End-of-life care and contingent vs. non-contingent duties
Ronald W. Pies
Boston, Massachusetts, United States
Mr. Joseph B, a 70-year-old widower and retired college professor, is hospitalized in the final stages of metastatic pancreatic carcinoma. His doctors estimate that he has “three or four weeks” to live. The patient is well aware of his prognosis, and, as he puts it, “I have come to terms with it. I just want to be kept comfortable.” Residing in a nursing home until about a month ago, Mr. B was admitted to the hospital’s palliative care unit when the home decided it could “no longer keep Mr. B comfortable.” The patient’s previous chemotherapy has been discontinued, and the treatment team’s primary goal is to control the patient’s substantial pain. The team decides against intravenous hydration, owing to the patient’s complaints of pain associated with needle insertion and the need for frequent needle site changes. However, the patient’s daughter adamantly insists that her father is “being denied appropriate care” and angrily demands that he receive IV fluids. She tells the hospital administrator, “Not giving my dad fluids is euthanasia!” One doctor on the treatment team is ambivalent, citing research showing that IV hydration helps reduce delirium and opioid-induced toxicity. The patient himself now tells his daughter and his doctors that he wishes no more fluids of any kind, and adds, “I just want to go peacefully. Please respect that.”
The hypothetical case of Mr. B raises a number of troubling medical and ethical questions. How “aggressive” should medical care and treatment be near the end of life? How do the traditional principles of medical ethics apply to the patient in the final stages of terminal illness? Are there important differences in how these principles are applied under ordinary conditions of medical treatment versus those situations encountered near the end of life? Are some of the physician’s medical-ethical duties absolute, and are others binding only under certain circumstances? Does discontinuation of intravenous fluids, dialysis, or respiratory support in a terminally-ill patient amount to euthanasia or physician-assisted suicide (PAS)? Do patients have a right to physician-assisted suicide? Are there ever instances in which PAS is ethically justified or compatible with medical ethics?
The physician’s general ethical duties
In general, the traditional ethical obligations of the physician include the four principles of autonomy, non-malfeasance (non-maleficence), beneficence, and justice (fairness). Autonomy refers to respect for the patient’s freedom of choice; non-malfeasance, to the time-honored concept of “do no harm”; beneficence, to the promotion of the patient’s well-being; and justice, to respect for the due process of law when considering the patient’s rights and prerogatives.
It is obvious, however, that these are very general and highly “elastic” constructs. They do not shed much light on the physician’s obligations in cases like that of Mr. B. For example, how should the physician weigh Mr. B’s autonomy (his requesting that all hydration cease) against the principle of non-malfeasance (which weighs against the discontinuation of fluids)? Put in more existential terms: how much weight should be given to the issue of respecting personhood (the patient’s philosophical or spiritual concerns) and how much to that of correcting pathophysiology (maintaining appropriate fluid and electrolyte balance)? This requires a more fine-grained analysis of the traditional four principles of medical ethics.
WD Ross’s ethics: absolute vs. prima facie duties
The Scottish philosopher William David Ross (1877–1971) provided a foundation to analyze the four principles of medical ethics more closely by distinguishing between “absolute” and “contingent” ethical duties. Ross is known for his emphasis on the “appreciation of moral distinctions.”1 Notably, he distinguished between what he termed prima facie and actual duties. Prima facie duties may be understood as conditional or “all other things being equal” obligations. These are duties we have “at first blush,” which may recede in light of new, overriding considerations. In contrast, actual (absolute) duties are those we must carry out after considering all our (often conflicting) prima facie duties.
For example, let’s say a friend has loaned you a shotgun for hunting. Suppose he awakens you at 3 am, clearly intoxicated, and says he wants the gun back so he can “take care of some bad guys.” All other things being equal, you have a prima facie duty to return your friend’s gun—but is that your actual duty under these circumstances? Most of us would probably defer returning the gun! Building on Ross’s distinction, I suggest we distinguish contingent from non-contingent duties (roughly analogous to Ross’s prima facie and absolute duties). Thus, the four ethical principles may be analyzed further, according to whether they yield duties that we are to fulfill “all other things being equal” or duties that we must fulfill under any and all circumstances.
