34 Palliative Medicine
The Massachusetts General Hospital Handbook of Pain Management
Andrew T. Putnam and J. Andrew Billings
You are outside life, you are above life, you are afflicted with ills the ordinary person does not know, you transcend the normal level and that is what people hold against you, you poison their quietude, you corrode their stability. You feel repeated and fugitive pain, insoluble pain, pain outside thought, pain which is neither the body, nor the mind, but which partakes of both. And I share your ills, I am asking: who should dare to restrict the means that bring us relief?
I. What is palliative care?
II. What is a good death?
III. Alternatives to an acute-care hospital
IV. Pain management at the end of life
V. Non–pain-symptom control
VI. Communication with patients and families
VII. Three key questions for patients facing a terminal illness
VIII. Family meetings
IX. Euthanasia and physician-assisted suicide
I. WHAT IS PALLIATIVE CARE?
Palliative care is comprehensive interdisciplinary care that focuses on quality of life for patients with a life-threatening or terminal disease and for their families. Care is comprehensive because it addresses all the sources of suffering, including not only pain and other disagreeable physical symptoms but also emotional, sociocultural, and spiritual distress. A core palliative care team typically consists of a physician, nurse, social worker, and chaplain, who meet regularly to provide coordinated, interdisciplinary care and promote a patient-derived concept of quality of life. Teams may also include a bereavement counselor, a volunteer coordinator, volunteers, a nutritionist, a physical therapist, and so on. The goals of care grow out of an exploration of the patient’s notions of living well until death comes—an individualized and changing view of a “good death”—and from a commitment to address all forms of suffering. Palliative medicine is the branch of medicine encompassing the physician’s knowledge, attitudes, and skills requisite for practicing on such a palliative care team.
Palliative care should not be confused with the dismissal or abandonment conveyed when physicians say to a dying person, erroneously and perhaps cruelly, “Nothing more can be done.” It is not “comfort measures only” but rather an aggressive, active, hopeful team approach to making the best of limited time. Palliative care is based on the hospice philosophy of care, but it is not subject to regulations that limit hospice care in the United States to persons with a prognosis of 6 months or less who have decided to forego all “aggressive,” expensive, and life-prolonging interventions. Indeed, palliative care has a role in the earliest phases of management of a life-threatening or terminal illness. It coexists comfortably with curative or life-prolonging measures, and it aggressively pursues comfort and support through all appropriate means in even the final days of life.
II. WHAT IS A GOOD DEATH?
Although all medical decisions should reflect patient values and goals, nowhere in medicine are individualized notions of good care—and of a good death—more important to recognize than in palliative medicine. Patient wishes for comfort, enjoying family, or being at home may be equally or more important than such common medical goals as prolonging life or trying to cure an illness. Weighing benefits and burdens often leads to a decision to discard a variety of “aggressive” interventions—diagnostic or therapeutic maneuvers that would seem appropriate at most times of life but not when one is dying. Simple measures (e.g., sublingual medication for a patient unable to swallow rather than an intravenous line) are often favored because they are easier to carry out at home, and they do not entail regular nursing attention or much discomfort or expense.
Areas typically addressed in a palliative care plan of care include the following:
Pain and symptom control: Dying patients often experience a bewildering array of disagreeable physical symptoms that interfere with their ability to function and enjoy remaining life. These symptoms include not only pain but also nausea, vomiting, anorexia, dyspnea, weakness, fatigue, and bowel and bladder problems. The fear of unrelieved pain and suffering can be enormously disturbing and make patients wish they were already dead. While studies of what patients and family members want most when they are terminally ill tend to identify primarily nonphysical matters, excellent pain and symptom control are essential substrates that allow good psychosocial and spiritual coping and a good death. Attention to all physical symptoms and meticulous treatment of even minor distress can contribute greatly to the patient’s overall well-being.
