Teaching how to deliver bad news: a systematic review

Camargo, Nicole Cavalari; Lima, Marcelo Gonçalves de; Brietzke, Elisa; Mucci, Samantha; Góis, Aécio Flávio Teixeira de

doi:10.1590/1983-80422019272317

person Nicole Cavalari Camargo

school. Departamento de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo/SP, Brasil.

0000-0003-4999-3384

person Marcelo Gonçalves de Lima

school. Departamento de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo/SP, Brasil.

0000-0001-6742-5845

person Elisa Brietzke

school. Programa Psiquiatria e Psicologia Médica, Departamento de Psiquiatria, Universidade Federal de São Paulo (Unifesp), São Paulo/SP, Brasil.

0000-0003-2697-1342

person Samantha Mucci

school. Programa Psiquiatria e Psicologia Médica, Departamento de Psiquiatria, Universidade Federal de São Paulo (Unifesp), São Paulo/SP, Brasil.

0000-0003-3809-8173

person Aécio Flávio Teixeira de Góis

school. Programa de Medicina de Urgência e Medicina baseada em evidências, Departamento de Medicina, Unifesp, São Paulo/SP, Brasil.

0000-0003-0217-1463

Correspondência Nicole Cavalari Camargo – Rua Botucatu, 720, Vila Clementino CEP 04023-062. São Paulo/SP, Brasil. Nicole Cavalari Camargo – Graduanda –
nicole.cavalari@huhsp.org.br Elisa Brietzke – Doutora –
elisabrietzke@hotmail.com Samantha Mucci – Doutora –
sammucci@gmail.com Aécio Flávio Teixeira de Góis – Doutor –
aecio.gois@unifesp.br

Declaram não haver conflito de interesse.

Participação dos autores

Nicole Cavalari Camargo and Marcelo Gonçalves de Lima elaborated the systematic review and tables. Elisa Brietzke developed the final revision of the text. Samantha Mucci together with Aécio Flávio Teixeira de Góis supervised the project.

Active learning
Author, year	Country	n	Population	Design	Instruments	Results
van Weel-Baumgarten and collaborators; 2012 ¹³	NL	1260	Medical students	Longitudinal	Basic knowledge in BBN was evaluated in a questionnaire before the intervention. Teaching methods used web lectures, home assignments, interactive DVD, practicing with SPs with Spikes communication protocol and ABCDE, in small groups of various sizes. Feedback was given to evaluate development.	Students believe the correct time to learn about communication is in year 3 of medical school
Lifchez, Redett; 2014 ³¹	USA	17	Plastic residents	Longitudinal	Residents were surveyed for their prior education in BBN (Orgel questionnaire) and took a personality test (Myers-Briggs) to understand how it influences their way to communicate.	Confidence increased after training and SP. Those who evaluated residents’ progress, reported on that too. The performance in delivering bad news was also better after training.
Arnold and collaborators; 2015 ³²	USA	38	Emergency residents	Longitudinal	Overview of basic knowledge followed by skills training and role-playing in groups.	After training, residents showed confidence in BBN and evaluated their progress with an increase of 23%. The WS was rated as good or excellent. On a follow up after 1 month, 100% of them declared themselves completely capable of delivering bad news
Greenberg and collaborators; 1999 ³³	USA	27	Pediatric residents	Longitudinal	Participants were videotaped and observed twice while communicating bad news to a trained SP. The first observation was followed by a feedback and the experience was repeated about 4 to 10 weeks later. The SP evaluated the progress in BBN before and after training, without knowing when the physician had training.	Before training, 79% believed they needed training in BBN and that they were not capable of delivering bad news or counseling parents. After the intervention, participants’ total score increased significantly. Content scores were correlated positively to counseling scores, showing that knowing the theory helps better practice.
Ju and collaborators; 2014 ³⁴	USA	11	Oncology residents	Longitudinal	Session with SPs and feedback regarding the communication	After the feedback, residents reported an increase in communication skills
Colletti and collaborators; 2001 ³⁵	USA	21	Medical students	Longitudinal	1-hour encounter in which students should discuss a new diagnosis of rectal cancer or a miscarriage. The student should discuss diagnosis, treatment and prognosis. After the experience, feedback showed the progress.	Students who had prior training had better scores and the scenarios did not change their performance, showing that previous classes are important.
Dikici, Yaris, Cubukcu; 2009 ³⁶	TR	146	Medical students	Longitudinal	Questionnaire evaluating effectiveness of the course and a 4-station OSCE to assess the impact of the approach objectively	54% of medical students perceived themselves as more competent at handling the task after the course and most of them classified the course as useful.
Ramaswamy and collaborators; 2014 ³⁷	USA	23	Internal medicine interns	Longitudinal	The curriculum uses small-group discussion, case-based learning, and repeated practice with role-playing exercises to engage learners and develop an active learning model. This approach was evaluated using questionnaires	Three months after the end of the workshop, respondents were using different techniques. 64% of participants felt that the Spikes mnemonic was the most helpful skill they learned in the workshop. After 3 months, 38% of them were using the Spikes method, even though 15% of the respondents felt that Spikes was easy to use in their interactions with patients. 91% of participants believed the amount of information disseminated in the curriculum to be “just right”. 81% of the participants felt that the opportunity for repeated practice was helpful in enhancing their communication skills
Vail and collaborators; 2011 ³⁸	UK	285	Physicians from 22 different specialties	Longitudinal	BBN scenarios, evaluating capability to communicate according to different specialties	Consultants mainly focused upon providing biomedical information and did not discuss life-style and psychosocial issues frequently. These approaches are not related to doctor gender, specialty, place of qualification or age.

