Background: Goals-of-Care Conversations in the ED Context

Initiating goals-of-care or “conversations-before-investigations” in the emergency department (ED) is critical for aligning acute care with patient values, especially for older patients with cognitive impairment, dementia, or other cognitive disabilities. These conversations – which explore a patient’s wishes, prognosis, and appropriate scope of interventions – can help prevent non-beneficial treatments and ensure care is goal-concordant (matching what the patient would want). ED clinicians, however, often find it challenging to pause and engage in such discussions amid the urgency of acute care. Patients with dementia or cognitive frailty pose additional complexity: their subjective preferences and experiences are difficult to ascertain, and they may lack decision-making capacity, necessitating the involvement of surrogates or reliance on prior directives. Without these conversations, clinicians may default to aggressive investigations and interventions that risk misalignment with the patient’s values, potentially causing harm or prolonging suffering in ways the patient would not have wanted. Recent literature highlights that while emergency physicians overwhelmingly agree that discussing goals of care is within their scope and essential, significant clinician-facing barriers remain that impede these discussions in practice. Below, we review key clinician-centred obstacles – such as uncertainty, discomfort, training gaps, cultural norms, hierarchy, risk perception, and role ambiguity – and how these manifest in the ED setting.

Clinician-Facing Barriers to Early Conversations in the ED

Prognostic Uncertainty and Epistemic Challenges

Emergency clinicians often face epistemic uncertainty about a critically ill patient’s trajectory, especially during brief ED encounters. It can be challenging to determine whether an aggressive intervention will help or merely prolong the dying process, particularly in patients with advanced dementia or multi-morbidity. This uncertainty can make physicians hesitant to initiate end-of-life conversations – they may fear “getting it wrong” or be unsure if the situation truly warrants a discussion about the goals of care. Research confirms that prognostic uncertainty is a recognised barrier: clinicians frequently report difficulty predicting illness course in the ED and thus struggle with the timing of goals-of-care discussions. In one survey, 66% of ED physicians admitted they had trouble initiating goals-of-care conversations, even if, in principle, they felt comfortable with the idea. Clinicians may err on the side of caution by proceeding with diagnostic tests or treatments “just in case” rather than pausing to discuss goals under uncertain circumstances. This dynamic is compounded in cognitively impaired patients whose baseline quality of life and wishes are not readily knowable – ED providers might default to maximal intervention in the absence of clear information to avoid the epistemic risk of undertreatment. Unfortunately, this uncertainty-driven caution can lead to overtreatment and non-beneficial interventions, as clinicians pursue every diagnostic avenue without first clarifying whether such interventions align with the patient’s goals. Developing better tools to assess prognosis and patient preferences in the ED (for example, rapid screening for advance directives or involving geriatric and palliative consultants) has been suggested as one way to mitigate this barrier; however, significant challenges remain.

Moral Discomfort and Ethical Dilemmas

ED clinicians may experience moral discomfort or distress when caring for patients with serious illness and cognitive impairment, especially if they sense that the aggressive default approach of emergency medicine might be causing harm. The culture of emergency care emphasises saving lives and “doing everything” in acute crises, which can sometimes conflict with the needs of a dying patient for comfort and dignity. This mismatch often gives rise to ethical dilemmas and inner conflict. For example, ED nurses have reported moral distress when forced to provide life-prolonging interventions that they feel are not in a patient’s best interest, describing aggressive emergency treatments at end-of-life as “absolutely at odds” with what compassionate care would entail. Physicians share similar struggles – they might feel a professional or moral duty to attempt every possible intervention, yet simultaneously feel it is wrong to subject a frail, cognitively impaired patient to invasive procedures that are unlikely to help. One qualitative study of Australian ED staff described clinicians feeling “caught in the middle” between the imperatives of emergency medicine and the principles of end-of-life ethics; participants encountered frequent ethical dilemmas about whether to initiate or forgo aggressive treatments. This moral tension can manifest as hesitation to begin candid conversations about poor prognosis or comfort-focused care. Some clinicians are uncomfortable acknowledging dying or worry that raising these issues will distress families (a “fear of causing distress” that is documented in the literature as a barrier). In other cases, moral discomfort may drive clinicians to avoid the conversation and continue with business-as-usual interventions, even as they inwardly question the benefit. Over time, unresolved moral distress in the ED is linked to provider burnout. It may perpetuate a cycle where clinicians “do everything” to avoid the guilt or uncertainty associated with limiting treatment. Addressing this barrier requires creating a safe environment for clinicians to voice ethical concerns and supporting them (through ethics consults or debriefings) in making ethically sound decisions that prioritise patient values.

