Introduction
Palliative care (PC) is associated with improved quality of life for patients (Bakitas et al. Reference Bakitas, Lyons and Hegel2009; Haroen et al. Reference Haroen, Maulana and Harun2025), patient and caregiver coping (Dionne-Odom et al. Reference Dionne-Odom, Azuero and Lyons2015; Naouma Siouta et al. Reference Siouta, Van Beek and van der Eerden2016; Greer et al. Reference Greer, Applebaum and Jacobsen2020), targeted treatment and care, and fewer hospital admissions and hospitalizations (Hui et al. Reference Hui, Kim and Roquemore2014; Senot and Chandrasekaran Reference Senot and Chandrasekaran2015; Naouma Siouta et al. Reference Siouta, Van Beek and van der Eerden2016). Additionally, PC consultations help educate patients about their health, prognosis, and treatment and to build supportive partnerships with patients and their relatives (Thomas et al. Reference Thomas, Jackson and Carlson2019).
Patient-centered communication is fundamental to PC and is a key mechanism by which it improves quality of life for patients with advanced, progressive disease (Ingersoll et al. Reference Ingersoll, Saeed and Ladwig2018). It includes understanding and validating the patient’s perspective, recognizing their psychological context, reaching a shared understanding of their problems, and involving them in health-related choices (Dean and Street Reference Dean, Street, Wittenberg, Ferrell, Goldsmith, Smith, Ragan, Glajchen and Handzo2015). This holistic approach addresses the psychosocial needs of patients and their families, contrasting with the limitations of a purely medical model (Mead and Bower Reference Mead and Bower2000; Kmetec et al. Reference Kmetec, Fekonja and Kolarič2022). PC clinicians use many principles common in patient-centered communication, such as assessing patients’ understanding of prognosis and treatment, and emphasizing effective coping, focusing on caregiver needs, and advance care planning (Thomas et al. Reference Thomas, Jackson and Carlson2019).
While communication is widely recognized as a core competency in PC and PC clinicians are often regarded among the best communicators in medicine (Wittenberg Reference Wittenberg, Wittenberg, Ferrell, Goldsmith, Smith, Ragan, Glajchen and Handzo2015), the communication tasks they perform are complex and broad (Buss et al. Reference Buss, Lessen and Sullivan2011; Clayton et al. Reference Clayton, Butow and Waters2012; Arnold et al. Reference Arnold, Back and Barnato2015; Anderson et al. Reference Anderson, Puntillo and Boyle2016, Reference Anderson, Bloch and Armstrong2019; Isaacson et al. Reference Isaacson, Minton and DaRosa2019). Over the last 30 years, numerous strategies and frameworks for communication with patients with serious illness have been developed and successfully used. Widely referenced approaches include the SPIKES 6-step protocol (Baile et al. Reference Baile, Buckman and Lenzi2000), the ABCDE plan (Rabow and McPhee Reference Rabow and McPhee1999), as well as resources like Buckman’s book “How to Break Bad News: A Guide for Health Care Professionals” (Buckman Reference Buckman1992), and “The Serious Illness Conversation Guide” (Ariadne Labs 2023). These approaches focus on informing the patient and family about the patient’s condition and prognosis and are typically designed to support specialists outside of PC (e.g., internists or oncologists) in specific, often time-limited conversations (Pfeifer and Head Reference Pfeifer and Head2018). While they offer useful guidance, they do not fully reflect the relational, emotional, and continuous nature of PC communication (Niederhauser et al. Reference Niederhauser, Egloff and Eychmueller2025), where PC clinicians face a variety of communication tasks (e.g., establishing illness understanding, eliciting patient values, or guiding decision making) over extended periods and within evolving clinical contexts.
Despite the centrality of communication skills in PC and advancements in communication training, clinicians’ comfort and perceived proficiency with specific communication tasks remain largely unknown. This is particularly relevant regarding the potential mismatch between feeling competent to conduct a conversation and feeling emotionally comfortable engaging in it, reflecting an emotional dimension of communication that might be distinct from perceived proficiency (Sorensen and Iedema Reference Sorensen and Iedema2009; Niederhauser et al. Reference Niederhauser, Egloff and Eychmueller2025). This paper aims to describe the learning experiences, comfort, and perceived proficiency of Swiss PC clinicians across specific communication tasks.
