Cynthia X. Pan M.D.1
1Assistant Professor of Palliative Care Education, Mount Sinai School of Medicine, and Staff Attending Physician, Hertzberg Insitute of Palliative Care, Sinai Hospital
The author has no commercial relationships with manufacturers of products or providers of services discussed in this chapter.
One unanticipated result of the advances in health care during the past century has been the emergence of chronic illness as the leading cause of death [see Table 1]. At the same time, the enhanced ability to significantly extend life for patients with chronic diseases has blurred the boundary between curable illnesses and illnesses that inevitably result in death. As a result, over the course of the 20th century, Western society increasingly attributed near-miraculous powers to medical science—and increasingly avoided the subject of death. Many patients and physicians came to regard the prolongation of life and the cure of disease as the fundamental and exclusive goals of modern medicine. Viewed from this perspective, death is a medical failure.
Table 1 Leading Causes of Death in the United States: 200046
Recent decades, however, have seen a growing recognition that this view is unrealistic and potentially harmful. This recognition has supported the emergence of the field of palliative care. Unlike curative care, which focuses on the disease process, palliative care focuses on the patient, striving to minimize the patient's burden and maximize the patient's quality of life. A distinguishing feature of palliative care is that it openly acknowledges dying as part of living and does not consider death an enemy.1
This chapter describes the general concepts of palliative care, reviews the clinical skills needed to provide competent palliative care to patients who are chronically ill or near the end of life, and discusses some of the challenging legal and ethical issues often encountered in palliative and terminal care.
History and Rationale
Palliative medicine was first recognized as a medical specialty in Great Britain in 1987. This discipline grew out of the hospice care movement, a special interdisciplinary system of comprehensive care for the dying and for their families.1 Over time, the palliative care model has been extended. It now applies not only to patients who are clearly at the end of life but also to those with chronic illnesses that, although not imminently fatal, cause significant impairment in function, quality of life, and independence. Palliative medicine for patients with serious illness thus should no longer be seen as the alternative to traditional life-prolonging care. Instead, it should be viewed as part of an integrated approach to medical care. Palliative care is not characterized by less care or by withdrawal of care. On the contrary, palliative care may involve intensive and highly sophisticated medical interventions, albeit ones intended to relieve suffering or improve quality of life.
Settings for Delivery of Palliative Care
Palliative care may be delivered in a variety of settings, including a hospital, nursing home, hospice, or private home. In some cases, the level of care required will dictate the choice of setting. For the most part, however, palliative care depends more on the attitude of the clinician than on the setting.
Increasing numbers of hospital-based palliative care programs have been developed in recent years to meet the needs of people who are chronically and critically ill and eventually die in hospitals.2 A national Center to Advance Palliative Care has been created to provide technical support and resources for hospitals that want to establish such programs [see Sidebar Palliative Care Information and Resources on the Internet].
Palliative Care Information and Resources on the Internet
American Academy of Hospice and Palliative Medicine (AAHPM)
Organization for physicians dedicated to the advancement of hospice/palliative medicine.
Americans for Better Care of the Dying (ABCD)
Nonprofit public charity dedicated to social, professional, and policy reform aimed to improve the care system for patients with serious illness and their families.
Center to Advance Palliative Care (CAPC)
For hospitals and health systems interested in developing palliative care programs.
Death and Dying: MEDLINEplus
Links from the U.S. National Library of Medicine and the National Institutes of Health.
Education for Physicians on End-of-life Care (EPEC)
Provides a core curriculum for physicians on the basic knowledge and skills needed to appropriately care for dying patients.
End of Life/Palliative Education Resource Center (EPERC)
Identifies and disseminates information on end-of-life care education and training materials, publications, conferences, and other resources.
End-of-life Nursing Education Consortium (ELNEC) Project
Provides a comprehensive national education program to develop a core of expert nursing educators and to coordinate national nursing efforts in end-of-life care.
Gateway to international resources for life-threatening illness and end-of-life care; intended to improve the quality of care for dying people through public education and global professional collaboration; includes links and search engine on reviewed resources for end-of-life care.
