General Guidelines

A guiding philosophy of medicine is that the health and well being of the individual patient is of paramount consideration. In keeping with this philosophy, health care providers have a legitimate moral and legal presumption in favor of preserving life and providing beneficial medical care with the patient's informed consent. Clearly, however, avoiding death should not always be the preeminent goal. Not all technologically possible means of prolonging life need be or should be used in every case. For the gravely ill patient and for his or her family, friends, and health care providers, decisions about the use of life-sustaining treatment have profound consequences. These decisions will, to some extent, hasten or forestall the time of death. They will also shape the patient's experience of his or her remaining life --where it is lived, with whom, and with what degree of comfort or suffering. This document deals with the moral and ethical aspects of withdrawing or withholding life-sustaining therapy when the patient does not desire the treatment or when continuing to treat is equivalent to prolonging the dying process rather than sustaining or preserving meaningful life.

Life-Sustaining Treatment

Life-sustaining treatment is any medical intervention, technology, procedure, or medication that forestalls the moment of death, whether or not the treatment affects the underlying life-threatening diseases or biological processes. Examples include mechanical ventilation, dialysis, cardiopulmonary resuscitation (CPR), antibiotics, transfusions, nutrition, and hydration. Discussions about forgoing life-sustaining treatment will often be raised when death is the predictable or unavoidable outcome of the patient's underlying medical condition. However, a patient need not be terminally ill or imminently dying for these discussions to be held.


The process of making treatment decisions is as important to the ethical quality of those decisions as the decision themselves. Effective communication among physicians, patients or surrogates, families, and the treatment team enhances trust and increases the likelihood of consensus in ethical decision-making. In a tertiary care and referral center, the importance of developing trusting relationships quickly is even greater because patient-physician relationships are generally short-term and other physicians often participate in the patient's care.

Physicians should inquire about the patient's values regarding quality of life and the patient's beliefs about postponing death. Physicians should avoid imposing their values on the patient. When physicians communicate their personal beliefs and values to a patient or surrogate, they should identify them as such.

Physicians should provide information that is relevant to treatment goals and decisions. The information should be communicated in language that a patient or surrogate can understand. When several staff persons provide information, they should consult one another and be consistent in their communications to the patient or surrogate.

Early in the process of treating a patient with a poor prognosis for whom the issue of forgoing life-sustaining or death-prolonging therapies might arise, the physician and patient or surrogate should reach a consensus on the goals to be achieved and the treatment(s) to be used to achieve those goals. When existing goals and treatment(s) are no longer achievable. A consensus should be redeveloped around new goals and treatment(s).

When death is likely to occur soon, the physician should inform the patient in advance (or a surrogate, if available, and when the patient lacks decision-making capacity). When possible, the physician should present the prognosis and recommendations before a life-threatening crisis occurs and before the patient loses decision-making capacity. When the physician determines that only comfort measures will benefit the patient, the physician should discuss with the patient or surrogate a recommendation to forgo all treatment except comfort measures. The physician should also convey the reasons for the prognosis and the rationale for the recommendation. The physician should present the information in a way that assures the patient or surrogate that every medically appropriate treatment option has been or will be tried, and tried long enough to know whether benefit would derive from it. The physician should make clear that forgoing treatment does not mean abandonment and should describe the comfort measures that are and will be given.

The physician should consider the extent of the patient's or surrogate's need for involvement in decision-making. Some persons want to exercise extensive controls in making decisions; others prefer to share decision-making; others prefer to have the physician make decisions for them. In the last case, a physician is ethically justified in taking a paternalistic role in decision-making. The physician should also be aware of persons in the patient's family who exert influence, and not assume the patient or the "next of kin" will be the one to exercise the exclusive role in decision-making. Patient wishes, expressed in advance directives or clearly communicated to the physician, family, or both, should be honored and protected from others' attempts to override them.

