Learning how to talk about approaching death: results from a pre-test post-test trial on the effectiveness of a blended-learning approach

Abstract

Background

Caring for dying individuals is a key aspect of many clinical and medical practices. However, training and reflection opportunities on conversations between healthcare professionals, dying patients, and caregivers are limited. To address this gap, we developed a blended-learning approach focusing on effective communication about approaching death, termed Talking About Dying (TAD). This study aimed to evaluate the impact of this approach on pre- and postgraduate medical learners’ knowledge, anxiety, self-efficacy, and skills related to TAD.

Methods

A pre-test post-test trial with n = 12 medical students and n = 11 nursing professionals was applied. Data were collected at three time points: (1) at baseline (pre-intervention) (2), after a 45-minute online TAD module (post-intervention I), and (3) after a 90-minute onsite workshop with experts (post-intervention II). Knowledge was tested with self-developed multiple-choice questions, while anxiety when discussing death and self-efficacy were assessed using published questionnaires, i.e., the 7-item Thanatophobia Scale and a 1-item scale according to Bandura. Participants demonstrated their TAD-related skills in encounters with simulated patients, which were scored by trained assessors using a self-developed scale. The TAD online module (intervention I) included text and video demonstrations with critical hints, while the onsite workshop (intervention II) involved group reflection, discussion, and role-play to deepen understanding.

Results

Participants demonstrated increased knowledge, reduced thanatophobia (amongst medical students but not nursing professionals), and enhanced self-efficacy and skills following both interventions, with these effects being both statistically and practically significant.

Conclusions

Combining an online module with an in-person workshop can meaningfully enhance learning outcomes across cognitive, affective, and behavioural dimensions related to TAD conversations for both medical students and professional nurses, though changes in the affective dimensions were observed only in students. We recommend integrating this method into medical and nursing education.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12909-025-07178-8.

Background

In the past decade, the importance of communication competencies in clinical practice, particularly within end-of-life care, has become increasingly recognized, resulting in a greater need for improved professional preparation and the integration of these competencies into educational programs [1–5]. Hence, to support acquisition of relevant communication skills, various communication models and educative tools have been developed to support (future) clinicians to effectively communicate with patients, e.g., with regards to breaking bad news [6, 7], the exploration of spiritual aspects [8], or serious illness conversations [9–11]. However, despite these advances, training in communicating with patients and relatives about approaching death, hereafter referred to as Talking About Dying (TAD), remains as a critical gap [12].

Although caring for dying individuals is an integral aspect of many clinical areas and medical practice, particularly in oncology care, training and reflections on conversations between health professionals and dying patients and their caregivers during medical and nursing education is almost nonexistent [13, 14]. As a result, physicians and nurses lack sufficient training in how to communicate about approaching death [15] and thus often feel unable to initiate corresponding TAD conversations [16]. Nurses and physicians also perceive having different roles in TAD conversations. Although TAD has traditionally been regarded as the responsibility of physicians [17, 18], nurses play a significant role due to their continuous interaction with patients at the end of life and their caregivers. However, just as physicians, nurses frequently lack training and clarity regarding their roles in TAD discussions [19].

In addition to the lack of training and their perceived role in TAD conversations, different barriers to these conversations have been identified. First, because of the possible impact on all persons involved and the uncertainties that remain, TAD conversations tend to be avoided [20, 21]. Second, discomfort with TAD conversations includes many aspects, as fears of causing negative psychological impact, distress to the patient and family members, confrontation with demanding intense emotions, uncertainties regarding prognosis, and not knowing what to say [22, 23]. Third, healthcare professionals as well as patients are described as having a hard time accepting the finality of life, generally due to the emphasis on a curative approach in medical education and practice [24]. Even oncologists have difficulties accepting that the limits of treatment effectiveness have been reached, and thus it is not unusual for patients to be unaware that they are dying. Instead, patients may risk spending their last days with burdensome interventions aimed at prolonging life [9, 25].