National Hospice and Palliative Care Organization (NHPCO)
Nonprofit membership organization representing hospice and palliative care programs and professionals in the United States. Offers information in Spanish and English on local hospice and palliative care programs across the country. Toll-free telephone number: 800-658-8898.
Project on Death in America (PDIA)
Strives to increase understanding and transform the culture and experience of dying and bereavement through initiatives in research, scholarship, the humanities, and the arts, and to foster innovations in the provision of care, public education, professional education, and public policy.
Hospice is one way to deliver palliative care [see Table 2]. Hospice care traditionally has been characterized as low tech, high touch. Hospice provides home nursing, support for the family, spiritual counseling, pain treatment, medications for the illness that prompted the referral, medical care, and some inpatient care. The National Hospice and Palliative Care Organization (NHPCO) estimates that in the United States, hospices admitted 775,000 patients in 2001 (compared with approximately 340,000 persons in 1994) and that, in 2000, one in four persons who died of all causes were receiving hospice care at the time of death.3
Table 2 Comparison of Hospice and Palliative Care
Palliative care and hospice share similar philosophies, and both are delivered by an interdisciplinary team of health care professionals. Palliative care differs from hospice care in that palliative care can be provided at any time during an illness and in a variety of settings, may be combined with curative treatments, and is independent of the third-party payer. In the United States, hospice is paid through the Medicare Hospice Benefit. Medicare requires that recipients spend 80% of hospice care days at home, which means that to qualify for hospice, the patient must have a home and have caregivers (e.g., family members) capable of providing care. In addition, primarily for financial reasons, Medicare requires that recipients have an estimated survival of 6 months or less and that their care be focused on comfort rather than cure.4 These eligibility rules were created at a time when hospice programs principally served patients with cancer or AIDS, in which the trajectory of dying is relatively predictable; in 1994, for example, 80% of hospice patients had cancer, and the average patient enrolled about 1 month before death.5 Because hospice increasingly serves patients with chronic conditions in which prognostication remains inaccurate, these eligibility rules now limit access to care.6,7 Asking patients and families to choose between curative care and palliative care is difficult for all concerned and is inconsistent with the current model of care, which views palliative care on a continuum with life-prolonging therapy [see Figure 1].8 Also, this either/or situation contributes to late referrals and underutilization of hospice services.9
Figure 1. Models of Curative and Palliative Care
(a) Formerly, curative care and palliative care were viewed as mutually exclusive; when death became inevitable, curative care was abandoned and palliative care begun. This model of care is now outdated, because prognosis is so difficult to determine. (b) The current model views palliative care on a continuum with life-prolonging therapy, with palliative care assuming increasing importance as the patient's illness progresses and curative options are exhausted. Also, many chronic and life-threatening illnesses have no cures; the goals of treatment are to contain the illness and maintain an acceptable level of function and quality of life.
Palliative care can be provided in nursing homes, and increasing numbers of nursing homes strive to do so. Policies governing the coordination of palliative care in nursing homes vary according to reimbursement venues and availability of trained staff. Many nursing homes coordinate palliative care through local hospices, taking advantage of the skilled hospice nurses and other health care professionals.