The physician should avoid language or expressions that could provoke reactions without considered judgment (such as, "Would you want us to do everything possible to save your life?") or place undue burden for decision making on the patient or surrogate (such as asking for a decision without having made recommendations, "What do you want us to do?")

The physician should notify other members of the health care team about the results of discussions with the patient or surrogate. The physician should make referrals for the emotional and spiritual support of the patient, family and/or surrogate.

Extraordinary versus Ordinary Treatment

The terms "extraordinary" and "ordinary" are often used in an attempt to distinguish a class of treatments that may be ethically withheld or withdrawn from a class of treatments that may not. Unfortunately, these terms are a source of great confusion. Sometimes extraordinary is distinguished from ordinary by appealing to the prevalence of a treatment or its level of technological complexity. This distinction is misleading because it focuses attention on factors that ethically are irrelevant to the decision to forgo treatment. Certain procedures for providing artificial nutrition and hydration, for example, are technologically rather complex, whereas administering chemotherapy is not. It clearly does not follow that the latter is ethically required because it is ordinary or technically simple, while the former is optional because it is extraordinary or technically complex. No treatment is intrinsically ordinary or extraordinary. More useful for decision-making is an assessment of the benefits and burdens for the patient that each procedure or treatment provides.

Withholding Versus Withdrawing Treatment


A great deal of confusion and anxiety surrounds the differences between withholding life-sustaining treatment, which many consider morally permissible, and withdrawing life-sustaining treatment that has already been instituted, which some consider morally wrong. Ethically, withholding a treatment does not differ from withdrawing a treatment. Treatment can be ethically withdrawn whenever it can ethically be withheld. However, there are psychological and sociological differences between withholding and withdrawing treatment. It is sometimes psychologically more difficult to withdraw treatment than to withhold it, for withdrawal may be perceived as violating a special commitment the health care professional has made to the patient. Likewise, society may find it less intrusive never to start a therapy than to withdraw a treatment, especially when the withdrawal may have more obvious consequences than those occurring if the treatment had never been started. However, from a medical standpoint, it is usually better to initiate a treatment provisionally with a plan for stopping it if the treatment proves ineffective or unduly burdensome to the patient than to withhold a treatment altogether for fear that stopping will be impossible. When it is unclear whether the burdens or benefits are greater, it is appropriate to choose on the side of life and provide the treatment. If a treatment is clearly futile (i.e., does not achieve its physiological objective, offers no benefit to the patient, and violates reasonable medical standards), there is no obligation to offer, begin, or maintain the treatment.

Autonomy, Obligation to Treat, and Medical Futility

Patients have a right to control what happens to their bodies, so the decision about whether to forgo life-sustaining treatment should, in the final analysis, is theirs. This IDEA is often referred to as the principle of self-determination or autonomy. In exercising their autonomy, patients at times evaluate the benefits and burdens of life-sustaining treatment differently than their physicians. Physicians, too, differ among themselves as to whether a treatment will benefit a specific patient. When physicians cannot agree or compromise, it may be appropriate to request another medical opinion, an ethics consultation, or the assistance of the Pastoral Care, Social Work or Psychiatry Departments. The physician has no obligation to render futile care and thereby violate reasonable medical standards. Rather, physicians should decide the extent of care in accordance with what they perceive is medically appropriate for that patient and inform the patient about their decisions.

There is a distinction between treatment a doctor believes to be detrimental to a patient's best interest, and treatment to which a physician has a conscientious objection. When the patient opts for a course of action that violates the health care professional's personal, ethical, or religious convictions, the professional should discuss the problem with the patient. In situations of unresolved conflict, the physician or the patient may initiate a transfer to another health care professional or health care facility. Continuity of the patient's care should be assured. If the physician decides to forgo a non-beneficial treatment, the patient should be informed.

Although patients do not have a right to insist on futile treatment, circumstances can arise in which providing life-sustaining futile treatment for a limited time is justifiable to achieve identified objectives, such as giving family members time to arrive at the health care facility before a patient's death, sustaining a conscious patient long enough to live for a special event or anniversary, or allowing time for reconciling relationships or for religious rituals.

