Education
Éducation

 

Teaching medical students how to break bad news

Amit Garg, MD; Robert Buckman, MB, PhD; Yvonne Kason, MD, MEd

CMAJ 1997;156:1159-64

[ résumé ]

Dr. Garg is a resident in internal medicine at McMaster University, Hamilton, Ont.; Dr. Buckman is Associate Professor of Medicine at the University of Toronto; and Dr. Kason is with the Department of Family and Community Medicine, University of Toronto, Toronto, Ont.

This article has been peer reviewed.

Reprint requests to: Dr. Robert Buckman, Toronto­Sunnybrook Regional Cancer Centre, 2075 Bayview Ave., Toronto ON M4N 3M5; fax 416 480-6002

© 1997 Canadian Medical Association (text and abstract/résumé)

Contents
Abstract

Program objective: To teach medical students to break bad news to patients and their families empathically and competently.

Setting: Seven teaching hospitals affiliated with the University of Toronto since 1987.

Participants: All medical students in their third preclinical year.

Program: The course presents a 6-point protocol to guide students in breaking bad news and comprises 2 half-day (3-hour) teaching sessions. Each session incorporates a video presentation, a discussion period and small-group teaching, consisting of exercises followed by 4 different role-playing scenarios conducted with the use of standardized patients. The course was evaluated through 2 questionnaires, 1 administered before and 1 after the course, which measured changes in the students' attitude and strategy. Questionnaires were administered during 5 of the years since the course was started. A total of 914 precourse and 503 postcourse questionnaires were completed, of which 359 matched pairs of precourse and postcourse questionnaires were analysed to study any changes due to the course.

Outcomes: Precourse questionnaires showed that 68% of the students had thought about the task of breaking bad news often or very often. Of the 56% of students who had seen clinicians performing this task, 75% felt that they had seen good examples. The proportion of the students who had a plan for how to conduct such an interview rose from 49% before the course to 92% after it, and the proportion who felt they might be reasonably competent in breaking bad news rose from 23% before the course to 74% after it.

Conclusions: The subject of breaking bad news is important to medical students, and it is practicable to design a course to teach the basic techniques involved. Most students perceive such a course as enjoyable and useful and find that it increases their sense of competence and their ability to formulate a strategy for such situations.

Résumé

Objectif du programme : Apprendre aux étudiants en médecine à annoncer avec empathie et compétence une mauvaise nouvelle aux patients et à leur famille.

Contexte : Sept hôpitaux d'enseignement affiliés à l'Université de Toronto depuis 1987.

Participants : Tous les étudiants de troisième année de médecine préclinique.

Programme : Le cours présente un protocole en 6 points pour aider les étudiants à annoncer une mauvaise nouvelle et comporte 2 séances d'une demi-journée (3 heures). Chaque séance comporte la présentation d'un vidéo, une période de discussion et une période d'enseignement en petits groupes comportant des
exercices. On joue ensuite 4 scénarios différents avec des patients normalisés. On a évalué le cours au moyen de 2 questionnaires, le premier a été administré avant le cours et le deuxième, après. Ces questionnaires ont permis de mesurer les changements d'attitude et de stratégie des étudiants. Les questionnaires ont été administrés pendant 5 des années écoulées depuis le lancement du cours. Au total, les participants ont rempli 914 questionnaires avant le cours et 503 après le cours. De ce nombre, on a analysé 359 paires jumelées de questionnaires administrés avant et après le cours pour étudier tout changement attribuable au cours.

Résultats : Les questionnaires administrés avant le cours ont démontré que 68 % des étudiants avaient pensé souvent ou très souvent à la tâche que constitue l'annonce d'une mauvaise nouvelle. Sur les 56 % d'étudiants qui ont vu des cliniciens à l'oeuvre, 75 % étaient d'avis qu'ils avaient été témoins de bons
exemples. La proportion des étudiants qui avaient pensé à une façon de s'y prendre dans un tel cas est passée de 49 % avant le cours à 92 % après celui-ci. La proportion de ceux qui étaient d'avis qu'ils pourraient être raisonnablement capables d'annoncer une mauvaise nouvelle est passée de 23 % avant le cours à 74 % après celui-ci.