Demographics of Death and Dying in the United States
Most people in the United States can now expect to die in old age. Of the over 2 million deaths per year in the United States, 73% occur in persons 65 years of age or older: 49% in persons 65 to 84 years of age, and 24% in persons 85 years of age or older. In 2001, the estimated life expectancy at birth reached 77.2 years, compared with less than 50 years in 1900.10
The median age of death in the United States is 77 years of age; of persons who survive to 65 years of age, median age at death is 84 years for women and 80 years for men.11 Persons 65 years of age or older constitute an increasingly large number and proportion of the United States population, and those persons 85 years of age or older constitute the most rapidly growing segment. In 1999, persons 65 years of age or older accounted for about 13% of the population; this proportion is projected to rise to 20% by the year 2030.12
The elderly population is extremely heterogeneous, varying in socioeconomic status, educational level, and cultural and ethnic background. This diversity is likely to increase in the coming years. For example, African Americans 65 years of age and older numbered 2.5 million in 1990 (constituting 8% of the population of persons older than 65 years), and their number is expected to more than triple, to 8.4 million (or 10.5% of that population) by 2030. Similarly, there were approximately 1.1 million Hispanic elderly persons in 1990 (3.5% of the population of persons older than 65 years), but by 2030 this number will skyrocket to 12.5 million (15.6% of that group).12
Compared with the current elderly population, elderly baby boomers will be far more knowledgeable about health care and far more demanding of health care providers. Their expectations are likely to lead them to challenge the health care profession to deliver high quality end-of-life care tailored to patients' individual need and to provide that care in a culturally sensitive manner.
Although most deaths occur in the elderly, people can become critically ill at any point in their lives and can die at any age. In fact, the persons whose cases were the basis for establishing important precedents for ethical and legal decisions related to death and dying were young adults: 26-year-old Nancy Cruzan,13 whose case involved the issue of artificial feeding of patients in a persistent vegetative state; and 21-year-old Karen Ann Quinlan,14 whose case involved the withdrawal of artificial ventilation from patients in a persistent vegetative state.
LEADING CAUSES AND SETTINGS OF DEATH
The three leading causes of death in adults in the United States in 2001 were heart disease, malignant neoplasm, and stroke.10 Chronic obstructive pulmonary disease (COPD), pneumonia, and accidents each accounted for less than 10% of all deaths. Most adult deaths in the United States occur in hospitals (56%), followed sequentially by deaths occurring at home (21%), in nursing homes (19%), and in other settings (4%).15 These statistics vary substantially according to geographic site, primarily because of regional variations in hospital, hospice, and nursing home bed supply.
Prognosis and Palliative Care
Traditionally, palliative care has been narrowly conceptualized as an alternative to standard life-prolonging therapy and has been provided to patients whose disease no longer responds to curative treatment. Although this model may be appropriate for patients dying of metastatic cancer, in which prognosis is relatively predictable and response to treatment is typically well defined, fewer than a quarter of persons in the United States die of cancer; the majority die of chronic diseases (e.g., heart disease) in which the prognosis is often uncertain, functional decline is nonlinear, and life-prolonging therapies coexist with or are identical to therapies directed at palliation and comfort (e.g., diuresis for fluid overload in heart failure).
One of the barriers to initiating palliative care is the uncertainty of predicting prognosis in these complex, chronic medical illnesses. For example, timing of death in heart failure is far less predictable than in many other fatal disorders. A patient dying of colon cancer usually has a long period of functional stability, then several months of progressive functional decline and weight loss just before death. In contrast, most heart failure patients experience a lengthy decline in daily function, with periodic bouts of severe symptoms and disability and multiple hospital admissions for exacerbation and for adjustment of therapy. Death may occur during a severe exacerbation but often occurs suddenly and relatively unpredictably from cardiac arrhythmia [see Figure 2]. In the SUPPORT project (Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments),16 heart failure patients were given surprisingly long prognoses even up to the day before death. The median prognosis on the day before death was a 50% chance of living 2 months.17 Dementia is another condition that often progresses over years rather than months.
Figure 2. Predictability of Prognosis
Prognosis is relatively predictable in metastatic cancer; these patients typically have a long period of functional stability, then several months of progressive functional decline and weight loss just before death. In contrast, prognosis can be difficult to predict in diseases such as chronic heart failure (CHF), chronic obstructive pulmonary disease (COPD), and Alzheimer disease; these patients typically experience a lengthy decline in daily function, with periodic bouts of severe symptoms and disability and multiple hospital admissions for exacerbation and for adjustment of therapy. Death may occur during a severe exacerbation, but—especially in CHF—often occurs suddenly and relatively unpredictably from cardiac arrhythmia.