Many of these barriers have also been observed among medical and nursing students. A focus group study with nursing and medical students revealed students reporting not knowing what to say, or how to deal with emotional responses, concerns about wasting the patient’s time, and concerns about their ability or how to cope with distressing experiences [23]. Consequently, TAD conversations remain one of the most challenging communication tasks for healthcare students and professionals [5, 26–28]. This may result in a strong desire to avoid addressing the issue of approaching death [29].

Even though health care professionals are expected to address approaching death in a timely, honest, and needs-oriented manner [5, 22], death often occurs rather “unexpectedly” potentially having a negative impact on all involved [30–33]. Leaving families and patients unprepared for death may contribute to profound distress shortly before the death and during bereavement [25, 30, 32, 33]. Health professionals who do not feel prepared for a TAD conversation, may experience dissatisfaction, increased stress, and may be at a higher risk of burnout or illness [16].

To address this gap in communication skills, we developed a TAD communication model that outlines precise steps related to the clinical TAD task (see [12]). Its development followed an integrative and systematic approach, incorporating input from key stakeholders, i.e., palliative care specialists, communication experts, and patient representatives. The resulting model provides a structured framework and communication aids, including practical tools to guide interactions through three distinct phases: before (diagnosing dying, preparation), during (the TAD task itself with specific conversational steps), and after the conversation (post-processing and debriefing). The model also details the conversation content, relational aspects, and essential dimensions for effectively discussing the approach of death, emphasizing the importance of self-awareness and self-care for healthcare professionals in the care of dying patients. We implemented the TAD model as an online learning module (see Fig. 1 and refer to Section “Learning intervention I: TAD online learning module” for further details).

Fig. 1.

Illustration of the TAD communication model (left) and the TAD online learning module (right). The learning module includes background information on TAD, learning objectives, text content outlining the precise steps of the TAD communication model, and audio-video demonstrations presented as worked examples that show how to apply the seven TAD steps during an actual conversation (played by actors). The videos are accompanied by brief text hints highlighting critical content without introducing new information

Evidence shows that challenging communication skills in the clinical context can be acquired using didactically sound blended-learning approaches [6–7, 34].

Blended learning accounts for a mix of online and onsite learning activities, scaffolding a learner’s development with carefully selected learning technologies, teaching methods, feedback and reflection activities that map out a learning journey for targeted learners [35]. There are three models for how to combine the online and onsite training in a blended-learning setting as outlined by Dakhi et al. [36]: (a) in the rotation model, learners rotate between onsite and online activities; (b) in the flex model the main learning content is normally offered online, and various learning activities are offered onsite, as a free choice (e.g. group work, discussion, short presentation); (c) the self-blend model is characterized by a combination of online and in-person teaching, typically in a sequential order. The latter model closely aligns with our implementation of the TAD-related blended-learning approach, in which learners first study theoretical background information and the specific steps of the TAD model by completing an online learning module individually, then come onsite to practice and refine these skills together with peers and experts. Within this framework and based on the mentioned lack of training, we formulated the following research questions.

RQ1: Cognitive dimension

to what extent does the TAD blended-learning offer enhance learners’ domain-specific knowledge about TAD?

RQ2: Affective dimension

to what extent does the blended-learning offer influence learners’ anxiety and self-efficacy towards TAD?

RQ3: Behavioural dimension

to what extent does the blended-learning offer impact learners’ skills performance related to TAD?

Methods

To investigate our research questions, a pre-test post-test trial was conducted. Data was recorded at three points in time, i.e., (i) at baseline (pre-intervention), (ii) after participants completed the TAD learning module (post-intervention I), and (iii) after they underwent the onsite workshop with peers and experts (post-intervention II). Figure 2 illustrates the study flowchart. At each of the three test points, a conversation with a simulated patient (SP) (see Appendices B-D) was included for two reasons: first, to assess the participants’ TAD skills, and second, to incorporate this type of training into the blended-learning concept.

Fig. 2.