Sense of control: Control of different aspects of life is very important to many individuals. From the time a terminal illness intrudes into one’s existence, more and more of that control is lost. People may feel diminished when they sense they are being treated as merely patients, subject to the orders of doctors and nurses and requiring family for assistance with everyday tasks. An individualized care plan reflecting sensitivity to concerns about control helps patients feel more fully themselves rather than passive victims. For example, using a patient-controlled analgesia (PCA) device rather than relying on nurses or family to bring medications may bolster a patient’s sense of active involvement in the care process.
Patient-centered decisionmaking: By current bioethical standards, competent adults can choose which medical treatment to accept. They have the right to refuse any proposed life-sustaining treatment. More important, to identify a plan of care that meets their personal values and goals, patients have the right to be helped to understand all reasonable diagnostic and treatment options and the likely consequences of carrying out such procedures.
Patients need professional help in making choices—shared decision making. If the patient is incapable of making decisions or prefers to delegate them, then a designated proxy or, by default, the closest family member, needs to consider carefully what the patient would want. For an incompetent patient, a process called substituted judgment is requested, asking the proxy to determine what the patient would choose if he or she were able to make a decision. The proxy might be asked, “Has the patient ever said anything about similar diagnostic or treatment options in the past? Would she want to live this way? How did she react to other people’s decisions in such situations or their deaths?” Answers to these sorts of questions provide clues to how substituted judgment can be carried out. Importantly, substituted judgment does not require the proxy or family to decide what they think is right or what they want; it asks them simply to express the patient’s values so that these values can be applied to decisions made by the healthcare team. Only when no such information about patient preferences is available is the proxy or family asked to help make a decision on behalf of the patient, based on what they think is right. Finally, when no proxy is available, decisions in the best interest of the patient are left to the medical team, sometimes assisted by a patient representative, an ethics consultant, or, occasionally, a court-appointed guardian.
Avoid inappropriate prolongation of dying when life is no longer enjoyable or meaningful: Patients and families fear painful, depersonalized, costly, prolonged dying. Many fear a medical system that uses technology to keep a patient alive but does not know when to stop. Patients and families should be reassured that wishes not to have the dying process prolonged will be honored. At the same time, they should never feel abandoned or that the intensity of support and attention to their comfort will be lessened because they have rejected any management options.
Great importance may be placed on protecting the patient from a bad death. Would it make more sense to let the patient die from a potentially reversible infection or metabolic disorder rather than to rescue him briefly from death but leave him to face great pain and suffering? Is an effective treatment, perhaps involving prolonged hospitalization or invasive procedures, less desirable than a simple, easily handled approach that can be carried out in the home? Neither excess reliance on medical technology nor crude and inflexible dismissal of the most elaborate technical approaches is appropriate.
Identify meaning and purpose in life and find hope for the remaining time: Paradoxically, facing death involves seeking meaning and purpose in life. Patients who recognize their terminal condition regularly ask spiritual questions: “Why me? Did I live a good life? What happens after I die? Will I be remembered?” Exploring these questions, perhaps in the context of religious beliefs, is part of making sense of mortality and facing death. The answers to these questions are also the basis of hopes for the future— not necessarily hopes to live forever, but to live longer or to make the best of whatever time is left. Patients often hope to fulfill key duties or wishes in their remaining time. Putting one’s affairs in order or surviving long enough to see a new grandchild or to attend a family wedding may be attainable goals. Reconciliation with loved ones may be important. Patients can be helped by asking about unfinished business: “If you died tonight, what would be left undone?” or “What do you still hope to accomplish?”
Assistance in preparation for death: Dying persons may want assistance with such practical issues as writing a will or making funeral arrangements. Many patients are hesitant to ask about their future or to share their worries about the dying process, although they harbor deep fears (e.g., of a crescendo of intractable pain, or of suffocation). Although clinicians cannot foresee the future, they can reassure patients about providing comfort and nonabandonment, and they often can reassure patients about unnecessary fears (e.g., “You really do not need to worry about suffocation with this kind of cancer.”) They can ask, “What are your concerns now? Have you had thoughts about what the future holds? Are you thinking about death? What are your worst fears?”
