Early in this past decade, a collaborative of critical care professional organizations convened a task force to create recommendations for intensivists to adhere to in order to maximize quality of care in a fiscally responsible manner (1). Of the top five recommendations in the Choosing Wisely Campaign’s task force report, the following was highest rated as a core tenet of ICU practice: “Do not continue life support for patients at high risk for death or severely impaired functional recovery without offering patients and their families the alternative of care focused entirely on comfort” (1). These goals-of-care situations are inherently value-laden and thus involve degrees of shared decision-making (2). Although the style in which a surrogate engages in shared decision-making with an ICU team may vary (3), it is important that surrogates are at a minimum clearly informed by the clinical team about their options, including comfort care, when patient prognosis is unfavorable (2).
There is certainly prognostic variability among physicians regarding when a particular patient might be viewed as “high risk” for death or poor outcome and thus appropriate for comfort care options (4). These provider-to-provider differences regarding prognostic estimates notwithstanding, there has been literature to suggest that intensivists could simply do a better job overall of making surrogate decision makers aware that comfort care is even an option. One study of 72 audio-recorded family meetings (regarding patients with poor prognosis) reported that families were only informed of comfort care as a treatment option roughly half of the time (5). The factors responsible for clinicians not informing patients’ families about comfort care options in appropriate situations are multifactorial (6), but one factor may be that, for the most part, clinicians are not necessarily prompted at any point in a typical daily ICU workflow to reflect on or even record each of their patients’ likely long-term prognosis. To this point, Turnbull et al (4) previously demonstrated in a randomized trial that intensivists who were provided hypothetical ICU case scenarios and asked to document their estimated 3-month functional outcomes for the provided cases were overall more likely to express an intention to discuss comfort care options with families for those scenarios than were intensivists who participated in the same exercise but who were not required to document their prognostic estimates explicitly.
In this issue of Critical Care Medicine, Turnbull et al (7) report the results of a new trial to examine whether this association between providers’ documenting prognostic estimates and “intending” to discuss comfort care with families would extend to providers’ “actually” discussing comfort care with families. Their so-called Simulated Communication with ICU Proxies (SCIP) study enrolled 116 U.S. physicians with recent ICU attending experience who were all asked, at a medical simulation center, to review the paper-based medical record of a hypothetical 81-year-old African American man who had developed acute respiratory distress syndrome and septic shock by ICU day 3. Afterward, all participants answered four questions about their management plan for the patient, but 63 of the 116 participants were randomized to answer an additional three questions about the patient’s in-hospital and 3-month prognosis (intervention group), whereas 53 were not (control group). After completing their assigned questions, participants in both groups then participated in a standardized, video- and audio-recorded meeting with a trained actor portraying the hypothetical patient’s daughter. The audio transcripts of the meetings were analyzed to obtain the proportion of intensivists in each group who presented to the patient’s daughter (1) the option of comfort care (i.e., the study’s primary outcome) and (2) survival and functional prognosis.
The authors expected that 50% of intensivists in the control group would present comfort care to the family member. However, the blinded reviewers of the audio transcripts found that only 13% of intensivists in both the intervention and control groups offered comfort care to the family member in the simulation. This finding contrasts with the study’s observations that (1) nearly 80% of the participants in the intervention group reported that they felt the patient would not survive to hospital discharge (2), no one in the intervention group indicated a belief that the patient would be able to live independently without any cognitive decline, and (3) nearly half of all participants reported after chart review that they felt a palliative care consult was appropriate. The intensivists in the intervention group were more likely to inform the patient’s daughter that the patient was likely to die (68% vs 43%; p = 0.01), but less than 5% of participants in either group spoke about possible functional impairments if the patient were to survive.
In the context of a study involving simulation, the study authors went through significant lengths to optimize their study design. In addition to executing a randomized trial, the authors obtained institutional review board approval to recruit participants with only a vague description of the study objectives. Blinded review of de-identified audio transcripts by multiple reviewers reduced possible bias. Participants rated the simulation with the family member as highly realistic. The authors even performed a sensitivity analysis removing those simulations in which an actor “deviated” from the script and found no difference in study outcomes.
With these study design strengths, the fact that the study found such a low percentage of participants offering comfort care raises the question of something systematic within the construct of the trial that gave participants pause in guiding their discussions in this direction. The hypothetical scenario presented to participants was that of a patient on ICU day 3. Although the patient was indeed rated by those in the intervention group as already having a very high likelihood of dying in the hospital, this timepoint is earlier than the typical timepoint at which many physicians may feel prognosis is more certain and interventions such as tracheostomy and gastrostomy placement are discussed with families, with palliative extubation presented as an alternative (8). Even regardless of the ICU day on which the hypothetical scenario is presented, it can be argued that in real-world scenarios practitioners may take the opportunity to hold several family meetings and build a trusting relationships before introducing comfort care as a treatment option, as opposed to during the very first family meeting (9). Although the authors were able to successfully simulate a family meeting, their study design could not allow for modeling this type of relationship building, even if they had tried to stipulate that participants had “met” the patient’s daughter before. The authors note that, although the actor was instructed not to be overly combative, she was instructed to show emotion upon hearing a poor prognosis for the patient. In real-world scenarios, one could imagine a clinician allowing a family member some time to allow news to “sink in” on ICU day 3 before offering the possibility of comfort care at a subsequent meeting.