Capacity and Surrogacy

Proper determination of capacity is crucial to an ethical decision-making process. Caregivers have a duty to respect the wishes of a patient with decision-making capacity. Decision-making capacity differs from competence. Competence and incompetence are legal terms and frequently refer to situations in which a formal judicial determination of mental status has been made. Decision-making capacity, on the other hand, refers to a patient's functional ability to make informed health care decisions in accordance with the patient's personal values. Decision-making capacity means the ability to comprehend information relevant to the decision, the ability to deliberate about the choices in accordance with personal values and goals, and the ability to communicate either verbally or non-verbally with the caregivers.

When a patient lacks decision-making capacity, and when attempts to restore that capacity fail (rectifying reversible causes such as over-medication, pain, or dehydration) or are not possible, then a surrogate may be identified who will be the ultimate source of consent or refusal to the health care team's plan or action. In identifying a surrogate, the physician should first honor any surrogate the patient has chosen in advance. A surrogate may be designated by a patient in a durable power of attorney for health care. A surrogate may also be identified as the first available person from the following list, in order: legal guardian, spouse, majority of adult children, parents, majority of adult siblings, or other nearest adult relative. The, primary function of the surrogate is to express those choices that the patient would express if he or she were able. Even when a surrogate is needed, any possible participation by the patient in the decision-making process is encouraged"

Legal guardians of a patient, whether natural parents or appointed by a court, have the legal right to make decisions on behalf of the patient. In the situation in which a patient does not have decision-making capacity and a legal guardian or surrogate does not exist, the physician should provide the course of treatment that the physician believes, in his or her best medical judgment, would be the course chosen by the patient. ;

Although parents are authorized by law to consent to and to refuse treatment for their minor children, the best interests of a patient who is a minor supersede a treatment decision requested by the parents (or legal guardian) when that decision conflicts with the best interests of the minor. Treatment decisions made by minors themselves should be given respectful consideration.

Advance directives can be useful in documenting a patient's desires regarding end-of-life treatment. The Cleveland Clinic firmly endorses the right of a patient with decision-making capacity or the patient's surrogate to forgo any life-sustaining medical treatment. Health care professionals in many ways must be advocates for life, even while they are willing to honor decisions to forgo life-sustaining therapy or to administer necessary treatment to alleviate pain that may at the same time hasten death. Health Care Professionals, however, should not participate in active euthanasia or assist in suicide.

In situations of unresolved conflict, the Department of Bioethics, the Ethics Committee, or the Office of General Counsel may also be consulted. Referral to a court of law should occur only as a last resort.

Specific Treatment Termination Guidelines

Intensive Care Unit Admission

The following types of patients are candidates for admission to intensive care units when it is consistent with their treatment preference and goals: critically ill patients who require life support for organ system failure that may be reversible; patients with irreversible organ system failure who cannot be treated appropriately in another setting; patients at risk of life-threatening complications who require monitoring or treatment; and patients who are receiving a trial period of monitoring or treatment when the prognosis or the effectiveness of therapy is in doubt. A decision to forgo some forms of life-sustaining treatment such as cardiopulmonary resuscitation should not automatically preclude other forms of treatment and admission to the ICU. Admissions should be subject to the constraints imposed by the availability of space, equipment, and personnel, the needs of the patients already in the unit, and the needs of others who are also candidates for admission. Patients who generally should not be admitted to the intensive care unit include: patients with documented irreversible cessation of all functions of the entire brain; patients who have been diagnosed as irreversibly unconscious; patients with irreversible illness who are near death; and patients who, while capable of making decisions, have requested that they not receive intensive care or its equivalent. Patients are entitled to refuse admission to an ICU, even when doing so puts them at risk of death. Patients should not, however, be able to demand admission to an ICU. A request by a patient or a surrogate for admission to an ICU may be denied if admission would be medically inappropriate for the patient, detrimental to patients already in the unit, or contrary to the admission criteria. These patients should be transferred from the ICU to another setting within the hospital or to another institution when intensive care will no longer benefit them, either because they have improved to a point where intensive care is no longer necessary, or because they have deteriorated to a point where it no longer offers reasonable promise of benefit. Such triage is ethically appropriate.