Consider the traditional medical goal of preserving bodily and organic functions (beneficence). I suggest that while this is a crucially important medical duty, it is a contingent one. For example, overriding factors such as terminal illness may render the preservation of bodily functions a subsidiary duty. In contrast, I believe that the relief of pain and suffering (beneficence) and the preservation of human dignity (autonomy) are non-contingent duties. From the ethical standpoint, these are always operative and preeminent duties. Under ordinary conditions of medical care, the physician may be most aware of the routine monitoring and maintenance of bodily functions because the typical healthy patient may not draw attention to a need for the relief of suffering or preservation of dignity. However, under end-of-life conditions the relief of pain and suffering and the preservation of human dignity may frequently become the predominant focus of care.
I believe that the duty of protecting the patient’s human dignity has been insufficiently emphasized in literature relating to end-of-life care. This duty must often supersede the traditional beneficent duty of preserving bodily functions, which is usually predominant during conditions of ordinary medical care. On the other hand, it is critical to appreciate that the contingent ethical duties of the physician never entirely disappear, even at the end of life. Conditions at that time may alter the expression, but not the essential nature of the physician’s ethical obligations. Thus, as we shall see, the duty of maintaining bodily integrity exerts a constraining effect on how far the physician may go in complying with a dying patient’s wish to terminate his own life.
The Judaic-rabbinical perspective
In several respects, the supersession of the preservation of dignity and the relief of suffering over the maintenance of bodily function dovetails with the ethics of the modern Judaic-rabbinical perspective on the physician’s duties at the end of life. The Judaic-rabbinical perspective places a very high premium on alleviating pain and preserving dignity in caring for a dying patient. These obligations are accorded such great importance in rabbinical teachings that I regard them as instantiations of Ross’ “absolute” ethical duties. Nonetheless, the Rabbis place clear limits on how far the physician may go with respect to alleviating pain and suffering. These carefully balanced principles may help provide ethical guidance for some physicians and patients faced with end-of-life decisions.
The concept of “removing impediments to death” forms the core of modern rabbinical beliefs about end-of-life care. According to Rabbi Joseph Telushkin, “Removing impediments to death involves removing treatments that prolong the dying of a terminal, suffering patient while doing nothing to actively bring about the patient’s death.”2 Trading a traditional focus on biological function for a focus on the alleviation of suffering, Rabbi Telushkin opines, “[t]he primary responsibility of a physician who is treating a patient with a terminal and painful illness is to treat the pain.”2 However, “it is permitted to give a sufficient dose to relieve the suffering even if doing so increases the likelihood of the patient’s demise.” Rabbi Telushkin stops short of promoting an active acceleration of the patient’s death, adding, “it is forbidden, for example, to purposely overdose and kill a patient in pain with morphine.”2 Telushkin advocates a dose sufficient to relieve suffering, but he draws the line at intentionally causing the patient’s death through an overdose or other means.
Similarly, Rabbi Moshe Feinstein emphasizes the relief of suffering over the maintenance of bodily function for patients at the end of life, writing:
If a patient is terminally ill and in intractable pain, so that there is no hope of his surviving in a condition free of pain, but it is possible, through medical or technological methods, to prolong his life, then it is improper to do so. Rather, the patient should be made as comfortable as possible and left without any further intervention.2
Like Telushkin, Feinstein advocates for the termination of life support that would prolong suffering, but likens intentional hastening of the patient’s death, such as overdosing a patient, to an act of “murder.”
