Chapter 17. Withholding or Withdrawing of Life-Sustaining Therapy and Administering of Palliative Care

• Many, if not most, patients who die in intensive care units do so during the withholding or withdrawing of life-sustaining therapy and the administering of palliative care.
• Life-sustaining therapy usually is withheld or withdrawn and palliative care is administered because physicians and patients or their surrogates agree that further restorative care would not be beneficial.
• Cardiopulmonary resuscitation, mechanical ventilation, and vasoactive drugs are the therapies most commonly withheld or withdrawn.
• Palliative care involves attention to the physical, emotional, and spiritual needs of patients and their families.
• Withholding or withdrawing of life-sustaining therapy and administering of palliative care are well supported by ethical pronouncements and case law.

Withholding or withdrawing of life-sustaining therapy or life support is a process by which various medical interventions either are not given to or are taken away from patients with the expectation that they will die from their underlying illnesses. This process is carried out in many medical settings but is especially common in the ICU, which offers an array of therapies capable of sustaining life. An example of withholding life-sustaining therapy is to not provide mechanical ventilation to a patient with chronic obstructive pulmonary disease (COPD) and acute respiratory failure who probably will die without the ventilator but is thought to be unweanable once mechanical ventilation is begun. An example of withdrawing life-sustaining therapy is the removal of mechanical ventilation from a patient with COPD with the provision that he or she will neither be ventilated again if acute respiratory failure recurs nor receive cardiopulmonary resuscitation (CPR) in the event of cardiopulmonary arrest. This second patient is different from a patient with COPD who is being weaned from mechanical ventilation and will be ventilated again or resuscitated if he or she deteriorates during the weaning process.

Strictly speaking, all patients who die while receiving close medical attention in an ICU or elsewhere do so as a direct result of the withholding or withdrawal of life-sustaining therapy. This is either because a decision has been made not to resuscitate the patient in advance of decompensation or because even vigorous resuscitation will not be provided indefinitely. For example, the second patient described earlier would receive CPR if he or she were to suffer an unexpected cardiopulmonary arrest during weaning, but CPR would be discontinued if it were not of benefit in restoring a viable cardiac rhythm in, say, an hour. The withdrawal of life support in this second patient would be comparable to the withholding of CPR in the first patient but for the fact that deliberate planning was more possible for the first patient before cardiopulmonary arrest. It is this more deliberate form of withholding and withdrawing of life-sustaining therapy from critically ill adults that is focused on in this chapter, in which the ethical, legal, clinical, and practical aspects of limiting care are discussed.

Note that the only patients who die in the ICU during the withholding or withdrawing of life-sustaining therapy who are not expected to die are those who are dead already owing to complete and irreversible loss of function of the cerebral hemisphere and brain stem. These patients frequently receive mechanical ventilation and other forms of life support either because the diagnosis of brain death has not been confirmed by apnea testing (after which the ventilator will be removed) or because vital organs (excluding the central nervous system) are being preserved prior to transplantation. Indeed, if transplantation is accomplished, the withholding or withdrawing of mechanical ventilation and other therapies usually takes place in the operating room, where the patient's organs are harvested. Because in brain-dead patients interventions such as mechanical ventilation are removed after death, these interventions are better thought of as organ support than as life support. Nevertheless, because patients and surrogates may not appreciate this distinction, and because the decisions and practical steps preceding the removal of interventions are similar in brain-dead and non-brain-dead patients, mention is made of brain-dead patients in this chapter.

Palliative care may be described as the prevention or treatment of pain, dyspnea, anxiety, and other kinds of suffering in dying patients. Palliative care represents an attitude in addition to a series of actions that ideally bring symptomatic relief and comfort as patients die. This attitude includes an emphasis on the family as well as the patient as the focus of treatment, a commitment to clear and continuous communication between the critical care team and these parties, and a concern for the emotional and spiritual, as well as the physical, well-being of patients and families. As in the withholding or withdrawing of life-sustaining treatment, the administering of palliative care is prompted by the appreciation that further restorative treatment would no longer be beneficial to patients who are dying.

