Breaking bad news and communication in the emergency department — the SPIKES framework

Breaking bad news is the single most feared clinical task in emergency medicine, and it is examined directly in every Fellowship OSCE. Bad news is any information that adversely and seriously alters a person's view of their future — a death, a terminal diagnosis, a catastrophic disability, or the disclosure that an error was made. The SPIKES protocol, first described by Baile and Buckman for the oncology consultation, is the structured, evidence-based framework that has carried into the resuscitation bay, and it is the model answer the examiner expects.[1] Communication in the ED differs from the clinic in three ways: the relationship is new and unformed, the news is often unexpected, and the time pressure is severe — yet the same six steps apply, compressed into the minutes a peri-arrest or death-notification conversation allows.[2][3]

When and why the framework matters

The Fellowship candidate must reach for a structured protocol whenever the news is serious, unexpected, or irreversible. The reasons are both ethical and practical. First, patients and families remember how they were told for the rest of their lives — the privacy, the empathy, and the silence are recalled long after the medical details fade. Second, a structured approach reduces the clinician's own anxiety and the risk of defensive, information-dumping, or jargon-laden delivery. Third, several ED conversations carry legal and regulatory weight: the death notification triggers coronial and organ-donation pathways, the medical-error disclosure triggers the open-disclosure obligation, and the goals-of-care discussion documents the ceiling of treatment. Reaching for the framework is not optional theatre — it is the standard of care. [1]

The SPIKES protocol — the six steps

SPIKES is the mnemonic that anchors the whole conversation. Each step has a single aim, and the steps are followed in order; skipping ahead to the knowledge before gauging the perception is the cardinal error, because the clinician then tells the patient what they already know or overwhelms someone who has not yet grasped the situation.[1]

The key principles — privacy, time, sitting, silence, empathy, the warning shot

Underneath the six letters sit six governing principles that make SPIKES work at the bedside. Privacy — a side room with the door closed, not a curtained bay where the next patient can hear; the bleep and the phone switched off; a chaperone or support person offered. Protected time — a conversation the clinician does not rush out of; ten unhurried minutes is more therapeutic than thirty distracted ones. Sitting down — sitting places the clinician at or below the eye level of the patient and signals that time is available; standing in the doorway signals the opposite. Silence — after the news is delivered, the clinician stops talking and waits; the grieving brain needs time to land, and most patients speak again within ten to twenty seconds if not interrupted. Empathy — articulated through the NURSE statements rather than left implicit, because an unspoken empathy is invisible empathy. The warning shot — a short phrase that prepares the listener for the blow, delivered before the news itself. [1]

Differential — the conversations distinguished

The four high-stakes ED conversations share the SPIKES skeleton but diverge in structure, legal overlay, and the questions the examiner will probe. Distinguishing them is itself a tested competency — a candidate who treats a death notification like a goals-of-care discussion, or an error disclosure like a new diagnosis, will fail the station.[2][3]

The death notification

Death notification is the conversation the family remembers forever, and it is a stand-alone OSCE station. The clinician collects the family from the waiting area, brings them to a private room, sits them down, and confirms who is present and their relationship to the deceased. The perception is gauged — "Tell me what you know about what happened to your mother today" — before the news is given. The warning shot is delivered, and then the news in plain, unambiguous words: "I am very sorry to have to tell you that your mother has died." The word died is used, not "passed away", "lost", or "gone to sleep" — euphemisms cause confusion and are culturally and linguistically unsafe.[2] The clinician then stops, and allows the grief to express itself — tears, silence, anger, denial, even laughter — without rushing to comfort or to fill the space.

