Beyond the rhetoric of bioethics, patients and families need help facing tragic diagnoses and prognoses and coping with traumatic treatment options. In this article, I will draw from developmental psychology, psychodynamic psychotherapy, and bioethics to show what we are still missing about shared decision-making and how we can do better. Before we bring in new models of shared decision-making we need to ensure that doctors know how to create the foundational conditions for safe emotional communication. For pediatricians, this requires knowing enough about how adolescents process cognitive, affective and sensory information to avoid traumatizing their patients, knowing enough about the doctor’s own fears not to project them onto the patient, and providing the supportive others that help the teenager tolerate and process information that is otherwise intolerable. To fail to do so can lead to tragic outcomes.
Typically, the call for shared decision-making (SDM) in medicine is to prevent the 2 extremes of physicians abandoning patients to face difficult decisions alone and of doctors paternalistically taking over such decisions. Although there is agreement on avoiding these extremes, the ongoing debate revolves around issues of responsibility, power and authorization: who really ought to be “making” the decision.1 The situation is more complex in pediatrics, especially in the care of adolescents, because there are often at least 3 parties involved, but the bioethics concerns are still framed in terms of a balancing act between too little and too much professional interference.
However, beyond the rhetoric of bioethics, patients and families need something else in the way of SDM. They need help hearing, processing, and facing difficult, even traumatic, diagnoses and prognoses and coping with poor treatment options. This is not news; experienced pediatricians, pediatric nurses, and social workers know patients and families need help hearing bad news and facing tragic choices. What has been less examined is what is necessary for physicians to provide the kind of safety, respect, and support necessary to enable patients and families to think through what is otherwise emotionally intolerable. In this article, I will examine what physicians need to do to create the communicative field and safe holding environment necessary to meet patients’ needs. Notably, it will not address specific communication skills; there is already much written about skillful empathic communication.2 Rather, the focus here is on the underlying conditions necessary for such communication, conditions that I will argue doctors are responsible for creating before carrying out SDM lest this devolve into a sham process.
First, physicians must communicate effectively in light of the distinct cognitive, emotional, and social world of the patient. This requires being well informed about the developmental capacities of adolescents. For example, how an adolescent hears information about life, death, pain, and suffering differs categorically rather than merely quantitatively from how either an adult or a younger child processes such information. (The “mature minor” label might be useful legally, but the person still does not process like a mini-adult.)
Second, pediatric oncologists, as doctors who deal with tragic choices, must become skillful in recognizing how much their own unconscious emotional projections frame the decisions they offer their patients in the first place.
Third, bioethics needs to better delineate the obligation to respect children and young adolescents as persons even if they are not yet fully capable of autonomy. And this obligation needs to be distinguished from and weighed appropriately against the responsibility to protect the best interests of children and young adolescents that arises from beneficence.
In this article, I will review a case from the late 1990s3 in which the medical team was hijacked by unconscious emotional projections and the bioethics reviewers were stymied by seeing the need to respect the patient too narrowly (and incorrectly) as the obligation to respect his autonomy. On the one hand, reviewing this case from almost 2 decades ago provides some critical perspective. We can glimpse how 20 years ago the norm of respecting autonomy was both overvalued and constricted to a negative obligation to avoid interfering with a patient’s stated choices and how this was used to justify abandoning patients in difficult situations. Perhaps this has changed since then, perhaps not so much. We can also judge how much more we have learned about the cognitive developmental needs of adolescents since then.
Yet, on the other hand, this case is quite contemporary insofar as physicians continue to be unreflective of the role of their own unconscious emotions in how they frame treatment options for seriously ill adolescents and their families. This inadvertent emotional “sharing” needs to be reflected on now as much as 20 years ago to build a solid foundation on which to add newer explicit models of SDM. Although some might argue that such attention to the emotional underpinnings of difficult decisions is not a component of the process of SDM, I would argue that it is impossible to approach SDM without it.