Cardiopulmonary Resuscitation

Cardiopulmonary resuscitation (CPR) refers to those measures used to restore ventilation and circulation in patients in whom these functions have been interrupted. Resuscitation techniques have no value in the management of irreversible or terminal disease states. They are intended to revive otherwise healthy individuals who experience some reversible catastrophe that interrupts breathing and circulation. Because of the emergency character of CPR, a patient or surrogate should ideally be consulted in advance about whether to begin resuscitation in the event of cardiac or respiratory arrest. Any patient who is at increased risk for cardiopulmonary arrest should be given the opportunity to make a decision about CPR while he or she is still capable of making the decision. In the absence of a Do Not Resuscitate (DNR) order, resuscitation should be attempted, and if any doubt exists as to whether a decision to forgo treatment has been properly made, treatment to preserve life should be given. Deceptive resuscitation efforts, known as "slow codes" and "walk, don't run" codes are not acceptable. Any code should be a full code unless a partial code or limited resuscitation effort has been consented to by an informed patient or surrogate. At the time of cardiac or respiratory arrest, if the physician summoned to direct resuscitation realizes that CPR cannot restore cardiac and respiratory function, the physician may call off the effort. If CPR is clearly futile (i.e., does not achieve its physiological objectives, offers no benefit to the patient, and violates reasonable medical standards), there is no obligation to offer or initiate CPR.

Mechanical Ventilation

It is important to emphasize again that treatment can ethically be withdrawn whenever it can be ethically withheld. It is not appropriate, however, to remove a ventilator-dependent patient from a ventilator without discussion with the patient or surrogate. - The responsible health care professional should not request other health care personnel to carry out a decision that he or she would not personally carry out. In the situation of a decision by a patient or a surrogate to forgo ventilation, it is ethically acceptable to sedate the patient if necessary to insure comfort. Supplemental oxygen can be used to relieve dyspnea from hypoxemia. If relieving the patient's dyspnea or other discomfort requires sedation to the point of unconsciousness, it is ethically acceptable to do so with the consent of the patient or surrogate. A patient may also decline to be weaned from a ventilator and may wish to be simply disconnected from mechanical ventilation as part of a fundamental decision to forgo any life-sustaining treatment. In such a case, it is permissible to disconnect the ventilator without weaning.


A vital part of the discussion of whether to forgo dialysis concerns the patient's transplantation options. To make an informed decision about whether to forgo dialysis, the patient or surrogate must receive an evaluation of whether the patient could receive a transplant and, if so, what the transplant possibilities are and what transplantation involves. Because dialysis is frequently supervised most directly by personnel other than the primary health care providers, it is important that all such personnel participate in the evaluation process. It is important to explore with patients already on dialysis why they wish to stop the treatment. It may be that their discomforts can be ameliorated without stopping the treatment entirely. Another important aspect of the discussion should be the question of where death will occur when the decision has been made to forgo dialysis. Often a patient will wish to die in the hospital, where supportive and palliative care is readily available. If the patient wishes to die at home, the health care professional should inform the patient and care givers of the risks and burdens. The patient's preference concerning the place of death should ordinarily prevail, as long as adequate care can be arranged. When dialysis treatment is clearly futile and of no benefit to the patient, there is no obligation to continue the therapy.