Flow of participants

The study took place in the skills lab of the Medical Faculty at the University of Bern, Switzerland. We prepared dedicated learning and test rooms, which were equipped either with computers featuring internet access, mouse, keyboards, and monitors for completing the cognitive and affective assessments (see Section “Cognitive assessment: knowledge test” and following), or with SPs, cameras and microphones for recording the behavioral assessments (see Section “Behavioral assessment: performance test”). In Learning Intervention I, in which participants studied the online TAD module, they learned and were assessed individually (see Section “Learning intervention I: TAD online learning module”). After completing Learning Intervention II, during which they engaged with onsite experts and peers to discuss the importance of TAD and share their experiences, participants were also asked to provide evaluations of the blended-learning approach.

Sampling1

A total of 24 participants agreed to participate and provided written informed consent for the collected data to be used for research purposes. One participant was omitted from analyses due to insufficient language skills. Thus, the data from a definitive sample of N = 23 participants were integrated into analyses, with n = 12 medical students from the 4th and 5th study year (10 females and 2 males, aged between 23 and 35 years (M = 24.7, SD = 3.3)) and n = 11 nurses with experience2 in caring for dying patients (11 females, aged between 30 and 49 years (M = 41.3, SD = 6.9)). All participants confirmed that they did not have contact with the online learning materials prior to participation.

Pre- and post-intervention assessments

The pre- and the two post-intervention assessments comprised tests that were related to cognitive, affective and behavioral dimensions. Each assessment lasted approximately 15 min and is described in the following sections.

Cognitive assessment: knowledge test

We assessed participants’ knowledge about TAD with a self-developed questionnaire with seven multiple true-false and two a-positive questions on how to care for dying patients and their relatives (Appendix A). These questions were aligned with the TAD communication model. The knowledge test was developed by a content expert in palliative care familiar with the TAD model and reviewed for technical flaws by two communication experts and two specialists in medical assessment.

The knowledge test was completed before and after each of the two learning interventions. The resulting three knowledge-test scores could contain between 0 and 36 points each (four points per items could be reached). We collected participants’ replies in the UNIPARK survey tool (http://www.unipark.info) from the computer in the dedicated learning and test rooms.

Affective assessments: thanatophobia and self-efficacy

We assessed thanatophobia, i.e., the participants’ anxiety towards caring for dying patients, with the German translation by Heintz et al. [37] of the 7-item Thanatophobia Scale [29]. It is described as a “valid and reliable assessment scale that may be of use when evaluating the impact of an education program” [38]. The items cover negative attitudes such as uncomfortable feelings and sense of helplessness, therefore, lower scores reflect lower anxiety or higher self-efficacy levels associated with the TAD communication task. Participants’ replies were collected in the UNIPARK survey tool.

In contrast to the other assessments, the Thanatophobia assessment was only provided twice, i.e., before and after the onsite workshop (intervention II; see Fig. 2). We decided to not present it in-between, as we assumed that it is related to a general anxiety-evoking state-trait, rather than an immediate reaction to each of the interventions (cf. [39]).

The assessment of the participants’ TAD-related self-efficacy was performed with a single item scale —based on Bandura [40, 41]— indicating how confident they currently felt in their ability to perform a TAD conversation (see Section “Behavioral assessment: performance test”). They used a slider in the UNIPARK survey tool, allowing their response to range from 0 to 100% confidence.

Behavioral assessment: performance test

Participants demonstrated their TAD-related communication skills performance in conversations during encounters with an SP. These performance tests took place at three points as indicated in Fig. 2. The specific task was to discuss the approaching death with the SP using the 7 steps of the TAD model, which had been taught during the learning intervention I (see Section ;“Learning intervention I: TAD online learning module”). The goal was to apply these steps as effectively as possible, such as by identifying the patient’s concerns and ensuring that appropriate care could begin immediately in the dying phase, including support for the relatives. Each SP was carefully trained to respond with statements such as, “What if I die and no one is there with me?”, “I’m afraid the pain will get worse as I get closer to the end,” and “I don’t know what to expect. I’m scared of what comes next.”

Each performance assessment applied one of three patient scenarios (Appendices B-D) that were counterbalanced in their order of appearance to avoid rank order effects. Performance assessments took place in rooms with two chairs or a bed respectively, which were arranged in the two preinstalled cameras’ focal point. Each performance test was video recorded.