After the initial reaction, the clinician offers the practical steps: the opportunity to see and touch the body (almost always offered, and culturally essential in many communities), the bereavement information, the named contact, and the timeframe. The coronial question is addressed — most unexpected, violent, or perioperative deaths, and any death within 24 hours of admission or after a procedure, are reportable. Organ donation is raised sensitively and in line with the local donor-service protocol, recognising that the decision must be decoupled from the moment of grief. A senior clinician signs the death certificate only when the criteria are met and the death is not coronial. [1]

The peri-arrest and goals-of-care discussion

The peri-arrest conversation is the most time-pressured in the ED, because a deteriorating patient may arrest within minutes, and the clinician must establish the ceiling of treatment before that happens. The setting is often imperfect — a side room off the resuscitation bay, the clock running — but the principles hold: privacy, sitting, and the perception question. The clinician asks what the patient or surrogate already knows of the prognosis ("What have the doctors told you about how serious this is?"), then introduces the conversation as one about goals and limits rather than abandonment. Structured serious-illness communication prompts improve alignment between care and values when time is short.[5]

The substance is a structured exploration of the patient's values and prior expressed wishes: what did they consider an acceptable outcome, what would they not want, and is there an advance care directive? Where the patient lacks capacity, the decision is made in the patient's best interests with the substitute decision-maker, drawing on the known values rather than the surrogate's own preference. The clinician states explicitly what will be done (ward-level care, symptom control, antibiotics, fluids, a trial of non-invasive ventilation) and what will not (cardiopulmonary resuscitation, intubation, ICU). The framing matters: a "not-for-resuscitation" order is framed as "we will focus entirely on your comfort and dignity, and we will not pursue treatments that we believe would cause you harm without benefit", never as "we will do nothing". The plan is documented, communicated to the team, and revisited as the clinical picture evolves. [1]

Medical-error disclosure and open disclosure

Open disclosure is the structured, truthful conversation with a patient and family after a healthcare-related harm, and it is now a regulatory and, in several jurisdictions, a statutory obligation.[3] The ED scenario is typically a missed diagnosis (a missed testicular cancer, an unrecognised subarachnoid haemorrhage) or a procedural complication, often discovered when the patient re-presents. The conversation is held as soon as the patient is clinically stable and the basic facts are established — delay erodes trust and is itself a harm.

The disclosure follows a sequence: an expression of regret for the harm and the experience ("I am sorry this happened, and I am sorry for what you have been through"), a factual account of what is known (without speculation beyond the established facts), an acknowledgement of what is not yet known and how it will be investigated, and a concrete plan for the ongoing care, the root-cause analysis, and the named follow-up contact. The apology is for the harm and the experience — it is not an admission of legal liability, and the clinician separates the facts from attribution. The error is documented factually in the record, an incident report is filed, and the conversation itself is noted. The second-victim phenomenon — the distress of the clinician who made or delivered the error — is recognised and addressed through the debrief and peer support. [1]

Symptom-control pharmacology — the peri-arrest and end-of-life drug doses

When the ceiling of care is comfort, the ED clinician must control the symptoms of the dying and the peri-arrest patient promptly, by the route that works, and without the fear of hastening death that paralyses untrained prescribers. The principle is that symptom relief in the dying is ethically and legally supported under the doctrine of double effect when the intent is comfort and the dose is proportionate. Four agents cover almost the entire peri-arrest symptom burden in the ED. Morphine 2.5 mg intravenously is first-line for pain and for the distressing dyspnoea of the failing circulation, titrated and repeated to effect. Midazolam 2.5 mg intravenously treats the anxiety, agitation and terminal restlessness that accompany hypoxia and hypoperfusion, also titrated. Glycopyrrolate 200 micrograms subcutaneously or intravenously reduces the respiratory secretions that produce the death rattle (it will not reverse secretions already present, and suction is both ineffective and distressing). Haloperidol 0.5 to 2 mg intravenously manages delirium and terminal agitation where a benzodiazepine alone is insufficient. These doses are subcutaneous-friendly for ongoing palliative care, and a syringe driver is arranged when the need is established. [1]

Cultural considerations

Culture shapes every layer of the conversation: who is told, who decides, how grief is shown, and what language is acceptable. The clinician does not assume a norm but asks. In many cultures the family, not the patient, is the legitimate recipient of serious news, and a direct disclosure to the patient against family wishes may be experienced as cruel; the clinician negotiates by exploring the family's reason ("What are you worried the news will do to her?"), by offering to be present for the disclosure, and by affirming that autonomy remains the patient's. A trained medical interpreter — not a family member, and never a child — is used whenever there is a language barrier; the clinician looks at the patient, not the interpreter, and speaks in short segments. Eye contact, physical touch, the gender of the clinician, and the rituals around death and the body vary widely and are respected. The organ-donation conversation is approached with particular cultural humility, recognising strong religious and community variation. [1]