It is crucial to note that, for virtually all rabbinical authorities, actively promoting the patient’s death is a murderous act. Many Orthodox rabbis also oppose the withdrawal of life-sustaining treatments.2 Nonetheless, while the withdrawal of life support remains a question of contention, Dr. Fred Rosner argues that, “All reputable commentators on . . . terminal care concur that analgesics and narcotics may be given to relieve pain and suffering even if these increase the danger of depressing respiration and of predisposing the patient to contracting pneumonia.”2 Identifying when a patient is actively dying remains a gray area for many, but it is important to keep in mind that, as Rosner notes, “To prolong life is a mitzvah [commandment], to prolong dying is not.”2
Further clarifying the issue is the Judaic concept of human dignity, which is based on the rabbinical principle known as k’vod ha-briot, or “respect for the dignity of all people.” One could argue that, while modern medical technology may prolong life, this same technology does not guarantee dignity to a dying patient. As Kenneth Prager MD, Director of Clinical Ethics at Columbia University, writes:
[There are] . . . an ever-increasing number of patients in [ICUs] who are being kept alive by ventilators, dialysis machines . . . potent antibiotics, and even artificial pumps . . . these patients are nearly always prevented from feeling pain by the administration of narcotics that render them unconscious. They cannot communicate with their families . . . [they] . . . develop horribly large and deep ulcers of their skin . . . [and] undergo repeated skin punctures for IVs and catheters . . . the [ICU] is clearly prolonging the misery of the dying process. These treatments . . . also violate the dignity of patients because of the disfigurement and breakdown of their bodies.2
From a rabbinical perspective, such a violation of the patient’s dignity at the end of life would contravene one of the physician’s primary and non-contingent ethical obligations—to preserve the dignity of the patient—while also prolonging the dying process. I would argue that the cessation of certain life-sustaining technology in terminal illness not only promotes the dignity of the patient by changing the focus from maintenance of bodily function to the relief of suffering, but also preserves patient dignity by allowing for the natural process of death to occur without the prolongation of suffering.
The ethical problem with so-called “physician-assisted suicide”
Preserving the patient’s dignity at the end of life inevitably draws questions about the term, “physician-assisted suicide” (PAS). Thomas Szasz, a psychiatrist and social critic of the moral foundations of psychiatry, has called attention to the semantic problems with the term “physician-assisted suicide” (PAS), which has been broadly used to characterize a wide range of interventions and resists an unambiguous definition. The term has been used to describe anything from providing a patient with information on “how to commit suicide” to the direct administration of lethal drugs by the physician. When the term is applied to a physician who actively and intentionally brings about the death of a terminal patient—albeit with the patient’s consent—Szasz argues that “physician assisted suicide” is merely a comforting euphemism for what really amounts to “medical killing” or “heterohomicide.” Consistent with Szasz’s perspective, I prefer the term “physician-actuated heterohomicide” (PAH) to describe a case in which a physician acts intentionally to end a patient’s life with the patient’s consent; for example, by administering or directly providing a lethal dose of morphine. Though morally and legally distinct from murder—in which case, the patient obviously does not consent to being killed—PAH is nonetheless a form of homicide (suicide is technically, “auto-homicide”).
Even at the end of life, the physician’s ethical obligations prohibit the intentional destruction of the patient’s biological and organic functions. Any direct action aimed at ending the patient’s life would violate the physician’s contingent ethical duty of preserving the patient’s bodily functions. This means that the patient has no “right” to ask the physician to carry out such an act of heterohomicide. Szasz points out that this is a different question than whether the patient ought to be “at liberty” to end his own life, without involving the physician.
Though prohibited from directly acting to end the patient’s life, the physician, in my view, may sometimes agree to the cessation of certain measures or devices that essentially prolong the dying process but provide no realistic hope of recovery—e.g. respirators, dialysis, and even IV hydration—provided the dying patient no longer wishes to be kept alive. This position—while controversial in both medical and theological circles—is consistent with the preponderance of rabbinical opinion and with the Judaic concept of “removing impediments to death.” Such an intervention does not fit the definition of PAH because the intent is not to end the patient’s life, even though this may be a highly probable outcome. (In fact, not all cases in which a respirator is removed result in the patient’s death). The intent is to avoid futile prolongation of dying and its attendant pain, suffering, and loss of human dignity. Neither does the “removal of impediments to death” fit Szasz’s concept of heterohomicide or “medical killing,” as I understand Szasz’s position. As Szasz himself observes, “The moral and legal difference between assisting a person to kill himself and desisting from keeping him alive against his will is quintessentially a difference of kind, not of degree.”3
Conclusion: personhood as the focus of end-of-life care
In my view, as we move from ordinary treatment to end-of-life conditions, there is a gradual metamorphosis in the physician’s primary ethical obligations. Specifically, in certain end-of-life conditions, the meaning and expression of beneficence and non-malfeasance change with respect to the physician’s obligations and behavior. Beneficence shifts from the preservation of organic function to the relief of pain and the preservation of the patient’s dignity. Non-malfeasance shifts from an emphasis on avoiding the compromise of biological functions to one of not prolonging pain or undermining the patient’s dignity. At no time do any of the physician’s ethical duties “disappear”—they merely wax or wane, according to circumstances. For example, a patient nearing the end of life might develop an abnormally low serum calcium level, with subsequent painful muscle spasms. While vigilantly maintaining a normal calcium level in all circumstances is no longer the physician’s primary duty, restoring the patient’s calcium level to normal is justified in so far as it relieves the patient’s pain and suffering—that is, to just the extent that the intervention serves the preeminent expression of beneficence at end-of-life.
