Communication about end-of-life care is increasingly recognized as a core clinical skill, but doctors are often unprepared to have these conversations. Crisis situations at the end of life, such as when a hospitalized patient with a poor prognosis requests that “everything” be done, pose even greater communicative challenges. Such decisions are often regarded as a demand for care that may be burdensome or even harmful, rather than the start of an important conversation about values and goals. Situations such as this have prompted the development of prescriptive approaches to communicating with patients, which have limitations since they focus more on the physicians’ message than on developing a relationship with the patient or eliciting the meaning of the illness and its treatment. Current models of communication are less about accurate transmission of information and more about the ways in which communication constitutes our social world. Thus, the goal of communication training for physicians should be less about tools and scripts, and more about doctors learning to engage patients in conversation. We suggest strategies for promoting dialogue to more effectively explore the concerns and misunderstandings that might underlie decisions to “do everything.”
Decisions about medical treatment at the end of life are complicated, emotional, and prone to misunderstandings. Our lived experience with health and medicine is characterized by a nuanced range of choices and alternatives, yet when faced with the need to make treatment decisions at the end of life, patients and families sometimes declare that “everything” should be done rather than explore alternatives. Anything less than “everything” may be perceived as frightening, socially unacceptable, or as giving up. “The word ‘everything’ conveys powerful symbolism to lay persons: fidelity, non-abandonment, respect for autonomy…. In the world of ‘do everything’ the opposite is doing nothing, which implies abandonment, neglect, surrender, and perhaps causing death” [1, pp. 276–277]. However, sometimes “doing everything” comes with a high burden and an uncertain benefit.
Physicians have largely retired the phrase, “there is nothing more I can do for you,” in favor of recommending hospice care for patients near death or palliative care for patients who want comfort care in addition to or instead of aggressive treatment. Over 1.5 million patients received hospice care in 2012, and over 65 % of US hospitals with over 50 beds have a palliative care team [2, 3]. Despite increased access to palliative care, physicians, family members, and even patients themselves are sometimes reluctant to accept comfort measures only as a plan of care. Patients may be unaware of their true prognosis, may feel their religious beliefs require they try every available treatment, or may be responding to family pressures . Palliative or hospice care is more likely to be acceptable if the patient tried aggressive treatment, if all agree improvement is very unlikely, or where the choice is consistent with the patient’s long-held and well-documented values and treatment preferences. In contrast, a patient’s or family member’s statement to “do everything” can be heard as a mandate, rather than the start of an important conversation.
Decisions to “Do Everything” at the end of life
Consider the following scenario, the subject of a recently published essay . A newly minted physician is caring for an 80-year-old patient admitted to the hospital with lung cancer and pneumonia. As the patient’s condition worsened, the doctor decided it would be prudent to ask the patient what care she would want if she got sicker. The doctor said, “Some patients ask us to do everything. That means electrical shocks, a breathing tube, the ICU. We could do that. Other patients tell us that they don’t want any of this and would prefer just to be comfortable.” The patient finally answered, “Well, I don’t know. If my heart stopped tonight? Do everything.” The young doctor left the patient’s room with her advance directive in hand—Full Code—contemplating how it felt “when ribs break under your hands during chest compressions.” The narrative ends with the doctor anticipating the likely end of the patient’s life:
The pneumonia would get worse. She would breathe faster and faster and then, tiring, slow gasps until the breaths stopped altogether. The monitor would alarm. We would run in, anesthesia, breathing tube, shocks and intravenous lines in her neck or groin. She would die in the ICU. But that was what she’d chosen, I told myself. After all, I had asked. Right? .
A patient’s or family members’ decision to “do everything” often brings subsequent conversations about treatment alternatives and goals of care to a halt, since it may be interpreted as a demand that any and all medical treatments be administered, regardless of the probability of effectiveness or outcomes. A predictable and unfortunate cascade of events can be set into motion: poor or diminished quality of life for the patient and/or a prolonged dying process, nearly unbearable stress for family members confronting the aftermath of their decisions, and moral distress for the healthcare team forced to provide futile or even harmful care.
Physicians whose patients ask for “everything” might reasonably conclude that the patient is asking for every imaginable treatment, no matter how harsh, invasive, or unlikely to be of benefit. One obvious choice of action is to write a “full code—no limitations” order in direct response to the request [4, pp. 345].