Psychosocial and spiritual care of the family: Because the patient’s medical condition usually takes center stage, professional caregivers may overlook the suffering of family and friends. While the patient is dying, the people standing around the bed may already be grieving, each struggling with his or her own loss. They also benefit from caring and support. Would a chaplain or priest be welcome at the bedside? Would a social worker be able to assist the family in coping better with their grief? How will they manage the financial burden of the illness or the needs of assisting the patient at home? Unhappy memories of a relative’s difficult death or the perception that an expected role was not fulfilled can cause suffering long after the patient dies. Family support should extend through the period of intense bereavement.
III. ALTERNATIVES TO AN ACUTE-CARE HOSPITAL
Patients and their families commonly wish to be home for most or all of the course of a terminal illness. In the United States, hospice programs specialize in offering appropriate end-of-life care in the home. Such programs have been shown to provide improved outcomes at lower cost for dying persons in the final 6 months of life. Unfortunately, government-mandated admission requirements, including a likely prognosis of 6 months or less and a rejection of costly, “aggressive” measures, have hindered appropriate referral of patients to hospice. Most patients use such services only in the last few weeks of life. Palliative care programs—well-established features of care throughout the United Kingdom, Australia, and Canada, and currently developing in many academic centers in the United States—are more flexible in serving patients and families earlier in the course of the illness and are willing to entertain all forms of diagnostic and therapeutic interventions that might benefit the patient.
Home care: At home, the patient often finds comfort, privacy, and security. The close involvement of family and friends and the convenience of being home are also appreciated. Most patients express the wish to be cared for and often even to die at home. Home hospice generally provides medications, durable medical equipment, occasional nursing visits, home health aide visits, and 24-hour on-call assistance, while offering support from volunteers, chaplains, social workers and bereavement coordinators. For the majority of the day, families are usually the direct caregivers in the home. When properly arranged, home hospice can be a rewarding experience and a powerful source of satisfaction for the family, leaving them with the feeling that they did everything they could for their loved one.
Institutional care: When the patient’s medical condition and comfort requires more attention or skilled care than is available in the home from trained health service providers and the informal assistance of family and friends, institutional sites of care are required. Inpatient hospices and palliative care units provide such an option, typically located in acute-care hospitals, chronic care hospitals, or nursing homes. These units ideally provide not only a homelike atmosphere (with ready access to family, friends, and pets) but also consistent professional assistance. By unburdening the family of difficult patient-care responsibilities that are often physically and psychologically burdensome, these units allow the family to support a loved one in a more comfortable, appropriate manner.
Discharge planning from an acute-care facility is a process, not a single event. Patients and their families, assisted by various health professionals, need to be educated about providing care and the common problems they might encounter. Arrangements for professional help and durable medical equipment need to be completed. Medication management is often tricky, but regimens can frequently be simplified. How can anxiety be lessened and crises averted by educating the patient and family about common problems they might face? How will they obtain assistance when new concerns arise? If the patient is going to die at home, whom does the family call at the time of death?
IV. PAIN MANAGEMENT AT THE END OF LIFE
Pain is frequently a problem at the end of life. Important principles in pain management in this situation include the following:
Opioids are the mainstay of cancer pain treatment. Many terminally ill patients with pain can be successfully managed with opioids alone, without resort to other analgesics or to interventional treatments. Some patients benefit from adjunctive therapies such as NSAIDs, neuropathic pain medications, and various psychotropic medications. A small number with unremitting, poorly controlled pain may benefit from an interventional procedure. The celiac plexus block is simple to perform and helpful for patients with intra-abdominal malignancies, particularly pancreatic cancer. Simplicity is the key in terminally patients, because many home caregivers are nonprofessionals who are easily overwhelmed by fancy treatments. Treatment choices are described in Chapter 32.