Insufficient Training and Communication Skills

Many emergency physicians and trainees report insufficient training in end-of-life communication and ethical decision-making, which undermines their confidence in starting “goals-of-care” conversations. Unlike oncologists or palliative care specialists, ED clinicians historically have had limited formal education in conducting delicate family meetings or discussing prognosis and code status. Studies indicate that inexperience and feeling unprepared are common clinician factors that impede end-of-life discussions. One review noted that doctors often feel they lack the appropriate phrases, language, and approach to discuss death, dying, or potentially futile interventions, especially in the heat of an emergency. This communication skill gap is evident in practice: a recent national survey found that while over half of emergency physicians see patients who could benefit from palliative care at least weekly, only 29% felt “very comfortable” initiating palliative or end-of-life conversations. Many are more at ease with concrete, procedure-focused questions (e.g., discussing whether to intubate or attempt resuscitation) than with broader explorations of the patient’s values and long-term outlook. In the survey, a lack of specialised training was rated as one of the top barriers to starting end-of-life discussions, second only to time constraints. The consequence of this training deficit is that clinicians may either avoid the conversation entirely or handle it suboptimally (for instance, using medical jargon or framing choices poorly), which can potentially lead to misunderstandings. In the context of cognitively impaired patients, where communication is already challenging, an unskilled conversation can easily be derailed or avoided. The literature suggests that enhancing ED provider education in palliative care communication and ethical reasoning would increase their comfort and competence; indeed, one study found that after targeted training, ED-initiated palliative consults for patients with advanced dementia modestly increased. Overall, limited training leaves clinicians ill-equipped to navigate these complex discussions, reinforcing the tendency to stick with familiar investigative and treatment routines rather than venturing into conversations that require a different skill set.

Institutional Norms and Hierarchy in the ED

The fast-paced, intervention-driven culture of many EDs creates institutional norms that can discourage taking the time for conversations. Emergency teams pride themselves on rapid diagnosis and treatment; pausing to discuss goals of care may be viewed (implicitly) as contrary to the mission of emergency medicine or as something that “slows down” patient throughput. Such norms can exert subtle pressure on clinicians to “treat first, talk later.” For example, ED physicians often work under the expectation that every acute problem must be aggressively addressed unless a formal “Not for CPR” order or advance directive explicitly limits intervention. This expectation can make clinicians feel that initiating a comfort-oriented discussion deviates from the standard of care, potentially inviting scrutiny from colleagues. Institutional inertia also plays a role: if an ED lacks protocols or support for goals-of-care discussions (e.g., no readily available palliative care consult service), clinicians may perceive that their role is to stabilise and admit, leaving the hard conversations to others. Indeed, a Canadian survey found that 54% of emergency physicians believed that admitting inpatient teams should conduct goals-of-care discussions, reflecting a typical attitude that the emergency department (ED) is not the ideal setting for these discussions. This points to a role ambiguityreinforced by institutional norms – even though most ED doctors in that study agreed that goals-of-care discussions are part of their scope, more than half still felt it wasn’t primarily their responsibility in practice. Such norms can lead to “missed conversations” in the ED, as opportunities for clarifying a patient’s wishes are often deferred and lost, especially if the patient deteriorates quickly or loses decision-making capacity by the time another team takes over.