Physicians who care for critically ill adults and are considering withholding or withdrawing life-sustaining therapy care can draw from several summaries of the ethical propriety of limiting care.1–5 One of the most succinct group statements is the “Consensus Report on the Ethics of Foregoing Life-Sustaining Treatments in the Critically Ill,” which was published by the Task Force on Ethics of the Society of Critical Care Medicine in 1990.6 This report noted that it is ethically appropriate to withhold or withdraw therapy either because a patient or surrogate decides to forego treatment or because a physician judges that further treatment is not likely to be beneficial. A decision to withdraw a treatment should not be more problematic than a decision not to initiate a treatment, particularly because starting a therapy may be necessary to adequately evaluate a patient's condition. Any treatment derives its medical justification from the benefits that a patient and physician hope to achieve by employing it. When the treatment has achieved those benefits or can no longer be expected to do so, it loses its justification and may be withdrawn.

The Society of Critical Care Medicine Task Force consensus statement also noted that there are no intrinsic moral differences between categories of treatment, such as CPR, ventilatory support, medications such as vasopressors and antibiotics, and the provision of hydration and nutrition by artificial means. Each treatment should be considered from the patient's perspective in light of the overall benefit that it may offer and the burdens it may entail, as well as the professional duties that are involved. Treatments that serve only to prolong the dying process should not be employed, and the indefinite maintenance of patients in a persistent vegetative state raises ethical concerns both for the dignity of the patient and for the appropriate use of health care resources.

A basic provision of the consensus statement was that the wishes of an informed adult patient with decision-making capacity should be the primary consideration in almost all decisions regarding treatment. When the patient (or surrogate in the case of a patient who is incapable of making decisions) and the physician (and other members of the critical care team) agree that therapy should be limited, it usually should be. When the patient or surrogate requests therapy that the physician considers nonbeneficial, the physician should clarify the goals of treatment with the patient or surrogate and may accede to their wishes. However, the physician is not ethically obligated to provide therapy and may elect to transfer the patient's care.

In 2001, the Ethics Committee of the Society of Critical Care Medicine published a statement entitled “Recommendations for End-of-Life Care in the Intensive Care Unit” that explored the ethical aspects of administering palliative care.7 The recommendations were based on recognition of the needs of patients, families, and providers during the dying process. The patients' needs are to receive adequate pain management, to avoid inappropriate prolongation of dying, to achieve a sense of control, to relieve themselves of burdens, and to strengthen their relationships with loved ones. The families' needs are to be with dying persons, to be helpful to them, to be informed of their changing condition, to understand what is being done to them and why, to be assured of the patients' comfort, to be comforted themselves, to express their own emotions, to be assured that their decisions regarding the patients were correct, to find meaning in the dying of their loved ones, and to be fed and rested. The needs of the critical care team are to establish consensus regarding the goals and strategies for providing palliative care, to gain knowledge and skills in palliation, to be supported in their tasks by their institutions, and to have opportunities for grieving and bereavement after patients die. In addition to articulating these needs, the recent statement by the Society of Critical Care Medicine offers a number of helpful recommendations in delivering end-of-life care.

In the United States, withholding and withdrawing life support are legally justified primarily by the principles of informed consent and informed refusal. Both principles have strong roots in the common law and reflect the American regard for self-determination. The principles hold that treatment may not be initiated without the approval of patients or their surrogates, except in emergency situations, and that patients or surrogates may refuse any and all therapies.8

The right of competent patients to refuse therapy was established directly in Bartling v. Superior Court,9 in which a California Court of Appeal held that a patient could be removed from a ventilator at his request over the objections of his physicians and the hospital where he was cared for. In Bouvia v. Superior Court,10 the California Court of Appeal held that a conscious and competent hospitalized patient with severe pain could refuse nutrition and hydration even if by doing so she believed she would die. Currently, all American courts recognize a competent adult's right to refuse life-sustaining treatment, although the constitutional, statutory, and common-law bases of the right may vary by jurisdiction. Most of the cases concerning that right have arisen, however, from situations involving patients who cannot make medical decisions for themselves.

The application of the principles of informed consent and informed refusal to the care of decisionally impaired critically ill patients began with In re Quinlan.11 In this case, the Supreme Court of New Jersey held that a patient had the right to refuse mechanical ventilation and that, because she was vegetative and could not exercise that right directly, her parents could act as surrogates and make a “substituted judgment” for her. The California Court of Appeal took a similar approach in Barber v. Superior Court,12 in which it held that physicians charged with murder had not committed an unlawful act when, with permission from a patient's family, they removed nutrition and hydration from a comatose patient. In Barber, the court noted that medical interventions ranging from mechanical ventilation to nutrition and hydration should not be distinguished by terms such as extraordinary and ordinary that had been used previously. Rather, the court stated, medical interventions should be evaluated by the benefits and burdens they offer patients; if the burdens of a particular treatment outweigh the benefits it confers, the treatment can be foregone.