Conclusions : L'obligation d'annoncer une mauvaise nouvelle est un sujet important pour les étudiants en médecine et il est pratique de concevoir un cours pour enseigner les techniques de base en cause. La plupart des étudiants estiment qu'un tel cours est utile et agréable et constatent qu'il augmente leur sentiment de compétence et leur capacité d'élaborer une stratégie dans de telles situations.

[ Top of document ]

Introduction

Communication is crucially important in the physician­patient relationship. Despite technologic advances, communication remains a determinant of patient satisfaction in the outcome[1] and of patient participation in and compliance with therapeutic planning.[2] Breaking bad news is one of the most difficult tasks a physician or any other member of the health care team has to do. The way it is done may change the nature of the relationship permanently -- strengthening it, undermining it, damaging it irreparably[3­8] or even leading to litigation.[9] For many years, studies have shown that physicians' communication skills could be improved[10,11] and that, in particular, deficiencies in breaking bad news are common.[12­14] In all cultures studied, patients and families have expressed an increasing desire to be knowledgeable about their diagnosis.[15,16] Consequently, the pressure on health care professionals to share that information has increased markedly. However, dissatisfaction with the process is common, and a predominant perception is that medical staff are impersonal and cold.[17] Junior staff have often stated that training in breaking bad news is insufficient[18] and that specific teaching would be appreciated.[19] Recently, several medical schools have begun to offer undergraduate teaching in breaking bad news,[20] but this is not yet universal. In terms of the feasibility of teaching the subject, there is much evidence that general communication skills can be taught and retained; however, some debates about aspects of that teaching are still unresolved.[21,22] The techniques involved in breaking bad news have been discussed and described in many articles. They have recently been expressed in the form of a 6-point protocol.[23] This protocol is practicable as a teaching tool and intelligible to medical students and junior physicians. The protocol has formed the basis for a course in breaking bad news that also incorporates standard teaching methods, including videos,[24] small-group discussion, role playing and simulated patient interviews.[25,26] This course has been taught at the University of Toronto since 1987 as part of the third-year component of the Medical Interviewing Skills Course, which spans all 3 preclinical years of medical school.

This article describes the course and the results of a survey to determine the students' perceptions of the course and of their competence in this area.

[ Top of document ]

Course structure and design

The course was designed to teach approximately 250 third-year medical students each year a practical approach to breaking bad news to patients and families. The course consists of 2 half-day sessions, each lasting 3 hours, given 1 or 2 weeks apart. The objectives of the course are to reinforce the medical interviewing skills learned in the previous 2 years, to teach the advanced medical interviewing skills required, to give students practice in using those skills and to provide individualized, constructive feedback. Each student is provided with a 40-page booklet that contains instructions for the role-playing and small-group exercises, all 12 role-playing scenarios and a recommended reading list.

Instructional videos

Two 30-minute videos were written and filmed specifically for teaching this subject.[24] One of the videos is shown at the start of each of the 2 sessions. The main content of each of these videos is described later.

Discussion period

After the presentation of each video, there is a 15-minute discussion with the entire group of students (approximately 30 students per teaching hospital).

Small-group exercises

After the discussion period, the students are divided into small groups of 4 to 6 with 1 clinician-supervisor per group. The small-group teaching starts with exercises; in the first session, these concentrate on the development of the empathic response and, in the second, they focus on how to respond to difficult questions and emotions. Techniques are outlined by the supervisor and then practised in the group.

Role playing

After the exercises, 4 10- to 15-minute role-playing scenarios are enacted, 3 of them with standardized patients and 1 with a student playing the role of the patient. The 4 scenarios are selected from a total of 12, including such situations as a young journalist with newly diagnosed rheumatoid arthritis, a woman with a previous mastectomy who presents with back pain and has a positive result of a bone scan for cancer, a young adult with myeloid leukemia, a patient wishing to pursue complementary medicine in the treatment of lymphoma, a person whose same-sex partner has just been diagnosed with AIDS, a woman whose husband has just died after he was brought to the emergency department with chest pain, and a daughter who wants to protect her mother from knowing that she (the mother) has a diagnosis of cancer. The written scenarios are included in the course material, and each scenario is read out before the role-playing exercise starts. It is suggested that each exercise last approximately 10 minutes, with a further 10 minutes allocated to feedback and discussion.

