Objectives The objectives of this study were to explore medical students' experiences of communicating with patients and their carers about death and dying, and to assess whether using high-fidelity simulation improved students' confidence in discussing cardiopulmonary resuscitation.
Methods This qualitative study was carried out at a hospital in the south of England. Participants were 7 final-year medical students. Tutorials were developed using high-fidelity simulation to teach communication skills regarding discussion of cardiopulmonary resuscitation with patients and carers. Scenarios involved a simulated ward environment, a high-fidelity simulation mannequin and actor playing the role of a carer. Data were collected through joint interviews carried out by one researcher which were audio recorded and transcribed verbatim. The same researcher analysed the data using framework analysis.
Results Students reported a lack of experience observing conversations with patients and carers about death and dying. They also reported a lack of opportunities to interact with dying patients during their training. Barriers reported by students included healthcare professionals' and patients' attitudes. Students reported a lack of confidence and preparedness to have consultations with patients and carers about death, dying and cardiopulmonary resuscitation as junior doctors. They perceived role-play scenarios observed by their peers to be stressful, and this detracted from the learning experience. Students reported that the high-fidelity simulation scenarios were more realistic than low-fidelity (‘role-play’) scenarios. This improved the learning gained from the sessions and improved confidence among some students.
Conclusions This study has suggested that high-fidelity simulation may be a useful adjunct for undergraduate communication skills training in palliative medicine. Further research is required to assess whether improvements in confidence described by students in this study translate to discernible improvements in competence, and whether using high-fidelity simulation in this setting is cost-effective.
Statistics from Altmetric.com
Effective communication regarding death and dying is a crucial skill for doctors to develop.1 As figures of continuity on the wards, it is often the most junior members of the team who are called on to have these conversations with patients and carers.2 However, there is evidence to suggest that junior doctors feel underprepared to discuss death and dying including cardiopulmonary resuscitation.3–6
Most medical school curricula rely on role-play to teach communication skills regarding death and dying.7–9 However, there is evidence to suggest that learning from these scenarios is limited by the presence of peers and by artificial classroom environments.10 ,11 This is reflected in the continued lack of confidence among junior doctors in this area, despite the widespread adoption of role-play based training methods in medical education.12 ,13
One potential alternative to role-play techniques may be high-fidelity simulation, which has been widely adopted as an effective method to teach technical and clinical skills. ‘High-fidelity’ simulation can be defined as ‘the use of technologically lifelike manikins with provision for a high level of realism and interactivity’.14 Several preliminary studies in the nursing literature have suggested that high-fidelity simulation may be a promising technique for use in palliative care communication skills training.15 ,16 However, there is currently a lack of evidence regarding the efficacy and acceptability of high-fidelity simulation in this context.16 ,17
The aims of this study were to explore students' experiences of communicating with patients and carers about death and dying, and to assess whether using high-fidelity simulation improved students' confidence in discussing cardiopulmonary resuscitation.
Criterion sampling was used to select final-year medical students from the same university on placement in a hospital in the south of England (n=36). This was chosen in order to ensure that students had a similar level of experience, so their narratives could be compared and contrasted. All students were at the same stage of their attachment, having completed their palliative care and oncology teaching week and associated clinical activities. These comprise lectures, seminars, ward rounds and oncology clinics. Students were recruited to the study via email from the undergraduate administrator. They signed up to the tutorials on a voluntary basis as an addition to the core curriculum. Initially, 16 students signed up to the sessions. However, only seven students took part in the sessions as the remainder of the students withdrew from the sessions before they took place. In total, there were three sessions with 2–3 students in each session. The sessions took place in September and October 2014.
The setting for the tutorials was a mock ward environment in the simulation suite. The scenarios were designed for one student, a professional actor playing the role of a carer and ‘SimMan’ (a high-fidelity simulation mannequin) as the patient. The students did not have to assess the patient clinically as the focus of the scenario was communication skills. Unlike a standard mannequin, SimMan was able to breathe, cough, make groaning noises and speak. The focus of the communication interaction was with the patient's carer, but in some of the scenarios the students also communicated with SimMan as the ‘patient’. The simulation suite was set up to feel like a ward environment (with a hospital bed, oxygen, resuscitation trolley, blood gas machine and other equipment), rather than a classroom environment. Each student participated in a different scenario involving discussion of a ‘do not attempt cardiopulmonary resuscitation’ (DNACPR) decision (see online supplementary appendix 1). Each scenario lasted for ∼15 min with ∼20 min for the debrief after each scenario. They were observed in a video-linked debrief room by 2–3 peers and 1–2 facilitators (one of whom was a palliative medicine consultant). The consultant helped to facilitate the debrief and discussion and answered students' questions. There was no formal teaching as part of the session.