Many patients prefer the oral route. The oral route for pain relief generally allows the greatest freedom and comfort for the patient and greatest ease for home management. Long-acting forms are convenient. In general, avoid (a) painful intramuscular injections; (b) intravenous injections that require skilled personnel, regular monitoring, and continuous access; and (c) epidural or intrathecal pumps that often intimidate families and home health nurses, that may fail at home, and that are unacceptable at some post–acute-care facilities.
If a patient loses the ability to swallow pills and liquids, concentrated opioid elixirs can be taken sublingually. Analgesia from sublingual administration of morphine or oxycodone is roughly equivalent to that from oral administration of the same dosage.
Use simple alternatives to the oral route when patients cannot swallow.
Transdermal administration of medications is commonly used for pain control at the end of life when patients cannot swallow reliably, or earlier in the course of an illness when the oral route is not feasible. The fentanyl patch may provide basal pain control, while the sublingual or rectal route is reserved for as-needed medication. The smallest patch (25 µg/hr) is roughly equivalent to 90 mg of oral morphine over 24 hours. Clinicians should avoid the temptation to start the patch when the patient’s opioid requirements are significantly lower than this daily dosage of morphine. Transdermal preparations of other medications, including opioids, are popular in some hospice practices, although little evidence supports their efficacy.
The rectal route may be useful. A few opioids are available commercially as suppositories, but a pharmacist can often prepare suppositories of a desired medication and dosage. Analgesic levels from rectal administration are generally comparable to that achieved with the same dosage given orally. Limited experience with long-acting opioids suggest that they can be used by the rectal route.
Subcutaneous infusion should be considered in place of intravenous or intramuscular medication for patients being managed at home, in a subacute care facility, or when intravenous access is problematic. Both intermittent and continuous infusions can be used, generally delivered through an intravenous infusion needle or a specially designed needle placed in the subcutaneous tissue. These needles are easily inserted, and accidental removal poses no hazard of bleeding. Absorption appears roughly equivalent to that achieved with an intravenous infusion, if somewhat delayed. Considerable experience has now accumulated in combining opioids with other agents in subcutaneous infusions. Subcutaneous administration of fluids (hypodermoclysis) can also be considered.
Minimization of tolerance. Many terminal patients take opioids for long periods of time. Tolerance and the need for high dosages may lead to inconvenient regimens with many pills or patches. Toxicity, including sedation or other side effects, particularly myoclonus, is not infrequent. Opioid rotation—switching from one drug to another—may allow for use of lower dosages or occasionally leads to better pain control, either from improved analgesia or from reduced toxicity. Methadone is often chosen when a component of neuropathic pain is suspected. The opioid equivalency found in equianalgesic tables (Appendix VIII) is not appropriate once a patient has developed opioid tolerance because there is incomplete cross-tolerance between the opioids. A dose that is one half to one quarter of the equivalent dose is normally used when a different opioid is started, (one tenth in the case of methadone), with slow upward titration, if necessary, to achieve the desired effect. The principle of opioid rotation and alternative ways to control tolerance are described in Chapter 32 (IV, 2, (ii)).
Appropriate use of opioid infusions. Continuous intravenous opioid infusions are regularly used in inpatient settings to ensure good analgesia, relieve breathlessness, and/or provide sedation in the last few days of life. Common indications include (a) inability to take oral medication (vomiting, dysphagia) or uncertain oral absorption, and (b) a need for fine-tuning or rapid titration of analgesics, including for breakthrough or incidental pain, dyspnea, or even agitation. For patients with severe distress, analgesic effects may need to be monitored every 5 to 15 minutes.