[ Top of document ]

Content of the course: the 6-point protocol

The content of the course is based on a 6-point protocol (Table 1). The techniques required for the protocol are demonstrated in the first video, which concludes with a complete and unrehearsed interview between a physician and a young woman with recently diagnosed acute myeloid leukemia. (A transcription of the interview is given in the textbook.)

The second video demonstrates how these basic interview techniques may be used in more difficult and challenging situations, including handling difficult questions ("How long have I got?"), avoiding blocking tactics such as normalizing (the physician responding immediately with "Everyone feels like that" instead of listening first) and dealing with anger.

Questionnaires

A precourse questionnaire was administered before teaching started and collected immediately after completion, so that it was not available to the students after the course. The students were asked to include some means of identification (initials or a number, for example) on the form to enable the organizers to match the postcourse questionnaire with the precourse form from the same student. However, not all students did so.

The questionnaire asked the students whether they had thought about the task of breaking bad news, whether they had seen an example of a physician performing such a task and, if so, whether the example was a good or bad one. The students were then asked whether they had any plan in mind for undertaking the task of breaking bad news, and how competent they felt to break such news.

The postcourse questionnaire was administered immediately after the end of the second session. It asked the students whether they found the course enjoyable and useful, and asked them to rank the components of the course in order of preference. The students were then asked whether they now had any plan in mind and how competent they felt.

For each question, the students were asked to select a response from a range of responses on a 5-point Likert scale. The wording of the responses varied, depending on the question; in general, -2 was a very negative response, -1 was negative, 0 was equivocal, 1 was positive and 2 was very positive. The Student's t-test was used to test the difference between mean response scores on matched pairs of precourse and postcourse questionnaires.

[ Top of document ]

Results

The questionnaire was given to the students attending the course in 5 of the years from the time the course was started in 1987. A total of 914 precourse questionnaires (from a total population of 1260 medical students) were completed. A total of 503 postcourse questionnaires were completed and collected. Because of scheduling changes, absenteeism and the lack of identification on some forms, not all postcourse questionnaires could be matched with precourse questionnaires. As a result, only 359 postcourse questionnaires could be matched satisfactorily. Information from these 359 matched forms was used in the analysis of the postcourse questionnaires.

Importance of the topic

Responses to the precourse questionnaire showed that 68% of the students had thought about breaking bad news often or very often (1 or 2 on the Likert scale, Table 2).

Perception of the course

Responses to the postcourse questionnaire showed that 82% of the students found the course enjoyable, 4% found it unenjoyable and 14% were equivocal. In terms of the course's usefulness, 88% found it useful, 4% found it useless and 9% were equivocal.

The students' ranking of the components of the course showed that the role-playing scenarios with the standardized patients were regarded as the most enjoyable and useful part of the course. Small-group exercises were regarded as the least enjoyable and useful.

Change in perceived competence and strategy

Before the course, 48% of the 914 students who responded felt that they had a firm plan in mind, 17% had little or no idea and 35% were equivocal. The mean score for all completed precourse questionnaires was 0.28 on the Likert scale. After the course, 92% of the 359 students for whom we had matched precourse and postcourse questionnaires felt that they had a firm plan, less than 1% had no plan and 8% were equivocal. The mean score for all completed postcourse questionnaires was 1.08. The difference in the mean score before and after the course was highly significant (p < 0.01), and the responses of 59% of the students changed by 1 or 2 points on the Likert scale (Table 2).

Before the course, 23% of the students felt that they were competent, 38% judged themselves incompetent and 39% were equivocal. The mean Likert score for all precourse questionnaire responses to this question was 0.23. After the course, 74% of the students felt that they were competent, 2% felt incompetent and 24% were equivocal (Table 2). The mean score for this response on all postcourse questionnaires was 0.78. The difference between precourse and postcourse responses was significant (p < 0.01).

