Empower Emergency Physicians to Make Patient-Centered Recommendations Regarding Code Status With Serious Illness Communication Training—Resident-Desired, Standard of Emergency Care in 2023

Intensive care is initiated daily for approximately 2000 older adults (age ≥65 years) in US emergency departments (EDs).¹ Unfortunately, most older adults do not fully comprehend the risk of developing a new disability as a result of their critical illness.² During the last 6 months of life, 75% of older adults visit the ED, and 30% of these patients start intensive care.^3,4 Intensive care use in this population increased from 18.5% in 1993 to 24.7% in 2002, resulting in an annual healthcare expenditure of $108 billion.^5,6 Although the decision to initiate intensive care often occurs in the ED and emergency physicians recognize the responsibility to provide patient-centered care, the communication tasks are often relegated to not-fully-trained resident physicians.^7,8 In addition, no standardized methods exist to guide physicians in leading critical shared decision-making regarding initiating intensive care in the ED.^9–11 In this issue of Annals of Emergency Medicine, Sanders et al¹² present a qualitative study describing the attitudes of Canadian emergency medicine residents about making recommendations for life-sustaining treatments during acute health decompensations. Their results are illuminating for our field.

Making a recommendation for treatment is a fundamental skill in clinical practice. Leveraging clinicians’ medical expertise, patients entrust physicians to explain the risks, benefits, and alternatives of treatment and make a recommendation. Physicians usually have no hesitation about the recommendations for routine conditions (eg, recommending wrist reduction in previously healthy patients with a broken wrist). For older adults with limited life expectancy, however, even seasoned emergency physicians struggle to make patient-centered recommendations because of prognostic uncertainty. In this study, the authors explored how emergency medicine residents think about making recommendations for life-sustaining treatments for seriously ill older adults. The authors asked them to review 2 cases consecutively (the first case with routine medical decision-making and the second with end-of-life decision-making) and explored their thoughts on making recommendations. However, the resulting study design was not devoid of weaknesses. Using the first case, the respondents recognized that clinicians routinely make recommendations for patients who are not at the end of life. Hence, contrasting case one (ie, the routine case) with case 2 (ie, the end-of-life case) likely introduced a priming informational bias.¹³

Strikingly, the end-of-life case did not contain a description of the patient’s values and goals. Rather, it highlighted the poor prognosis (eg, advanced age) and clinician-perceived futility (eg, worsening despite full resuscitation) and then asked the respondents to do something that no palliative care clinician would likely ever do, namely to make a recommendation about life-sustaining treatment without knowing patients’ values and goals. Not surprisingly, the respondents’ answers did not reveal how to incorporate patients’ values and goals into a recommendation, signaling a lack of familiarity with what constitutes patient values in the first place (eg, trade-offs, worries, etc).¹⁴

Unlike prior literature, this study observed that emergency medicine residents “felt a sense of duty” to make recommendations.¹⁵ The authors report that many residents experienced moral distress during these conversations because of a perceived duty to provide intensive care despite the potential futility of intensive treatment. The residents’ moral distress is understandable, given that they are trying to make recommendations for end-of-life decisions based solely on anticipated poor prognosis and physician-perceived futility of care. This finding is a clear illustration of residents’ paucity of clinical experience and lack of formal training on how to incorporate patients’ values and goals into end-of-life decision-making. A strength of the study is that it has exposed the nature of a crucial gap in emergency medicine training.

Patient-centered recommendations require the following 3 key steps: 1) evaluate the functional prognosis based on the patient’s baseline quality of life and treatment options; 2) establish the patient’s priorities based on the prognosis; and 3) formulate the recommendation most suitable to patient’s priorities, likely functional prognosis, and available treatment options. Even if 2 patients have the same baseline function, quality of life, prognosis, and treatment effectiveness, the patient’s values and goals alter what the patient-centered recommendations would be (Figure). Although these methods are well established, the current evidence suggests low adoption of these practices by experienced emergency physicians.^14–17 In a recent cross-sectional study of self-reported practice patterns among emergency physicians (N=161), only one-third reported making recommendations about life-sustaining treatment for patients who were seriously ill in their clinical practice. Most emergency physicians focus their conversations on the risks of procedures (eg, intubation) rather than on asking about patients’ values.¹⁷ Emergency medicine residents’ approach to such conversations has not been well studied, and the findings stand in contrast to what we know about practice patterns of more experienced emergency physicians, most of whom do not report making such recommendations.¹⁷ In other words, a critical gap in clinical practice exists between emergency medicine residents’ desire to provide patient-centered recommendations and what current attending emergency physicians report doing.