It is not clear, though, if the patient knew what “everything” might entail, or if she understood the quality-of-life implications if treatment succeeded in prolonging life.
Communication as a clinical skill
Communication at the end of life is being increasingly recognized as a core clinical skill. Communication skills training for medical students and practicing physicians tends to take the form of tools, checklists, or loose scripts for delivering “bad news” to patients, which are sometimes practiced in simulated situations. A recent study of the effectiveness of simulation-based communication training with 391 internal medicine residents and 81 nurse practitioner trainees revealed no differences between participants randomized to participate in the training and those not enrolled in patient-reported quality of communication . The effectiveness of the scripts and tools themselves has not been systematically studied, many of which rely on some form of the SPIKES model [7–12].
The SPIKES model
The SPIKES model uses acronyms for each of the recommended steps for delivering bad news to patients. “S” stands for setting, and suggests that physicians rehearse, choose an appropriate time and place, and involve significant others. “P” stands for perception, and encourages physicians to check the patient’s perception of his or her health situation and obtain an invitation (“I”) to deliver the information. “K” stands for knowledge, and physicians are advised to tell patients about their medical condition while avoiding bluntness. Once the news is given to patients, the physician is encouraged to attend to the patient’s emotional reactions (“E” for emotions), and to summarize (“S”) and strategize about next steps.
In the hands of a skilled and compassionate physician, tools such as the SPIKES model can yield impressive results, even in cases where the patient and family start with a demand that “everything” be done. Consider the case of a 57-year-old man with incurable cancer admitted to the ICU after a catastrophic event . The patient was comatose with no hope of survival, but his family members insisted that everything be done for him. The SPIKES model helped to structure a productive family conference, and what initially looked like a demand for futile care was revealed instead as a desire to control the timing of death, a death the family understood was inevitable. All agreed to remove the patient from life support the following day, thus ensuring that he did not die on his wife’s birthday. The physician’s excellent communication skills were the key to defusing the situation, and the patient died peacefully with his family present.
Prescriptive approaches are preferable to “avoiding the elephant in the room,” but they have significant limitations . The SPIKES model focuses physicians’ attention on the message to be delivered (i.e., the “bad news”) rather than on ways to develop a relationship with the patient. This model is based on an outdated transactional model of communication and fails to account for other salient aspects of the communicative situation such as patients’ expectations, framing, or the impact of default options. Rather than planning a scripted communication encounter as described by SPIKES and similar models, patients’ needs for knowledge might be better served if their physicians learned and practiced ways to promote dialogue. Thin communication practices, i.e., those that transmit information but do not promote dialogue, can lead to a kind of collusion between doctors and patients . Patients want to believe their physicians will tell them what they need to know, and physicians expect patients to ask questions when they need more information; what needs to be discussed for informed decision making may be overlooked. This collusion between doctors and patients can make the short-term more bearable, but may be regretted later when the reality of the patient’s impending death can no longer be avoided. Only later might they lament, “if only I had known to ask….” . For example, in the scenario discussed earlier, the possibility of palliative or hospice care was never mentioned by the physician. A colleague whose father recently died of end-stage renal failure after years of aggressive treatment said the physician never mentioned hospice care. “Was I supposed to know this?” she asked, as if she failed because she and her father were unaware of the option to forgo aggressive treatment.
Patient and family expectations about communication
A patient’s or family member’s inexperience or lack of knowledge of a particular illness affects their understanding of the disease process, treatment options, and likely outcomes. In other words, patients do not know what they do not know, and their capacity to comprehend medical information and ask pertinent questions may be compromised by a range of factors, including emotions, education level, and deference to the physician as an authority figure . Right or wrong, patients and their family members make decisions under the assumption that doctors will tell them what they need to know, and it is essential that doctors have the skills to help promote rich communication and encourage dialogue to assess understanding of diagnoses, prognoses, and treatments. Direct, yet open-ended questions are more effective in promoting dialogue than are closed-ended questions such as, “Do you have any questions?” or “Do you understand?” Indications of comprehension, such as nodding or saying yes, or a lack of questions, do not necessarily indicate understanding.