Mixed strategies
Author, year	Country	n	Population	Design	Instruments	Results
Fujimori and collaborators; 2014 ²¹; Fujimori and collaborators; 2014 ²²	JP	30 oncologists and 580 patients	Oncologists e patients	Longitudinal	Questionnaires evaluating knowledge were given to oncologists. A group of oncologists attended workshop training on communication, while another group did not receive training. Follow up with patients took place.	Simulation with patients was taped and evaluated by patients on an 11-point scale. Oncologists who received training in BBN had better results than those who did not. The workshop was well evaluated by oncologists and they felt more confident in BBN after the intervention.
Setubal and collaborators; 2018 ³⁹	BR	61	Perinatology residents	Longitudinal	Residents met a SP and were allocated to control and intervention (Spikes) groups. After the intervention, both groups met a SP again.	There was no difference between intervention and control groups. The second session with the SP was better, showing practice leads to success. The experience was well-rated.
Karam and collaborators; 2017 ⁴⁰	USA	16	Anesthesiology residents	Longitudinal	The BBN experiences with GRIEV_ING check-list were obtained using simulators with a high fidelity to immersive experiences. After that, role-playing with SP complemented the experiences.	Performance in case 1 did not vary in relation to the year of training. Pre-test scores were lower than post-test scores on the GRIEV_ING check-list. Also, before the workshop, 52% rated their competence in BBN as good and very good. This number increased to 93% after the workshop, with an increase in confidence as well.
Cohen and collaborators; 2016 ⁴¹	USA	26	Nephrology fellows	Longitudinal	Fellows were assigned to three cases about treatment, how to break bad news and to discuss prognosis	Respondents affirmed their development went from “not prepared” to “very well prepared”. Also, they rated the course as excellent.
von Lengerke, Kursch, Lange; 2011 ⁴²	DE	267 (100)	Medical students	Longitudinal	Students were assigned to seven sessions lasting four hours each, containing theory about physician-patient communication, practical exercises, nonverbal communication, video-classes, cases, disclosure of a diagnosis and role-playing.	Most students fully agreed with the teaching methods and subjects, affirming progress in communication.
Abel and collaborators; 2001 ⁴³	UK	140	Senior doctors and nurses	Longitudinal	A course was developed for doctors and nurses with previous training in communication. Areas that needed improvement were investigated and implemented in the project	They all agreed the course was important and most of them thought BBN was the most useful part. Also, there was an increase in confidence on how to deliver bad news.
Alexander and collaborators; 2006 ⁴⁴	USA	56	Medical residents	Longitudinal	16-hour curriculum that included control of pain and symptoms, communication skills (BBN) and ethics. Groups were divided into control and intervention, and they were evaluated before and after the approach.	Both groups had the same prior training. After the training, the intervention group had much higher scores.
Sombra Neto and collaborators; 2017 ⁴⁵	BR	119	Medical students	Longitudinal	Students watched theoretical weekly classes and participated in practical BBN simulations, with simulated patients and training in pairs	67% of students presented an excellent score (>90%), 7% was considered regular or bad. 16% concluded the test with the maximum score and the lowest score was 68%.
Burn and collaborators; 2014 ⁴⁶	CH	225	Medical students	Longitudinal	In this cohort trial, the intervention group received training in ethics, truth-telling and BBN. Students were evaluated by questionnaires before and after the approach.	Students’ ethical attitudes regarding truth-telling remained stable, but they started to feel more comfortable about those situations, and two thirds of students who did not feel confident before felt confident afterwards.
Hurst and collaborators; 2015 ⁴⁷	CH	225	Medical students	Longitudinal	Students participated in encounters with 3 simulated patients and received feedback to understand their progress. Intervention was composed of a 90-minute talk with SP, a 15-minute ethical discussion and a 60-minute communication skills practice	Students ethical attitudes remained stable. They developed new skills following the intervention, and increased their awareness of the difficulties and challenges raised by BBN situations, allowing them to resolve their weaknesses
Skye and collaborators; 2014 ⁴⁸	USA	451	Medical students	Longitudinal	Students watched a simulation about a woman with colon cancer. After that, they discussed empathy, doctors, gender, dynamics, power and support	The intervention was well received (94% agreed or strongly agreed with the method). They valued the strategy.
Parikh and collaborators; 2017 ⁴⁹	USA	105	Medical students	Longitudinal	Questionnaires about how participants rated the intervention which was composed of discussions and role-playing, and whether they retained the training after 1 year	Students rated the intervention as essential for medical education and affirmed feeling more capable of discussing life and death with patients. Results were evaluated after 1 year (12-24 months)
Ellman, Fortin; 2012 ⁵⁰	USA	1455	Medical students	Cross-sectional	Communicating Difficult News Workshop and Ward-Based End-of-Life Care Assignment, which was developed at Yale Medical School.	Not evaluated
Szmuilowicz and collaborators; 2010 ⁵¹	USA	56	Internal medicine residents	Longitudinal	Residents were assigned to a control or intervention group, which was addressed with a combination of teaching styles and skills practice	Residents in the intervention group showed an improvement in communication (treatment options and prognosis) and ability to show emotional support. Also, they felt more confident. These changes were not seen in the control group.