To help clinicians assess prognosis in various nonneoplastic conditions, the NHPCO has compiled guidelines describing factors associated with poor prognosis; these guidelines can promote discussion about preferences for care and advance care planning.18 However, SUPPORT data indicate that for seriously ill, hospitalized patients with advanced COPD, heart failure, or end-stage liver disease, recommended clinical-prediction criteria cannot reliably identify those patients whose survival prognosis is 6 months or less.19 Because it is not possible to consistently and accurately predict the timing of death, palliative care interventions should be incorporated early in a patient's course of illness, even in the face of substantial uncertainty about prognosis. As disease progresses, the goals of care should change accordingly, with the balance shifting from curative to palliative.
Caring for dying patients and for those who suffer from chronic and severe illnesses with uncertain prognoses requires an interdisciplinary approach and specific clinical skills. In particular, the clinician who provides palliative care must be competent in clinical communication, management of symptoms (physical, emotional, and psychological), and planning for continuity of care.
The ability to communicate well with both patient and family is paramount in palliative care. In the beginning, it enables the physician to deliver bad news, assess the patient's and the family's knowledge and understanding of the disease process, determine the factors that they consider important to quality of life, and discuss goals and preferences for future care. As the illness progresses, regular communication about the course of illness and the patient's needs and expectations enables the physician to provide the most appropriate care for the patient and support for the family. Communication continues to be important after the patient's death, because the period of bereavement poses major challenges and increased risks of medical and psychiatric illness for family members.1
Patients whose cultural background and language differ from that of the physician present special challenges and rewards and need to be approached in a culturally sensitive manner [see Accounting for Cultural Differences, below]. Physicians also need to communicate effectively with colleagues and interdisciplinary team members to achieve optimal care for their patients. Communication in palliative care is discussed in detail elsewhere [see CE:XI Management of Psychosocial Issues in Terminal Illness].
Symptom management in palliative care encompasses the assessment and treatment not only of physical symptoms but of emotional and psychological symptoms as well. Physical symptoms that can contribute to discomfort, disability, and dependence include pain, dyspnea, constipation, nausea and vomiting, delirium, fatigue, and anorexia. Emotional and psychological symptoms include depression, anxiety, delirium, cognitive impairment, fear, and agitation or sedation, as well as spiritual and existential angst.
Pain is the most common symptom of terminal illness, reported by 84% of patients with cancer and 67% of patients dying of other causes.11Surprisingly, the leading cause of physical distress in patients dying of heart failure is also pain, followed by fatigue and shortness of breath. Other common symptoms reported by dying patients include trouble with breathing (49%), nausea and vomiting (33%), sleeplessness (40%), depression (36%), loss of appetite (47%), and constipation (36%). Apart from illness, symptoms that become more prevalent with increasing age include mental confusion, loss of bladder and bowel control, difficulty seeing and hearing, and dizziness.
At present, the identification and management of many symptoms, including pain, remain suboptimal. Undertreatment of symptoms is common in elderly patients whether they have cancer20 or other chronic conditions, whether they reside in long-term care settings (45% to 80% prevalence)21,22 or in the community, and whether they are white or are members of minority groups.20,23 Undertreatment of pain in the elderly may be more common in patients who are women, are members of minority groups, or have mild to moderate cognitive impairment.24 Clinicians may contribute to undertreatment of pain through lack of proper pain-assessment procedures, misconceptions regarding both the efficacy of nonpharmacologic painmanagement strategies and the attitudes of the elderly toward such treatments,25 and legitimate concerns about drug interactions and side effects.
Education and involvement of the patient and family as partners in care are key to the successful management of symptoms. Specific strategies for symptom management include both pharmacologic and nonpharmacologic measures. These strategies are discussed in detail elsewhere [see CE:IX Symptom Management in Palliative Medicine and 11:XIV Pain].
ACCOUNTING FOR CULTURAL DIFFERENCES
The United States is a culturally heterogeneous country. Culture can broadly include race and ethnicity, as well as country of origin, religion, spirituality, and profession.26 Medicine has its culture as well, with its attendant values, beliefs, behaviors, and language. Thus, in some cases, patients may be encountering two unfamiliar cultures: that of the United States mainstream and that of Western medicine.