Although the principles of informed consent and refusal articulated in Quinlan, Barber, and other cases are well established, states vary in their processes and rules for vindicating these principles for decisionally impaired patients. In general, these variations fall into two categories: rules for selecting surrogate decision makers and rules determining what surrogates can decide. As for the latter rules, all states agree that medical decisions should be guided by the wishes of the patient whenever possible. Some states require a high level of certainty before a surrogate may refuse life-sustaining treatment. For example, in the O'Conner case,13 a New York court ruled that treatment is obligatory unless evidence shows that it is almost certain that the patient would have refused treatment. Missouri also requires “clear and convincing evidence” of a decisionally impaired patient's wishes, a position that was contested in Cruzan v. Director, Missouri Department of Health,14 as will be discussed.

In most other states, however, surrogates may consider other factors, such as the patient's “best interests,” when a patient has not expressed a prior unequivocal preference. The most influential case to describe the best interests standard is the New Jersey Supreme Court's decision in In re Conroy,15 which was made 9 years after that court decided Quinlan. The Conroy court would apply the best interests standard only when there was no reliable evidence of the patient's wishes. The test of this standard has two elements. First, the burdens of the patient's life with treatment must outweigh the benefits to the patient of continued life. Second, the recurring, unavoidable pain of the patient's life with treatment must be such that administering the treatment is inhumane.

The issue of withholding or withdrawing life-sustaining treatment was first addressed by the U. S. Supreme Court in the aforementioned Cruzan case,14 which involved a request by parents to have a feeding tube removed from their daughter in a Missouri nursing home. In its decision, the Court allowed Missouri and other states to require “clear and convincing evidence” of patients' wishes and thereby potentially limited the role of surrogates in making decisions for decisionally impaired patients without advance directives. Nevertheless, the Court accepted the principle that a competent person's right to forego treatment, including nutrition and hydration, is a liberty interest protected under the Fourteenth Amendment to the Constitution. Thus, although Cruzan did not set a single standard for surrogate decision making for patients who cannot make decisions for themselves, it did affirm the legal principles of informed consent and refusal that are used in all the United States.

After Cruzan, the Supreme Court confirmed its approval of the foregoing of life-sustaining treatment in Washington v. Glucksberg16 and Vacco v. Quill17 and also furnished guidelines for administering palliative care. These last two cases dealt with the constitutionality of laws prohibiting physician-assisted suicide in Washington and New York. In Glucksberg, the court decided that terminally ill patients did not have a liberty interest in committing suicide or in receiving physician assistance in suicide because of both the long tradition of prohibiting suicide in the United States and the states' legitimate reasons for continuing to make suicide illegal. In Vacco, the court drew further distinctions between assisted suicide and withholding or withdrawing life-sustaining therapy. “Everyone, regardless of physical condition, is entitled, if competent, to refuse lifesaving medical treatment; no one is permitted to assist a suicide,” the court wrote. “When a patient refuses life-sustaining medical treatment, he dies from an underlying fatal disease or pathology; but if a patient ingests lethal medication prescribed by a physician, he is killed by that medication.”

In Glucksberg and Vacco, five justices of the Supreme Court reasoned in concurring opinions that Washington and New York could prohibit assisted suicide because these states had no legal barriers that prevented patients from obtaining medications to relieve pain and suffering and therefore had no need for assisted suicide. However, as Justice Breyer wrote in his opinion concurring in the judgment in the two cases, “Were state law to prevent the provision of palliative care, including the administration of drugs as needed to avoid pain at the end of life, an action against such law might be called for by the Supreme Court.” Through this and other statements, a majority of Supreme Court justices suggested that the Constitution's guarantee of individual liberty included the liberty to be free of unnecessary pain during the dying process.18 Indeed, the justices' opinions have been interpreted as indicating a willingness on the part of the Supreme Court to mandate patient access to palliative care.19