When prescribing continuous opioid infusion for a dying patient, write orders for a generous dose range so that the patient’s fluctuating symptoms can be quickly relieved. In addition to the constant infusion, there should be orders for boluses “as necessary” (e.g., morphine sulfate 1 to 8 mg every 15 minutes) to be used whenever the patient complains of severe pain. Providing relief immediately with a bolus injection is preferable to waiting for the effects of an increase in infusion rate. Boluses can be via a PCA pump or by injection, remembering that very ill patients may not have the capacity to control PCA.
Inappropriate fears of hastening death. All health professionals are concerned about causing a patient’s death. Unfortunately, some clinicians avoid opioids or prescribe them at ineffective dosages to avoid any hint of hastening death. In reality, opioids prescribed and titrated appropriately for physical distress rarely produce life-threatening toxicity, except perhaps constipation. Dosages required to alleviate severe pain or dyspnea generally are not associated with serious sedation.
In unusual situations, clinicians are faced with a difficult choice in prescribing the right dose of an opioid. Higher opioid dosages that produce good symptom control may cause serious sedation (generally when nonsedating analgesics and psychostimulants have been ineffective), whereas lower opioid dosages allow greater alertness but do not control symptoms well. Patient goals and values should play an important role in deciding the appropriate therapy.
In situations where sedation or, more rarely, respiratory depression is a desirable goal and may conceivably hasten death, clinicians should be familiar with the principle or rule of double effect, which makes a distinction between intended effects and foreseeable but unintended effects. This principle, widely accepted by bioethicists and medicolegal experts, indicates that when comfort is the primary goal of medical management, a treatment for achieving comfort is justifiable even if that treatment has the potential for causing serious, predictable, but unintended side effects. The principle of double effect justifies the use of opioids or the upward titration of opioids even if the doses necessary for the patient to be comfortable cause unintended but foreseeable harm to the patient. The importance of keeping a patient comfortable at the end of life is arguably the most important responsibility of the clinician and justifies the risks of medications necessary for that comfort.
In very unusual circumstances, physical or emotional suffering near the end of life is so severe and refractory to usual medications that “sedation for intractable suffering in the dying patient” (often called by the misleading term, terminal sedation) is necessary. In these situations, continuous infusions of barbiturates (e.g., pentobarbital) or anesthetics (e.g., propofol) may be used alone or added to opioid regimens. Opioid dosages should be continued to ensure adequate pain relief.
V. NON–PAIN-SYMPTOM CONTROL
To convey the complexity and richness of non–pain-symptom control, Table 1 lists a few of the major interventions that are regularly employed by palliative medicine practitioners. Details of evaluation and treatment are included in some of the texts in the Selected Reading list at the end of this chapter. In all cases, consideration of the underlying etiology, diagnostic testing, and potential specific treatments is appropriate, but purely symptomatic approaches make sense for many patients nearing death.
Table 1. Non–pain-symptom control
VI. COMMUNICATION WITH PATIENTS AND FAMILIES
Taking care of a patient with a terminal disease can be stressful for us as clinicians. We naturally feel distress when someone is suffering greatly from an incurable disease. Patients whom we particularly like or who are similar to ourselves or to our loved ones can make us particularly uncomfortable, reminding us of our mortality and that of persons close to us. Some clinicians believe that they or the medical system have failed dying patients, and so they feel ashamed or guilty. Others believe that since we cannot cure the illness, there is “nothing more to do.” Consequently, clinicians often unconsciously avoid spending time with dying patients or do not talk with them about emotionally charged matters. Yet many terminally ill persons wish to share their concerns about dying and find few people who will really listen. They turn to their physicians for help with their fears and concerns.
Talking with a dying patient is similar to talking with any patient. Confusional states are very common, particularly near the end of life, so the mental status should be assessed carefully. Rather than knowing what to say, communication primarily requires basic listening skills.
Introduce yourself. Patients meet many clinicians in the hospital. They cannot always read nametags or remember names. The clinician should always introduce herself and explain her role in the patient’s care.