Three outcome raters (Masters-level psychology students) independently rated every participant’s skills performance. The raters used a self-developed scale for this purpose (TAD Assessment Scale; Appendix E). In addition, they used the ‘global Breaking Bad News Assessment Scale’ (glBAS; [42]). Following the procedure from Schmitz et al. [34], the three raters were given adequate training in how to use the instruments. They were blinded to the point of measurement (pre- vs. post-intervention I and II), the learning interventions, and our research questions. The three raters were only informed about the study purpose when they had completed the entire scoring. Each participant’s final TAD and glBAS scores were calculated as grand means based on the assessors’ ratings. Mean scores between 1 and 5 points could be achieved on both scales, the higher the scores the better the skills performance.

Learning intervention I: TAD online learning module

We gave participants 45 minutes to complete the online TAD learning module, during which they individually studied the entire TAD model, including the steps to be performed before, during, and after each phase of a respective conversation (see Fig. 1). The main learning objective of the module was to introduce the TAD model and lay the cognitive foundation for the practical application of its seven steps during patient encounters. The specific task was to carefully read the content presented in a linear fashion and attentively watch the worked examples provided at the end of the module.

In line with this, the seven conversational steps of the TAD model were demonstrated through audio-video demonstrations, structured as worked examples. Worked examples typically include an introduction to the given problem (in this case, a patient suffering from a severe illness), the task (a TAD conversation), and a demonstration of each step required to achieve the task (i.e., applying the TAD steps as defined in the model; see [43]). Our worked examples were accompanied by text hints highlighting critical actions. According to Lorch [44], hints must not introduce new content but should only reference information already provided. Educational tools scaffolded in this manner have been shown to support the initial learning of complex clinical skills (e.g., [6, 8, 45]). We provided two worked examples, each with a duration of approximately 10 minutes. The first example featured an actor in the physician role and another in the patient role, while the second example showed an actor in the nursing role and another in the patient role. Both examples were accompanied by nine hints each. All content presented in the module was in German.

Learning intervention II: TAD onsite workshop

The learning intervention II included an onsite workshop, addressing the importance of TAD, also aiming at reflecting the experiences of the participants. The main aims of this part were to encourage participants to reflect on the end of life in contemporary medicine and on their own communication style. The workshop lasted 1.5 h and was based on the following learning objectives: the participants (1) reflected on the conceptual framework of dying as a potential failure of medicine, (2) their personal exposure to and experience of dying and death, (3) their personal communication style (in private and as a health care professional), and (4) their role in the health care team to identify team coping strategies in relation to TAD conversations. The methods used for this part consisted mainly of group reflections, plenary discussions and short role plays. At the end of this session participants were asked to indicate the impact it had on their overall course experience.

Statistical analysis

The statistical analysis was carried out using the Statistical Package for Social Sciences (SPSS) version 28.

To assess the reliability of the knowledge test, we calculated bivariate correlations using Pearson’s correlation coefficient (r) to determine test-retest reliability. Correlations below 0.5 indicate weak reliability, values between 0.5 and 0.7 suggest moderate reliability, and correlations above 0.7 are considered acceptable.

To determine the three raters’ agreement in performance ratings, we calculated intraclass correlations (ICC). According to Koo et al. [46], ICC measures below 0.5 imply low, from 0.5 moderate, from 0.75 good, and from 0.9 excellent agreements.

To determine differences across students and nurses, we either used Friedman or Wilcoxon tests for paired samples and, if applicable, appropriate pairwise comparisons according to Dunn [47]. To determine differences between students and nurses, Mann-Whitney U tests were calculated. Statistical significance was determined at the 5% level. Effect sizes were determined using Pearson’s r. According to Cohen [48], coefficients between|r| = 0.1 and|r| = 0.3 imply small, between|r| = 0.3 and|r| = 0.5 medium, and|r| > 0.5 large effects.