The debrief and clinician self-care

The clinician who delivers the news, performs the failed resuscitation, or makes the error is a second victim, and the psychological aftermath — intrusive imagery, self-doubt, sleep disturbance, and in some cases post-traumatic stress — is common and under-recognised. The structured debrief is therefore a clinical obligation, not a courtesy. The hot debrief occurs immediately after the event, at the bedside or in the corridor, and is led by the team leader: it confirms what happened, addresses immediate clinical questions, checks on the team members, and identifies anyone who needs follow-up. The cold debrief, held hours to days later, reviews the case in detail, extracts the learning, and is the forum for the emotional as well as the technical review. Peer support, employee-assistance access, and a culture that normalises seeking help are the protective factors; their absence predicts burnout and attrition. [1]

Common errors and pitfalls

The recurring errors are those SPIKES was designed to prevent. Information dumping — delivering the news in a single, rapid, jargon-laden monologue — overwhelms and is recalled as callousness. False reassurance ("I am sure it will all be fine") corrodes trust the moment reality arrives and is never justified. Medical jargon ("she has had an infarct with a poor ejection fraction and we are concerned about her prognosis") is unintelligible to most families. Defensiveness in the error-disclosure setting, or the temptation to minimise, is the fastest route to litigation and loss of trust. Filling the silence after the news robs the patient of the time to process. Skipping the warning shot delivers the blow unannounced. Euphemisms for death ("passed", "lost", "gone") cause confusion and are avoided. Finally, failing to debrief compounds the harm to the clinician themselves. [1]

Special populations

The paediatric death notification is the most difficult conversation in medicine and demands a senior clinician, a private space, both parents where possible, and honesty pitched to the developmental age of any siblings. The cognitively impaired patient receives the news at the level their capacity allows, with a surrogate present, and never excluded by default. The non-English-speaking patient receives the news through a trained interpreter, with the clinician speaking to the patient directly. The angry or threatening relative is moved to a safe environment, de-escalated with empathy and a senior presence, and never met alone. The clinician-patient is told with the same rigour, with the added discipline of not minimising because the news is uncomfortable to give. [1]

Evidence and regional guidelines

SPIKES is the global default, with the addition of the NURSE empathy statements from the oncology and palliative-care literature.[1] The death-notification structure and the missed-diagnosis disclosure scenario are codified in the contemporary emergency-medicine teaching cases.[2][3] The regulatory overlay varies by region.

The SPIKES protocol — a structured bedside checklist

SPIKES is most useful as a checklist run silently before and during the conversation. The Fellowship candidate who can recite the six letters but cannot execute them at the bedside has not yet mastered the station. Below is the operational checklist that maps each letter to observable behaviours the examiner scores. [1]

Managing anger, distress and denial

Anger, distress and denial are normal and predictable responses to bad news; they are not pathology, and they are not personal. The Fellowship candidate who pathologises grief, argues with denial, or matches anger with defensiveness will fail the station and harm the patient. The structured response is de-escalation, validation, and a return to the plan. [1]

Anger is often the first emotion the family expresses after a death, and it is frequently displaced — at the ambulance, the triage nurse, the clinician who arrived "too late". The response is empathic, not defensive: the clinician acknowledges the emotion ("I can hear how angry and frustrated you are — that is completely understandable"), does not argue the timeline, does not justify the team's actions in that moment, and brings a senior colleague into the room if the anger is escalating. Safety is paramount — the clinician never interviews an angry relative alone, positions themselves between the relative and the door, and calls security if there is any threat. The angry relative is given a private space (a public corridor amplifies the escalation), water, and time. [1]