Sit down for a discussion of any length. Studies show that patients feel the seated clinician has given them more attention and has been present for a longer time. Try to sit at roughly the same eye level as the patient.
Body language conveys interest in what is being said. For example, lean towards the speaker, make good eye contact, and nod to show you are listening.
Physical contact can be positive if the clinician and patient feel comfortable with it. A light touch on the arm or shoulder for emphasis or to comfort the patient or family member can help form an important connection.
Avoid jargon. Use words and expressions that the patient knows. Explain complicated matters in clear, simple language. The clinician is responsible for being sure that the patient understands the substance of the conversation.
Open-ended questions allow patients to tell their stories and to feel that they are being heard. More directed and closed-ended questions are useful for clarification after the patient has laid out a broad response to open-ended questions. Clinicians tend to cut off the patient early and concentrate on topics they consider important, thus missing what the patient wants to convey.
Avoid interruptions and tolerate silence. Asking another question before the patient has finished answering a previous one can give the impression that you are not interested in the answer. Let the patient tell his or her story at his or her own pace. Avoid the normal tendency to fill up silences with your own words or new questions. Tolerating silence can be especially difficult for busy clinicians who generally prefer quick, fact-oriented history-taking. But silence encourages the patient to lead a conversation, and can also convey tolerance of difficult feelings.
Listen for and respond to the affect. Patients quickly learn from their medical interviewers whether feelings are supposed to be part of an interview. History-taking should not be just about “the facts.” Acknowledge affect (“So that really upset you” or “I can see how troubling that must have been”) and encourage its expression (“How did that make you feel?”).
Facilitate conversation and support the patient. Statements of encouragement (“Yes, good, go on” or “Say more”) or nonverbal encouragement through head-nodding also help patients feel attended to and encourage them to deepen the conversation.
Encourage the patient to talk with clinicians, family, and friends. Many dying persons do not talk unless encouraged, feeling that others do not wish to listen. Families often have a hard time talking about death and may need gentle encouragement. Giving permission to both family and patient to discuss the dying process can be among the most important interventions a clinician makes.
VII. THREE KEY QUESTIONS FOR PATIENTS FACING A TERMINAL ILLNESS
If you become temporarily or even permanently unable to make decisions for yourself—perhaps from a stroke or because of medications—whom would you like to make decisions on your behalf? Does this person know you want him or her to take responsibility? Does he or she accept the role? Have you created a formal healthcare proxy document?
Have you discussed the preferences you would have if various situations arose and if decisions need to be made about using cardiopulmonary resuscitation or artificial ventilation or similar lifesustaining procedures? I strongly advise you to review advanced care planning documents with your proxy and family, and to consider writing out a living will or similar document.
How much information about your illness would you like to have? Do you want the frank truth or do you prefer to be shielded from distressing information? Would you like to be told everything first or would you prefer that we talk with someone else of your choosing who would decide what to share with you?
Strong pain medicines can sometimes make people sleepy or confused. In general, we can control pain without causing serious mental clouding. However, in some circumstances, pain treatment can interfere with alertness. Given the choice, would you rather have some pain and a clear mind, or would you rather have no pain even if that required enough medicine to make you sleepy or confused?
VIII. FAMILY MEETINGS
Meetings between family members and health professionals, either with or without the patient, provide an important means for sharing key information, assessing family coping, and providing family support. The communication strategies described earlier are also important when talking with families. Table 2 lists guidelines for breaking bad news.
Table 2. Guidelines for breaking bad news
Begin a meeting by making sure that everyone knows everyone else present. State the general purpose of the meeting such as “to help determine how Mr. Smith would wish to be taken care of now.” Next, find out what the family knows by asking them their understanding of the patient’s current medical condition and how they think he is doing. Some of the health professionals may then give their own views on how the patient is doing. Consider inviting a nurse or physical therapist who has had close contact with the patient. Clear up any important misconceptions. Ask the family to express their concerns. Listen for questions that have not been fully expressed, and speak about worries that are not being said out loud. If possible, do not ask that any important decisions be made immediately after bad news is delivered.