Methods
Study design and recruitment
In this study, a cross-sectional online questionnaire focusing on a variety of different PC communication tasks was employed. The survey was administered using the online platform SurveyMonkey. Participants of the Swiss National PC Congress were contacted via email and invited to participate. Although all congress registrants received the invitation (a maximum of 769 registrants), only individuals working clinically were eligible; students, researchers without clinical roles, and administrative participants were not included. In addition, palliative.ch (the Swiss Society for Palliative Medicine, Nursing and Support) distributed the survey to its Medical Section, inviting clinically active physicians throughout Switzerland. Because the exact proportion of congress attendees with clinical roles remains unknown, and due to potential overlap with the palliative.ch mailing list, a precise response rate cannot be calculated. Of 293 initial responses, 35 were excluded due to missing data. In addition, 27 responses from clinicians other than nurses or physicians (e.g., physiotherapists or spiritual care workers) were also excluded, resulting in a final sample of 231 participants. Data were collected from October 1 until November 11, 2021.
Participation did not require any specific PC certification, so that the full spectrum of clinicians engaging in PC could be represented. In Switzerland, PC is recognized as a postgraduate area of specialization. For physicians, the formal qualification is the Interdisciplinary Specialty Focus in Palliative Medicine, awarded by the SIWF (the Swiss Institute for Postgraduate Medical Training), which is built on an existing medical specialty (e.g., internal medicine, oncology, and family medicine) and requires additional supervised clinical training and postgraduate theoretical education (SIWF, 2023). For nurses, qualifications exist at 3 levels (Basic, Intermediate, and Advanced) ranging from a minimum of 40 learning hours at the Basic level to a Master of Advanced Studies at the Advanced level (Iris et al. Reference Iris, Elke and Renate Gurtner2021). Theoretical content is often delivered in an interdisciplinary format, so that physicians, nurses, and other clinicians can attend the same courses (e.g., at a Certificate of Advanced Studies level).
Measures
The questionnaire, developed by the researchers, covered 22 specific PC communication tasks with 1 item each and was based on already existing studies or questionnaires on PC communication (Buss et al. Reference Buss, Lessen and Sullivan2011; Clayton et al. Reference Clayton, Butow and Waters2012; Arnold et al. Reference Arnold, Back and Barnato2015; Anderson et al. Reference Anderson, Puntillo and Boyle2016; Anderson et al. Reference Anderson, Bloch and Armstrong2019; Isaacson and Minton Reference Isaacson and Minton2018). The questionnaire was made available in German, French, and English. The original English questionnaire was translated into German and French by academic staff, and each translation was checked for accuracy and clinical appropriateness by a bilingual PC clinician. The final versions in all 3 languages were then reviewed by a PC expert fluent in all 3 languages. Prior to distribution, the survey was pilot tested with clinicians from the PC team; no adaptations were required.
For each task, participants rated their perceived proficiency and comfort, and reported the frequency of use and whether they had received training in the specific task. All tasks can be found in Table 2. Perceived proficiency with tasks was rated on a 5-point Likert scale, with the verbal anchors “Not proficient at all” to “Extremely proficient,” and was defined as “the degree of expertise, preparation, and knowledge that one perceives one needs to deal effectively with each of the situations described,” capturing perceived competence and preparedness. Perceived comfort with tasks was rated on a 5-point Likert scale, with the verbal anchors “Not comfortable at all” to “Extremely comfortable,” and was defined as “the feeling of being free from stress or worry,” reflecting a more emotional dimension of communication. Frequency of tasks was rated on a 5-point Likert scale, with the verbal anchors “Never” to “Several times per day” indicating how often the clinician engaged with the specific communication task in a working week. Level of training for each task was rated as either “No specific training received” or “Yes specific training received,” referring to an “organized instructional process (whether in a classroom or on the job) aimed at improving specific skills needed for a particular activity, for which the learner also received feedback.”
Statistical analysis
Responses were analyzed using SPSS version 27 (IBM Analytics). For each of the communication tasks, mean scores were calculated separately for perceived proficiency, comfort, and frequency. Learning experiences were dummy-coded as no training (0) and received training (1), and therefore, frequencies are reported. Scores for perceived proficiency and comfort were categorized as low (1–2.34), moderate (2.35–3.67), or high (3.68–5), and for frequency as rarely (1–2.34), sometimes (2.35–3.67), and a lot (3.68–5).