Even when patients ask questions, the SPIKES model offers little guidance about how physicians might answer effectively. Consider a patient who asks the dreaded question, “how long do I have to live?” Physicians are reluctant to “being pinned down to any sort of number” . When physicians do offer prognostic information to patients, they are often overly optimistic; recent statistics revealed that doctors are accurate only about 20 % of the time . It is also possible that the knowledge (“K” in the SPIKES model) patients seek is not only scientific. Patients may also want their humanity to be acknowledged through a relationship with their doctor, as “the angst of facing mortality has no remedy in probability” . Or as Anatole Broyard famously said, “To most physicians, my illness is a routine incident in their rounds, while for me it’s the crisis of my life. I would feel better if I had a doctor who, at least, perceived this incongruity” .
When full recovery is not possible, patients and their loved ones need someone, ideally their physician, to map the landscape of care. The potential for patients to stay on the path of aggressive treatment is great, especially absent information about the range of treatment options. If doctors only give patients information they need to make the next decision, patients are deprived of the knowledge that would allow them to determine the route they might have wanted to take from the point of diagnosis until death. The “GPS” approach, which only signals the next turn, does not easily allow for alternative routes; a “Google Earth” view might prove helpful at some point so patients can decide how to negotiate their illness trajectory in a way that supports their values and preferences . Not all patients want to do everything, yet they may do so if they are unaware that each incremental decision places them further along a particular approach to treating their illness. Once treatment starts, it is often difficult for patients and physicians to move away from aggressive treatment until all options have been exhausted, even those with extremely small odds of success.
Contextual features of the communicative situation
End-of-life treatment decisions are also influenced by contextual features of the communicative situation. For example, framing, defined as the selective use of words or phrases that influence listeners, or in other words, what information is emphasized or foregrounded, is important to consider . For example, patients viewed surgery for lung cancer as a substantially better treatment choice when the situation was framed in terms of the probability of living rather than the probability of dying . Patients also tended to agree more often with the default treatment option (expressed as what is normally done). For example, patients completing advance directives were significantly more likely to defer to the default option, whether it was continued treatment or comfort measures . Respondents to a hypothetical vignette about care for a premature infant were also significantly more likely to choose the default option, whether it was resuscitation or comfort care . Thus, the information provided, the clinical situation, and the pressure family members feel to make the right choice all influence the ways in which patients and family members interpret medical information and make decisions about treatment, including the decision to “do everything.”
Promoting dialogue between patients and physicians
The implications of framing, the impact of default options, and the potential for misinterpretation and misunderstanding make communication about end-of-life treatment extremely complicated; a demand that everything be done for a patient unlikely to recover or significantly improve adds further challenges. We need new ways to effectively promote dialogue between physicians, patients and family members at the end of life; prescriptive approaches to communication have limitations as does some simulation training for communication effectiveness . A commitment to more effective communication can lead to better understanding the concerns underlying decisions to “do everything.” The SPIKES model and its more recent elaborations are helpful ; however, “it is also possible that, as we devote more time to teaching students and doctors effective communication techniques, we risk muting their authentic human voices, and instead of learning to connect they apply rote tools and scripts” .
Prescriptive protocols for delivering “bad news” to patients are based in large part on a transmission or at best a transactional model of communication.
In the simplistic transmission model of communication that is so often taken for granted in everyday discourse, communication is conceptualized as a process in which meanings, packaged in symbolic messages like bananas in crates, are transported from sender to receiver. Too often the bananas are bruised or spoiled in transport so we have the ubiquitous problem of miscommunication: The message sent is not the message received, the sender’s meaning does not come across. In order to improve communication, according to this model, we need better packaging and speedier transportation of messages. Good communication is basically a technical problem [24, 1].
It is easy to see why a technical approach to a complicated communication situation would appeal to physicians who are used to relying on technical solutions to solve problems. Prescriptive approaches to patient–physician communication assume that the important parts of communication—the information to be conveyed, specific individuals to be included, the technical means of communication, and the desired outcome of the interaction—must all be fixed in place before communication occurs. The steps in the SPIKES model, for example, embody these assumptions, namely that a specific patient and physician will meet so that pertinent information (the bad news) can be delivered from one to the other using a shared code and identified channel of transmission, and that the outcome will include a plan of action chosen from predetermined treatment alternatives. When understood this way, the nature and function of communication preserve the power differential between physicians and patients, which influences patients’ understanding of information they receive. If more treatment is suggested by their doctor, a patient may assume that this is the only or best choice. Equally troubling is the focus on the short term that a transactional model of communication reinforces; the emphasis is on the accurate and efficient transmission of information rather than on the context in which the information is being shared.