Of note, the authors did ask participants about their self-perceptions of whether they disclosed prognosis and offered comfort care to the family member, in addition to the study team’s own-blinded assessments of audio transcripts. Across the board, a higher proportion of participants self-reported disclosure of risk of death, risk for future functional impairment, and the option of comfort care during the exercise than the blinded reviewers ascertained. The authors note that the reviewers of the audio transcripts were strict in their metrics for judging whether these disclosures were made clear by participants. In the context of prior research on nonverbal cues during end-of-life discussions (10), one might wonder whether a blinded analysis of nonverbal behaviors between the intervention and control group using the video recordings would have yielded additional insight into the potential impact of priming physicians with documentation of prognosis before meeting with ICU family members.
Nevertheless, this study’s finding that physicians offered comfort care at a very low rate during a one-time simulated ICU family meeting—irrespective of being told to document prognostic estimates beforehand—emphasizes the need for more real-world research in how best to encourage physicians to make families aware of comfort care as an option in appropriate situations. The study by Turnbull et al (7) was able to show that participants in the intervention group were at least more likely to disclose prognostic information regarding patient survival during the simulation exercise. One could hypothesize for a future experiment that, in addition to being asked to document patient prognosis, a clinician might be asked directly to consider and/or document his or her opinion regarding comfort care itself as a reasonable option to offer during a family meeting. Perhaps being asked to answer such a question might increase the likelihood that a clinician would explicitly offer comfort care during a family meeting. If this were shown to be true, it would provide additional evidence for the impact of incorporating such a question into a formal daily ICU checklist (11), or at the very least into a clinician’s routine mental workflow when caring for patients with poor prognosis and their families.
1. Halpern SD, Becker D, Curtis JR, et al.; Choosing Wisely Taskforce; American Thoracic Society; American Association of Critical-Care Nurses; Society of Critical Care
Medicine: An official American Thoracic Society/American Association of Critical-Care Nurses/American College of Chest Physicians/Society of Critical Care
Medicine policy statement: The Choosing Wisely® Top 5 list in Critical Care
Medicine. Am J Respir Crit Care Med 2014; 190:818–826
2. Kon AA, Davidson JE, Morrison W, et al.; American College of Critical Care
Medicine; American Thoracic Society: Shared decision making in ICUs: An American College of Critical Care
Medicine and American Thoracic Society Policy Statement. Crit Care Med 2016; 44:188–201
3. Johnson SK, Bautista CA, Hong SY, et al. An empirical study of surrogates’ preferred level of control over value-laden life support decisions in intensive care
units. Am J Respir Crit Care Med 2011; 183:915–921
4. Turnbull AE, Krall JR, Ruhl AP, et al. A scenario-based, randomized trial
of patient values and functional prognosis
on intensivist intent to discuss withdrawing life support. Crit Care Med 2014; 42:1455–1462
5. Schenker Y, Tiver GA, Hong SY, et al. Association between physicians’ beliefs and the option of comfort care for critically ill patients. Intensive Care
Med 2012; 38:1607–1615
6. Visser M, Deliens L, Houttekier D. Physician-related barriers to communication
and patient- and family
-centred decision-making towards the end of life in intensive care
: A systematic review. Crit Care 2014; 18:604
7. Turnbull AE, Hayes MM, Brower RG, et al. Effect of Documenting Prognosis
on the Information Provided to ICU Proxies: A Randomized Trial
. Crit Care Med 2019; 47:757–764
8. Nelson JE, Cox CE, Hope AA, et al. Chronic critical illness. Am J Respir Crit Care Med 2010; 182:446–454
9. Meeker MA, Jezewski MA. Metasynthesis: Withdrawing life-sustaining treatments: The experience of family
decision-makers. J Clin Nurs 2009; 18:163–173
10. Elliott AM, Alexander SC, Mescher CA, et al. Differences in physicians’ verbal and nonverbal communication
with black and white patients at the end of life. J Pain Symptom Manage 2016; 51:1–8
11. Creutzfeldt CJ, Engelberg RA, Healey L, et al. Palliative care needs in the neuro-ICU. Crit Care Med 2015; 43:1677–1684