The Problem Case
What is the clinical problem that the call for SDM is trying to solve? Because I am not a pediatrician, I sought to address this question by reading the classic pediatric cases that were taken to represent unsatisfactory decision-making. In the 1990s and early 2000s, there were cases of young teenagers (ages 13, 14, and 15)4 whose refusals of life-saving therapy were accepted (or nearly so) under the banner of respect for autonomy. This led to a growing concern that doctors were abandoning patients under the guise of respect for autonomy. But then reading more recent cases like that of Cassie C5 suggests just the opposite, that pediatricians now were going to great lengths to override patients and families who were refusing medically necessary care. As I reviewed the bioethics literature, I experienced a certain whiplash between calls for doctors to interfere more and to interfere less. This led me to question whether the current call for SDM itself obfuscates rather than elucidates what we should learn from carefully reading how these communicative impasses arose in the first place.
Whether doctors were too removed or too paternalistic, it seemed to me that all of these cases had in common the fact that there was actually too much sharing, but it was all unconscious. Doctors, parents, and adolescents shared unmetabolized fear and anger, which manifested as collusion in catastrophic and wishful beliefs that hijacked all of their decision-making.6 They were enmeshed in psychodramas in the clinical setting without recognizing it. My goal, if I had been called in as a psychiatric consultant, would have been to help them differentiate their individual views rather than share them more with each other, which would have then hopefully created a space for empathic curiosity, listening, and, eventually, better explicit discussions about the treatment choices. And, as I hope to show in the following case, this enmeshment occurs even when it seems that doctors are taking an explicitly hands-off, respect-for-autonomy approach.
The following tragic case of 13-year-old Jorge appeared in Pediatrics in 2001 with commentary by 2 pediatricians and an ethicist about what the doctors should do when an adolescent refuses treatment for a life-threatening illness.3
Jorge, a 13-year-old Hispanic male, was admitted and treated for acute lymphocytic leukemia 9 months ago. He promptly experienced remission in the first month of therapy. This was followed by 6 months of intensive consolidation that required numerous hospitalizations. He then was placed on maintenance chemotherapy.
Jorge is the oldest son of five children born to working-class parents who were both high-school graduates. While undergoing induction, Jorge and his parents seemed to be aware of his treatment course and prognosis. The family related well to the treatment team, and Jorge developed a close relationship with the oncology social worker. At first, he appeared depressed. At times he would state that he was “cursed” and would never live to graduate college, a long-term goal he shared with his family. However, despite his age, Jorge appeared to be quite mature and was always cooperative with medical procedures.
During his second month in maintenance therapy, Jorge’s hair was growing back well, and he had returned to school for partial day attendance. His mood was pleasant, and he seemed to be adjusting to resuming life. During a routine clinic visit, a bone marrow aspiration detected the resurgence of his acute lymphocytic leukemia, and he and his family were told about this by the oncologist who worked closely with them. Jorge appeared devastated, as did his parents and the two siblings who were with them that day. The oncologist explained that Jorge’s only real chance at survival would require a bone marrow transplant, and because they did not have a matched related donor, he would have to be placed on a list for an unrelated donor. Jorge immediately asked what his chances of living were, and he was told that approximately 40% of children would survive and be disease-free for 5 years. Jorge and his family were devastated. After a prolonged consultation during which various options were discussed, his family agreed to visit the transplantation center to get additional information.
At the center, they were informed about the lengthy time he would need to be in seclusion, the amount of pain and invasiveness involved in the procedure, the long-term issues about graft-versus-host disease, and the likelihood of success using a partially matched nonrelated donor. His parents report that Jorge asked very few questions and did not get along well with the supervising oncologist at the transplant referral center.