Alternatively, we support a model of patient–physician interaction at the end of life that encourages physicians to create a communicative space where the meaning of the patient’s condition can be explored. Patients need a safe communicative space in which a full range of treatment options can be considered, including hospice and palliative care (which can be provided in addition to, or instead of, aggressive treatments). Only then can physicians and patients co-construct a plan of care that takes into account the current state of medical science, as well as the patient’s values, expectations, preferences, and family situation. Contemporary conceptions of communication acknowledge that the act of communicating is constitutive: talk, information exchange, and dialogue construct or constitute a healthcare encounter, and shape the experience and meaning of the illness .
The constitutive model posits instead that the elements of communication, rather than being fixed in advance, are reflexively constituted within the act of communication itself. Communication is defined as an ongoing process that symbolically forms and re-forms our personal identities, our social relations, our common world of meaningful objects and events, our ideas and feelings, and our routine ways of expressing these socially constructed realities .
Such a model is more than a technical equation that supports how to get one’s message across without distortion, and can account for the moral and political aspects of exchanges between physicians and patients.
A constitutive model of communication sees communication as incorporating time, creating a context, being influenced by a context, history, culture, establishing a relationship between humans, that it is in essence human and symbolic since it acknowledges and respects the essence of the other. This may mean that communication can be seen as the process whereby one interprets in order to make sense of the world, in other words, to understand the world or our reality [26, emphasis original].
The constitutive nature of communication is a salient idea for patients and physicians discussing end-of-life treatment preferences because how health information is conveyed, who participates in the interaction, and the meaning of the information to the patient may be as important as the content of the message. Communication about treatment at the end of life can be a dynamic and participative information exchange between patients, family members, and physicians, rather than a simplistic transmission of information . Participating in such a process exerts mutual influence on attitudes, behaviors, and relationships and encourages all parties to attend to verbal and nonverbal, explicit and implicit, and obvious and subtle behaviors.
Rather than prescribing a series of steps for physicians to use when delivering bad news, a constitutive model of communication prompts physicians to participate in a dialogue or conversation with patients where information is openly shared, including the long-term perspective. Physicians in this scenario are still empowered by the information and specialized expertise they possess, but patients are also empowered to bring their preferences and feelings as important factors in a treatment plan. We propose a series of considerations that might allow this more robust model of communication to facilitate an exploration of what “everything” may mean and how treatment decisions may proceed.
Considerations for improving communication
The first step is to regard the decision to “do everything” as the beginning of a conversation about prognosis and goals of treatment, not as a mandate . Patients and family members should be asked specifically what “everything” means to them. A decision to do everything for a loved one might be “an expression of powerful feelings and desperate hope” rather than “a claim on every tool of critical care” [1, pp. 276]. Understanding the patient’s and family members’ point of view is essential in trying to resolve the issue of how to continue with medical care. Physicians should attempt to explain what “everything” entails from a medical perspective. The new physician described earlier  told her patient about “breathing tubes and ICU,” but stopped short of discussing the likelihood of a poor outcome, and never broached the topic of death, a possibility the patient never considered when stating she wanted everything.
Second, keep in mind that doctors and patients live in different communicative worlds, and are unlikely to interpret information in the same way. Patients live in the lifeworld and physicians live in the world of medicine [28, 29]. A study of doctor–patient interaction with end-stage head and neck cancer patients provides an example . Despite eligibility criteria requiring patients to have been told by their physicians that their condition was terminal, one participant was surprised by interview questions about whether he and his physician had discussed his impending death. He said, “My doctor told me my disease can’t be cured, but that doesn’t mean I’m dying!” It is possible he was in denial or overly optimistic, but equally plausible he did not understand the implications of his doctor’s euphemistic communication or framing of his situation.
Third, asking patients if they have questions and assuming that having no further questions indicates that comprehension is—as pointed out earlier—problematic. Instead, physicians have to answer the important questions that patients and families are afraid to ask, or do not know to ask. Exploring what “everything” means should involve telling the patient and family what quality of life they might expect if doing everything succeeds in prolonging life. Wolff described the decision he and his siblings made about whether or not their 84-year-old mother with progressive dementia should have heart surgery for aortic stenosis, which was causing potentially lethal symptoms after 15 years of successful management . “The truth is,” he writes, “you’re so relieved that someone else has a plan, and that the professionals with the plan seem matter-of-fact and unconcerned, that you disregard even the obvious fallacies of logic: that the choice is between life as it was before the operation and death, instead of between life after the operation and death.” The operation was successful in that his mother’s heart was repaired, but he and his siblings were not prepared for the terrified creature their mother became post-surgery, unable to speak, extremely agitated, and requiring numerous return trips to the hospital’s emergency department.
