SPIKES protocol training scenario — delivering a serious diagnosis to a standardized patient
Delivering serious medical news is one of the most difficult and consequential tasks in clinical practice. In 2000, Walter Baile and colleagues at MD Anderson Cancer Center published the SPIKES protocol in The Oncologist — a six-step, evidence-informed framework that remains the most widely taught model for breaking bad news in medical education worldwide.
Before SPIKES, breaking bad news was treated as an innate personality trait — some clinicians were assumed to be "naturally good" at it, others were not, and almost no one was formally taught how. Ptacek and Eberhardt's influential 1996 JAMA review of the literature found that physicians receive strikingly little structured training in this skill despite performing it thousands of times across a career, and that poor delivery is strongly associated with patient dissatisfaction, distrust, and increased risk of complaints or litigation.
SPIKES reframed breaking bad news as a teachable clinical procedure with discrete, checkable steps — much like a physical exam or a resuscitation algorithm. Each letter names both a goal and a set of concrete communication behaviors, making the skill assessable in simulation with standardized patients, exactly as this training scenario models.
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." The Oncologist. 2000;5(4):302–311 — one of the most cited papers in communication-skills education.
The "S" step is deceptively simple but strongly predicts how the rest of the conversation unfolds:
• Privacy: a closed door, a private room — never a hallway or a shared bay • Involve significant others: ask the patient whether they would like a family member or support person present • Sit down: standing communicates rushedness and creates a power gradient; sitting at eye level signals the clinician has time • Eliminate interruptions: pagers and phones silenced, a "do not disturb" arrangement made with colleagues in advance • Manage time realistically: block enough time that the conversation does not feel truncated • Prepare mentally: review the chart, the results, and rehearse the opening line before entering the room
Studies of patient satisfaction consistently rank the physical setting and the clinician's composure as highly salient — patients remember whether the clinician seemed rushed far more clearly than they remember the technical details discussed.
Since the early 2000s, most North American and European medical schools have incorporated standardized-patient (SP) encounters — trained actors portraying patients receiving a new diagnosis — as a core Objective Structured Clinical Examination (OSCE) station. Learners are scored against behavioral checklists derived directly from SPIKES.
This training scenario mirrors that format: the "News Severity" control simulates how grave the diagnosis is (a borderline finding versus a terminal prognosis), and the "Learner Empathy Skill" control simulates how well the trainee executes the emotion-handling steps (E) — the two variables shown by research to most strongly predict the standardized patient's emotional trajectory and resulting satisfaction score.
Before disclosing anything, SPIKES instructs the clinician to first find out what the patient already believes. The "P" step — sometimes remembered as "before you tell, ask" — surfaces denial, unrealistic optimism, or accurate understanding, all of which change how the news should be framed.
The step is operationalized with open-ended questions such as: "What have the doctors told you about your illness so far?" or "What is your understanding of why we did this scan?" These are deliberately non-leading — a closed question like "Do you know you might have cancer?" pre-loads the answer and forfeits diagnostic information about the patient's baseline understanding.
Listening carefully at this stage reveals three common patterns: accurate understanding (the patient has correctly pieced together the clinical picture from prior visits), informational gaps (the patient genuinely was never told key facts), and denial or wishful thinking (the patient minimizes or reframes prior warnings). Each pattern calls for a different pacing and vocabulary in the Knowledge step that follows.
Denial is not simply an obstacle to be corrected; it is a psychological buffer that allows patients to titrate how much threatening information they process at once. Mild, transient denial ("I'm sure it's nothing") early in an encounter is common and often adaptive — it lets the patient continue functioning while an unbearable possibility slowly becomes cognitively available.
Problems arise when denial is rigid and persists after clear, repeated disclosure, interfering with treatment decisions. The Perception step allows the clinician to calibrate: a patient who is already frightened and accurate needs a gentler, more validating Knowledge step, while a patient in confident denial needs the warning shot to be more explicit to avoid the news being unconsciously filtered out.
Communication researchers note that skipping the Perception step is one of the most common protocol violations among trainees under time pressure — and one of the strongest predictors of patients feeling "blindsided" rather than informed.
Not every patient wants the same depth of information at the same time. The "I" step secures explicit or implicit permission to proceed with disclosure, respecting patient autonomy over how much detail they receive and when — a principle rooted in informed-consent ethics as much as in communication technique.
Typical invitation phrasing: "Would it be all right if I go through the results with you now?" or "Some people want every detail, others prefer the big picture first — how would you like me to go through this?" This step converts disclosure from something done to the patient into something done with the patient.
