Grief Connects Us

by Joseph D. Stern

  Many patients “described a mismatch between their informational needs and the type of information physicians provided,” which often resulted in their feeling dissatisfied. The authors conclude that “in an era where patients are supposed to be participating in ‘shared decision making,’ we found that many patients with AML do not even have a functional understanding of their disease or of the risks and benefits of available therapies,” making shared decision-making “a particularly difficult challenge in the AML setting.” According to the Duke study:

  Patients often described “abduction by the illness” and a subsequent surrendering of control to the medical system…. Older patients with AML often did not know their prognosis and many grossly overestimated their likelihood of survival at 6 months…. Many patients reported an expected cure rate of over 90%, wherein their oncologists estimated it at below 10%…. Many patients described blunt communication that lacked emotional support, resulting in feelings of disappointment, fear, or even abandonment. Oncologists often miss opportunities to empathize with their patients, and our data suggest that this finding is true in the AML setting as well.

  In this context, Victoria’s reaction to her diagnosis and illness seem entirely consistent with this reported framework of how patients react to a new diagnosis of AML. In When Blood Breaks Down, Mikkael Sekeres supports this finding, citing “a separate study in 260 patients undergoing a bone marrow transplant found that those with higher-risk cancers—meaning that their expected survival following the transplant was lower—tended to be much more optimistic about their chances than their doctors.” (p. 87) Prolonged hospitalizations and in-hospital deaths in patients with AML were the norm, rather than the exception. In one study of 330 people diagnosed with AML from 2005 to 2011, “those who died spent over 28 percent of their lives from the moment of diagnosis in the hospital and almost 14 percent in the clinic. Within 30 days of death, 85 percent were hospitalized, 45 percent had received some type of chemotherapy, and 61 percent died in the hospital.” (p. 257)

  The difference between the ways in which many doctors choose to end their own lives and how they prescribe treatment to others is telling. This speaks to difficulties with trust and the management of expectations of patients, as well as to better understanding of risks and benefits of treatment and comfort with medical decision-making among physicians. These inconsistencies are emblematic of the problems we have with discussing and understanding death in our society.

  Several articles have highlighted these differences. Ken Murray’s “How Doctors Die,” looks at the choices some physicians made when they became patients and faced terminal illnesses. They opted for less care, rather than more, and usually chose to stay home, rather than go to the hospital. They accepted the consequences of their diagnosis and preferred to maximize quality-of-life considerations over quantity of life. Another article, “Association of Occupation as a Physician with Likelihood of Dying in a Hospital,” by Saul Blecker et al., published in JAMA in 2016, analyzed the death certificates of physicians and similarly educated professionals and concluded, “Physicians were slightly less likely to die in a hospital than the general population, but equally as likely to die in a hospital as others in health care or with similar educational attainment. In addition, physicians were the least likely group to die at any facility.” The authors state that this area is poorly studied and that an analysis of death certificates and place of death does not provide understanding of nuance or the choices patients make in their end-of-life care.

  Ironically, by declining care, many terminally ill patients extended their remaining time longer than their treating physicians had anticipated they would live. This doesn’t surprise me, because we tend to overestimate the benefits of the interventions we provide toward the end stages of a disease process. We downplay the risks and suffering so that avoidance of some care options, particularly when the likelihood of benefit is low, may end up helping the patient rather than causing harm, at least in the short-term. Certainly, the quality of a patient’s remaining time is better with fewer drugs and fewer invasive treatments, and patients are often energized by doing things that give them pleasure and meaning. They surround themselves with family and friends, minimizing time in the hospital.

  Dr. Sekeres provides data to support the observations that patients and doctors “exaggerate the benefits of interventions and minimize the harms. In one systematic review of 35 studies that enrolled over 27,000 patients, most participants overestimated benefit for 65 percent of outcomes, and underestimated harm for 67 percent … Doctors don’t fare much better. In another systematic review of 48 studies that included more than 13,000 clinicians … more than half of the participants overestimated benefits, and underestimated harm, for almost one-third of outcomes.” (p. 82)

  Patients and their families often ask me, “What would you do if this were you or someone you love?” I am well aware that thinking about a situation in the abstract is different from living with a poor prognosis as a daily reality. The implication of a cancer diagnosis in my patients always gives me pause, yet when my sister was diagnosed with leukemia, my reaction was intense and raw. The distance I normally rely upon in order to make decisions and put plans into action had vanished, replaced by fear and emotional pain.

  Should I become gravely ill, provided there were good treatment options, I would almost certainly pursue them. I would be willing to undergo surgery and accept physical pain if there were a high likelihood of success. Had I been in Victoria’s position, I would also have willingly undergone a bone marrow transplant.