Fourth, clear and comprehensive information is essential. When there is little or no hope for recovery it is not realistic to assume that patients and families will figure out that they or their loved one is not getting better and eventually opt to discontinue treatment. It is often difficult for families to understand their loved one’s condition in the face of fragmented care; “None of the platoon of physicians seeing a critically ill patient is eager to step forward and assume command of a complicated, time-consuming situation” [1, pp. 277]. Even when more coordinated care is delivered by interdisciplinary teams, the job of giving information about a poor prognosis often becomes a shared responsibility in which accountability belongs to everyone and to no one . It is appropriate to say the word “death” when it is inevitable, and to tell patients and families that certain medical events are not survivable or may limit quality of life .
Fifth, whenever possible, patients themselves should be involved in discussions about prognosis, treatment, and quality of life. Most cancer patients are found to have decisional capacity upon admission to their terminal hospitalization, but a recent study found that less than half participated in their own end-of-life conversation . The same study found that patients who participated in documented discussion about end-of-life care were less likely to receive life-sustaining treatments, less likely to be admitted or transferred to ICUs, and more likely to avoid prolonged hospitalization. Family members are important, but take care that their concerns do not overwhelm those of the patient if at all possible.
Communication is essential to good decisions at the end of life. Scripted methods of interacting with patients are not only sometimes inadequate, but also limited in their ability to establish trust, promote dialogue, and explore treatment options. Situations in which patients and/or family members decide that “everything” must be done without understanding what such a decision entails create undue burden on all concerned. While the conclusions of this essay are admittedly not innovative or especially new, our intention is to create a space for conversation between patients, family members, potential patients, and healthcare providers about what doing everything at the end of life means—what procedures and routine monitoring are involved, what is the cost in terms of comfort and quality of life, what is the patient’s prognosis for increased life expectancy of an acceptable quality. The anecdotal evidence presented here points to a difference in perceptions between patients and physicians; future studies should tackle the measurement of magnitude and specific dimensions of this perceptual split.
The end of a patient’s life presents complicated communicative challenges for all concerned. We live in a death-denying culture and are encouraged by medicine’s constant technological advances to believe that death cannot only be denied, but avoided entirely. “For a society that is arguably death obsessed, with reports of violent and unexpected death paraded across our media every day, it is still very difficult to talk about the one great certainty: the fact that all of us one day will die. Death is regarded as an illness that with further medical knowledge will soon be avoidable… [32, pp. 118]. It is not surprising that some patients want “everything” to be done, and believe treatment will restore their health and vitality, or that physicians believe in the value of treatments they are trained to provide. We all must face the reality of our mortality, and must learn to communicate about death and dying in ways that promote our humanity and our humility. Our reliance on medicine to manage the end of life casts physicians in a powerful and difficult role to help give meaning and purpose to terminal illness. We all live in the social worlds our communication with one another creates, and learning to talk about difficult issues in ways that promote dialogue is an essential step in allowing us all to make the best decisions we can about the end of life.”