Distress — tears, sobbing, hyperventilation, the stunned silence of shock — is met with presence, not intervention. The clinician sits, is quiet, offers tissues and water, and waits. Comforting gestures (a hand on a shoulder) are offered cautiously and only when cultural and relational cues support them. The clinician does not medicate acute grief. [1]

Denial — "this can't be happening", "there must be some mistake", "I want a second opinion" — is respected, not argued. The clinician reflects ("I can see this is impossible to take in"), restates the facts gently without forcing acceptance, and offers the second opinion or the repeat imaging that the clinical situation supports. Denial is a protective psychological mechanism; it dissolves in its own time, and confronting it head-on produces resistance and damages the relationship. [1]

Family-witnessed resuscitation

Family-witnessed resuscitation (FWR) — the presence of family members during cardiopulmonary resuscitation, invasive procedures, and the dying moments of their relative — has shifted over two decades from controversial to recommended practice. The evidence is consistent: families who are offered the opportunity to be present report lower rates of post-traumatic stress, greater satisfaction with care, an easier bereavement, and a clearer understanding that everything possible was done. The critical finding is that the offer matters more than whether the family accepts — even families who decline report benefit from having been given the choice. [1]

Family-witnessed resuscitation is therefore offered as standard where feasible, with a dedicated liaison and team consent before family enter the bay.[4]

The practical execution is structured. A dedicated family liaison — usually a senior nurse or social worker — stays with the family throughout, explains what they are seeing in plain language, answers questions, and escorts them out if they become distressed or if the team requests it. The team leader consents to FWR before the family enters; team members are briefed; the family is positioned so they can see their relative (typically near the head, holding the hand) without obstructing the resuscitation. If the resuscitation is failing, the team leader may invite the family closer for the final moments. Termination of resuscitation is then explained in the room, not after the family has been removed — this is the most powerful antidote to the persistent fear that "they gave up". [1]

The organ donation conversation

The organ donation conversation is among the most delicate in the ED, and it is almost always held in the context of a catastrophic brain injury where death by neurological criteria is anticipated or has just been confirmed. The ED clinician's role is recognition and referral — the formal consent conversation is held by the trained donor-service coordinator, who is decoupled from the treating team to remove any conflict of interest and to bring specialist expertise. The clinician does not raise donation at the moment of death notification; it is raised in a separate conversation, after the family has had time to absorb the news, and after brain death has been confirmed or is being formally assessed. [1]

The decoupling principle is critical. Conflating the death notification with the donation request — "I'm sorry to tell you your husband has died; have you considered organ donation?" — is the cardinal error: it creates the impression that the team's interest in the organs influenced the declaration of death, and it profoundly damages trust. The treating team confirms death, supports the family, and then introduces the donor coordinator as a separate person with a separate role. Where the patient is a registered donor (on the driver's licence, the registry, or an advance directive), the clinician can frame the conversation around the patient's own recorded wishes, which substantially reduces the family's decisional burden. [1]

Time is the enemy of donation after neurological death, because the haemodynamic instability of brain death rapidly compromises organ viability. The early call to the donor service — at the point of recognising a devastating neurological injury with a poor prognosis, before death is declared — allows the coordinator to assess suitability, approach the family at the right moment, and organise the retrieval if consent is given. The ED clinician does not pre-empt the family's decision and does not record a personal opinion on donation in the medical record. [1]

Cultural and religious sensitivity — beyond the general principles

The general principle — ask, do not assume — applies with particular force to the high-stakes ED conversations, where the consequences of a cultural misstep are magnified. Several specific scenarios recur in the Fellowship exam and in practice. [1]

Jehovah's Witnesses and blood products. The refusal of whole blood, packed red cells, platelets, fresh frozen plasma and primary components is a deeply held article of faith for most Jehovah's Witnesses, and accepting blood is understood as a sin that threatens eternal salvation. The clinician respects this, documents the specific refusal (which products, in what circumstances, by whom — patient or surrogate), and explores the alternatives: cell salvage in a continuous circuit, acute normovolaemic haemodilution, fractionated products (albumin, immunoglobulins, clotting factor concentrates) where the individual accepts them, and the early use of tranexamic acid, iron and erythropoietin. The individual's conscience varies — the clinician does not assume what a particular Witness will accept, but asks. The Hospital Liaison Committee network is a valuable resource and the patient may welcome their involvement. The ED clinician does not obtain a court order to override a competent adult's refusal, even to save a life; the right to refuse treatment is absolute in capacitous adults. [1]