To retain conceptual clarity and interpretability, we performed analyses at the item level, but organized them into conceptually related communication domains based on thematic content and relevance to clinical PC practice, as determined by the authors. The 6 domains were used solely to structure the presentation of results and comprise the following: (1) understanding the other and building the relationship; (2) understanding the illness and decision-making; (3) communication about symptoms; (4) special situations; (5) the dying phase and the family after death; and (6) recognition and discussion of own emotions and self-care.
To assess associations between the aforementioned constructs, Pearson and Spearman correlations were calculated, and the strength of correlation was rated as low, moderate, or strong according to Cohen (Reference Cohen2013). Task-specific correlations deviating from the mean scores are highlighted. To compare physicians and nurses and between perceived comfort and proficiency scores, independent and paired samples t-tests were applied (Norman Reference Norman2010). The significance level in this study was alpha <0.05.
Results
Participants
A total of 231 participants (78.8% women), nurses (65%), and physicians (35%), aged 24–75 years (M = 50.11, SD = 10.31) completed the survey. The majority was German-Speaking (82.3%). Most participants came from Canton Zurich (20.8%), Bern (19.0%), and St Gallen (6.9%). Table 1 provides a summary of sociodemographic characteristics of the total sample.
Sociodemographic characteristics
Table 1 Long description
Sociodemographic and work characteristics are summarized using averages with standard deviations and counts with percentages. Participants averaged 50.11 years of age, with 24.88 years of health care experience and 12.39 years in palliative care. Gender was mostly female at 182 participants (78.8%), with 49 male (21.2%). Language was mainly German at 190 (82.3%), followed by French at 37 (16.0%) and English at 4 (1.7%). Workplace canton was most often Zurich at 48 (20.8%) and Berne at 44 (19.0%), with smaller shares across other listed cantons and a large “Other” category at 175 (49.5%). Occupation was more often nursing at 150 (64.9%) than medicine at 81 (35.1%). Most worked full-time at 151 (65.4%) versus part-time at 79 (34.2%). Patient population was primarily adults at 202 (87.4%), with smaller proportions in paediatrics and other groups. Clinical settings were varied, with “Other” the largest category at 73 (31.6%), and notable representation from hospital specialist palliative care units (46, 19.9%) and mobile palliative care teams (42, 18.2%). Percentages for workplace canton should be interpreted cautiously because only cantons with more than 10 people are listed and participants may work in more than one canton.
a Only Cantons are shown with more than 10 people; participants might be able to work in more than one canton.
Task-specific descriptive statistics
Overall, participants reported high perceived proficiency for 9 tasks, moderate perceived proficiency for 11 tasks, and low perceived proficiency for 1 task (M = 3.55; SD = 0.82). In terms of comfort, they also reported high comfort for 9 tasks, moderate comfort for 12 tasks, and low comfort for 1 task (M = 3.47; SD = 0.62). Across all tasks, clinicians reported the greatest perceived proficiency and comfort in communication about what PC is/does and the least perceived proficiency and comfort in advising parents what to tell their underage children when one parent is receiving PC. Perceived proficiency scores were, on average, slightly higher than comfort scores (t = 3.25; p = 0.001). Tasks were mostly performed a few times a week. Overall, training levels across tasks were rather low; fewer than half of the participants reported receiving formal training for most tasks, with only a few tasks surpassing 50%. The great majority reported that they had learnt their skills by practicing them (89%) or by seeing others (70%). Clinicians’ ratings for all tasks regarding perceived proficiency, comfort, frequency, and training can be found in Table 2, with the highest and lowest scores highlighted.
Descriptive results for perceived proficiency, comfort, frequency, training, and perceived proficiency vs. comfort
Table 2 Long description
Ratings summarize clinicians’ perceived proficiency, comfort, and how often they perform 19 palliative-care communication tasks, plus the share reporting training for each task. Scores are generally high for explaining what palliative care is and for discussing pain or other physical symptoms, with proficiency and comfort both around four on a five-point scale and training near two thirds. The lowest proficiency and comfort occur for advising parents what to tell underage children when a parent is receiving palliative care, with training reported by about one in ten. Other low-comfort, low-training areas include responding to requests for non-indicated treatments and communicating with patients in denial, each with training under one third and comfort below three. Several tasks show statistically significant gaps between proficiency and comfort, especially in challenging conversations such as non-indicated treatments, denial, requests to hasten death, and communication around impending death and immediate bereavement. Frequency tends to be moderate overall, highest for symptom communication and lower for special situations and bereavement follow-up. Across all tasks, average proficiency is about three and a half, comfort slightly lower, frequency under three, and training is reported for roughly nine tasks on average.