Influence of role models

Fifty-six percent of the students had seen clinicians breaking bad news. Of these students, 75% scored the best interviews they had observed as 1 or 2 on the Likert scale. According to the precourse questionnaire, the students who had seen examples of clinicians breaking bad news had a better idea of how they would undertake the task than those who had not seen examples (mean scores 0.36 and 0.12, respectively). The students with role models also had a better sense of competence than their peers (mean scores -0.12 and -0.34, respectively).

According to the postcourse questionnaire, more of the students who had seen clinicians breaking bad news had a definite plan in mind than those who had not seen any role models (mean scores 1.16 and 1.05, respectively). The students who had seen role models also had a greater sense of competence than their peers (mean scores 0.82 and 0.74, respectively).

Cases of loss of perceived competence or strategy

In 11 pairs of forms, the students recorded a decrease in perceived competence or a loss of a strategic plan or both. In 10 of these cases, the students recorded that they found the course both useful and enjoyable. However, in 1 case, the student recorded that the course was neither enjoyable nor useful. It is difficult to determine whether some or any of these cases truly represent "de-skilling"; that is, a loss of competence as a result of the course.

[ Top of document ]

Discussion

Our primary objective was to show that it is possible and practicable to offer medical students a course in the techniques of breaking bad news to patients and their families. Such a course is clearly practicable, and our course has run without any problems since 1987. The course can be implemented on a wider scale in other medical schools and nursing colleges. The written material and videos are available to anyone interested in starting such a course. Many medical schools already have standardized patient programs; training standardized patients for the particular roles in this course is not difficult or arduous. Once trained, the standardized patients have participated in the course each year and have found it enjoyable and valuable.

The results of the questionnaires show that this topic is an important one for medical students. However, the results are from a relatively small proportion of the students who attended the course. The main reason for this low response rate is that a large number of the students did not return both the precourse and postcourse questionnaires. As well, many of the forms that were returned had no identification and so could not be matched. Hence, the sample is small and may represent only those students who were highly motivated and were therefore more likely to enjoy the course and find it useful.

With these limitations in mind, however, several interesting points emerge. First, it is valuable to note that the influence of a role model is not a major one. Students who have not seen a physician break bad news well may still have a good idea of how they would do it. However, a role model does influence learning to some extent; students who have seen a good role model seem to gain a little more from the course than the those who have not.

The course is clearly perceived as enjoyable and useful by the great majority of students, and all but 11 students thought that their attitude and strategy had improved as a result of the course. In 11 cases, the questionnaires showed an apparent loss of perceived competence or strategy or both. In 10 of those cases, however, the students reported that they enjoyed the course and found it useful. Although it is possible that some of these students actually lost a feeling of competence (for example, as a result of losing preconceptions about how to break bad news and gaining a more realistic view of the difficulty of the task), it is also possible that some students simply did not recall the score they had assigned to each question in the previous session. It is clear, however, that 1 student found the course to be of no value.

We can conclude only that the great majority of the students who completed and returned the questionnaires gained from the course.

We cannot yet draw any conclusions about the students' actual performance in the task of breaking bad news. We have not seen or devised an adequate and validated method of testing performance. Although feedback from the standardized patients is very valuable (and is used as part of the course), we do not feel that this is a truly objective measure of performance. The next stage of this project is to incorporate into the course some form of evaluation -- a short written test or videotaping of a small sample of students.

[ Top of document ]

Conclusion

These findings show that teaching of the communication skills needed for breaking bad news can be incorporated into undergraduate curricula and that students find the resulting course both enjoyable and useful. The questionnaires showed that most students felt that the course helped them with their approach to the topic and improved their perception of how they would perform the task. Whether these perceptions are translated into objective improvements in performance is the subject of the next phase of this project. We suggest that videotaped material can be useful in courses of this nature and that standardized patients, under the supervision of experienced instructors, should be included in such courses.

We thank Dr. Barbara Stubbs, Ms. Anja Robb and the Standardized Patient Programme of the University of Toronto for organizing the involvement of standardized patients in this course.

