education in pulmonology or mental healthcare.

‘Of course, there are those questions you can ask from the Diamant model [Ars Moriendi model [26]], for example, which I don’t always have laid out on the table, but they are in my head.’ (participant 15)

‘I do believe that stress has a significant impact on someone’s illness, as well as their perception of illness. Perhaps that should be higher on the list of things we should address than medication for some people.’ (participant 11)

Besides the expansion of knowledge, skills are also necessary to put this knowledge into practice. Almost all participants mentioned the importance of communication skills alongside care and treatment skills. However, there is a range of confidence levels among participants, ranging between completely confident to not confident enough and maybe avoiding discussions about end-of-life. Others felt generally confident but acknowledged the need for additional training.

’Am I causing them more sorrow than necessary? ‘ (participant 12)

Furthermore, participants identify prerequisites for effective palliative care and end-of-life conversations, including adequate time, appropriate patient information, and suitable locations. While these prerequisites are recognized by most participants, there is a lack of action taken to optimize them.

‘So I’ve actually stopped doing that [giving out brochures]… Because I don’t find the information helpful. At one point, we received a brochure. It again had a sentence that was incorrect. And we have a very simplistic brochure... I really think it’s a baby brochure. You should also take your patient somewhat seriously. I’ve kind of stopped giving everything out, unless people ask for it themselves.’ (participant 4)

The individual professional

Besides the organizational and collaborative part of care, the individual professional influences conversations and interactions with the patient and his or her partner about end-of-life. This is reflected in two sub-themes: professional and personal experiences, and tailoring to the patient.

Professional and personal experiences

Experiences of HCPs directly impact the care they deliver. Positive professional experiences facilitate effective end-of-life communication and increase confidence. Conversely, negative experiences may breed hesitation. Personal experiences also play a pivotal role, often serving as a source of motivation. HCPs reflect on their own experiences of having been a patient themselves or on experiences involving loved ones who required palliative care.

‘It also touches on personal experiences, for example, my father with whom I experience the last phase… But it can also be inhibiting, and therein it’s also very important that you know yourself very well. ‘ (participant 15)

However, a few participants noted that personal experiences can sometimes cloud judgment, and impede end-of-life conversations due to emotional involvement.

Moreover, while experience in end-of-life conversations is undoubtedly valuable, it is not universally perceived as the sole essential factor. Some participants state that possessing ‘guts’ and ‘trust on their skills’ suffice to address and follow up on end-of-life topics with patients.

‘When I compare it to five years ago when I just started on the ward, I found it much more difficult back then. You think: oh, you want to discuss the topic of palliative care, but then you wonder if it’s already time. You doubt yourself. I don’t really have that anymore now… And I find it much easier to broach the subject… It’s through that experience that you know better to ask the right questions. ‘ (participant 3)

Tailoring to the patient

To ensure appropriate care for all patients, care needs to be tailored to the coping process and needs of individual patients. Therefore, the HCP needs to, for example, involve patients and partners in development of a care plan, and check preconceptions with the patient. One-size-fits-all care does not exist, and the fact that palliative care has a personalized character is mentioned by almost every participant. However, HCPs admitted that they do not always verify their assumptions with patients, and they appear to struggle when patients have different coping strategies or hold values and norms that differ from their own. Part of tailoring care to the patient involves considering intercultural differences and health literacy, however this was mentioned by only a few participants.

‘That [to avoid disturbing the patient and their loved ones] is why I sometimes don’t have a conversation. Even though we’re still following a code B [limited treatment; no reanimation and intubation] I estimate it won’t last longer than a week. Especially the time for them together… and now that I’m saying this, I think, maybe you should just ask instead of assuming that it’s desired for them to be together.‘ (participant 14)

Strengths and weaknesses

One of the strengths of this study is its qualitative nature, where the combination of FGs and interviews gives a better understanding of the participants’ perspectives and experiences than other research methods. The combination of disciplines led to interesting discussions. The interviews were used to validate the results from the FGs and gave more depth and exploration of personal experiences. All steps of analysis were performed by two researchers (LD, LT), with a senior researcher (EdG) for guidance. This improved the interview quality during the study and the analysis of the data, as the senior researcher’s expertise ensured a more thorough and nuanced interpretation of the findings. Results were discussed within the entire inter-professional team.

A limitation is the professional relationship between interviewers and interviewees. While this facilitated initiating interviews and building rapport, it may have hindered deeper exploration of personal or sensitive topics, potentially leading to socially desirable responses. Interviewers attempted to address this by probing further into the topics.

Additionally, the interviewers’ initial lack of experience affected the quality of the early interviews, reflected in a larger-than-anticipated sample size. However, through reflection and guidance from experienced interviewers, their skills improved over the course of the study. As a result, later interviews were more in-depth and focused on key areas identified earlier, ultimately yielding higher-quality data.

This study provides insights into barriers and facilitators in end-of-life communication with patients with PF, considering both its strengths and limitations.

Implications for practice

Combining the COM-B model with the results of this study supports the development of implementation strategies necessary to implement palliative care in clinical PF care. The implementation strategies should include establishing a shared vision on palliative care, encouraging HCPs to address not only physical but also psychological, social, and existential dimensions of patients’ lives. Clear guidelines and agreement in roles and responsibilities are also essential.

Improving end-of-life communication requires fostering a learning culture that promotes continuous professional development and supports HCPs. Education, training, and on-the-job coaching—guided by implementation champions—can help strengthen the professional competencies of all members of the multiprofessional team. All strategies are equally important in supporting behavior change and successful implementation of palliative care [35]. Management and champions must actively stimulate HCPs and facilitate their growth.

Further research should explore patients’ perspective on barriers and facilitators in conversations exploring values, preferences, and needs for end-of-life care. Additionally, gaining deeper insights into effective learning strategies for palliative care among HCPs in centers of excellence for rare diseases could improve implementation efforts.