Joint interviews between 2 and 3 participants were used for data collection. A topic guide was used to structure the interviews, with initial questions to start the discussion, and further questions to stimulate further discussion if required. Each joint interview lasted between 20 and 30 min. Interviews were audio recorded and the data were transcribed verbatim by a single researcher. The data were transcribed anonymously and the audio data were subsequently deleted. Despite the small sample size, many of the same themes and ideas were mentioned in each of the joint interviews with few new themes emerging, suggesting that data saturation had been reached.
Framework analysis was used, which consists of a matrix or ‘frame’ to display cases and codes of summarised data.18 The four titles for the framework grids were chosen on the basis of key concepts identified from a literature review performed by the authors before the start of the study. Within each grid, each student formed a row, and columns contained subthemes. ‘Nvivo’ software was used to code data according to the themes and subthemes.
Ethical approval was sought from the University of Bristol ethics committee; the decision was made by the committee that ethical approval for the study was not required. Written consent was obtained from the students to participate in the study prior to the start of the study.
Of the seven students who took part in the study, six were female and the mean age was 22.1 years. There were four key themes, namely lack of experience of discussions about death and dying, lack of confidence and preparedness for practice, the impact of stress on learning and realism in role-play and simulation scenarios.
Theme 1: Lack of experience and barriers to participation in discussions about death and dying
Students described a lack of opportunity to observe consultations about death and dying including cardiopulmonary resuscitation with patients and carers. Several of the students described situations in which they had been discouraged by supervising doctors from observing or participating in consultations in which death and dying were discussed. One student commented on an experience during a hospital placement:
‘I remember saying you know like I'll ask like can I go and see that patient and they're like no, no, no, they're dying…’. (Student 7)
Other students suggested that they had felt uncomfortable sitting in on consultations regarding cardiopulmonary resuscitation and end-of-life care, and that they had been asked to leave the room by patients and carers.
Students described little dedicated time spent in specialist palliative care settings, except for one student who had arranged time in a hospice during Student Selected Component (SSC) time. The majority of their experiences of palliative medicine were during general practice and emergency department settings, and these experiences were largely ad hoc, as one student described:
‘It becomes almost kind of almost by chance that you'd come across someone that was dying’. (Student 6)
Three of the students (students 4, 5 and 7) reported that they had never witnessed a conversation between a doctor and patient or carer in which dying, cardiopulmonary resuscitation or end-of-life care were discussed.
Theme 2: Lack of confidence and preparedness for practice
Five of the students (students 2, 3, 4, 6 and 7) described a lack of confidence in communicating with patients and relatives about death and dying including cardiopulmonary resuscitation.
I think I am more scared and it's kind of a fear of the unknown I suppose’. (Student 7)
Six of the students (students 1, 3, 4, 5, 6 and 7) commented on the implications of their lack of experience of end-of-life communication skills for their future practice as doctors. One student explained:
‘It's not something you want to get wrong. Everyone has a hideous story of someone who they know who got told something in a horrendous way like you don't want to be that person who told someone something awful’. (Student 5)
Students 1, 3, 4 and 5 mentioned that the simulation-based sessions improved their preparedness to have conversations with patients and carers about death and dying. One student described her experience of the simulation session:
‘I feel like these sessions prepare you for like so one day you will actually be walking into a room with someone there and it's more okay so I've been here before’. (Student 4)
The remaining three students did not make explicit comments regarding the impact of the tutorials on their preparedness for clinical practice. None of the students stated that the tutorials were unhelpful or adversely impacted on their confidence.