Results

Cognitive assessment outcomes

The test-retest reliability of the knowledge test was moderate (r = 0.63, p < 0.01). The knowledge-test results are shown in Fig. 3. The Friedman test showed that they differed significantly between measurement points (χ² = 29.66, df = 2; p < 0.001). The Dunn-Bonferroni tests showed that, compared to the pre-test, participants attained significantly higher knowledge-test scores after undergoing learning intervention I (|z| = 4.01, p < 0.001; r = 0.84) and learning intervention II (|z| = 3.61, p < 0.001; r = 0.75). No further effects were detected.

Fig. 3.

Boxplot of the knowledge-test results as derived from the cognitive assessments

Affective assessment outcomes

The thanatophobia scores are presented in Fig. 4. The Wilcoxon and the Mann-Whitney U tests did neither reveal effects across the groups (W = 65, p = 0.135) nor between the groups, respectively (U_{pre-interventions} = 36.5, p = 0.069; U_{post interventions} = 86, p = 0.235). However, when focusing on student participants only, they showed a statistically significant decrease in thanatophobia towards the task after the onsite workshop (post-intervention II) (W = 3.5, p < 0.01; r = 0.76). This effect was not found for the nurses (W = 30.5; p = 0.342), whose thanatophobia scores were rather low throughout the whole experiment.

Fig. 4.

Boxplot of the thanatophobia scores as derived from the affective assessments

The TAD-related self-efficacy scores are shown in Fig. 5. The Friedman test revealed that they differed significantly over time (χ² = 24.87, df = 2; p < 0.001). The Dunn-Bonferroni tests showed that participants reported significantly higher self-efficacy scores after undergoing learning intervention I (|z| = 2.51, p < 0.05; r = 0.52) and learning intervention II (|z| = 4.69, p < 0.001; r = 0.98 and|z| = 2.29, p < 0.05; r = 0.48). Notably, the increase among the student participants was steeper, while the nursing participants started with comparably higher self-confidence. Thus, the target populations differed significantly from each other at the pre-intervention time point (U = 110.0; p < 0.01, r = 0.57).

Fig. 5.

Boxplot of the self-efficacy scores as derived from the affective assessments

Behavioral assessment outcomes

The three outcome assessors rated the participants’ skills performances similarly; their rater agreement was excellent with regards to both the TAD scale (ICC = 0.94) and the glBAS (ICC = 0.91).

Figure 6 illustrates the resultant performance-test outcomes. The Friedman test showed that the TAD scores differed significantly between measurement points (χ² = 18.63, df = 2; p < 0.001). This was also true for the glBAS scores (χ² = 18.36, df = 2; p < 0.001). The Dunn-Bonferroni tests showed that, compared to the pre-tests, participants attained significantly higher TAD and glBAS scores after undergoing learning intervention I (|z| = 3.91, p < 0.001; r = 0.82 and|z| = 4.06, p < 0.001; r = 0.85) and learning intervention II (|z| = 3.39, p < 0.001; r = 0.71 and|z| = 3.02, p < 0.001; r = 0.63). No further effects were detected.

Fig. 6.

Boxplot of the performance-test scores as derived from the behavioral assessments. Top: TAD score. Bottom: glBAS score

Onsite workshop: participant feedback

The participants’ evaluation on the blended-learning approach is shown in Table 1.

Table 1.

