Breaking bad news: the SPIKES protocol in teaching

The fifth-year intern walks into the room with the biopsy report in her hand. The patient is sitting with her daughter, the two of them silent. The preceptor asked her to lead the conversation, and the intern has already rehearsed the first sentence in the hallway. But as she sits down, she realizes she does not know what the patient already imagines, does not know how much she wants to know, and does not know what to do if she cries. She speaks fast, uses two technical words, hands over the paper and waits. The daughter asks: "Is this cancer?". The intern looks at the preceptor.

Giving bad news is one of the hardest tasks in medicine and one of the least taught. The SPIKES protocol was created precisely to give structure to this conversation. This guide explains what SPIKES is, what each step asks for, what the research shows about teaching the protocol, what its limits are, and how students, residents and preceptors can train this skill with simulation, cases and structured feedback.

Key points

• SPIKES has six steps and organizes four goals: gathering information from the patient, conveying the medical information, offering support and building the strategy for the future together with the patient (Baile et al., 2000).
• In a systematic review, teaching SPIKES was associated with more satisfaction, knowledge and performance among learners, with no studies measuring outcomes in the health system (Mahendiran et al., 2023).
• Teaching interventions improved skills assessed by observers and physicians' confidence in a meta-analysis (Johnson and Panagioti, 2018).
• In a randomized trial, patients seen by trained oncologists were less depressed than those in the control group (Fujimori et al., 2014).
• Simulated patients are the most used resource in teaching, and virtual patients do not appear to be inferior (Dale MacLaine et al., 2021).
• In Brazil, a survey of neurology residency programs found deficient structured training and little feedback on communication (Lima et al., 2023).

Summary

• Why bad news is a skill, not a gift
• The six steps of SPIKES
• What the evidence says about teaching SPIKES
• How to teach giving bad news
• From recommendation to practice
• The limits of the protocol
• Training the difficult conversation with NEXO
• Frequently asked questions
• Checklist for the next bad news

Why bad news is a skill, not a gift

For a long time, communicating bad news was treated as a personal talent or as something learned by watching the more experienced. The problem is that many of the more experienced never learned it either. A review in the Lancet recalled that physicians find these conversations stressful and that, without effective training, they may adopt inadequate ways of giving the news and of handling the emotional burden that follows (Fallowfield and Jenkins, 2004). The effect hits both sides of the table.

A randomized trial with 160 oncologists in the United Kingdom showed that the communication problems of experienced physicians are not solved by time and clinical experience alone, and that an intensive course significantly improved skills such as asking open questions, expressing empathy and responding to the patient's cues (Fallowfield et al., 2002). In other words, the skill improves with practice, and experience alone does not guarantee that improvement.

What bad news is

Bad news is any information that negatively changes the way a person sees their own future. It can be a cancer diagnosis, but also a chronic disease, a functional limitation, the result of a genetic test, the failure of a treatment or the news that there is no curative option left. What defines the severity is the impact for that patient, not the medical classification. That is why the physician's first task is to find out what the patient already knows and what they expect.

Context also changes the conversation. The Lancet review paid special attention to three scenarios: parents receiving news in obstetrics and pediatrics, people in acute trauma situations in emergency services, and patients with cancer (Fallowfield and Jenkins, 2004). In the emergency department, for example, there is almost never a prior relationship, time is short and the family arrives unprepared. In pediatrics, the person receiving the news is not the patient. In each scenario, the steps of the protocol remain useful, but the weight of each one changes, and it is worth training with varied situations, not only with the cancer diagnosis in the office.

What patients say matters

At a cancer center in the United Kingdom, 244 patients answered how they preferred to receive the diagnosis. Most gave more importance to the content of the message than to supportive aspects, and among the dissatisfied the most common complaint was the doctor's unempathetic or pessimistic manner (Brown et al., 2011). Most also wanted to take part in decisions. A German qualitative study reinforced this picture and recommended including the patient perspective in quality management of these conversations (Krieger et al., 2023).