Brain death contested in some faith traditions. Brain death (death by neurological criteria) is the legal standard of death in most jurisdictions, but it is theologically contested in some traditions — certain Orthodox Jewish, some Buddhist, and some Islamic scholars hold that death occurs only with cardiopulmonary cessation. Where a family contests the declaration on religious grounds, the clinician engages chaplaincy, the hospital ethics committee, and where the law allows, accommodation such as a time-limited continuation of support pending further discussion. The organ donation conversation in this setting is held only after the family has accepted the declaration. [1]

Family as decision-maker. In many South Asian, East Asian, Middle Eastern and African communities, the family — not the patient — is the legitimate recipient of serious news, and a direct disclosure to the patient may be experienced as cruel. The clinician negotiates: explores the family's reason, offers to be present, affirms that the patient's autonomy remains, and where the patient is capacitous, ultimately respects the patient's own preference (which may be to delegate to the family). [1]

Rituals around the body. The timing of viewing, who may touch the body, the direction the body faces, the presence of a religious leader, and the requirement that the body be released promptly for ritual washing and burial (within 24 hours in Orthodox Jewish and Islamic practice) are all respected and facilitated where possible. The death certificate and coronial processes are explained and expedited; the ED works with chaplaincy and the family's religious community. [1]

The paediatric death notification

The death of a child is the most devastating news a clinician can deliver, and it produces the longest-lasting grief in any family. The Fellowship candidate must approach this station with a senior clinician, a private and quiet room, both parents where humanly possible, and unhurried time — the standard ED pressures do not apply here. The words used are unambiguous ("your son has died"), but the framing is gentler and the silence is longer. Parents almost universally want to see and hold their child after death, and this is facilitated — the child is cleaned and dressed where possible, lines and tubes are removed where it is safe to do so, and the family is given private time. Memory-making (handprints, footprints, a lock of hair, photographs) is offered and is associated with better bereavement outcomes. [1]

Siblings are told at their developmental level — young children need concrete, literal language ("his body stopped working and he died"), not euphemisms ("he has gone to sleep") which generate fears of sleep and separation. Parents are coached on what to say to siblings at home. The developmental honesty extends to the cause: parents are told what is known, what is not yet known, and how the investigation (including the coronial process, mandatory in most child deaths) will proceed. Blame — self-blame by parents is near-universal, particularly in sudden infant death and in accidents — is anticipated and addressed gently, with factual reassurance where it is warranted and without false reassurance where it is not. [1]

The paediatric bereavement supports — the SIDS and Kids hotline, the child bereavement services, the hospital social work and chaplaincy — are offered and the family is given a named contact and a follow-up appointment. The clinician's own wellbeing after a paediatric death is addressed below. [1]

Vicarious trauma, moral injury and clinician wellbeing after a death

The clinician who delivers the death notification, leads the failed resuscitation, or cares for the dying patient across a long shift is exposed to cumulative psychological harm that the literature now frames as vicarious trauma (the absorption of the patient's trauma through empathy and exposure) and moral injury (the distress of acting, or being unable to act, in ways that conflict with one's moral framework — the patient who could not be saved, the family who blamed the team, the system that did not provide the resource). These are distinct from burnout (which is a function of chronic workload and control), though they overlap and co-exist. [1]

The protective factors are well-described and within the ED's control. The structured hot debrief (immediately, led by the team leader — what happened, is everyone safe, who needs follow-up) and cold debrief (days later, the case reviewed in detail, learning extracted, the emotional dimension explicitly invited) are the cornerstone. Peer support — a colleague who has been there, who listens without fixing — is the most consistently protective factor in the literature. Access to employee-assistance programs, a workplace culture that normalises seeking help, regular breaks, hydration, sleep, and the deliberate separation of work from home are the daily practices. The unhelpful responses — alcohol, isolation, presenteeism, minimisation — are recognised and named. [1]