Closely tied to norms is the influence of clinical hierarchies in the ED. Emergency departments are team environments where specialists supervise registrars, residents, interns, nurses, and other healthcare professionals. Hierarchical dynamics may inhibit junior clinicians from initiating end-of-life conversations if their supervisors have not signalled it is appropriate. For instance, a junior doctor may recognise that an elderly patient with advanced dementia is unlikely to benefit from invasive treatment yet feel unable to broach limiting interventions without the specialist’s lead – fearing overstepping their role or contradicting a senior’s more aggressive plan. Nurses, too, can feel caught in the hierarchy: they often spend more time at the bedside. They may sense when care is futile or misaligned with patient comfort, but they might struggle to convince a physician to consider palliative measures. One qualitative study noted that conflicts about end-of-life treatment decisions in the ED tended to occur between staff of different seniority – for example, tension between senior nurses and junior doctors when less experienced physicians were hesitant to deviate from maximum treatment protocols.

In contrast, senior doctors with more experience in end-of-life care were more likely to collaborate with nurses and acknowledge when comfort-focused care was appropriate. These hierarchical patterns suggest that younger clinicians may experience considerable role confusion and pressure to conform to a perceived “heroic medicine” ethos rather than initiating a goals-of-care discussion that might be met with disapproval. The result can be overtreatment or delayed decision-making (waiting for a senior physician or another service to address goals of care). During this time, the patient may undergo burdensome interventions by default.

Risk Aversion and Fear of Repercussions

Emergency clinicians also operate under a degree of risk aversion that can discourage the use of limiting interventions. Emergency medicine is a specialty where missing a treatable condition or making an incorrect disposition can have immediate consequences, and providers are trained to err on the side of intervention. This mentality extends to end-of-life situations. Some physicians worry about legal or professional repercussions if they do not pursue every available test or treatment. For example, there may be a lingering fear (even if infrequent in reality) of malpractice allegations for “withholding treatment” or concern about being seen by peers as rationing care. Interestingly, recent data suggest that this fear of litigation or backlash is not as prominent as traditionally thought. In a 2023 survey, emergency physicians ranked “fear of negative reactions from patients/families or fear of malpractice” very low as barriers to end-of-life discussions. This indicates that explicit legal fear may be diminishing, possibly due to the growing acceptance of palliative approaches. However, risk aversion, in a broader sense, still pervades decision-making. Clinicians may be psychologically averse to the clinical risk of not intervening – for instance; they might think, “If I don’t intubate this fragile patient and they die, will I have failed them? Better to do it just in case.” Such defensive reasoning favours action over conversation. Likewise, if a patient’s code status or wishes are unknown (common in cognitively impaired individuals arriving without family or documentation), the safest medicolegal course is often perceived as full resuscitation. This default approach, driven by “better safe than sorry” logic, leads to many patients with advanced illnesses receiving aggressive, ultimately non-beneficial treatments before anyone has paused to consider if that aligns with their values. In sum, while outright fear of malpractice may not be the top concern, emergency clinicians often have a general risk-averse mindset that biases them toward intervention and may inhibit them from initiating conversations that could lead to a decision not to intervene. Overcoming this barrier might involve emphasising that goal-concordant care is not “doing nothing” but rather doing what is right for the patient – a reframing that could reduce the perceived professional risk of choosing comfort and communication over tests and procedures.