On returning home, Jorge announced to his parents that he would refuse the transplantation and did not want to be placed on the list. His parents called the oncologist for help. A meeting was arranged the next day, and Jorge, his parents, doctor, social worker, and primary nurse met to discuss the options. His doctor noted that Jorge appeared to be very firm in his decision that he would refuse to cooperate. Jorge had read about other teenagers who had also chosen to die, rather than accept additional painful treatments. His parents were absolutely firm that every possible treatment should be pursued. They felt that a 40% chance was better than no chance at all. They asked his doctor to place Jorge on the waiting list, despite Jorge’s strenuous objections.3
This is where the case ends and the commentaries by 2 pediatric oncologists and 1 bioethicist begin. All 3 focus solely on what to do next. Reflecting on what to do after a breakdown in communication rather than asking how it happened in the first place is typical in inpatient bioethics; given the pace of events and the urgent need to decide the next step, there is rarely attention to how things got to where they are. The commentators make a few suggestions of how to try to change Jorge’s mind and then quickly move on to assert the need to respect Jorge’s autonomy by accepting his refusal of the transplant.
Free of the medical team’s time urgency, we can unpack the events that led up to the impasse. Consider the way that each physician communicated with Jorge. First, the oncologist immediately answered Jorge’s question about his prognosis with a statistic: 40% survival at 5 years. Other details convey that this oncologist had a good relationship with Jorge, so it seems odd that he would begin with the most difficult fact, especially knowing that he was dealing with a young teenager. Why not first elicit Jorge’s understanding and then frame the information in a way that Jorge would be able to make use of?7
This physician was doing what the American bioethics movement was calling for at the time, respecting patient autonomy by fully disclosing all requested information. Doctors were being coached not to avoid talking about death with dying children. Not telling them the truth, by this view, could rob them of the opportunity to have meaningful conversations with their parents and the medical team.
Second, soon after Jorge heard this news, he and his family mustered the hope to go and meet with the transplant team. The transplant doctor emphasized all the negative facts: Jorge would face prolonged seclusion and pain, the risk of graft-versus-host disease, as well as lower odds of success with an unrelated donor.
Although both doctors were dutifully fulfilling their understanding of what respecting Jorge required, such vivid and frightening information clearly terrified the 13-year-old boy. In other words, this approach had benefits, but it also had harms.
Three Sources of the Problem
To reflect on what could have been done better, I will focus on 3 clinical and research areas: developmental psychology, psychodynamic psychotherapy, and the philosophical foundations of bioethics.
First, developmentally, how can we expect a 13-year-old boy, even one labeled as somewhat “mature,” to understand and process information delivered to him this way?8 How does a young adolescent think about invasive procedures, pain, suffering, isolation, and death itself or even what he might be giving up by not living past 13? We know that most 13-year-old boys have immature executive function, with insufficient ability to regulate emotions like fear, and difficulties envisioning prospective states.9 Many such children seem, in their decision-making, to be able to essentially only picture 2 temporal points: now and never. If one cannot envision the future in terms of shades of gray, and if one’s thinking is readily hijacked by fear, what does it mean to hear that one’s chance of survival 5 years from now is 40%, and only if one goes through a noxious treatment right now? Jorge likely heard that treatment would certainly cause suffering and that he was more likely to die than live anyway.
All of this was then likely made much worse by the messages that Jorge received from the transplant doctor. Teenagers have a magnified sensory imagination, so although the detailed description of all the pain and suffering associated with the bone marrow transplant (BMT) might preoccupy any adult, the odds are high that such details would hijack the attention and imagination of a teenager. So, it is likely that Jorge was picturing terrible pain whenever he tried to deliberate about the value of the BMT. What I think he imagined was an equation like, “die now or go through torture, then wait to die?”
Sometimes, even for an adult, emotions overwhelm reason. When a person’s emotional view of his or her situation is so vivid and compelling that it becomes impossible to genuinely weigh the risks and benefits, the person is temporarily incapable of the basic deliberation central to decisional capacity.6 Although we would need to have learned more details to assess Jorge’s thinking, his age and the terrifying details emphasized by the transplant doctor make it likely that he was temporarily hijacked from being capable of such deliberation.