  Because I am a physician, my relationship to my caregivers would be inherently different than my sister’s. I lack her degree of faith and trust. Decisions would be made based on the knowledge and facts at our disposal, and I would want to be part of the process. I don’t believe I would agree to treatment for lack of anything better; for example, I would likely have drawn a line after early recurrence of the leukemia posttransplant, since the likelihood of cure at that point was essentially nil.

  Our culture has systematically removed death from everyday life. We avoid talking about it. As a result of this avoidance, 80 percent of Americans do not put their affairs in order before they die. The vast majority of Americans say they wish to die at home, but 75 percent die in a hospital or in a nursing home. This has begun to change: in 2017, 29.8 percent of deaths by natural causes occurred in hospitals and 30.7 percent occurred at home (Kolata, 2019). People who live alone, isolated from family and friends, have higher rates of admission and in-hospital death, even when those patients express the same desire to die at home as do patients with greater social support. Our relationships with others are a crucial part of the end of life. We need to strengthen these connections.

  In their last month alone, half of Medicare patients go to an emergency department, one-third are admitted to an ICU, and one-fifth will have surgery—even though 80 percent of patients say they hope to avoid hospitalization and intensive care at the end of life. This is largely because patients are ill-prepared to confront their mortality and medical professionals are not willing to tackle the business of dying. Medicare spending for patients in the last year of life is six times what it is for other patients and accounts for a quarter of the total Medicare budget. This proportion has remained essentially unchanged for the past three decades, despite increasing awareness of the need for earlier discussion of issues surrounding end-of-life care.

  Often, the first time people talk about death is in the doctor’s office, when they may discover they are critically ill. If we engage in conversations about death long before we must face it, perhaps this will alleviate some of the fear, encourage us to live more fully, and allow us to prepare for our inevitable mortality. People want to go down swinging. I respect that desire, but not at the cost of denying reality. If Victoria had been able to acknowledge her predicament after her transplant failure, perhaps she might have done things differently those last few weeks of her life. So often,
an unwillingness to confront our mortality is a kind of magical thinking from which no good comes. We miss opportunities to say and do what we really want, and instead, we may spend our remaining days stranded in a disorienting fog of treatments and their side effects.

  It is important to know a patient’s goals, fears, and expectations. Atul Gawande addresses this constellation in Being Mortal. A patient may be married, but the marriage may be an unhappy one. They may be fearful of the financial consequences of prolonged illness or treatment. They may be scared of death or accepting of dying. Some patients say they are ready to die, and older patients in particular, those who have already lost loved ones and friends, may themselves have lost interest in living and welcome death, rather than wishing desperately to prolong their lives.

  The opposite is true of someone like my sister, who was in the prime of her life and desperately wanted to live. The degree of acceptance of a terminal diagnosis is often colored by the quality of life that the patient has lived to date and the level of personal satisfaction they feel. It matters greatly whether the patient is in a critical, time-sensitive episode of illness, such as a brain hemorrhage, in which decisions need to be made immediately or in a slower-moving scenario, such as prostate cancer.

  Of great importance is whether the patient is conscious and competent. I spend a lot of time with families who must suddenly care for loved ones after a severe injury resulting in impaired levels of consciousness. In these situations, families need to speak on behalf of the patient. They often struggle to reach consensus regarding care, particularly when it comes to withdrawal of treatment when the likelihood of meaningful survival is negligible. Rifts can build between family members, and long-seething resentments surface in strange and often unexpected ways. Estranged family members may be the most uncomfortable with discontinuing care; the loudest protests can come from those who do not actually show up at the hospital.

  This is one reason that living wills are crucial, yet only approximately one in three adults in the United States has one. Honest discussions about your end-of-life wishes, sharing these wishes with loved ones now in case you become unable to express them later, make these excruciating circumstances easier to navigate. Of course, it is not possible to project every eventuality in a written document, but by talking about it beforehand, family members have a far better idea what their loved one would want and can act on those wishes. I spend considerable time explaining to families that the emphasis should not be on what they would desire for themselves but on the understanding that they are acting as the voice of the impaired patient. Their duty is to honor the patient’s wishes as best they can.

  One of the ironies that frequently plays out in the ICU is that patients and family members who have lived unhappy or unfulfilling lives will cling to life and demand that “everything be done,” not realizing that by making these demands they are ensuring that the patient experiences a bitter end, remaining on a ventilator for a prolonged period of time or moving to a nursing home for the remainder of his or her days.

  Sometimes, families of patients who have felt disenfranchised in their lives will demand heroic, aggressive intervention in the ICU and remain suspicious of any efforts made by the medical team to back off on treatments, no matter how futile those treatments may be. By contrast, those who have led fulfilling lives and have announced their wishes for the end of life in advance of a critical illness seem to anguish less over the medical decisions. More often, they choose graceful exits rather than prolonged and painful ones. In both instances, open communication with families is imperative. There is clearly a racial disparity between patient and family willingness to involve palliative care or hospice services and transition from more aggressive to less aggressive care. This includes frequency of in-hospital death, with African American patients more likely to die in the hospital than white patients. African American patients are more likely to mistrust the medical system and to avoid enrolling in hospice care than white patients. They are less likely than white patients to know about these care options, such as home hospice. Efforts are being made to ameliorate these disparities.