[ Top of document ]

References

  1. Korsh BM, Gozzie K, Francis V. Gaps in doctor­patient communication. Pediatrics 1968;42:855-71.
  2. Ley P. Satisfaction, compliance and communication. Br J Clin Psychol 1982;21:241-54.
  3. Professional development -- breaking bad news. Nurs Times 1994;90(11):5-8.
  4. Michaels E. Deliver bad news tactfully. CMAJ 1983;129:1307-8.
  5. Atkinson JM. To tell or not to tell the diagnosis of schizophrenia. J Med Ethics 1989;15:21-4.
  6. Roberts CS, Cox CE, Reintgen DS, Baile WF, Gibertini M. Influence of physician communication on newly diagnosed breast patients' psychologic adjustment and decision-making. Cancer 1994;74:336-41.
  7. McLauchlan CAJ. Handling distressed relatives and breaking bad news. BMJ 1990;301:1145-9.
  8. Slavin LA, O'Malley JE, Koocher GP, Foster DJ. Communication of the cancer diagnosis to pediatric patients: impact on long-term adjustment. Am J Psychiatry 1982;139:179-83.
  9. Shapiro RS, Simpson DE, Lawrence SL, Palsky AM, Sobocinski KA, Schiedermayer DL, et al. A survey of sued and nonsued physicians and suing patients. Arch Intern Med 1989;149:2190-6.
  10. Maguire P, Rutter D. Training medical stuents to communicate. In: Bennet AE, editor. Communication between doctors and patients. Oxford: Oxford University Press; 1976:45-74.
  11. Simpson M, Buckman R, Stewart M, Maguire P, Lipkin M, Novack D. Doctor­patient communication: the Toronto consensus statement. BMJ 1991;303:1385-7.
  12. Girgis A, Sanson-Fisher W. Breaking bad news: consensus guidelines for medical practitioners. J Clin Oncol 1995;13:2449-56.
  13. Reynolds PM, Sanson-Fisher RW, Poole AD, Harker J, Byrne MJ. Cancer and communication: information-giving in an oncology clinic. BMJ 1981;282:1449-51.
  14. Lind SE, Delvacchio MJ, Seidel S, Csordas T, Good B. Telling the diagnosis of cancer. J Clin Oncol 1989;7:583-9.
  15. Charlton RC. Breaking bad news. Med J Aust 1992;157:615-21.
  16. Charlton R, Mizushima Y. National differences in breaking bad news. Med J Aust 1993;159:72.
  17. Quill T, Townesend P. Bad news: delivery, dialogue and dilemmas. Arch Intern Med 1991;151:463-8.
  18. Marteau TM, Johnston M, Wynne G, Evans TR. Cognitive factors in the explanation of the mismatch between confidence and competence in performing basic life support. Psychol Health 1989;3:173-82.
  19. Marteau TM, Humphrey C, Matoon G, Kodd J, Lloyd M, Horder J. Factors influencing the communication skills of first-year clinical medical students. Med Educ 1991;25:127-34.
  20. Knox JD, Thomson GM. Breaking bad news: medical undergraduate communication skills, teaching and learning. Med Educ 1989;23:258-61.
  21. Maguire P, Fairburn S, Fletcher C. Consultation skills of young doctors.
    1. Benefits of feedback training in interviewing as students persist. BMJ 1986;292:1573-6.
  22. Sanson-Fisher RW, Poole AD. Teaching medical students communication skills: an experimental appraisal of the short- and long-term benefits. In: Osborne DJ, Bruneberg M, Eiser J, editors. Research in psychology and medicine, London (UK): Academic Press; 1979:228-94.
  23. Buckman R, Kason Y. How to break bad news -- a practical protocol for healthcare professionals. Toronto: University of Toronto Press; 1992.
  24. Buckman R. How to break bad news: the University of Toronto course (video series). Mississauga: Medical-Audio Visual Communications; 1986.
  25. Vu NV, Barrows HS, Marcy ML, Verhulst SJ, Colliver JA, Travis T. Six years of comprehensive, clinical, performance-based assessment using standardized patients at the Southern Illinois University School of Medicine. Acad Med 1992;67:42-50.
  26. Rothman AI, Cohen R, Bilan S. A comparison of short- and long-case stations in a multiple-station test of clinical skills. Acad Med 1996;71:S110-2.

Comments Send a letter to the editor responding to this article
Envoyez une lettre à la rédaction au sujet de cet article

| CMAJ April 15, 1997 (vol 156, no 8) / JAMC le 15 avril 1997 (vol 156, no 8) |