The six steps of SPIKES

The protocol was published in 2000 by Walter Baile, Robert Buckman and colleagues, designed for the patient with cancer but applicable to other contexts (Baile et al., 2000). The name comes from the initials, in English, of each step. The table summarizes what each one calls for and an example sentence.

Letter	Step	What to do	Possible sentence
S	Setting (prepare the environment)	Private place, sit down, turn off distractions, ask who should be present	"Would you prefer your daughter to stay with us?"
P	Perception	Find out what the patient already knows or imagines	"What have you understood from the tests so far?"
I	Invitation	Ask how much he wants to know and in what form	"Would you like me to explain the details or would you prefer the main points first?"
K	Knowledge	Give the news with a warning shot, in plain language, in parts	"Unfortunately, the result was not what we hoped for."
E	Emotions	Acknowledge and name the emotion, respond with empathy	"I see this was a shock. It is a lot to hear."
S	Strategy and summary	Summarize, agree on next steps, check understanding	"Let's agree on what we do from now on."

Step by step, with the most common mistakes

Each step has a typical pitfall, and knowing them helps more than memorizing the acronym:

• Setting: giving the news standing up, in the corridor, or with the phone ringing.
• Perception: jumping straight to the explanation without knowing what the patient already imagines.
• Invitation: assuming that every patient wants all the details at once.
• Knowledge: using jargon, giving everything at once or softening it so much that the news is not understood.
• Emotions: filling the silence with more technical information instead of receiving the reaction.
• Strategy: ending without agreeing on the next step, leaving the patient with no ground under their feet.

The knowledge step deserves special attention. In a study with the Oncotalk program, only 16% of participants used the word "cancer" when giving the news before the training, and 54% began to use it afterwards (Back et al., 2007). Avoiding the word seems kind, but it often leaves the patient confused about what is actually happening.

Silence is also part of the technique

After the news, many doctors feel an urge to talk. The patient, however, often needs a few seconds to absorb what they heard. Staying silent, keeping eye contact and waiting for the first reaction is an empathetic response, not a void. When the reaction comes, name the emotion before going back to the information. The technical explanation that comes too early is almost always lost.

The scene from the beginning, redone with SPIKES

Go back to the intern with the report in hand. Before entering, she could have agreed with the preceptor who leads and at what moment the preceptor comes in, if needed. Once in the room, sitting beside the patient, not behind the desk, and asking whether the patient wants her daughter present fulfills the first step in less than a minute.

The second step changes everything. A question such as "what do you already know about the reason for the biopsy?" shows whether the patient imagines something serious, thinks it is a routine test or has already heard from another doctor the word no one said. The answer defines the starting point. If she says "the doctor said it could be a tumor", the intern does not start from zero. If she says "I think it is just an inflammation", the distance between expectation and news is large, and the warning shot needs to be more careful.

In the third step, the intern asks how the patient prefers to receive the information. Some people want the full picture, others just the essentials at first. In the fourth, she gives a warning first ("I have difficult news about the result"), pauses and says it clearly, using the word the daughter asked about. In the fifth, when the patient goes quiet or cries, the intern waits, offers a tissue if it makes sense and names what she sees: "I imagine this is very hard to hear." Only when the patient starts asking again does the technical explanation make sense.

In the last step, the intern sums up what was said, asks what was understood and agrees on the concrete next step: the appointment with the specialist, the date, who the patient can turn to if a question comes up before then. The whole conversation may be no longer than the rushed version. What changes is the order, and the order changes what the patient takes home.

The preceptor's role in the room

When a student leads the conversation, the preceptor has two tasks. The first is to protect the patient: step into the conversation if something important is being said wrongly or if the emotion is going unanswered. The second is to protect the learning: not take over the conversation without need and save the observations for the feedback right afterwards. Useful feedback starts by asking how the student felt, goes through one or two things that worked and ends with a single concrete change for next time.