The second-victim phenomenon, first described by Albert Wu, is the predictable psychological injury to the clinician who was involved in (or who delivered the news of) an adverse event. The trajectory is well-mapped: an initial impact phase (shock, rumination, self-doubt), an ensuing phase (insecurity, fear of judgement, withdrawal), and either a recovery (with support) or a trajectory toward burnout, attrition, or chronic post-traumatic stress. The ED clinician who recognises these phases in themselves or a colleague has a professional obligation to act on them — to seek the debrief, the peer, the employee-assistance referral, and where needed, formal psychological care. This is not weakness; it is the standard of self-care. [1]

Structured handover — ISBAR and SBAR

Communication failure is the single most common contributor to adverse events in emergency care, and the structured handover is the evidence-based mitigation. The ED uses ISBAR (Identity, Situation, Background, Assessment, Recommendation) — an extension of the original SBAR — for shift changeovers, escalations to senior staff, transfers of care, and the pre-hospital-to-ED handover. The structure forces a logical sequence, reduces omission, and equalises the hierarchy (a junior can escalate to a senior using ISBAR with confidence). The Fellowship candidate is expected to use ISBAR fluently in the OSCE and to model it on the floor. [1]

End-of-life care in the emergency department

End-of-life care has shifted from the ward and the hospice into the ED over the last two decades, because patients present in the last weeks of life with symptoms they and their families cannot manage at home, and because the ED is increasingly the gateway for the unselected dying patient. The Fellowship candidate is expected to deliver active, compassionate, evidence-based end-of-life care within the time-pressured ED environment — this is not a failure of curative medicine, it is a distinct clinical domain. [1]

The recognition of the dying patient is the first step, and it is under-taught. The trajectories are well-described: the cancer trajectory (a slow decline with a sharp terminal drop), the organ-failure trajectory (a stepwise decline with acute exacerbations, each closer to death), the frailty and dementia trajectory (a slow progressive decline), and the sudden death. The clinician asks the "surprise question" — "would I be surprised if this patient died in the next 12 months?" — and a "no" answer triggers a palliative overlay on the ED care. [1]

The goals-of-care conversation (above) is the centrepiece. Where the patient is recognised as dying and the goals are comfort, the ED delivers: prompt symptom control (morphine 2.5 mg IV for pain and dyspnoea, midazolam 2.5 mg IV for agitation, glycopyrrolamine 200 micrograms for secretions, haloperidol 0.5 to 2 mg for terminal delirium — see the symptom-control section), a side room if possible, the family invited in, the noise and the monitors reduced, and the chaplain and social work engaged. The fast-track discharge home or to a hospice, with a community palliative plan in place within hours, is the gold standard for the patient who wants to die at home and is clinically stable enough to transfer. [1]

The bereavement conversation at the bedside of the patient who has just died — distinct from the formal death notification to relatives who were not present — acknowledges the family's loss in real time, allows private time with the body, and offers the practical next steps. The verification of death by the ED clinician follows the structured sequence (no response to pain, no central pulse, no heart sounds for one minute, no respiratory effort, fixed and dilated pupils), and the death certificate is completed when the death is not coronial. [1]

SAQs [1]

Exam pearls

• SPIKES in order: Setting, Perception, Invitation, Knowledge, Emotion, Strategy/Summary — skipping Perception is the most common station failure.
• The warning shot, then the silence — say "I am afraid I have some serious news", deliver it, then stop talking.
• Use the words "died" and "dead" in death notification — never euphemisms.
• Autonomy overrides a family request for non-disclosure — assess capacity, explore, offer presence, do not deceive.
• Open disclosure apologises for the harm and the experience without admitting legal liability beyond the facts.
• Peri-arrest comfort dosing: morphine 2.5 mg IV, midazolam 2.5 mg IV, glycopyrrolate 200 micrograms, haloperidol 0.5 to 2 mg.
• The clinician is a second victim — the hot and cold debrief are clinical obligations. [1]

Red flags