Most patients, across large surveys in North America, Europe, and Australia, report wanting comprehensive information about diagnosis and prognosis — but a meaningful minority prefer a staged approach, wanting the diagnosis first and prognostic detail (e.g., survival statistics) only on a later visit, or only if a family member is present. The Invitation step is what allows the clinician to detect and honor this variation rather than assuming a one-size-fits-all disclosure style.
In many cultural contexts, full autonomous disclosure directly to the patient is not the prevailing norm; family members may request that certain information — particularly a terminal prognosis — be filtered or withheld from the patient to protect them ("protective truth-telling"). SPIKES does not mandate a single culturally universal approach; instead, the Invitation step is precisely the mechanism by which the clinician negotiates, in real time, how disclosure should be structured for this particular patient and family.
Ethically, the clinician still owes the patient the opportunity to receive information if they want it — the Invitation step protects against both the paternalistic extreme (withholding information the patient wants) and the mechanistic extreme (delivering a standardized information dump regardless of readiness).
A patient who declines full detail today may want it at a future visit. SPIKES treats Invitation as a repeatable checkpoint, not a single irreversible gate at the start of the relationship.
The Knowledge step is the moment the actual bad news is spoken aloud. SPIKES prescribes a specific technique: a brief verbal "warning shot" that signals difficult information is coming, followed by clear, jargon-free disclosure delivered in small chunks with frequent checks for understanding.
A warning shot is a brief statement that telegraphs the emotional weight of what follows without yet stating the diagnosis: "I'm afraid the news is not what we were hoping for," or "I wish I had better news to share with you today." This one sentence gives the patient's nervous system a fraction of a second to brace — measurably reducing the shock of the subsequent statement and improving later recall of what was said, because a small anticipatory startle response primes attention rather than triggering a dissociative freeze.
Skipping the warning shot and stating a diagnosis cold ("Your biopsy showed adenocarcinoma") is one of the most reliable ways to trigger acute dissociation — patients frequently report that everything said after the first unexpected sentence became a blur, undermining every subsequent step of the conversation regardless of how well it was executed.
Once the news is stated, SPIKES calls for:
• Avoiding jargon: "the tumor has spread to your liver" rather than "we found hepatic metastases" • Avoiding excessive bluntness: technical accuracy without warmth reads as callous even when factually correct • Chunk and check: deliver one idea, pause, and ask "Does that make sense so far?" before continuing — cognitive load research shows that comprehension collapses when multiple complex facts are delivered in rapid succession under emotional load • Aligning with the patient's vocabulary from the Perception step: if the patient used their own words for the illness, echoing that language builds rapport and comprehension • Silence: allowing pauses rather than rushing to fill them lets the patient process rather than simply receive
Studies of recorded oncology consultations show that once emotional arousal spikes, working-memory capacity for new verbal information drops sharply — patients frequently cannot accurately recall the specific words used minutes later, which is why written summaries and recorded encounters are increasingly recommended.
In this simulation, "News Severity" scales how abrupt and threatening the disclosed information is — a borderline biopsy result sits at the low end, while a metastatic, life-limiting diagnosis sits at the high end. Higher severity produces a steeper spike in patient distress at this exact stage, visualized as the aura shifting sharply toward grey-white shock. How well the learner then executes Step 5 (Emotion) determines whether that spike resolves toward sadness and eventual engagement, or hardens into unaddressed anger or withdrawal.
The Emotion step is widely regarded as the most difficult — and most consequential — part of SPIKES. Rather than retreating into more facts when a patient becomes tearful or angry, the clinician is trained to pause the information stream and respond directly to the emotion using the NURSE taxonomy of empathic statements.
Developed and popularized by Anthony Back, Robert Arnold, Walter Baile, James Tulsky and colleagues, NURSE gives clinicians five concrete sentence-level tools for responding to emotion:
• Name: "It sounds like this is very frightening news." — labeling the emotion out loud, tentatively, so the patient can confirm or correct it • Understand: "I can only imagine how overwhelming this must feel." — validating without claiming to fully know the patient's experience • Respect: "You have been asking exactly the right questions — that takes real strength." — acknowledging the patient's or family's coping efforts • Support: "Whatever happens, I will be with you through this." — pledging non-abandonment, one of the most reassuring statements a patient can hear • Explore: "Can you tell me more about what worries you most?" — inviting the patient to elaborate rather than assuming what they mean
Critically, NURSE statements are used instead of immediately resuming factual disclosure. Returning to data while a patient is crying is experienced as the clinician avoiding the emotional reality of the moment.
Back AL, Arnold RM, Baile WF, Tulsky JA, Fryer-Edwards K. "Approaching difficult communication tasks in oncology." CA Cancer J Clin. 2005;55(3):164–177 — the primary source formalizing the NURSE mnemonic for oncology communication training.