The Supreme Court distinguished assisted suicide from palliative care in Glucksberg and Vacco by accepting the principle of double effect. Under this principle, acts such as giving sedatives and analgesics that lead to morally good effects, such as the relief of pain, are permissible even if they produce morally bad effects, such as the hastening of death, provided that only the good effect is intended.20 The Supreme Court's approval of palliative care included sanctioning the practice of terminal sedation, in which patients are rendered comatose and then may have nutrition and hydration withdrawn. The term terminal sedation appears only in a footnote concerning sedation while withdrawing ventilatory support or other treatment in the Vacco majority opinion. Nevertheless, under Vacco, a state may allow terminal sedation if it is “based on informed consent and the double effect. Just as a state may prohibit assisting suicide while permitting patients to refuse unwanted lifesaving treatment, it may permit palliative care related to that refusal, which may have the foreseen but unintended ‘double effect’ of hastening the patient's death.”

Information about the clinical aspects of withholding or withdrawing life support has come primarily from surveys of physicians and other health professionals and from observational studies of the limiting of life-sustaining care. For example, several surveys21–23 have suggested that the great majority of critical care physicians have withheld or withdrawn life support at one time or another and that many have done so frequently. Similarly, retrospective studies24–26 have revealed that approximately one-half of the patients who die in ICUs do so during the withholding or withdrawal of life-sustaining therapy. Prospective studies27,28 conducted in two ICUs in hospitals affiliated with the University of California, San Francisco, revealed that withholding or withdrawal of life support preceded 90% of deaths in 1992 and 1993 compared with 51% of deaths in 1988 and 1989. To determine whether limiting care is as commonplace as suggested by the aforementioned studies, a prospective survey29 was conducted in 1994 and 1995 of 107 ICUs associated with all U.S. training programs in critical care medicine or pulmonary and critical care medicine. It revealed that only 25% of deaths followed full ICU care and failed CPR.

Similar observations have been reported recently concerning the practice of critical care medicine in Europe.29a In a prospective observational study in 37 ICUs in 17 European countries from 1999–2000, of the patients who died, 72.6% had a limitation of therapy, although substantial intercountry variability in practice was noted. Limitation of therapy versus continuation of life-sustaining therapy was associated with patient age, acute and chronic diagnoses, number of days in the ICU, region, and religion.

Most surveys21–23,30 and studies24–28,30,31 have indicated that physicians recommend that life support be withheld or withdrawn either because patients or surrogates request this action or because physicians believe that further care is not beneficial. This belief stems from prognostication on the physicians' part. They generally arrive at estimates of prognosis through their personal experience and knowledge of the medical literature and not with the help of prognostic scoring systems such as the Acute Physiology and Chronic Health Evaluation (APACHE) system, although use of such systems is increasing in some ICUs. If present, brain death, persistent coma, other unacceptable neurologic prognoses, and multiple-organ-system failure are important rationales for care withdrawal. The need for ICU beds, patients' social worth, and financial cost-benefit analyses are not important rationales.

Most patients have not prepared medical directives in advance of suffering a critical illness, and most patients are too neurologically depressed by their disease or by therapeutic drugs to make medical decisions when they become critically ill. Surrogates usually were available and willing to take part in the decision-making process in the prospective studies27,28 of withholding or withdrawing life support. When surrogates were not available, physicians were willing to make decisions for the patients. Although consultation from other physicians often was sought before deciding to limit care, hospital ethics committees were involved infrequently. Furthermore, probate courts or judicial hearings were never used.

The surrogates who were available in these studies27,28 either recommended by themselves that support be withheld or withdrawn when patients were not improving or accepted physician recommendations to that effect. These recommendations were not always accepted immediately, but in most cases the surrogates agreed within a few days. Only rarely did surrogates insist on continued care against the advice of physicians, and in all cases care was continued. In certain instances, however, the physicians stopped short of providing CPR in patients who were otherwise supported.

Although differences between physicians and surrogates do occur over the issue of limiting care, their incidence is uncertain. In one survey22 of self-identified members of the critical care section of the American Thoracic Society, 34% of respondents reported that they had refused surrogate requests to withdraw care either because they believed that the patient still had a reasonable chance of recovery or because they believed that the surrogates might not be acting in the best interest of the patient. At the same time, 83% of physicians reported that they had unilaterally withheld life-sustaining treatment on the basis of futility, often without patient or surrogate knowledge or consent, and 82% had withdrawn treatment on the same basis.