Among the treatments a patient may choose to forgo is the administration of blood and blood products. This refusal arises most frequently on religious grounds, usually asserted by a Jehovah's Witness. An individual's freedom to act in accord with personal religious values is one aspect of autonomy, and the right of Jehovah's Witnesses to refuse blood should be recognized. However, as with non-religious aspects of autonomy, the right of self-determination is not absolute. For example, the right to forgo treatment may sometimes be restricted on the grounds that it will cause harm to specific others. These exceptions include parents making decisions for a child, or when the patient is pregnant or has dependent children. The decision-making process for refusal of blood and blood products may occur when serious bleeding is expected but has not yet started, when such bleeding occurs if there is time to go through the entire decision-making process, or when treatment for bleeding has started and the question is whether to continue. If medically indicated, patients should receive treatment for bleeding in an emergency except when the patient, while capable of making decisions, has given direction refusing blood and blood products.

Antibiotics and Other Medications

Some patients who are terminally ill or in a severely debilitated and irreversible condition may determine that treatment with antibiotics or other medications will only prolong their pain and suffering. Decisions about using antibiotics and other medications, similar to decisions about other forms of life-sustaining treatment, require patients or surrogates to balance carefully the potential burdens against the benefits. Respecting the considered choice of the patient or the patient's surrogate to forgo life-sustaining medication does not violate the ethical mandates of the health care professionals. Only in cases where it is necessary to override a patient's refusal of antibiotics or other life-sustaining medications for public health reasons should a patient's wishes not be upheld.

Nutrition and Hydration

Medical procedures for supplying nutrition and hydration treat malnutrition and dehydration; they mayor may not relieve the hunger and thirst that can occur, Conversely, hunger and thirst can be treated without necessarily using medical nutrition and hydration techniques. For instance, dehydrated patients may have their thirst relieved by having their lips and mouth moistened with ice chips or lubricants. Patients in their last days before death may spontaneously reduce their intake without experiencing hunger or thirst. Indeed, clinical experience indicates that dehydration may offer benefits for certain dying patients. Dehydration can reduce secretions and excretions, thus decreasing breathing problems, vomiting, and incontinence. Dehydration can also produce a sedative effect on the brain, making the dying process more tolerable. Forgoing nutrition and hydration is one of the most difficult treatment-termination decisions because of the association of nutrition and hydration with basic needs and human caring. Individual cases should be decided by balancing these basic human needs and their potential benefits with the burden to the individual of the technology needed to provide artificial nutrition and intravenous hydration. The artificial provision of nutrition and hydration is a form of medical treatment and may be forgone when requested by a patient or a surrogate. Provision of nutrition and hydration may be technologically possible but not an ethically mandatory means of prolonging life. Like other life supports, decisions about starting or stopping nutrition and hydration require carefully balancing the potential burdens against the benefits.

Pain Relief

Pain relief is an extremely important aspect of providing humane care to dying individuals. Although the majority of dying patients do not feel substantial pain, most fear the possibility of pain --perhaps more than death itself. Because the primary goal of caring for dying patients is to relieve pain and suffering, unless the patient chooses otherwise, measures involving substantial risk may be considered, although they might not be undertaken to relieve the discomfort of patients with a reasonable chance of survival. Examples of such measures include percutaneous cordotomy or neurolytic blocks.

The proper and adequate use of analgesics, especially narcotics, is critically important to alleviate pain for patients who are dying. Concerns about addiction or physical dependence are irrelevant to the dying patient. Likewise, psychological dependence on narcotics is most often the result of under-medicating rather than appropriately medicating. Patients are less likely to become psychologically dependent when narcotic agents are given on a prophylactic schedule to prevent pain, rather than in response to request after pain is experienced. The health care professional should ordinarily seek to give sufficient medication to relieve pain while enabling the patient to remain as mentally alert as the patient wishes. The continuous intravenous infusion of narcotics is appropriate therapy to alleviate pain and suffering in a dying patient, even to the point of unconsciousness, with the consent of the patient or surrogate, and even though alleviation of the pain and suffering may hasten death.


Back to Ethics Information for Patients

© 2005 The Cleveland Clinic Department of Bioethics | Contact Us | Privacy | Disclaimer | Last modified: 08/16/2005