a For training across all tasks, average number of tasks in which training was received is given; perceived proficiency, comfort, and frequency on 5-point Likert scale ranging from 1 to 5; training percentage of people reporting specific training; lowest value per domain is highlighted in bold; 2-sided p-value for perceived proficiency vs. comfort, significant highlighted with *p < 0.05 and **p < 0.01.
Differences between perceived proficiency and comfort and between professions
Several significant differences were observed between comfort and proficiency. For example, perceived proficiency was significantly higher than comfort in communication about prognosis or life expectancy (t = 2.90, p = 0.004), with patients who are in denial about the gravity of their illness (t = 6.56, p < 0.001), and responding to patients who want treatments that one believes are not indicated (t = 8.08, p < 0.001), and in discussions about the patient’s request to hasten death (t = 6.28, p < 0.001). Comfort was significantly higher than perceived proficiency in communication about ACP (t = −2.40, p = 0.017). All comparisons between perceived proficiency and comfort can be found in Table 2.
Several profession-specific differences emerged across communication domains. Physicians generally reported higher perceived proficiency and comfort in illness- and treatment-related communication tasks, particularly discussions about prognosis, future treatments, treatment limitations, and communication with patients in denial (all p ≤ 0.043). Physicians also reported performing certain prognosis- and treatment-related discussions more frequently than nurses (both p < 0.001).
In contrast, nurses more frequently reported receiving training in emotionally oriented communication tasks, including discussions of patients’ or families’ feelings, advance care planning, spiritual issues, communication with families after death, and self-care-related communication (all p ≤ 0.041). Nurses also reported higher perceived proficiency in discussing spiritual issues (p = 0.046) and greater perceived proficiency and comfort regarding self-care practices to prevent burnout and compassion fatigue (p ≤ 0.029). Significant differences between professions across all tasks can be found in Table 3.
Significant differences between nurses and physicians across all tasks
Table 3 Long description
Significance tests compare nurses and physicians across palliative-care communication tasks for proficiency, comfort, frequency, and training. Physicians score higher on complex medical discussions, especially future treatments and stopping nonbeneficial interventions, and on prognosis or life expectancy, with higher proficiency, comfort, and frequency and also higher training for prognosis. Physicians also report higher proficiency and comfort when addressing denial, and higher proficiency when responding to requests for treatments they consider not indicated. Nurses report more training for explaining what palliative care is, advance care planning or goals of care, discussing feelings, psychosocial issues, spiritual issues, and family communication after death, including immediate and later follow-up. Nurses also report higher comfort about pain or other physical symptoms and more training for pain and spiritual communication. For patient values, physicians report higher comfort and frequency. Self-care differs: nurses report higher proficiency and comfort using self-care practices, while physicians report higher frequency of discussing their own feelings with colleagues. Only tasks with statistically significant differences are listed; direction indicates which profession scored higher .
Note: Positive t-test indicates physicians > nurses, negative t-test indicates nurses > physicians, non-significant findings appear in grey.
Associations between perceived proficiency, comfort, frequency, training, and participant characteristics
Perceived proficiency across all tasks was strongly correlated with comfort across all tasks. There was also a moderate and significant association between frequency, perceived proficiency, and comfort. Training had a moderate positive association with perceived proficiency, comfort, and frequency. Correlation coefficients can be found in Table 4, and correlations across all tasks can be found in Appendix Table A1.
Correlation between comfort, perceived proficiency, frequency, and training
Table 4 Long description
Correlations compare perceived proficiency, comfort, frequency of use, and training. Perceived proficiency has a strong positive relationship with comfort, about 0.79. Perceived proficiency is moderately related to frequency, about 0.40, and weakly related to training, about 0.28. Comfort is moderately related to frequency, about 0.38, and weakly related to training, about 0.22. Frequency and training also show a weak positive relationship, about 0.22. All reported relationships are positive, with the strongest pairing between proficiency and comfort and the weakest pairings involving training. Interpret results as associations rather than cause and effect, and note that training uses a rank-based correlation while the other variables use a standard Pearson correlation.