Ambiguity in Role and Responsibility

Finally, ambiguity about whose role it is to hold end-of-life conversations can impede timely discussions in the ED. As mentioned, many ED doctors believe that while they can discuss goals of care, the primary responsibility lies with outpatient providers or admitting hospital teams, who know the patient better. This can lead to a diffusion of responsibility, where each clinician assumes someone else (the general physician, the ICU team, the primary care doctor, etc.) will address resuscitation status and goals of care. In the fast turnover of emergency care, such conversations then go unaddressed. This role ambiguity is especially problematic for patients with cognitive impairment who may present repeatedly to the ED. If each episode is treated in isolation with urgent investigations and no one claims ownership of the goals-of-care discussion, the patient may endure fragmented care and potentially burdensome interventions that are never contextualised within a broader care plan. The literature highlights this gap – for example, in one study, a significant portion of ED providers acknowledged that no prior goals-of-care discussionshad occurred for the patient (21% cited this as a barrier). Yet, they did not feel it was appropriate or feasible to initiate one on the spot. Without clarity in roles, essential conversations are “kicked down the road,” resulting in missed opportunities to align care with the patient’s values. Clarifying that ED providers should initiate at least a preliminary goals-of-care dialogue (when feasible) might improve outcomes. Still, it would require a cultural shift and possibly systemic support (such as prompts in the electronic record or involvement of ED-based palliative care liaisons). Researchers have suggested that establishing ED-specific protocols or triggers for goals-of-care discussions (akin to how trauma teams have protocols) could help normalise these conversations as part of the ED workflow. Absent clear responsibility, however, the inertia of “not my role” perpetuates, to the detriment of patients whose care continues in a goal-discordant default mode.

In summary, the clinician-facing barriers in the ED – from prognostic ambiguity and moral unease to lack of training, entrenched norms, hierarchy, risk aversion, and role confusion – all converge to make it less likely that an emergency clinician will initiate a “conversation-before-investigation.” These barriers often reinforce each other. For instance, an ED doctor who is uncertain about prognosis (epistemic uncertainty) and lacks confidence in communication skills may readily yield to institutional norms of aggressive care, especially if no senior colleague encourages a different approach. The outcome, as documented in the literature, is that many ED patients with advanced age or dementia receive high-intensity, non-beneficial interventions (intubations, CPR, unnecessary tests) that do not improve their quality of life. Meanwhile, opportunities for goal-concordant decision-making – such as focusing on comfort or hospice referrals – are missed, leading to care that is misaligned with patient preferences and potentially avoidable suffering. Recognising these clinician-facing obstacles is the first step toward designing interventions to overcome them. The next step is to explore targeted research questions that can deepen our understanding of these barriers and inform effective strategies for addressing them.

Research Questions and Future Directions

Building on the above review, we propose three research questions to guide further investigation into clinician-facing barriers and their impact on the quality of care. Each question is formulated to spur both empirical inquiry and theoretical reflection on how to mitigate the harms of non-beneficial treatment, missed conversations, and misaligned care in the ED setting:

  1. How does prognostic and epistemic uncertainty influence ED clinicians’ decision-making for cognitively impaired patients, and in what ways might this uncertainty lead to overtreatment or avoidance of goals-of-care conversations?

Rationale: Uncertainty about a patient’s prognosis or wishes was identified as a significant barrier (emergency physicians often struggle to predict outcomes during brief encounters). This question calls for research into the cognitive processes and risk perceptions of ED providers. For example, qualitative interviews or decision-making simulations could reveal how uncertainty about a dementia patient’s trajectory or a lack of clear advance directives influences clinicians’ decisions, tipping them towards immediate investigations rather than conversations. By elucidating this, we can understand how often uncertainty results in non-beneficial interventions (like extensive workups or resuscitation that may not align with patient goals). Such research could also explore interventions (decision aids, prognostic tools, communication of uncertainty strategies) to see if clarifying uncertainty or increasing tolerance for it can reduce the inclination to default to maximal treatment. Ultimately, answering this question will illuminate the mechanism by which epistemic uncertainty contributes to goal-discordant care and suggest ways to support clinicians in managing uncertainty more ethically.

  1. What are the sources and impacts of moral discomfort (or ethical distress) among ED clinicians when treating patients with cognitive impairment at the end of life, and how does this moral discomfort affect the likelihood of initiating goals-of-care discussions versus continuing potentially non-beneficial treatments?