Given those developmental considerations, let me turn to a psychodynamic inquiry into the communicative field between the physicians and Jorge and his family. Why would a pediatric oncologist and transplant doctor, both of whom likely knew a great deal about young teenagers and how they process bad news, deliver a message that would be received this way by Jorge?
My guess is that, on some psychodynamic level, the transplant doctor and team were protecting themselves from feeling guilty if Jorge died of the BMT. They may have felt guilty about inflicting suffering on the children who went through terrible treatment and then died anyway. In some unconscious sense, their bluntness might have been an unconscious attempt to decrease feeling guilty insofar as they had prepared Jorge and his family for a possible tragic outcome. This dynamic may be at work in all informed consent discussions. They may be medicine’s way of dealing with its own guilt.
Third, let me turn to the bioethical assumptions made in this case and the philosophical misunderstandings that underlie them. The doctors believed that their full disclosure of the facts was called for as part of respect for autonomy. Much is made about Jorge being “mature,” and all 3 commentators said that his choice to refuse the BMT needed to be respected to respect his autonomy.
To me, these judgments were problematic at 2 levels. First, even if respect for autonomy was the right goal, Jorge’s fear is likely to have hijacked his ability to exercise whatever capacity he had for autonomy (regardless of how mature he was). So, if respect for autonomy was the goal, Jorge’s decision could not simply be accepted at face value; rather there was an obligation to address (and not provoke) his fear and anxiety to make deliberation possible.
Second, I believe that respect for autonomy by respecting Jorge’s independent decision was not the right goal in the first place. Why? This particular decision had to do with how valuable a 40% chance of having a future was to him. In general, 13-year-olds have great difficulty weighing and balancing future-oriented considerations, which are critical to the decision-making task he was facing.10
On the other hand, we know that children with chronic diseases often know much more about what it is like living with those diseases and going through noxious treatments than do their parents or physicians. We rightfully do respect their judgments about what they can or cannot tolerate. Jorge’s experiential knowledge from having already gone through chemotherapy was crucially important to his decision. He needed to be an active participant in deciding if he felt that he could tolerate the BMT, even if he could not envision the future life that the treatment might help him gain. Thus, Jorge was an ideal candidate for an SDM process, one in which emotional scaffolding from others enabled him to become less afraid and in which he was able to trust and rely on his parents’ and physicians’ views of his future. What his doctors owed Jorge was not respect for his autonomy but respecting his actual 13-year-old mind by providing the scaffolding for Jorge, his parents, and his team to engage in SDM.
The Ethical Problem
The ethical mistake was to reduce respect for Jorge as a person to respecting his autonomy and miss the other crucial ways in which he was being disrespected. Respect for persons is a much broader and deeper obligation than respect for autonomy. We respect the dignity of people who lack autonomy, even those in persistent vegetative states. And this respect for dignity is distinct from the obligation of beneficence; thus, we respect deceased bodies although we cannot provide the dead person any benefit. Even after death we are obligated not only to respect people’s bodies but also their reputations. I have argued, for example, that psychotherapeutic records should stay confidential after death not only to protect relatives but out of respect for the narrative person.11
Broadening our focus from respect for autonomy to respect for persons including respect for dignity helps us in this case. We fail to respect someone’s dignity when we undermine them psychologically. We can readily see how treating people manipulatively or cruelly can undermine their sense of secure personhood and ongoing-ness in the world (think of how torture or brainwashing undermine narrative personhood). Sadly, for a 13-year-old boy facing a terminal diagnosis and horrible treatment options, such a rupture of personhood can readily occur, and the first obligation of doctors, nurses, and social workers is to prevent or help repair such ruptures. This includes addressing the emotional context of the information being discussed. To discuss traumatic facts with a patient without empathic attention to what these facts mean to this particular patient can in and of itself trigger depersonalization. For example, holocaust survivor Primo Levi recounts how if he were asked to tell about what it was like in a concentration camp and the listener was polite but emotionally neutral, he felt himself to be disappearing.12
So where does this leave us with Jorge? It leaves us locating the bioethical drama earlier in the case to when his physicians and especially the transplant doctor communicated in a way that failed to respect Jorge as a person. This poor communication was a serious ethical problem, which then created a cascade of other problems, and it should be the rightful focus of the commentaries on this case but is instead ignored because bioethics sees respectful communication (as opposed to respecting autonomy) as discretionary rather than obligatory.