Cultural traits may have a far-reaching impact on a patient's views on illness, preferences, and ultimate decisions.27 Compared with patients from mainstream United States culture (most of whom are whites of European descent), people from other cultural backgrounds may be less willing to discuss resuscitation status,27 less likely to forgo life-sustaining treatment,27,28,29 and more reluctant to complete advance directives.30
Although many individual variations exist, some frequently encountered examples of cultural differences include the following:
With patients who do not speak English, it is extremely helpful to have access to a trained interpreter who can provide an objective translation and shift the translation burden from family members. This can prevent awkward and inappropriate situations, such as having to ask a male teenager to interpret for his mother who has cervical cancer. Translators may also be able to provide valuable information about patients' cultural attitudes and expectations.
Although it is important to learn about and respect different cultural practices, it is even more important not to stereotype or assume that membership in a group determines preferences. Instead, the physician should treat each patient as an individual. When in doubt, ask: “I have had patients from your cultural group who told me…. Does this apply to you?” Or, “I don't know much about medical practices or beliefs in your culture; can you tell me more about this?”
Public opinion polls have revealed that close to 90% of adults in the United States would not want to be maintained on life-support systems without prospect of recovery. Yet a survey of emergency department patients found that 77% did not have advance directives, and of those patients who had one, only 5% had discussed their advance directive with their primary care physician.32 A survey of community-dwelling older adults found that only about 16% had written advance directives.33 In a survey of adult outpatients, most felt that discussions about advance directives should take place at an earlier age, earlier in the course of the disease, and earlier in the patient-physician relationship; most subjects also agreed that it was the physician's responsibility to initiate the discussion.34
Primary care physicians are in an excellent position to speak with patients about their care preferences because of the therapeutic relationship that already exists between patient and doctor. Conversations about preferences of care should be a routine aspect of care, even in healthy older patients. Determination of the patient's preferences can be made over two or three visits and then updated on a regular basis (e.g., annually). Reevaluation is indicated if the patient's condition changes acutely. In general, it is preferable that a close family member or friend accompany the patient during these discussions, so that these care preferences can be witnessed and any potential surprises or conflicts can be explored with the family.
Such discussions have particular urgency in patients who are showing early signs of cognitive impairment, because advancing impairment may render these discussions impossible. In older persons with existing cognitive impairment, it is important to assess both their current degree of decision-making capacity and any evidence, written or verbal, of previously stated preferences.
Decision-making capacity refers to the capacity to provide informed consent to treatment. This is different from competence, which is a legal term; competence is determined by a court. Any physician who has adequate training can determine capacity. It does not need to be determined by a psychiatrist. Primary care physicians often have more insight and knowledge about their patients than a psychiatrist, who might be seeing the patient for the first time.35 The more complicated and serious the decision, the more stringent the requirements for understanding. For instance, a demented patient may have the capacity to appoint a trusted family member to serve as health care proxy but may not have the capacity to decide whether to have a permanent feeding tube placed.
A patient must meet three key criteria to demonstrate decision-making capacity: (1) the ability to understand information about diagnosis and treatment; (2) the ability to evaluate, deliberate, weigh alternatives, and compare risks and benefits; and (3) the ability to communicate a choice, whether verbally, in writing, or with a nod or gesture.
In eliciting patient preferences, the clinician should explore the patient's values—what is important to the patient and what makes life worth living or what makes life intolerable. The clinician should help the patient identify and set realistic goals, then direct treatment decisions according to these goals. More specifically, it is important to evaluate whether the patient would prefer to focus on length of life or quality of life if faced with a serious illness. In older persons who have chronic conditions that are not immediately life threatening, there is more time to explore these issues and to modify decisions over time. Outlining the available treatment options (e.g., probability and extent of response to treatment, duration and quality of extended life, anticipated side effects), identifying patients' short- and long-term goals and needs, uncovering their expectations about therapy, evaluating their coping strategies, and identifying their support networks are critical components of this discussion.