What the evidence says about teaching SPIKES

The original article itself reported that oncologists, residents and students who learned the protocol felt more confident about communicating unfavorable information (Baile et al., 2000). Confidence is a start, but it does not prove the conversation got better for the patient. Later studies tried to measure more than that.

A systematic review of SPIKES found interventions ranging from lectures to role-play with simulated patients, videos, debriefing sessions and computer simulations, and concluded that the protocol is associated with greater satisfaction, knowledge and performance among learners (Mahendiran et al., 2023). None of the included studies assessed outcomes in the health system.

A meta-analysis of teaching interventions in general, not only SPIKES, found moderate and significant improvement in breaking-bad-news skills assessed by observers, with a standardized mean difference of 0.74 across 19 comparisons, and improvement in physicians' confidence, with 0.60 across 9 comparisons (Johnson and Panagioti, 2018). Only one study assessed patient depression or anxiety.

The effect on the patient

That study that measured the patient is Japanese. In a randomized trial, oncologists went through training based on patients' preferences, and patients seen by them after the course were significantly less depressed than those seen by physicians in the control group (Fujimori et al., 2014). Patients' satisfaction with the communication style, however, did not change in the same study. It is an important result and still an isolated one.

Does the effect last?

A follow-up of the British trial, 12 to 15 months after the course, showed that the physicians had incorporated the main skills into clinical practice, with 81% fewer interruptions and more summaries of the information (Fallowfield et al., 2003). At the same time, expressions of empathy decreased compared with the assessment at three months. The lesson is that training has a lasting effect on some skills, and that others need reinforcement.

How to teach breaking bad news

The most used resource in teaching is the simulated patient, a person trained to portray the patient in a planned conversation. In a systematic review with medical students, most models used actors, and the impact was described as uniformly positive, although quality comparisons between formats are lacking (Dale MacLaine et al., 2021). The same review found some evidence that virtual simulated patients were as useful as in-person ones.

A quick review of virtual patients reached a similar conclusion: they are valuable when compared with no intervention and not inferior to simulated patients, with the extra benefit of recording, but with the barrier of lack of emotion and interaction (Kang, 2024). The post about virtual patients discusses when this format works best.

Worked examples with hints

Not all teaching needs an actor. In an experiment with students, worked examples of bad news conversations, in video or text, prepared them better for the simulation when they included hints marking each step of SPIKES (Schmitz et al., 2020). The format, video or text, made no difference. The hint did. This opens up a cheap alternative: annotated transcripts of conversations, discussed in a group before each simulation.

A possible sequence for the class

One way to organize teaching over a few weeks:

1. Short reading on the protocol and patient preferences.
2. Worked example in text or video, with the steps marked.
3. Role-play among colleagues, with an observer using a checklist.
4. Simulation with a simulated patient or virtual patient, recorded when possible.
5. Structured debriefing, starting with the emotion of the person who led the conversation.
6. Supervised real conversation, with feedback right afterwards.

Debriefing is the step that is most often missing. Without it, the simulation becomes just an experience, not a learning moment. The post about simulation of a consultation offers a debriefing script.

Teaching as a team

Bad news is rarely given by one person alone. Nurses, psychologists and social workers are often in the room or follow the patient afterwards. A scoping review of interprofessional approaches to teaching how to give bad news found few initiatives and called for more research on how to incorporate collaboration skills into this teaching (Lackie et al., 2024). The text about interprofessional clinical reasoning shows how to organize the conversation between professions.

From recommendation to practice

Data from residency programs show the gap between what is recommended and what is practiced. In Brazil, for example, in a survey with 172 responses from residents and directors of 47 neurology programs, 59% of the directors acknowledged that feedback on communication was not standard practice and about 32% reported the absence of any specific training (Lima et al., 2023). The residents felt capable of giving bad news, but most had not had classes, simulations or feedback, and reported sadness and helplessness in these conversations.