  Ideally, both the patient and the physician should be united in a desire for authenticity and understanding. Familiarity with the real possibility of dying allows for trust and open communication, so that both parties can strive for honest connection. This is why it is important that we physicians understand a patient’s idea of appropriate and compassionate care and why it is similarly essential that the physician be honest about the disease process, its likely outcomes, and the implications of treatment.

  As physicians, we need to communicate technical information to patients—biology, physiology, and treatment issues—including side effects and complications, whether they be related to surgical procedures, chemotherapies, drugs, or radiation therapy. We have to review tests, interpret data, counsel patients, and chart these interactions in excruciating detail. Adding on emotional support, such as grief counseling and details of palliative care, seems altogether too much for one individual to manage. We need to work seamlessly as a team, and health systems should be organized with an eye to this approach, with the goal of decreasing the burden on physicians and simultaneously improving patient care. Bringing palliative care services into play earlier in patients’ illnesses, rather than waiting until the end of life, will not only improve the patient’s experience but also off-load physicians. This needs to be strongly encouraged both in hospitals and medical practices. Not only will patients receive better care, but physicians will also feel less burdened, more supported by their health systems, and as Stacy Wentworth and Sean Fischer suggest, freed to address the many necessary aspects of patient care they are currently expected to manage.

  Palliative care nurses have been an enormous benefit to brain tumor patients in our multidisciplinary cancer treatment program. Their involvement should be the rule, rather than the exception. In the Neuro-ICU at my hospital, involvement of palliative care is inconsistent and depends on who is attending a given patient, as well as frequently complex family situations. These observations are supported by Susan Shapiro’s Speaking for the Dying: Life-and-Death Decisions in Intensive Care, in which she systematically observes on a daily basis how decisions are made in two ICUs in a diverse urban hospital. These decisions, and the quality of communication with patients and families, are highly variable and frequently unpredictable. Physicians who welcome the involvement of palliative care are often more sensitive to the issues that are affecting patients and their families, while others are less receptive to these concerns and may be unwilling to involve palliative care services in the first place. We have tried to eliminate this variability in our hospital by embedding palliative care services in the ICU, just as we have done so in the brain tumor clinic. While this effort has only recently been initiated, we believe this will make for greater patient and family satisfaction and is also likely to curtail unnecessary procedures and suffering. These interventions have been shown in other institutions to save money and to increase satisfaction.

  Psychologists are also extremely useful members of the treatment team. Knowing that the emotional concerns of patients are being handled well allows me to concentrate on their medical care. This is an enormous benefit to the patient as well as to the staff. But many medical systems do not provide these services, as they do not see a direct financial benefit but only a cost, measured in full-time equivalents (FTEs), to these added staff. The cost-effectiveness of these services is changing as organizations transition from fee-for-service models to those based on population health, or to value-based care, in which health systems are responsible for the overall costs of care (earlier palliative care drives these down), as opposed to billing for interventions that are not necessarily efficacious but almost always expensive and profitable.

  Physicians receive little support in the form of counseling, grief management, or ongoing professional training. Stacy Wentworth suggests that we return for training after completi
ng residency and formal medical education. This could be accomplished through workshops and continuing education programs for physicians designed to reinforce skills and supplement our earlier formal medical training. These programs already exist, but they need to be made more widely available. Included should be training in interviewing techniques, becoming comfortable speaking with patients and their families about end-of-life care, and sharing the insights of palliative medicine with other specialties. In my practice, a significant portion of my time is spent conducting interviews, yet I have had no formal training in this and receive no feedback on my skills in this area. Most doctors learn as they go, with no training after medical school and residency have been completed, and because of this we are likely to repeat mistakes or miss opportunities to improve.

  It is only after we have been in practice for several years that we begin to formulate new questions about, and insights into, patient care. There are few avenues for coaching or professional development available to physicians in practice. Perhaps some physicians do not see the need for ongoing training or grief counseling, but I think these services are beneficial, especially when we think of a doctor’s exposure to tragedy and loss as cumulative. Partly, this is a cultural issue because just as patients and people in general don’t discuss death or dying, physicians don’t readily choose to discuss their failures and sadness. Medical systems don’t invite this type of disclosure, and often doctors do so at their own risk. Administrators and peers may perceive such behavior as a sign of weakness or maybe “frailty,” rather than a healthy way to grow as a practitioner and as a human being. Substance abuse, withdrawal, and failed marriages are well-known symptoms of burnout; these need to be prevented where possible and treated readily and without recrimination when they do occur.