Discussions of care preferences should cover specific life-sustaining treatments such as cardiopulmonary resuscitation (CPR), artificial nutrition and hydration, and mechanical intubation and ventilation. Physicians should review with the patient the potential indications for such therapies and, if possible, offer an explicit appraisal of the outcome in their situation. A helpful strategy is to ask patients how long they think they would want a particular treatment to be continued if it did not seem to be helping. For example, the physician might ask, “If you had a brain injury that left you in a coma and the neurologists determined that only a miracle would restore your brain function, how long would you want to stay on treatments that were keeping you alive?” Some patients may specify a week, some a month, and still others, a year. Such discussions help clarify the patient's preferences and tolerance for uncertainty.
In eliciting patient preferences for care, it is critical to consider the person's cultural, ethnic, and religious background.26 For example, it is fairly well known that Jehovah's Witnesses will not accept blood transfusions, even in the face of life-threatening conditions, but may want all other invasive treatments. Such differences can make a patient unwilling to accept a physician's recommendations and can make a physician angry at the patient's resistance to those recommendations. With patience and training, however, it is usually possible to uncover these beliefs and negotiate treatment plans that are acceptable to all concerned.
TYPES OF ADVANCE DIRECTIVES
There are three types of advance directives: (1) do not resuscitate (DNR) orders, (2) directives involving health care proxies, and (3) living wills. All are legal instruments. The federal Patient Self-Determination Act of 1990 requires hospitals, skilled nursing facilities, home health agencies, hospice programs, and health maintenance organizations to maintain written policies and procedures guaranteeing that every adult receiving medical care be given written information concerning advance directives. Although forms for designating health care proxies and living wills are completed by patients themselves, physicians may wish to secure copies of the forms used in their state and assist their patients in completing these forms.
A health care proxy is a person appointed by the patient to make health care decisions in the patient's stead, in the event that the patient loses the capacity to make those decisions. In general, it is preferable for the physician to speak to the patient first, ask the patient to think about appointing a health care proxy, and then ask the patient to bring the potential proxy to a follow-up meeting. The proxy should be aware of and advised about the patient's goals of care and preferences and should be able and willing to assume the responsibilities of serving as proxy. Typically, an alternative proxy is also appointed.
A living will is a document that directs health care personnel to withhold or withdraw life-sustaining treatment in the event that the patient is in an incurable or irreversible condition with no reasonable expectation of recovery. Not all states have statutes recognizing living wills. However, courts have recognized and upheld the use of living wills as long as these documents provide “clear and convincing evidence” of a competent patient's wishes.
Ethical Issues in Palliative Care
Chronic illness and end-of-life care bring into focus some compelling ethical issues. These include limiting life-sustaining treatments, physician-assisted suicide, and euthanasia. Guidelines and principles on these issues have been established to enable patients, families, and physicians to reach medically sound, ethical treatment decisions in cases of irreversible illness. As a result, and despite widespread physician feeling to the contrary, these treatment decisions are almost devoid of litigation danger. Nevertheless, physicians should work with their hospital attorneys to clarify the status of legislation and case law on these issues in their particular jurisdiction.
LIMITING TREATMENT (REFOCUSING GOALS OF CARE)
In the discussion of treatment goals and plans with patients or family members, the language a physician uses can make a tremendous difference. If the physician says, “It is time to stop [or limit] the treatments,” the patient or family will likely feel abandoned and hopeless and therefore ask for more interventions that may not be appropriate or useful. However, if the physician says, “It is time to refocus our efforts; let's strive to maximize comfort and dignity rather than prolong the dying process,” the patient and family are more likely to feel validated and reassured. Similarly, if the physician refers to mechanical ventilation, dialysis, or artificial nutrition as “life-sustaining” treatments, it is a rare individual who will elect to forgo them. Rather, the physician should refer to them as medical interventions used to achieve specific goals. For example, one might speak of instituting mechanical ventilation to support breathing, in the hope that the patient will regain spontaneous breathing; if this hope is not realized, it is then time to discuss what the goals of care are and whether they need to be modified.