Note: Pearson correlation for perceived proficiency, comfort, and frequency, Spearman correlation for training, *p < 0.05 and **p < 0.01.
Regarding demographic variables, positive associations were found for age and experience in PC with perceived proficiency. Perceived comfort was positively associated with experience in the job, experience in PC, and working full-time. Frequency was significantly associated with the medical profession. Training was only significantly associated with experience in the job. A full overview of the correlations can be found in Table 5.
Correlation between comfort, perceived proficiency, frequency, training, and other sociodemographic characteristics
Table 5 Long description
Correlations link perceived proficiency, comfort, frequency, and training with age, gender, profession, job experience, PC experience, and work dedication. PC experience shows the strongest positive relationships, with higher PC experience associated with higher perceived proficiency and higher comfort. Frequency is most negatively related to profession and to experience in the job, indicating those groups report using it less often. Comfort rises slightly with job experience and with work dedication, while age and gender show only small relationships with all four outcomes. Training has small positive relationships with age and job experience and is near zero for gender and work dedication. Statistical significance is marked for some relationships, so small coefficients without significance markers should be interpreted cautiously.
Note: Pearson correlation for perceived proficiency, comfort, frequency, age, experience in job and experience in PC, Spearman correlation for training, gender, profession, and work dedication, *p < 0.05 and **p < 0.01.
Despite these overall patterns, several task-level associations deviated from the general trend, particularly regarding the relationship with training. In the domain understanding of the other and building the relationship neither perceived proficiency nor comfort in introducing one’s role was significantly associated with training (r = 0.12, p = 0.066; r = 0.13, p = 0.055), nor were perceived proficiency and comfort in communication about what PC is (r = 0.10, p = 0.161; r = 0.13, p = 0.062). In the domain of understanding the illness and decision-making, comfort in communication about early PC (r = 0.05, p = 0.458), advance care planning (r = 0.11, p = 0.108), and communication with patients who are in denial about the gravity of their illness (r = 0.11, p = 0.105) were not significantly associated with training. Similarly, perceived proficiency in responding to patients who request treatment which one thinks is not indicated had no significant association with training (r = 0.13, p = 0.051). When examining communication about symptoms, perceived proficiency and comfort in discussing pain or physical symptoms were also unrelated to training (r = 0.04, p = 0.573; r = 0.04, p = 0.532). As for special situations, perceived proficiency in discussion of a patient’s request to hasten death was not significantly associated with training (r = 0.11, p = 0.101). Finally, in the dying phase and the family after death, neither perceived proficiency nor comfort in communication about the patient’s impeding death (r = 0.09, p = 0.178; r = 0.12, p = 0.093), nor communication with families immediately after the patient has died (r = 0.09, p = 0.211; r = 0.11, p = 0.118) showed significant associations with training.
Discussion
We identified significant variability in Swiss PC clinicians’ comfort and perceived proficiency levels across different communication tasks. Clinicians reported higher perceived proficiency and comfort when communicating their roles and discussing physical symptoms, whereas notable gaps were seen particularly in guiding discussions between parents with cancer and their young children and addressing spirituality. This variability across tasks suggests that clinicians’ communication competence in PC is task-specific rather than a uniform skill.
The marked variability in comfort, perceived proficiency, and training across communication tasks underscores the need for targeted improvements. For example, while over 85% of clinicians felt comfortable discussing the nature of PC and physical symptoms, less than 36% were comfortable addressing spiritual aspects, and fewer than 20% felt competent or comfortable in guiding discussions about what patients can tell their children, as well as in other areas requiring similar high emotional engagement. Spiritual care remains a neglected component in patient care (Gijsberts et al. Reference Gijsberts, Liefbroer and Otten2019), despite the calls and approaches developed to facilitate discussions about spirituality. Similarly, the needs of parents with cancer who have young children for additional information and support highlight the importance for clinicians to proactively address the impact of the disease on their patients’ children (Inhestern et al. Reference Inhestern, Johannsen and Bergelt2021). However, many of our participants expressed not feeling comfortable or proficient with such discussions. This omission has been strongly associated with insufficient training, which underscores the need for enhanced educational efforts in this area (Balboni et al. Reference Balboni, Sullivan and Amobi2013), as well as for greater attention to clinicians’ emotional preparedness for these tasks, beyond content-level training.