Figure.

Simplified examples of patient-centered recommendations incorporating patient’s values and goals.

This study demonstrates that when prompted, emergency medicine residents recognize the opportunities for skill improvement and want to make patient-centered recommendations. It builds on previous work examining the state of goals of care conversations in the ED.¹⁸ Goals of care conversations are procedural skills that can be taught, learned, and deliberately practiced, which have the potential to result in improved patient-centered outcomes and reduced nonbeneficial treatments.^19–22 Yet, recent studies suggest a clear gap in clinical practice: emergency physicians, especially resident physicians, want to provide patient-centered recommendations about life-sustaining treatment, but most of them are not familiar with how to incorporate patients’ values and goals into such recommendations. Furthermore, the study findings raise questions about our field’s current understanding of how to formulate patient-centered recommendations for life-sustaining treatment in patients near the end of life.

So why is this happening in our field? Several potential reasons exist. First, for experienced emergency physicians, focusing on patients’ values and goals to make patient-centered recommendations regarding life-sustaining treatments is easier said than done. Time-pressured environment, prognostic uncertainty, and years of moral distress from humbling end-of-life cases may prevent many experienced emergency physicians from making patient-centered recommendations. As a result, they often make recommendations based on the prognosis for survival rather than the anticipated quality of life after intensive care. However, this approach fails to account for what is important to most older adults. More than 70% of older adults report that at the end of life, they prefer to focus on the quality of life rather than life extension and would even give up one year of life to avoid dying in the ICU.^23,24 Second, emergency physicians are often unfamiliar with long-term functional prognosis after intensive care for patients who are seriously ill. For example, among older adults who are intubated in the ED, one-third will die in the hospital, one-third will return home, and one-third will be placed in rehabilitation and nursing facilities—their one-year mortality may be as high as 70%.^25,26 Moreover, emergency physicians may not be familiar with physical and cognitive declines for older adults after ED visits alone or with intensive care use.^27,28 Without such familiarity, emergency physicians cannot make a calculated estimation of long-term outcomes for patients who are seriously ill faced with resuscitation decisions. Third, emergency physicians are often not trained to make patient-centered recommendations based on the patient’s values and goals, baseline function, quality of life, and prognosis. Perhaps younger generations of emergency physicians have benefited from improved education in serious illness care and now recognizing this shortcoming in our field. To address this issue, robust studies investigating these topics are currently underway in our field.^29–31

During acute health decompensations for patients who are seriously ill, emergency physicians often bear the responsibility of steering treatment trajectories. Patients and families entrust us to make recommendations aligned with their values and preferences and are likely to achieve the best outcomes individualized for them. Although emergency physicians, including residents, aspire to provide patient-centered clinical recommendations, as a specialty, we simply have not garnered the serious illness communication skills to do it, which generated the education gap reflected in this study’s findings. The COVID-19 pandemic forced us to think harder about how to execute these decisions quickly. Innovative approaches emerged to partner with palliative care teams in the ED: embedded, in-person (ie, palliative care clinicians stationed in the ED) or telemedicine (ie, out-of-institution palliative care clinicians providing consults) palliative care clinicians actively screening for patients who may benefit from their consultations.^32–35 These innovations resulted in markedly different goals of care, and some institutions have permanently adopted these approaches (eg, Massachusetts General Hospital).^33–35 Yet, as a specialty, we need to ensure that we are well positioned to carry out patient-centered goals of care conversations that address values and goals through robust training and environmental support. Even for trained palliative care specialists, these high-acuity conversations in the ED remain challenging—we must tailor our serious illness communication training to be most feasible to our uniquely challenging settings (eg, EMTalk, an evidence-based serious illness communication training tailored for emergency physicians).³⁶ Near the end of life, serious illness communication skills cannot guarantee patient-centered care concordant with patients’ wishes every time. It can, however, make this process as humane as it can be for both emergency physicians and patients and families.