Patient emotional reactions to bad news vary widely and are not linear. Some patients move rapidly through visible shock into sadness and tears; others externalize distress as anger — at the disease, at delays in diagnosis, or, most commonly, displaced anger directed at the messenger. A well-trained clinician does not personalize anger; it is treated as a symptom of distress requiring the same NURSE response as tears.
Audio-recorded studies of real oncology visits have repeatedly found that clinicians recognize and empathically respond to only a minority of the emotional cues patients actually offer — often estimated at roughly one in ten identifiable cues — with most emotional moments instead met with a reflexive return to facts, reassurance that shuts down further disclosure ("try not to worry"), or silence that the patient experiences as abandonment.
In this scenario, "Learner Empathy Skill" determines how many effective NURSE statements the trainee deploys and how well-timed they are. At high empathy skill, the patient's aura transitions from grey shock through blue sadness toward green acceptance within this stage — distress measurably falls as the patient feels heard. At low empathy skill, the simulation keeps distress elevated or lets it curdle into red anger, because unaddressed emotional cues compound rather than dissipate — mirroring what communication-skills training research consistently documents in real encounters.
The final step closes the encounter with a clear plan: summarizing what was discussed, confirming the patient's understanding, and outlining concrete next steps. Getting this far well is not merely a communication nicety — it measurably affects patient trust, adherence, and even clinician burnout.
The Strategy step accomplishes three things: it summarizes the key points covered so the patient leaves with a coherent narrative rather than fragments; it checks retained understanding ("Can you tell me in your own words what we've discussed?"); and it establishes a concrete next step — a follow-up appointment, a treatment-planning meeting, a phone number to call with questions, written materials to take home.
Ending without a plan is strongly associated with patients describing the encounter as abandonment — even when the clinical facts were delivered accurately. A stated next step, however small, restores a sense of agency and reduces the acute helplessness that follows a serious diagnosis.
Multiple observational studies link SPIKES-consistent delivery to measurably better patient outcomes: higher satisfaction with the consultation, greater trust in the clinician, improved short-term psychological adjustment, and better treatment adherence in the weeks following diagnosis. Conversely, poorly delivered bad news is associated with prolonged anxiety and depressive symptoms, mistrust of the medical team, and delayed engagement with treatment.
In the United States, analyses of malpractice litigation have found that a large majority of communication-related claims stem not from technical errors but from patients or families feeling that they were not listened to, not informed adequately, or treated without empathy — underscoring that the Emotion and Strategy steps are not "soft skills" layered on top of medicine, but load-bearing components of safe clinical care.
Fujimori M, Uchitomi Y. "Preferences of cancer patients regarding communication of bad news: a systematic literature review." Jpn J Clin Oncol. 2009;39(4):201–216 — patients across multiple cultures consistently rank emotional support and a clear plan alongside honesty as top priorities.
Repeatedly delivering bad news without adequate skills or support extracts a measurable toll on clinicians themselves. Ramirez and colleagues' landmark 1996 study in the British Journal of Cancer linked hospital consultants' burnout directly to communication-skills deficits and the emotional strain of breaking bad news without adequate training — clinicians who felt unprepared for these conversations reported significantly higher emotional exhaustion.
Surveys of oncologists in the following decades have consistently found burnout prevalence in a similar range, driven substantially by the cumulative weight of unsupported emotional labor. Structured training such as SPIKES-based simulation with standardized patients has been shown in multiple studies to increase clinician confidence and reduce self-reported distress associated with these conversations — reframing communication training as a form of both patient safety and clinician wellbeing intervention.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| S — Setting | Private space, eye level, no interruptions | Prepares the physical and relational context before any disclosure begins | "Let's sit down together — I've set aside time to talk properly." |
| P — Perception | Establish what the patient already believes | Open-ended probe reveals denial, accuracy, or informational gaps | "What have you been told about your condition so far?" |
| I — Invitation | Obtain permission for the level of detail shared | Respects autonomy over pacing and depth of disclosure | "Would it be okay if I go through the results with you now?" |
| K — Knowledge | Deliver the news clearly, in digestible chunks | Warning shot + plain language + chunk-and-check | "I'm afraid I have some serious news — the biopsy shows cancer." |
| E — Emotion | Respond to the emotional reaction, not just the facts | NURSE: Name, Understand, Respect, Support, Explore | "It sounds like this is very frightening news to take in." |
| S — Strategy/Summary | Summarize, check understanding, plan next steps | Converts a diagnosis into an actionable, less helpless path forward | "Let's go over what we've covered, and talk about what happens next." |