The mismatch between perceived proficiency and comfort across several tasks, such as communication with patients who are in denial, discussion of the patient’s or family’s feelings, or the use of self-care practices to prevent burnout and compassion fatigue, suggests that communication competence in PC may encompass an emotional dimension in addition to knowledge and skills. Clinicians may thus know what to say and how to structure a discussion, yet still experience discomfort when engaging in it. This aligns with definitions of emotional labor in healthcare (Sorensen and Iedema Reference Sorensen and Iedema2009) and with work highlighting that emotional complexity is inherent to PC communication, highlighting how clinicians’ comfort and competence evolve through ongoing adaptation to complex emotions and learning through different means, including learning by doing, drawing on personal experiences, connecting with peers, and reflecting on their own and patients’ emotions (Niederhauser et al. Reference Niederhauser, Egloff and Eychmueller2025). Importantly, deciding to engage emotionally in clinical encounters requires not only skill but also a sense of safety and emotional confidence, which may help explain why clinicians can feel proficient yet remain uncomfortable in specific tasks (Moens et al. Reference Moens, Bilsen and Zambrano2026). Recognizing and supporting this emotional dimension in training may therefore be critical for fostering both communication competence and clinician wellbeing.
Building on this, the finding that the majority of clinicians reported learning communication skills primarily through practice or observation, with few reporting formal training for most tasks, underscores the need for structured educational interventions. While experiential learning clearly has value and our findings suggest that feeling comfortable evolves through experience, structured training remains essential. The moderate positive association between comfort, perceived proficiency, and training reinforces the potential of educational interventions and formalized training to strengthen communication competence in PC settings. However, in contrast to some reviews which show that there is only a small association between training and comfort (Chung et al. Reference Chung, Oczkowski and Hanvey2016), our study found a moderate association, suggesting that the type and quality of training might be crucial factors. Similar results have been observed in a sample of internal medicine and family practice physicians, where training in PC was linked to better perceived performance (Farber et al. Reference Farber, Urban and Collier2004). Therefore, integrating advanced communication training into medical, nursing, and interdisciplinary education and professional development is essential (Perron et al. Reference Perron, Sommer and Louis-simonet2015). Training methods such as simulation, role-play, and structured mentoring programs could be especially effective, particularly for the emotional dimension of communication (Rasenberg et al. Reference Rasenberg, Brand and van Weel-baumgarten2023). Additionally, previously investigated programs, such as the COMFORT communication curriculum, have been shown to be effective resources for teaching discipline-specific PC communication (Wittenberg et al. Reference Wittenberg, Ferrell and Goldsmith2016, Reference Wittenberg, Reb and Kanter2018, Reference Wittenberg, Goldsmith and Williams2020). Nevertheless, we also identified that for tasks such as communication about what PC is/does or communication about the patient’s impending death, training was not associated with perceived proficiency or comfort, which remains unclear. One possible explanation is that the training occurred a long time before participation, and its perceived benefits might have diminished over time, particularly for tasks not performed frequently. Alternatively, the content of the training may not have aligned well with the specific demands of the tasks in clinical practice, limiting its effectiveness (Berglund et al. Reference Berglund, von Knorring and McGrath2023). This result may also signal how certain skills develop more informally through reflection, observation, and accumulated experience, highlighting the complementary role of formal training and experiential learning in fostering communication competence (Berglund et al. Reference Berglund, von Knorring and McGrath2023; Niederhauser et al. Reference Niederhauser, Egloff and Eychmueller2025; Goloff et al. Reference Goloff, Sultanem and Fong-leboeuf2026).