Greene HR (2013) Do everything. Narrat Inq Bioeth 3:275–280
Center to Advance Palliative Care (CAPC) (2012) Growth of palliative care in US hospitals 2012 snapshot. http://www.capc.org. Accessed 1 March 2014
NHPCO Facts and Figures: Hospice Care in America (2013) National Hospice and Palliative Care Organization, Alexandria. http://www.nhpco.org/hospice-statistics-research-press-room/facts-hospice-and-palliative-care. Accessed 1 March 2014
Quill TE, Arnold R, Back AL (2009) Discussing treatment preferences with patients who want “everything”. Ann Intern Med 151:345–349
Lama DJ (2013) If your heart stopped tonight. The Atlantic. 25 October 2013
Curtis JR, Back AL, Ford DW, Downey L, Shannon SE et al (2013) Effect of communication skills training for residents and nurse practitioners on quality of communication with patients with serious illness. JAMA 310:2271–2281
Emanuel LL, von Gunten CF, Ferris FD (1999) Education for physicians on end-of-life care (EPEC) curriculum. Institute for Ethics at the American Medical Association and The Robert Wood Johnson Foundation, Princeton
Back A, Arnold RM, Baile WF, Tulsky JA, Fryer-Edwards K (2005) Approaching difficult communication tasks in oncology. CA: Cancer J Clin 55:164–177
Eggly SS, Penner L, Albrecht TL, Cline RJ, Foster T, Naughton M et al (2006) Discussing bad news in the outpatient oncology clinic: rethinking current communication guidelines. J Clin Oncol 24:716–719
Eggly SS, Albrecht TL, Kelly K, Prigerson HG, Sheldon LK, Studts J (2009) The role of the clinician in cancer communication. J Health Commun 14:66–75
Baile WF, Buckman R, Lenzi R, Glober G, Beale EA, Kudelka AP (2000) SPIKES—a six step protocol for delivering bad news: application to the patient with cancer. Oncologist 5:302–311
Buckman R (1992) How to break bad news: a guide for health care professionals. The Johns Hopkins University Press, Baltimore
Quill TE (2000) Perspectives on care at the close of life: initiating end-of-life discussions with seriously ill patients—addressing the “elephant in the room”. JAMA 284:2502–2507
The A (2002) Palliative care and communication: Experiences in the clinic. Open University Press, Buckingham
Roscoe LA, Tullis JA, Reich RR, McCaffrey JC (2013) Beyond good intentions and patient perceptions: competing definitions of effective communication in head and neck cancer care at the end of life. Health Commun 28:183–192
Kalanthi P (2014) How long have I got left? The New York Times. 26 January 2014
Christakis NA (2001) Death foretold: prophecy and prognosis in medical care. University of Chicago Press, Chicago
Broyard, A. (1990) Doctor talk to me. The New York Times. 26 August 1990
Tannen D (2005) Conversational style: analyzing talk among friends. Oxford University Press, New York
McNeil B, Pauke S, Sox H, Tversky A (1982) On the elicitation of preferences for alternative therapies. New Engl J Med 306:1259–1262
Quill CM, Halpern S (2012) Deciphering the appropriateness of defaults: the need for domain-specific evidence. J Med Ethics 38:713–718
Kressel LM, Chapman GB, Leventhal E (2007) The influence of default options on the expression of end of life treatment preferences in advance directives. J Gen Int Med 22:1007–1010
Gilligan TD, Sekeres MA (2014) Can doctors be taught how to talk to patients? The New York Times. http://well.blogs.nytimes.com/2014/02/27/can-doctors-be-taught-how-to-talk-to-patients? Accessed 27 Feb 2014
Craig RT (2001) Communication. In: Sloane TO (ed) Encyclopedia of rhetoric. Oxford University Press, New York, pp 125–137
Craig RT (1999) Communication theory as a field. Commun Theory 9:119–161
Burger M (2004) To think about research is to think about humans. Afr Journal Stud 25:373–379
Albrecht TL, Penner LA, Cline RJW, Eggly SS, Ruckdeschel JC (2009) Studying the process of clinical communication: issues of context, concepts, and research directions. J Health Commun 14:47–56
Mishler EG (1984) The discourse of medicine: dialectics of medical interviews, vol 3. Greenwood Publishing Group, Santa Barbara
Barton E (2007) Situating end-of-life decision making in a hybrid ethical frame. Commun Med 4:131–140
Wolff, M. (2013) A life worth ending. The New York Times Sunday Magazine. 20 May 2013
Zaros MC, Curtis JR, Silveira MJ, Elmore JG (2013) Opportunity lost: end-of-life discussions in cancer patients who die in the hospital. J Hosp Med 8:334–340
Seymour J (2007) Commentary 1. ‘Quality of life to the end’. Commun Med 4:117–118
We wish to thank Dr. Eric M. Eisenberg, Dean of the College of Arts & Sciences, University of South Florida, for his helpful comments on this manuscript.
Conflict of interest
About this article
Cite this article
Roscoe, L.A., Tullis, J.A. The meaning of everything: communication at the end of life. J Med Pers 13, 75–81 (2015). https://doi.org/10.1007/s12682-015-0205-x
- End of life
- Treatment preferences
- Doctor–patient communication
- Clinical communication training