And so, at last, what is the ethical course of action in caring for Mr. B at the end of his life? Conscientious physicians and ethicists will undoubtedly proffer widely differing answers to this vexing question, and not all physicians or patients will feel comfortable drawing upon either Ross’ philosophy or the ethics of rabbinical Judaism. However, based on these two intertwined perspectives, I believe that a competently-rendered decision on Mr. B’s part to discontinue all fluids ought to be respected. Attempts to prolong Mr. B’s life against his expressed wishes would violate not only the patient’s autonomy, but also his dignity.
The ethical perspectives discussed rule out any direct form of so-called “physician-assisted suicide,” such as knowingly administering a lethal dose of morphine to a dying patient, even with the patient’s consent. As I have argued, such an action is more accurately described as physician-actuated heterohomicide. In addition to violating the contingent duty of preserving bodily function, I believe that PAH breaks down the “social contract” between physician and patient, confusing the roles of healer and killer. In the long run, this will serve only to undermine society’s trust in the medical profession. Nonetheless, removing impediments to death, under certain carefully-defined conditions and with the patient’s expressed, informed consent, may be within the purview of ethical medical practice.
- Skelton, A, William David Ross. In: Zalta EN, editor, The Stanford encyclopedia of philosophy [Internet]. 2010 [updated Fall 2010; cited 9 Mar 2011]. Available from: http://plato.stanford.edu/archives/fall2010/entries/william-david-ross/.
- Telushkin J. A code of Jewish ethics (vol. 2). New York: Bell Tower; 2009.
- Szasz T. Fatal freedom: the ethics and politics of suicide. Syracuse: Syracuse University Press; 1999.
- Battin MP, Rhodes R, Silvers A, editors. Physician assisted suicide: expanding the debate. New York: Routledge; 1998.
- Brescia FJ. Palliative care in pancreatic cancer. Cancer Control [Internet] 2004 [cited 2011 Feb 6];11(1):39-45. Available from: http://www.medscape.com/viewarticle/468138_10
- Dalal S, Bruera E. Dehydration in cancer patients: to treat or not to treat. J Support Oncol 2004;2:467–487.
- Rosner F. Biomedical Ethics and Jewish Law. Hoboken, NJ. Ktav, 2001.
- Ross WD. The right and the good. Oxford: Oxford University Press; 1930.
- Tendler M, Rosner F.“Quality and sanctity of life in the Talmud and Midrash,” in Tendler M: Responsa of Rav Moshe Feinstein, vol. 1, Care of the Critically Ill. Hoboken NJ, Ktav, 1996.
RONALD W. PIES, MD, is Professor of Psychiatry and Lecturer on Bioethics & Humanities at SUNY Upstate Medical University. He is also Clinical Professor of Psychiatry at Tufts University School of Medicine and Editor-in-Chief Emeritus of Psychiatric Times. He is the author of several textbooks, including Clinical Manual of Psychiatric Diagnosis and Treatment and the Handbook of Essential Psychopharmacology. Dr. Pies has also written A Consumer’s Guide to Choosing the Right Psychotherapist, a book of poetry called Creeping Thyme, and Everything Has Two Handles: The Stoic’s Guide to the Art of Living. His most recent books are Becoming a Mensch and The Judaic Foundations of Cognitive-Behavioral Therapy.
Highlighted in Frontispiece Spring 2011 – Volume 3, Issue 2
Spring 2011 | Sections | Ethics