Three crucial things were not done in this case. First, the information was not framed in a way that allowed the patient to process it. Second, the doctors were unaware of their own unconscious motivations. Third, the doctors did not recognize their own limitations and thus failed to bring in people who could have helped the youth and his parents work through and tolerate their otherwise unbearable choice.
On this last point, there was a social worker who helped Jorge earlier in his cancer treatment. She was brought in again later in the case after things fell apart, but she was notably absent at the critical moments when the oncologist and then the transplant doctor each delivered the toxic information to Jorge. But just as a burn victim might need to hold a trusted nurse’s hand during the most difficult moment of treatment, Jorge likely needed accompaniment during those most traumatic moments, especially because Jorge’s parents were just as traumatized by the news.
Once the damage was done, what should the team have done next?
Jorge was owed the scaffolding necessary to restore his sense of ongoing-ness and security as a person. He was owed this regardless of whether he then accepted the BMT. That is, he was owed it out of respect for him as a person, not beneficence.
What are some practical interventions that can provide such scaffolding? Jorge and his parents probably needed more time alone with a trusted therapeutic other like the social worker to ask their questions and address their fears without worrying about the physician’s presence or feeling rushed.13
Note that after Jorge refused treatment, his oncologist did engage in a lengthy meeting with Jorge that included the social worker and others. Presumably the team tried to provide emotional support and a chance for him to express his concerns, but Jorge did not engage. Perhaps it would have helped if his doctor had directly addressed how frightening it must have been to hear such scary messages from the transplant team and himself, but this does not appear to have occurred.
But what if none of that helped? One specific intervention that has proven useful is introducing the patient to another person who has been through the feared scenario (in this case, a teenager or young adult who went through a BMT and could describe how he or she got through it and help Jorge see what quality of life afterward was like) (J.H., unpublished observations).
Finally, what if Jorge had persisted in refusing transplant after all this? We have already disagreed with the 3 commentators who say that we ought to accept his decision out of respect for his autonomy because we do not believe that he could exercise autonomy by deliberating (weighing and balancing both sides) in the case of this particular decision. When we cannot respect the autonomy of a minor we shift some weight to our obligation to serve their best interests. But at the same time, we have argued that we still owe Jorge respect as a person. How does this actually translate into action? In the case of research subjects, the Belmont Report translates respecting people who lack autonomy into protecting them from possible harm, which then often (but not always) leads to excluding them from the research. There is no such option in clinical care. In addition, this practical collapse of respect into nonmaleficence fails to distinguish the 2 separate obligations of respect for persons and beneficence sufficiently.
Perhaps the clearest distinction between protecting a person out of respect versus beneficence and/or nonmaleficence arises in cases that we might think of as performing battery for the person’s longer-term vital interests, such as when we restrain an actively suicidal but apparently competent patient. (Here I am treating the obligations of beneficence and nonmaleficence as part of 1 continuum of strictly ends-oriented obligations that can be distinguished from deontological obligations to respect patients as persons. But, of course, there are also deontological obligations to promote well-being.) Such restraint itself is not an expression of respect for the patient as a person; it is rather an expression of the state obligation of parens patriae, which is rooted purely in beneficence. However, there are many safeguards and restrictions on such restraints that arise out of respect for persons, not beneficence, including using the least restrictive means possible for as short a time as possible, holding a hearing within 72 hours, etc. Being clear on the difference between the obligations from beneficence and those from respect for persons keeps us attentive to the moral obligations that persist when we must infringe on a person’s rights in order to promote their best interests.