Participant’s’ feedback on the blended-learning offering

Comment	Group	Online module	Onsite workshop	Whole learning offer
I found the combination of theory and practical part with SPs very instructive, as the reading could be applied right away—not like usual, where theory is some days before practice.	Student	x	x	x
Many thanks for the practical day with great insights into this important topic.	Student			x
Conversations are very helpful for applying what was learned. Exchange with peers and experts were exciting and educational.	Student		x
Very good teaching, including practical application. Important topic!	Student		x	x
A great and educational module on an important topic that is rarely discussed in studies. It reduced my fear of discussing this topic.	Student	x	x	x
Great offering! Varied with theory, discussions, solo, and group exercises.	Student	x	x	x
Excellent course with self-assessment and SP sequences. More communication training would be beneficial.	Nurse	x	x	x
A great setting and so important! I really enjoyed it! The SP setting requires courage, but it’s worth it! The theoretical input was helpful and manageable. Please provide a handout.	Nurse	x	x
It was a fantastic day, from organization to thoughtful details, treats, and, of course, the content. Big thanks to everyone!	Nurse	x	x	x
It was a fascinating, varied, and intensive day. I felt tired by the third SP session with too much information to process. As I digest the new knowledge, I think it will greatly benefit my practice.	Nurse		x	x
Very interesting and educational! Thank you! Great topic and structure. I gained more confidence in discussing death from the SP conversations. It was well-organized and enjoyable. I also found it exciting that a communication expert was involved.	Nurse		x	x
Excellent course! I learned a lot, and the instructors were extremely empathetic, giving emotions the right amount of space. Only suggestion: more time for the learning sequence.	Nurse			x

Discussion

Our findings yielded valuable insights that will be used to answer the research questions (RQs), which are discussed in turn. The RQ 1 addresses whether the blended-learning approach results in increased domain specific knowledge. Domain specific knowledge is the pre-condition for the skills performance, i.e., applying the steps of the TAD model. We found both, an increase of TAD-related knowledge scores between the pre-intervention and after participants completed the online learning module (Intervention I), and after they underwent the onsite workshop with peers and experts (Intervention II). The quality of onlinelearning content presentation is known to be a paramount condition for effective learning. This is best achieved when it is rooted in a conceptual framework [49]. Online learning content must be logically structured, and within the cognitive capacity of the learners [43]. It is confirmed in uncountable studies (e.g., [50]) that learning is hampered, when such principles are violated. The present results show that it was possible to internalize the steps of the TAD model in a relatively short learning time (i.e., 45 min), which indicates that the quality of the presented material have met the above criteria.

The RQ2 addresses another important aspect, also pre-conditional to be able to successfully engage in a TAD conversation: the thanatophobia level needs to be sufficiently low to, first, enact in a conversation at all, and second, to be able to interact and communicate in a meaningful way. It is known that anxiety is an effective block to retain and apply information from memory [51]. So, in combination with the beneficial outcome on the cognitive level (as addressed in RQ 1), we reason that the significantly decrease in thanatophobia for the students, was an important pre-condition for their skills performance. The nurses, on the other hand, did not show a significant change in this score over the measurement points. This is not surprising since they were already familiar with the task based on their work experience. Even if they needed to apply a new model for the TAD conversations with the SPs, it was not experienced as threatening. In line with this finding, we also found that their self-efficacy score was significantly higher at baseline compared to the students. These related measures strengthen the validity of our results, as the data pointed in the same direction. Also, other researchers show that these two measures are related [52, 53].

The RQ3 addresses the important transfer of theoretical knowledge to practical application. This is a cornerstone in a blended -learning setting and particularly in the clinical setting addressed in this study. Our results clearly show that the increase in the skills performance from pre- to the post tests was significant. This has several implications. First, participants could adapt to and apply a new communication model appropriately. Obviously, the internalization of the new learning content did not intervene with their previous experience in an unintended manner. Second, the theoretical knowledge could be transferred and applied to the practical tasks. There were no detected differences between the post-interventional assessment points (i.e., between the online and the onsite part of the blended approach). Third, the self-developed TAD assessment scale, which is aligned with the TAD model, elicited comparable results as the glBAS scale. The TAD model captures underlying principles that were measurable with this validated, pre-existing scale that addresses less immediate fatal braking bad news tasks (however a similar task).This supports the validity [54] of our developed TAD Assessment Scale.

This study implemented blended learning on different levels. The online learning content was represented with the online implemented module of the TAD model. The onsite activities were represented by the (a) three SP-communication tasks, which are repeated practical onsite training units throughout the course; and (b) with the workshop including a final discussion session addressing personal experiences and emotional aspects. Our results demonstrate how important the steps from theoretical knowledge into practice and application of knowledge is [55]. The lack of an explicit or “boostering” effect of the onsite workshop on assessment scores does not indicate that it has low impact in general, since the onsite part was always provided after the online part. On an emotional level, participants indicated in the discussions that they found this session important for reflecting on the training and their experiences before leaving the course with all its potentially emotionally challenging content.