Ordinarily, discussions of goals of care involving limitation of life-sustaining treatment occur in three categories of patients. The first category includes patients whose illness is judged irreversible and who are moribund; these patients usually do not benefit from aggressive medical interventions, which can become invasive and burdensome. For patients who will die with or without treatment, such as a patient with advanced metastatic cancer or a patient with end-stage cardiomyopathy for whom a transplant is not possible, interventions often pose more burdens and risks than benefit. The second category consists of patients with capacity who are not moribund but have an irreversible illness, such as amyotrophic lateral sclerosis or multiple sclerosis. These patients often wish to discuss their ultimate goals of care and their right to refuse medical treatments so as to retain control over their health care as their disease progresses. The third category includes patients with capacity who have a reversible illness. As with any patient with capacity, the principle of autonomy guarantees these patients the right to refuse any treatment, even a lifesaving one, although physicians obviously will question these refusals much more vigorously than refusals in cases of irreversible and progressive illnesses.
An important caveat here is that although supreme autonomy of the patient is valued by mainstream culture in the United States, it is not the guiding value of many other cultures. In fact, most ethnic groups in the United States (e.g., Hispanic Americans, Asian Americans, Orthodox Jews, African Americans) favor a family-based decision-making process. Furthermore, autonomy does not always mean that the patient must be informed or must participate in decision making. Autonomy means that patients should be asked whether they wish to be informed or participate in decision making; they may refuse to do either.
In some cases, a limited trial of life-prolonging treatment may clarify the patient's chances of recovery. The treatment can be stopped if it becomes clear that health (or the extent of recovery acceptable to the patient) cannot be restored. However, sometimes it is psychologically more difficult to stop such a treatment once it has been started, even if its original justification no longer applies.
When the patient does not have capacity, there are several ways to resolve treatment decisions. Advance directives are the most helpful. Otherwise, common sense should be followed, and the patient's next of kin should be asked to provide a substituted judgment about what the patient would have wanted or what decision would be in the patient's best interest.
Ethically and legally, there is no difference between forgoing or withholding a medical treatment (such as mechanical ventilation) and stopping or withdrawing it. However, family members and health care providers may feel that withholding and withdrawing interventions are emotionally different. It is therefore critical to counsel families and health care professionals that the decisions about any medical treatment should be guided by overall goals of care. Consultation with the hospital's palliative care service or ethics committee may be valuable for resolving conflicts over life-prolonging treatments.
Conflicts that require arbitration often center on treatments that either the family or the treatment team regard as futile or ineffective. Futility is a narrowly applied term that is used in the setting of CPR to describe a resuscitation attempt that would not succeed in resuscitating the patient or that, if successful, would likely be followed shortly afterward by another arrest. In many cases, when a patient is irreversibly ill and dying, CPR would be futile. Application of it is contrary to the standards of medical practice; it is unethical and inhumane. In such a case, the physician does not have a duty to consult anyone before writing a DNR order but should inform the family that a DNR order is being implemented. This is an opportunity for the physician to remind the family just how severe the illness is and to refocus attention to meeting other needs of the patient and the family.
Defining futility is currently a major goal of medical ethics.36 The negative right of refusal has become transformed by some into a positive right to demand of physicians any life-sustaining treatment. Others argue that physicians have a duty not to offer or provide treatments that are ineffective.37 Because most risk-versus-benefit considerations of life-sustaining treatment involve value judgments and because the principle of autonomy requires that the patient's values come first, some argue that so-called objective standards for futility are impossible to formulate and that physicians should make no such judgments.38 However, for patients in an irreversible coma and, increasingly, for those in a persistent vegetative state, life-sustaining treatments are seen to be futile.