At the end of the meeting, quickly review the discussion. If some decisions will have to be made soon, make that explicit. Also, plan how to be in contact again. A family spokesperson is helpful for disseminating information in larger families.
IX. EUTHANASIA AND PHYSICIAN-ASSISTED SUICIDE
“Doctor, I do not want to live anymore. Can you help me?” Physicians usually dread hearing those words, and they often manage to cut off patients from expressing such common sentiments near the end of life. Society and physicians are divided about whether a physician may, in certain circumstances, be the agent of a patient’s death. Regardless of how one feels about helping a patient take his own life, clinicians should foster discussions about hastening death and even ask patients directly if they are entertaining such thoughts.
Once the topic is broached, further inquiry is productive. What are the factors that make life intolerable? What sort of relief for physical and emotional suffering might make life bearable or even worthwhile? Is the patient troubled by current treatable pain or symptoms, or perhaps frightened by the possibility of future, accelerating symptoms? Can we treat a depression or a delirium? Is dependency on others intolerable? Are there worries about burdening the family with physical or emotional demands of personal care or its financial implications? Sit with the patient and try to explore the fear and sadness that led to this call for help. How can we maximize the quality of the life that is left to the patient? What does the patient value that may lead to a wish to live? In many instances, psychiatric consultation can be helpful.
Once the clinician explores and responds to these issues, few patients persistently request for death to be hastened. Data from Oregon and the Netherlands, where physician-assisted suicide is legalized, suggest that difficulty tolerating dependency and lack of control is the major reason that patients seek to hasten death. A very small portion (around 0.1% of dying patients) request and are offered assisted suicide and about half of them actually commit suicide.
In the context of caring for terminally ill patients, pain physicians are called upon to offer expert advice about pain management, possibly but rarely involving interventional therapy. The guidance of the palliative care team is indispensable for formulating reasonable treatment plans that fit the patients’ needs as well as those of families and other caregivers, often within the constraints of caring for patients at home, with or without the aid of hospice.
Abrahm JL. A physician’s guide to pain and symptom management in cancer patients. Baltimore: Johns Hopkins University Press, 2000.
Berger A, Portenoy RK, Weissman DE. Principles and practice of supportive oncology. Philadelphia: Lippincott-Raven, 1997.
Block SD for the ACP-ASIM End-of-Life Care Consensus Panel. Assessing and managing depression in the terminally ill patient. Ann Intern Med 2000;132:209–218.
Block SD, Billings JA. Patient requests to hasten death: Evaluation and management in terminal care. Arch Intern Med 1994; 154:2039–2047.
Cassell ES. The nature of suffering and the goals of medicine. N Engl J Med 1982;306:639–645.
Cassem EH. The person confronting death. In: Nicholi AM Jr. The new Harvard guide to psychiatry. Cambridge, MA: Harvard University Press, Belknap Press, 1998:728–758.
Doyle D, Hanks GWC, MacDonald N, eds. Oxford textbook of palliative medicine, 2nd ed. New York: Oxford University Press, 1998.
Farr WC. The use of corticosteroids for symptom management in terminally ill patients. Am J Hospice Care 1990;7:41–46.
MacDonald N, Boisvert M, Dudgeon D, Hagen N, eds. Palliative medicine: A case-based manual. Oxford: Oxford University Press, 1997.
Waller A, Caroline NL. Handbook of palliative care in cancer. Boston: Butterworth-Heinemann, 2000.
Wanzer SH, Federman DD, Adelstein SJ, et al. The physician’s responsibility toward hopelessly ill patients: A second look. N Engl J Med 1989;320:844–849.
Woodruff R. Palliative medicine: Symptomatic and supportive care for patients with advanced cancer and AIDS, 3rd ed. Melbourne: Oxford University Press, 1999.