Despite the interesting results, potential limitations exist regarding the validity of the self-developed instruments. The knowledge-test items and the TAD Assessment Scale have yet not undergo validation across different healthcare settings or among further professional groups, potentially limiting their generalizability. Future research should prove the applicability of these instruments used across various clinical contexts. A further limitation of our study is that we could not include a control group. Adding a control group, however, would have allowed us to control for even more threats to validity. Another limitation is that we were unable to examine long-term effects and therefore cannot provide insights into optimal learning retention. The post-tests were administered immediately after the learning module’s and after the onsite workshop’s completion, primarily for practical reasons. This timing accommodated the busy schedules especially of the nursing professionals’, ensuring full participation and allowing us to assess immediate learning outcomes. We acknowledge that this approach captures only short-term effects, without addressing the retention of knowledge and skills over time. Future studies should therefore focus on investigating long-term effects.

Conclusions

Our blended-learning TAD intervention was implemented with an online learning module based on an evidence-based model for TAD conversations and an onsite workshop with experts. We showed that theoretical knowledge could be successfully applied in practical TAD conversations with SPs (onsite part). Increase in knowledge and competence was found for two different target groups (master students in medicine and professional nurses). The blended-learning approach seemed beneficial for both professions most relevant in the care of dying patients and their caregivers. Furthermore, the onsite workshop supported reflections and discussions about the profession’s role in TAD. Our results show that training on this challenging topic is possible across various levels of expertise and that theoretical knowledge with worked example consistently can be transferred into practical application. We also found that our intervention had positive impact on TAD-related affective dimensions, i.e., thanatophobia and self-efficacy, but only in the student group, as initial thanatophobia levels were comparatively low and self-efficacy levels relatively high in the nursing group. Our study encourages health profession-educators to proactively address and teach TAD in both pre- and postgraduate education.

The blended-learning approach can be integrated into healthcare education, also in an interprofessional setting, to better prepare (future) healthcare professionals for TAD conversations. By combining online modules with in-person workshops, this method improves not only knowledge and skills but also helps reduce anxiety when discussing death and increase confidence in handling sensitive TAD discussions. Future studies should prove, whether this approach can be adapted for different healthcare settings to enhance compassionate communication with dying patient and their caregivers.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

TAD

Talking About Dying

SP

Simulated Patient

Fig.

Figure

glBAS

global Breaking Bad News Assessment Scale

SPSS

Statistical Package for Social Sciences

ICC

Intraclass Correlation

Author contributions

FMS led the conceptualization and methodological analysis of the study, managed data recording, and conducted data analysis, interpretation, and visualization. He drafted the manuscript with SJF, SCZ, and SG. ALB assisted with planning and overseeing data recording. She was responsible for the overall scheduling of the study in the skills lab (project administration). SJF played a key role in developing the TAD online learning module (intervention I) and conducting the onsite workshop (intervention II), led participant recruitment, and helped draft the SP cases. SCZ played a key role in developing the TAD online learning module (intervention I) and was responsible for the theoretical background of the paper. TH contributed to study planning and implementation in the skills lab, managing SP training, scheduling, and debriefing. KPS selected and co-trained the SPs. SE managed funding acquisition, developed and conducted the onsite workshop (intervention II) and supported manuscript validation. SG supervised all phases, contributed to funding acquisition, and ensured rigor in the final draft. All authors reviewed and approved the final manuscript.

Funding

This work was supported by the Swiss Cancer Research Foundation (KFS-4522-08-2018).

Data availability

The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study does not fall under Article 2, Paragraph 1 (Scope) of the Swiss Federal Act on Research Involving Human Beings (Human Research Act, HRA). Thus, the independent cantonal Bernese ethics committee declared that authorization was unnecessary and deemed the study exempt from full ethical review (REQ-2019-00977). We complied with international Good Practice Guidelines (GPG) for research involving human beings. The participants provided written informed consent for the material to be used for research purposes.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.