This mismatch between feeling capable and having been trained is a warning. Confidence without feedback can consolidate bad habits. For the student, the practical consequence is to actively seek training and feedback opportunities, even when the curriculum does not offer them: academic leagues, simulations, conversations with preceptors willing to observe.

The limits of the protocol

SPIKES is a support, not a rigid script. A literature review prior to the protocol already found consistent recommendations, such as giving the news at the patient's pace, conveying some hope and speaking with empathy, but it noted that empirical evidence on what actually makes a difference was lacking (Ptacek and Eberhardt, 1996). The criticism remains valid: the protocol came from consensus and practice, and research is still measuring its effects.

Another criticism is treating bad news as a single event, when in practice it is a process, with several conversations over days or months. Two authors proposed a revised version, called S-P-w-ICE-S, which treats bad news communication as a process, from preparation to documentation and reflection after the conversation (Meitar and Karnieli-Miller, 2022). The name change matters less than the idea: the conversation continues after you leave the room.

Culture, family and individual preference

The protocol starts from a view in which the patient decides how much they want to know. In many cultures, however, it is common for relatives to ask the doctor not to tell the patient. The invitation step helps with this dilemma: asking the patient how they want to receive the information and who they want beside them respects autonomy without overriding the family. When the patient says they prefer their daughter to decide, that is also their choice. The post about empathy and clinical reasoning in the consultation explores how listening changes the decision.

The person who gives the news also feels it

Giving bad news weighs on the person speaking. Brazilian residents reported sadness and helplessness (Lima et al., 2023), and the Lancet review describes the stress of these conversations for doctors (Fallowfield and Jenkins, 2004). It is worth talking to someone after a difficult conversation, and recognizing that feeling something is not a lack of professionalism. When the load builds up, the topic touches on burnout in medical school.

Practicing the difficult conversation with NEXO

NEXO does not simulate the bad news conversation. What it trains is the part before: reaching the diagnosis with confidence. In the case of the day, the diagnosis arrives within up to six clues, and the proximity in ICD-10 shows how far each guess was from the answer. The cases are fictional with a real clinical presentation, and many end in a diagnosis that, in real life, would require a difficult conversation.

This allows a simple exercise for study groups and student leagues. After playing the case of the day, each participant writes how they would open the conversation with that patient, following the SPIKES steps, and the group compares the versions. The exercise combines diagnostic reasoning and communication, which happen together in the consulting room. The guide on what NEXO is explains the whole game, and the NEXO blog gathers other texts about medical education.

If you coordinate a medical school, a student league or a residency and want cases for communication workshops, see NEXO Closer. Researchers in medical education can propose studies through NEXO Researcher. Doctors who review cases take part through NEXO Curator, and those who create health content on social media can take part through NEXO Creators.

Frequently asked questions

What is the SPIKES protocol?

It is a six-step protocol for communicating bad news, published in 2000 by Baile, Buckman and colleagues, which organizes the conversation into setting, perception, invitation, knowledge, emotions and strategy (Baile et al., 2000).

Does SPIKES work?

Teaching the protocol is associated with greater satisfaction, knowledge and performance among learners, but studies on the effects on patients and the health system are lacking (Mahendiran et al., 2023).

Does the training change anything for the patient?

In a randomized trial, patients of trained oncologists were less depressed (Fujimori et al., 2014). It is a result that has still been little replicated.

How can it be trained in medical school?

With worked examples, role plays, simulated or virtual patients and structured debriefing (Dale MacLaine et al., 2021).

Should I use the word cancer?

In general, yes. Avoiding the word can leave the patient without understanding the diagnosis, and training increases its use (Back et al., 2007).

What should I do if the family asks me not to tell?

Ask the patient how much they want to know and who they want beside them, as the invitation step calls for. This respects their autonomy and the family's presence.