CPR probably is the therapy most often withheld from ICU patients, as suggested by the finding that do-not-resuscitate (DNR) orders preceded 60% of all in-unit deaths in a large observational study32 in 40 U.S. hospitals. Antibiotics, vasoactive drugs, renal dialysis, and mechanical ventilation were withheld commonly in the two studies27,28 from ICUs in hospitals affiliated with the University of California, San Francisco. Mechanical ventilation was the therapy most commonly withdrawn in these studies, followed by vasoactive drugs. In the survey22 of American Thoracic Society members, 89% of the respondents had withdrawn mechanical ventilation, 88% had withdrawn vasoactive drugs, and 80% had withdrawn blood or blood products.

In a study33 conducted in the two ICUs at hospitals affiliated with the University of California, San Francisco, analgesics and sedatives were given to 75% of non–brain-dead patients during the withholding or withdrawal of life support. Patients who did not receive medication were comatose and considered incapable of benefiting from the drugs. Physicians ordered analgesics and sedatives to decrease pain in 88% of patients, to decrease anxiety in 85%, to comfort families in 82%, to decrease air hunger in 76%, and to hasten death in 39%; in no instance was hastening death the only reason cited. The amounts of opiates and benzodiazepines averaged 3.3 mg/h of morphine and 2.2 mg/h of diazepam in the 24 hours before withholding or withdrawing of life support and 11.2 and 9.8 mg/h, respectively, in the 24 hours thereafter. The median time until death following the initiation of withholding or withdrawal of life support was 3.5 hours in the patients who received drugs and 1.3 hours in patients who did not. Thus the administration of sedatives and analgesics did not appear to hasten death in this study.

It would appear from these surveys and studies that withholding or withdrawal of life-sustaining therapy from adults occurs frequently in most U.S. ICUs and that these processes commonly follow a similar course. Furthermore, professional attitudes and behaviors on the part of physicians are consistent with ethical pronouncements and judicial decisions regarding the withholding or withdrawal of life support. The only major area in which consistency may be lacking concerns physician refusal to follow surrogate demands either to limit care in certain circumstances or to provide care that the physicians consider nonbeneficial. This finding is not surprising given the lack of ethical and legal clarification of the issue; hopefully, such clarification will occur over time.

The surveys and studies discussed in the preceding section suggest that physicians, patients, and surrogates generally reach consensus regarding when and how withholding or withdrawal of life support should take place. Consensus is not reached in all cases, at least not initially, and the process of foregoing life-sustaining treatment can be quite strained. Most of the differences that exist between the parties involved in this process result from difficulties in communication and can be breached if communication is improved. There is much interest in the critical care community for enhancing the process of withholding or withdrawing care and providing palliative care, as well as enhancing communication generally in the ICU. Models for improved communication have been proposed and at least preliminarily tested.34

The following 12 suggestions are designed to facilitate communication inside and outside the ICU34a:

1. Regular meetings should be held with physicians and patients or their surrogates from the time of ICU admission to discuss all medical matters, including prognosis. The meetings should be held in a quiet, comfortable, and private place conducive to conversation. The timing of the meetings should allow the participation of the attending physician, the primary physician (if different from the attending), consultants, bedside nurses, and facilitators such as family counselors or clergy. House staff may be included, but a physician in training should not serve as a substitute for an experienced attending in directing the course of the meetings.

2. Consensus should be reached by the physicians regarding prognosis and other medical matters before they communicate with the patients or their surrogates. Although physicians may disagree privately among themselves, they should be prepared to present an opinion to the family that they can publicly agree on. This includes neurologists or neurosurgeons who may be asked to consult regarding functional outcome. Ideally, the viewpoints of all physicians can be summed up and communicated by the attending physician to patients or surrogates. Nurses and other caregivers should be included in the consensus-building process.

3. The role of surrogates in decision making should be clarified from the outset of ICU admission.35 Surrogates should be informed that they can help the patient best by articulating his or her wishes and determining what he or she would want done in a given situation instead of expressing the surrogates' wishes and what the surrogates themselves would want done. If a certain family member carries a durable power of attorney for the patient or only one family member is available, this individual will act for the family as a whole. Otherwise, and especially if the family is divided in opinion, it should be asked to designate a spokesperson.