Beyond formal training, our findings further underscore the importance of experience in developing communication skills. This is supported by previous research, such as Isaacson et al.’s (Reference Isaacson, Minton and DaRosa2019), which demonstrated a link between years of experience and greater comfort with end-of-life communication in nurses. These results suggest that greater clinical experience may contribute to higher comfort and perceived proficiency to navigate sensitive discussions effectively. This highlights the need for ongoing professional development and mentorship opportunities for less experienced clinicians, corroborating previous research that emphasizes the role of continued practice and mentorship in skill development (Kilminster and Jolly Reference Kilminster and Jolly2000; Hauer et al. Reference Hauer, Teherani and Dechet2005; Keinänen et al. Reference Keinänen, Lähdesmäki and Juntunen2023). The significant differences in comfort and perceived proficiency levels according to profession indicate a need for tailored training that addresses the unique challenges faced by different professional groups within PC or their distinct communication roles. For instance, physicians showed higher perceived proficiency and comfort in illness- and treatment-related discussions, while nurses reported more training and higher perceived proficiency in emotional, spiritual, and self-care communication, reflecting their distinct roles and responsibilities. Tailoring training to reinforce these strengths while addressing each profession’s gaps could strengthen interdisciplinary collaboration and communication competence.
Strengths and limitations
The main strength of this study is its detailed assessment of perceived proficiency and comfort, training, and frequency across 22 PC communication tasks. This provides a broad view of clinicians’ self-reported communication experiences, perceived proficiency and comfort, and challenges across a large and diverse sample spanning professions and linguistic regions in Switzerland. The study also has several limitations. The sample is specific to the Swiss PC context, and given the limited Swiss-specific literature on PC communication training, comparisons with other local studies are not possible. Additionally, the reliance on self-reported data using a new, not yet validated survey could lead to over- or underestimation of actual perceived proficiency and comfort levels. In addition, self-reported frequency ratings captured how often clinicians performed each task, not how often the task was required. Therefore, avoidance of certain conversations could lead to an underestimation of the actual demand for those communication tasks. This distinction is particularly relevant for emotionally demanding or ethically sensitive topics, where lower reported frequency may reflect avoidance rather than actual need. Finally, given the cross-sectional design and the number of statistical tests conducted, the results remain exploratory and descriptive rather than causal or confirmatory. Future research should focus on using longitudinal designs to track changes in the development of communication competence over time, the effectiveness of targeted interventions, and comparative studies across different countries or healthcare systems. Further qualitative studies could help us extend the knowledge on barriers impairing healthcare workers' perceived proficiency, competence, and comfort. Such research could provide deeper insights into how best to support PC clinicians in their communication roles and ultimately improve patient care.
Conclusion
PC clinicians perceive themselves as more comfortable and proficient communicating about their own role and physical symptoms, while training gaps remain around guiding discussions with young children or addressing spirituality. Differences between perceived proficiency and comfort levels across tasks, and their mismatch within specific tasks, underscore the important role of emotions in the development of communication competence in PC. These findings highlight the need for targeted training and ongoing support to enhance PC clinicians’ communication skills in sensitive and emotionally complex contexts. Our findings emphasize the importance of integrating both technical and emotional communication training into PC education. They may also inform training in other healthcare areas where similarly complex and difficult topics may need to be approached. Targeted interventions, including simulation, role-play, and mentoring, could strengthen clinicians’ confidence, emotional readiness, and ability to navigate sensitive conversations (Berglund et al. Reference Berglund, von Knorring and McGrath2023; Niederhauser et al. Reference Niederhauser, Egloff and Eychmueller2025; Goloff et al. Reference Goloff, Sultanem and Fong-leboeuf2026). Tailoring training to specific tasks and professional roles may further enhance communication competence and improve patient-centered care.
Data availability statement
The datasets generated and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.
Acknowledgments
We would like to thank all the clinicians who participated in the survey, as well as palliative.ch (the Swiss Society for Palliative Medicine, Nursing and Support) and Medworld for their support with participant recruitment. We also want to thank Prof. Steffen Eychmüller for his feedback and support during survey development and the data collection process.
Author contributions
R.S.: Methodology, formal analysis, data curation, writing – original draft, and writing – review and editing; M.E.: Methodology, recruitment, and writing – review and editing; S.C.Z.: Conceptualization, methodology, investigation, supervision, project administration, writing – original draft, writing – review and editing, and resources.
Funding
This work was funded by an Eccellenza Professorial Fellowship from the Swiss National Science Foundation (grant number: PCEFP1_194177) awarded to S.C.Z.
Competing interests
The authors declare that they have no competing interests.
Ethical approval
The Bernese Cantonal Ethical commission exempted the study from ethical review (Req-2020-00611).
Ethics and consents
Not applicable.
Appendix
Correlation across all Tasks and Proficiency, Comfort, Frequency and Training
Open access