Checklist for the next piece of bad news

• Prepare the setting: a private place, seated, without interruptions.
• Ask before explaining: what the patient already knows and how much they want to know.
• Give notice and say it clearly: short sentences, no jargon, in parts.
• Wait and acknowledge: silence, naming the emotion, responding with empathy.
• Agree on the next step: a summary, next steps and when the next conversation will be.
• Take care of yourself afterward: talk to someone and ask for feedback from whoever observed.

Want to train the reasoning that comes before the conversation? Play the case of the day on NEXO.

References

1. Baile WF, Buckman R, Lenzi R, Glober G, Beale EA, Kudelka AP. SPIKES: a six-step protocol for delivering bad news: application to the patient with cancer. Oncologist. 2000. PubMed
2. Fallowfield L, Jenkins V. Communicating sad, bad, and difficult news in medicine. Lancet. 2004. PubMed
3. Ptacek JT, Eberhardt TL. Breaking bad news: a review of the literature. JAMA. 1996. PubMed
4. Fallowfield L, Jenkins V, Farewell V, Saul J, Duffy A, Eves R. Efficacy of a Cancer Research UK communication skills training model for oncologists: a randomised controlled trial. Lancet. 2002. PubMed
5. Fallowfield L, Jenkins V, Farewell V, Solis-Trapala I. Enduring impact of communication skills training: results of a 12-month follow-up. Br J Cancer. 2003. PubMed
6. Brown VA, Parker PA, Furber L, Thomas AL. Patient preferences for the delivery of bad news: the experience of a UK Cancer Centre. Eur J Cancer Care (Engl). 2011. PubMed
7. Krieger T, Salm S, Dresen A, Cecon N. Cancer patients' experiences and preferences when receiving bad news: a qualitative study. J Cancer Res Clin Oncol. 2023. PubMed
8. Back AL, Arnold RM, Baile WF, Fryer-Edwards KA, Alexander SC, Barley GE, et al. Efficacy of communication skills training for giving bad news and discussing transitions to palliative care. Arch Intern Med. 2007. PubMed
9. Mahendiran M, Yeung H, Rossi S, Khosravani H, Perri GA. Evaluating the effectiveness of the SPIKES model to break bad news: a systematic review. Am J Hosp Palliat Care. 2023. PubMed
10. Johnson J, Panagioti M. Interventions to improve the breaking of bad or difficult news by physicians, medical students, and interns/residents: a systematic review and meta-analysis. Acad Med. 2018. PubMed
11. Fujimori M, Shirai Y, Asai M, Kubota K, Katsumata N, Uchitomi Y. Effect of communication skills training program for oncologists based on patient preferences for communication when receiving bad news: a randomized controlled trial. J Clin Oncol. 2014. PubMed
12. Dale MacLaine T, Lowe N, Dale J. The use of simulation in medical student education on the topic of breaking bad news: a systematic review. Patient Educ Couns. 2021. PubMed
13. Kang W. The use of virtual patients for breaking bad news: a rapid review. Clin Teach. 2024. PubMed
14. Schmitz FM, Schnabel KP, Bauer D, Woermann U, Guttormsen S. Learning how to break bad news from worked examples: does the presentation format matter when hints are embedded? Results from randomised and blinded field trials. Patient Educ Couns. 2020. PubMed
15. Lackie K, Miller S, Brown M, Mireault A, Helwig M, Beatty L, et al. Interprofessional collaboration between health professional learners when breaking bad news: a scoping review of teaching approaches. JBI Evid Synth. 2024. PubMed
16. Lima TAC, Bruno FP, Gushken F, Degani-Costa LH, Novaes NP. Breaking bad news in neurology: assessing training, perceptions, and preparedness among residency programs in Brazil. Einstein (Sao Paulo). 2023. PubMed
17. Meitar D, Karnieli-Miller O. Twelve tips to manage a breaking bad news process: using S-P-w-ICE-S, a revised version of the SPIKES protocol. Med Teach. 2022. PubMed