Passive learning and online forum
Author, year	Country	n	Population	Design	Instruments	Results
Brown and collaborators; 2014 ⁵²	USA	109	Radiology residents	Longitudinal	Questionnaires applied immediately before and after communication workshop, containing mediated discussions, lectures and media content	After completing the workshop, more trainees reported feeling comfortable when communicating bad news to patients. They desired additional communication training on error disclosure, general communication and radiation risks
Levi, Green; 2003 ⁵³	USA	20	Medical residents	Longitudinal	There were didactic lectures, discussions, sessions with poetry and depositions, to understand how the patient receives bad news. There was also a didactic lesson about effective ways of communicating bad news	Residents enjoyed the retreat and valued it as a professional experience. They appreciated it and rethought the purpose of becoming a physician and dealing with patients
Coutinho, Ramessur; 2016 ⁵⁴	PT	260	Medical students	Cross-sectional	45-minute lecture that aimed to provide basic theory on how to deliver bad news (for example, Spikes).	21% responded to the survey. 92% of them felt the subject should be included in the normal curriculum. 83% of them felt the lecture was appropriate and effective, efficient for the duration. Nevertheless, they thought that only lectures were not enough and practical activities would be useful too.
Makoul and collaborators; 2010 ⁵⁵	USA	315	Medical students	Longitudinal	The DC Online Forum uses virtual dialectic design. It makes possible for students to respond to each other and to reflect. Also, it can be accessed at different times and locations	Students submitted stories about patients and concluded that the interface was important and made them more capable of planning conversations and reactions. Also, they commented about their experiences in a group with attendings and said how much influence they had.

[TFN1] n: sample number; USA: United States of America; NL: Netherlands; UK: United Kingdom; TR: Turkey; SP: Standardized Patients; BBN: Breaking Bad News; WS: Workshop; Spikes: Oncology/Bad news communication protocol (“S=setting up; P=perception; I=invitation; K=knowledge; E=emotions; S=strategy”)

Teaching how to deliver bad news: a systematic review

Abstract