Controversial questions about defining treatments as futile will most likely be resolved city by city by a panel of experts set up to judge whether a treatment is futile after hearing all evidence presented by the family, the medical team, and others. This was the approach used by the Houston citywide consortium of hospitals.39
THE REQUEST FOR ASSISTED SUICIDE OR EUTHANASIA
In the United States, the public increasingly accepts physician-assisted suicide and euthanasia as moral practices and believes that these practices should be legal.40 These views can be seen as the public's condemnation of at least two things: the way hospitals and physicians overtreat sick patients in their last days, making death a painful journey; and medicine's inadequate and ineffective treatment of suffering. These views also reflect a demand for more control in decisions about the end of life. Some people equate physician-assisted suicide with euthanasia, but they are different concepts. In physician-assisted suicide, the patient requests the physician's help in dying, usually in the form of a prescription of a lethal dose of medication to be taken at home. Euthanasia occurs when there is no patient request but the physician (or other health care professional) decides to hasten the patient's dying process in order to relieve suffering (the patient's or the physician's).
In June 1997, the United States Supreme Court ruled that there is no constitutional right to physician-assisted suicide.41 This opinion did not remove the authority of individual states to outlaw or decriminalize physician-assisted suicide, however; and in November 1997, Oregon voters confirmed their acceptance of the Death with Dignity Act, which allows terminally ill Oregon residents to obtain from their physicians and to use prescriptions for self-administered, lethal medications. The act states that ending one's life in accordance with the law does not constitute suicide, and it specifically prohibits euthanasia (i.e., direct administration of a medication to end another person's life).
Many, if not most, of those patients who want physician-assisted suicide want it not to relieve suffering as ordinarily understood but to maintain control over their dying.42 As of 1999 (2 years after legalization of physician-assisted suicide in Oregon), a survey of Oregon physicians found that they granted about one in six requests for a prescription for a lethal medication and that one in 10 requests actually resulted in suicide. Substantive palliative interventions led some—but not all—patients to change their minds about assisted suicide.43,44 As of 2002, a total of 129 people had committed physician-assisted suicide in Oregon, corresponding to a rate of 8.8 per 10,000 deaths from any cause in the state.45 Compared with Oregon residents who died of similar underlying causes, rates of physician-assisted suicide decreased with age and were higher among those who had been divorced and among those with higher levels of education. The rate of physician-assisted suicide was also higher among those afflicted with amyotrophic lateral sclerosis and cancer. The majority of patients using physician-assisted suicide were non-Hispanic whites, but a significant minority were Asian Americans. Overall, the number of Oregon residents using physician-assisted suicide has increased over the years, but it remains a very small minority relative to overall deaths.
Regardless of legal issues, however, when a patient requests a prescription for enough medication to commit suicide or to hasten death, the physician has the ethical responsibility to try to learn why. What is it that now makes death seem a better option than life? What is it that the patient feels must be avoided? From what is the patient trying to escape? Is the patient depressed? Why does the patient feel that that he or she can no longer be someone who matters? Are there financial considerations—that is, does the patient fear becoming a financial burden, a burden to care for, or both? Has any of this been discussed with the family? How would the family understand the patient's requests and be affected by them? If the patient considers life to be devoid of value and meaning, does the patient's life still have meaning for other persons? Does this affect the patient? Has the patient made any effort to achieve consensus so that his or her death can be a meaningful, shared family experience?
FEAR OF LEGAL REPRISAL
When a physician makes a reasonable clinical judgment of irreversible illness and decides to forgo or stop life-sustaining treatment—whether on the wish of the patient or, if the patient is incompetent, with the agreement of the patient's proxy or surrogates—fear of litigation is neither a reasonable nor a legitimate excuse not to proceed. The courts have made it clear that these decisions are valid and that the persons involved in those decisions should not be brought to trial. It is irrational to demand guarantees that no litigation will follow, however. It is hoped that physicians' energies will be spent doing the best they can for the patient, in accord with the patient's wishes. Should litigation follow an action taken in accord with the above guidelines, the physician will be well prepared to defend the decisions in court.
Editors: Dale, David C.; Federman, Daniel D.