Rationale: The review highlighted that many ED providers feel moral unease when the care they are providing is not in the patient’s best interest (for instance, performing invasive procedures on a severely demented, terminally ill patient). However, this moral discomfort can be cut both ways – it might motivate some clinicians to speak up and initiate a goals-of-care conversation, or it might, conversely, lead to moral distress, where they “suffer in silence” and continue aggressive care due to external pressures. This research question invites mixed-methods studies: surveys could quantify how prevalent moral distress is among ED physicians and nurses regarding end-of-life care (and whether higher moral distress correlates with more burnout or defensive practice). In contrast, in-depth interviews or ethics case studies could uncover why clinicians feel uncomfortable (e.g. internal values conflict, fear of harming the patient or upsetting the family) and how they respond to that discomfort. We also need to examine the consequences: Does moral discomfort without institutional support result in more missed conversations and continued futile interventions? Conversely, are there instances where moral discomfort has prompted positive action, such as an ED clinician arranging a palliative care consultation or a family meeting? Understanding this will clarify how emotional and ethical factors contribute to the persistence of non-beneficial treatments. It will also help in designing support systems (like moral distress debriefings or ethics training) to channel clinicians’ ethical concerns into proactive communication rather than silent moral suffering.

  1. How do emergency department culture and team dynamics – including institutional norms, clinical hierarchies, and unclear role responsibilities – hinder clinicians from engaging in “conversations before investigations,” and what strategies could redefine these norms to support goal-concordant care better?

Rationale: This question addresses the complex social environment of the ED that our review found to be influential. We observed that many clinicians defer goals-of-care discussions due to cultural expectations (e.g., “the ED is not the place for that”) and hierarchy (junior staff deferring to senior staff or everyone deferring to inpatient teams). To investigate this, ethnographic studies or observational research in EDs could be employed – for instance, observing how often goals of care are discussed, in what circumstances, and who initiates them. Focus groups with ED staff could reveal perceptions like “It’s frowned upon to slow down for a family meeting in our ED” or “Residents don’t feel it’s their place to make DNR decisions in the ED.” Armed with such data, we can identify specific cultural or structural modifications, such as whether an ED policy that explicitly encourages palliative discussions in specific scenarios would change the norm. Would involving a palliative care nurse or social worker on the ED team empower junior clinicians to speak up more effectively? This question also allows exploration of the solutions – interventions like training attending physicians to mentor residents in real-time on having these conversations or instituting trigger criteria for when a goals-of-care discussion must be considered (such as advanced age with critical illness or known dementia with acute decline). The goal is to find actionable ways to shift the institutional mindset from one that inadvertently perpetuates non-beneficial interventions to one that normalises aligning treatment with patient goals, even in the emergency department (ED). By researching and answering this question, we can justify and design interventions (from simulation training to protocol changes) that directly address the normative and organisational barriers, thereby reducing the frequency of misaligned, harmful care resulting from “missed” conversations.

In conclusion, tackling these research questions will deepen our understanding of why ED clinicians often struggle to initiate critical conversations with or on behalf of cognitively impaired patients, even when such conversations could prevent harm. Each question targets a different facet – cognitive (uncertainty), emotional (moral distress), and social (culture and hierarchy) – reflecting the multifaceted nature of the problem. Addressing clinician-facing barriers is crucial for enhancing goal-concordant decision-making in emergency settings. By systematically investigating these questions, future research can inform interventions to support ED providers in overcoming their uncertainties, ethical qualms, and systemic constraints. The ultimate aim is to ensure that even in the urgency of the ED, patients with dementia or cognitive disability receive care that is compassionate, appropriate, and aligned with their values – rather than care that defaults to the maximal intervention by the absence of discussion. This alignment could reduce non-beneficial treatments and improve end-of-life experiences for some of our most vulnerable patients, transforming the ED from a site of potential overtreatment into an environment where timely, values-driven conversations guide the care delivered.