4. Discussions with surrogates should be frank, informative, and as consistent as possible. The major purpose should be a regular updating of the family about the patient's condition, including his or her chances of recovery. Physicians should be aware that the ICU is a foreign and often frightening place for most family members. Bedside explanations, drawings, radiographs, CT scans, and other visual aids may help families understand the patient's status and appreciate the full extent of supportive therapy.

5. Withholding or withdrawal of life support should be recommended, not merely listed as a therapeutic option, once physicians and nurses agree that life support should be withheld or withdrawn. This agreement should be based on disease-specific data, prognostic indices, and clinical experience, all of which should be explained to the family in detail. The family then should be asked to respond to the recommendation as the patient would if he or she were able to do so.

6. CPR should not be the focus of discussions regarding the withholding or withdrawing of life support. Some physicians who intend to recommend the forgoing of further life-sustaining therapy instead advise that the patient not be resuscitated in the event of cardiopulmonary arrest. Yet such events are uncommon in the ICU unless another form of life support such as mechanical ventilation is withheld or withdrawn. Furthermore, by focusing on CPR, the physician implies that the patient actually may benefit from the procedure despite medical evidence to the contrary36 and puts surrogates in the uncomfortable position of having to refuse what is presented as a potentially beneficial therapy. A better approach is to recommend that all therapies considered incapable of benefiting the patient, including CPR, will be withheld or withdrawn.

7. Surrogates should be allowed to accept the recommendation to forgo life-sustaining therapy. Some families actually precede physicians in their realization that further care would be futile for and unwanted by a given patient. Others are willing to accept the recommendation to withhold or withdraw life support once they have accepted the prognosis offered them. Even if the patient is brain dead, surrogates may have difficulty realizing that the patient is not alive despite the presence of a heart beat and respiratory excursions caused by a mechanical ventilator. Such acceptance may take time, particularly if the patient was entirely well prior to a critical illness. Only a few surrogates are unwilling to accept the prognosis or demand that the patient be supported indefinitely because he or she would prefer life under any circumstances.

8. Physicians should inform surrogates that physicians are not obligated to provide nonbeneficial care if the issue arises. In general, a physician who cannot in good conscience continue therapy he or she considers nonbeneficial should transfer responsibility for the patient's care to another physician in the same or another institution. If transfer is not possible, life support may be withdrawn in accordance with institutional policies. However, such policies are uncommon, and the courts may require that life support be continued when a family insists that a patient would so desire.

9. Surrogates should be told that life support will be withheld or withdrawn as humanely and expeditiously as possible once they have accepted the recommendation to forgo further life-sustaining therapy. They should be given time to spend with the patient and may, if they wish, be present at the patient's death. For many ICU patients, death will come only after removal of supplemental oxygen, positive end-expiratory pressure (PEEP), and the mechanical ventilator. Terminal weaning of this sort should be long enough to allow the family to visit with the patient but should not be prolonged beyond several hours or, in rare cases, a few days. However, early extubation may produce signs of upper airway restriction that are uncomfortable to surrogates and caregivers alike and therefore should be avoided unless sedatives are provided.

10. Whatever process of withholding or withdrawing of life support is followed, surrogates should be assured that the patient will receive sedatives and analgesics to relieve pain and suffering but that death will not be hastened intentionally, in keeping with the principle of double effect. Drugs are best given unobtrusively by constant infusion at doses sufficient to suppress grimacing and signs of air hunger. Formal protocols that dictate how sedatives and analgesics are to be given may help nurses who give the drugs feel that they are doing so properly and to avoid the impression of hastening death. Muscle relaxants should be avoided if at all possible because they provide no therapeutic benefit in the dying patient. The same is true for agents such as potassium chloride that do not relieve pain and suffering but merely hasten death.

11. Physicians, nurses, and other caregivers should meet after the patient dies to discuss how his or her death was managed. Discussions are particularly helpful to the ICU team when conflicts have existed with the patient and his or her surrogates. These discussions provide emotional support for the team and can generate ideas about how to improve the dying process for future patients. They may even inspire the development of guidelines and standing orders used during the dying process.

12. Physicians should be as available to surrogates after the patient dies as they were beforehand. In particular, the attending or primary physician should be accessible by telephone or in person to answer medical questions and report the results of an autopsy or other postmortem procedure. The bedside nurse or family counselor may have been helpful to the family in the ICU and may continue to be a source of emotional support. Through such actions, the caregivers communicate their ongoing concern for the patient and his or her family.