In Jorge’s case, given that he is only 13, there is an even stronger argument for parens patriae, or beneficence outweighing, but not eliminating, the obligation to respect his wishes. That argument is that he is not yet responsible for his own life in the way an adult is, but rather his parents and his doctors, and (in the case of neglectful parents), the state, are strongly responsible for his well-being. Children require support not only for the practical aspects of life but also for their emotional, social, and educational needs. Thus, Ross,14 Salter,8 and others have shown that the issue of decisional responsibility for adolescents is important and distinct from the issue of decisional capacity.10 In their work, the emphasis is on parental authority for such decisions, given that parents are responsible for shepherding adolescents into adulthood. Although I agree with their arguments, my focus is somewhat broader, not on parents per se but on how the fact that children require physical, emotional, and social resources to have a decent life trajectory creates a societal responsibility to protect their best interests. Usually, parents appropriately take on much of this responsibility, as Jorge’s parents did, but under parens patriae physicians too have direct obligations to protect the child’s best interests, just as schools have direct obligations to protect children from bullying regardless of parental preferences.15
In this case, his parents and medical team might have, as a last resort, decided to override Jorge’s decision (out of beneficence, not respect). In that case, they should convey to Jorge that they are overriding him but will defer to him for the smaller decisions about the circumstances of his undergoing the BMT, when to return home and to school, etc. It is possible, even likely, that Jorge would have ultimately accepted this and felt adequately safe and respected and loved. Thirteen-year-olds often feel safer when their more dangerous choices are overridden. Thus, the balancing of respect and beneficence would lean toward beneficence, but some respect for him as a person would remain.
Alternatively, in the worst-case scenario, telling him that he will have the transplant could trigger some lasting negative reaction on his part. Unfortunately, a BMT differs from a 1-time surgery or even 72-hour psychiatric hold in that it can involve months of suffering. If Jorge were to experience this suffering as forced on him, he could experience an ongoing state of terror akin to what people experience when they are prisoners subject to torture, and it would be difficult to justify imposing this on him even for his long-term best interests. Possibly we could shift his fear level by promising (truthfully) that if he could not tolerate the situation after the procedure and there was no other way of alleviating his suffering, he could be given comfort measures only and enabled to pass away quickly rather than stay in a prolonged state of involuntary suffering. This is of course an ethically fraught situation, but it does seem unjustifiable to force him to go through months of fear and suffering against his will.
Insights from Jorge’s case may not be relevant to all situations in which adolescents face difficult decisions. Prognosis, the burdens of therapies, and the urgency of the decision could all affect how respectful communication can best be structured to provide a supportive and safe environment for both the adolescent and the family. Even more importantly, however, creating such a space is necessary even when there does not appear to be disagreement with a recommended plan. Jorge’s refusal of the transplant made his fears apparent to the medical team; other patients who do not object may be suffering in silence.
The terrible ethical dilemmas just laid out should be strenuously avoided by ensuring better communication in the first place. For pediatricians with seriously ill patients, communicating in ways that create security and trust is not optional. Before we bring in new models of SDM, we need to ensure that doctors know how to create the foundational conditions for good communication. For pediatricians, this requires knowing enough about how adolescents process cognitive, affective, and sensory information to avoid traumatizing their patients; knowing enough about the doctor’s own fears not to project them onto the patient; and providing the supportive others that help the teenager tolerate and process information that is otherwise intolerable. To fail to do so can lead to tragic outcomes.
Dr Halpern conceptualized, drafted, and revised the manuscript and approves the final manuscript as submitted and agrees to be accountable for all aspects of the work.
FUNDING: No external funding.
Thank you to Lainie Ross MD, PhD, for her help in refining the arguments central to this paper.
POTENTIAL CONFLICT OF INTEREST: The author has indicated she has no potential conflicts of interest to disclose.
FINANCIAL DISCLOSURE: The author has indicated she has no financial relationships relevant to this article to disclose.