Theme 3: The impact of stress on learning during role-play and simulation scenarios
Four of the students (students 3, 4, 5 and 6) described feeling stressed by the large number of observers in the room during role-play-based scenarios. As one student described:
‘Everyone dreads communication skills (pause) every time it comes around we all question whether it would be acceptable to miss it because um personally I'm thinking more of the fact there's like 8 people watching me rather than what I'm actually saying’. (Student 4)
Three of the students (3, 4 and 6) agreed that some stress in a scenario (both role-play and high-fidelity simulation) was necessary for their learning. As student 3 explained:
‘It is good to be put under some stress I think, and you learn a lot from that’. (Student 3)
Four students (students 1, 3, 5 and 7) agreed that although the simulation-based session was daunting initially, they learnt more from the experience. As one student explained:
‘In that situation [high-fidelity simulation] you don't have a ‘get out of jail free’ card (laughs). When we've done role-play in the past if you've got stuck, you say I'm stuck, I don't know what to say’. (Student 1)
Theme 4: Realism in role-play compared with simulation
The primary explanation for this from all students was that high-fidelity simulation sessions were perceived to be more realistic than role-play-based scenarios. As one student explained:
‘[High-fidelity simulation scenarios are] more realistic in terms of it's you and the person. As opposed to you, a person and 8 other people watching you’. (Student 4)
Students explained that smaller class sizes and use of the debriefing room so that peers were not in the room observing the scenario improved the realism of the simulation-based scenarios. All of the students commented that it was easier to suspend their disbelief during the simulation scenario compared with the role-play-based teaching. As one student explained:
‘I think it was a good thing that you kind of got lost in the situation and it made it feel like it was real’. (Student 1)
Students also commented that role-play-based scenarios relied on students' own acting skills. As one student commented:
‘I don't find the communication skills that we get given at university that useful because it's too informal, it's too unrealistic. I'm not a good actor (laughs), most of my friends are not good actors, therefore when we are doing these scenarios it's very unreal, it's very fake’. (Student 2)
The main findings of this study were as follows. Students lacked experience in discussing death and dying with patients and carers, including discussion of DNACPR decisions. They lacked confidence and did not feel prepared to have these discussions as junior doctors. Stress played an interesting role in students' learning experiences; role-play scenarios with peers observing were considered particularly stressful, which detracted from the learning experience. Simulation-based scenarios were considered to be more realistic than role-play-based scenarios, which improved the learning gained from the sessions.
Several students reported a lack of confidence in communicating with patients and carers about death and dying including cardiopulmonary resuscitation, which was in part fuelled by a lack of exposure to these consultations during their training. This reflects findings of existing studies highlighting students' lack of experience and consequent lack of preparedness to look after dying patients.6 ,19 ,20
Students in this study discussed the barriers to observing consultations about death and dying including DNACPR decision-making, and stated that patients and clinicians sometimes suggested that it was not appropriate for them to observe in this setting. This is a novel finding compared with Gibbins' qualitative study of 39 Foundation Year 1 Doctors.6 While both studies have highlighted a lack of preparedness, Gibbins et al found that students themselves stayed away from dying patients, rather than patients with staff limiting their experience. It may be that either students in this study did not shy away from experience with dying patients, or that they did not feel able to admit to doing this.
This study builds on the findings of a study of nursing students which found improved learning outcomes using high-fidelity simulation-based sessions due to improved realism,15 along with a study by Nestel et al which found that a classroom environment was a barrier to engagement in role-play scenarios.11 This study also reflects the results of a systematic review of the nursing literature, which has highlighted the benefits of high-fidelity simulation in improving students' self-confidence and communication skills.17
There are several limitations to this study. The sample size was less than initially planned due to students choosing not to attend the sessions. This may therefore have reduced the breadth of opinions included in the study, and reduces our ability to generalise the findings of this study to the student population as a whole. Second, it is not possible to state whether students' descriptions of their confidence relates to their competence in clinical practice. Despite this, a meta-analysis has suggested that self-assessed confidence can be used as a moderately reliable proxy measure of competence and preparedness, particularly among more senior medical students.21 Third, students in the study sample were essentially self-selecting in terms of their interest in participating in the tutorials. Students who did not volunteer to participate may have been more confident in their communication skills than the students in the study sample, or possibly less confident and therefore more reticent about practising their communication skills. This may therefore reduce the generalisability of this study to other populations.
This study has suggested that high-fidelity simulation may be a useful adjunct to communication skills training surrounding death and dying, specifically discussions regarding cardiopulmonary resuscitation. Further studies will be required to ascertain whether the improvements in confidence described by students in this study translate to discernible improvements in competence. Moreover, given the increased costs related to the set-up and running of simulation-based sessions, the cost-effectiveness of simulation-based training in this setting needs to be established.
The results of this study add to the limited body of evidence regarding the potential impact of high-fidelity simulation training on teaching end-of-life communication skills. Greater understanding of these issues may help to improve students' learning experiences and develop their skills and confidence in communicating about death and dying, thereby improving the experiences of patients and carers.
The authors would like to thank the students who took part in the study and the hospital staff for their assistance with recruitment for the study. They would also like to thank Dr Liz Anderson for her help with the educational aspects of the project.
Contributors AH designed the project, developed the tutorials, helped facilitate the simulation scenarios, carried out the data analysis and wrote the report. KT was involved in editing the simulation scenarios and facilitating the debrief sessions.
Competing interests None declared.
Ethics approval The University of Bristol Ethics committee reviewed the study and deemed that ethical approval was not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Raw anonymised data have been stored securely and are available for review by the journal if requested.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.