Surrogate Decision Making

Modern medicine has embraced the concept of shared decision making between patients and their physicians based on the principle of autonomy.^1,2 This approach is often more complicated in the intensive care unit (ICU), because patients are frequently too ill or otherwise impaired to make meaningful contributions to decisions about their care. Increasingly, decisions are made in the ICU to withdraw care,³ and conflicts are common between physicians’ practices and patients’ wishes.⁴ In the ICU, as in other medical situations, patients have an ethical (and in many places, a legal) right to determine the goals of their medical care. An individual patient’s wishes regarding future care in the case of his or her incapacity may be made known in advance of a serious medical illness. The process by which patients, with or without the assistance and participation of their physicians, family members, or other close personal relations, plan for future medical care is called advance care planning.⁵ In general, the results of these deliberations are known as advance directives; defined broadly, they may be verbal or written and may be quite specific or very general. In this process, the patient determines what kind of care he or she would want in the setting of some hypothetical (or anticipated) situation and makes known his or her wishes regarding future medical care. The advance directive helps direct medical care in case of the patient’s incapacity and comes into play only if the patient is unable to make his or her current wishes known.⁶ For example, a patient who awakens after a surgical procedure and is deemed competent (see later) is asked outright about his or her wishes, and the advance directive is no longer necessary.

Advance directives have ethical authority in whatever form (including verbal), as long as the directive was promulgated within the requirements of informed consent (see later). Unfortunately, the reliability of a specific advance directive as “authentic representations of autonomous patient choices” is often suspect.⁷ Advance directives specific enough to guide day-to-day clinical decision making in the ICU are rare; more commonly, the ICU physician is left to work with a surrogate to make decisions for a patient who is too sick to participate in decisions.

For medical decisions in which patient factors play a large role, the physician must have a surrogate decision maker with whom to discuss goals of care and treatment options. There are two questions that must be answered: Who may and should act as surrogate? How should the surrogate make decisions for the ill patient?

In some cultures, physicians often turn to the “next of kin” for surrogate decision making. However, the legal status of surrogates varies from country to country, and this individual may have no legal or ethical grounds for assuming this role. Even in cultures in which surrogate decision making is valued, there is often no designated hierarchy of surrogates. In those cultures in which such a hierarchy has been determined by law, a typical sequence might be (1) spouse, (2) eldest child, (3) next child, (4) parent, (5) sibling. In addition to legal standing, the surrogate should have some moral standing to act as such. For example, a surrogate specifically named in an advance directive document or verbally designated by the patient as the preferred surrogate would have this standing. In fact, some would argue that this is the single most important question for a patient who is sick enough to warrant ICU care (“If you become too sick to speak for yourself, who would you want to make medical decisions for you?”).⁸ In surveys about advance directives and surrogates, patients and well individuals typically name their spouses or other immediate family members as their preferred surrogates. These individuals frequently (though not always) have a shared value system. Interestingly, when asked whether they would prefer that their advance directives be followed no matter what or that their care be discussed with their chosen surrogate, a majority of patients would cede authority to the surrogate.⁹

In many cultures, surrogate decision making is not considered acceptable. Even in this paternalistic approach, it is incumbent on the physician to collect information from those who know the patient well in an attempt to collectively determine what this patient would prefer in terms of medical care and then balance that information with the physician’s judgment as to the best course of therapy. This shared decision-making model is now viewed as the most appropriate in many cultures, including North America and Europe.

In the United States, advance directives allow patients to make their wishes for future care known, either formally or informally. These directives may also designate a specific surrogate decision maker who then has ethical and possibly legal standing (if the appropriate statutory document is properly executed) to make medical decisions for the patient. In the absence of advance directives, the legally appointed surrogate—or, in the absence of such a surrogate, those who know the patient well—make decisions for the patient using substituted judgment based on their knowledge of the patient. When no specific information is available about a patient, the decision makers apply a “reasonable-person” standard—that is, what a reasonable person would prefer in the clinical situation at hand—and sometimes resort to a “best-interest” standard.

Advance Directives

As noted, in the United States, advance directives are formal or informal instructions to healthcare providers, family members, or others involved in a patient’s care regarding treatment that may be required while the patient is unable to participate in medical decision making. The earliest form of advance directive was the “living will.” Classically, the living will is restricted in terms of both scope and applicability. Living wills are usually reserved for patients with terminal illnesses and are typically restricted to statements about forgoing medical treatments that would “only prolong my dying”; they typically make explicit statements about the acceptability of discontinuing intravenous fluids and artificial nutrition if death is imminent and there is no significant hope for recovery. They usually do not provide instructions in case of nonterminal illness and typically do not name a surrogate. A more generally useful legal document is one that gives statutory authority to an individual to make medical decisions for a patient in case of incapacity. This document is sometimes referred to as a durable power of attorney for health care. Similar to a durable power of attorney that provides legal decision-making authority for financial and other matters in case of incapacity, this document provides legal standing to a named surrogate with regard to healthcare decisions. These documents typically provide an opportunity for an individual to give general information about healthcare preferences in a variety of situations. Some also provide an opportunity for the person to make a statement about quality of life and the kind of life that would and would not be worth living. Preferences for organ donation, wishes for spiritual care, and even funeral arrangements are sometimes included.

Additionally, a number of advisory documents have been developed, including “values histories” and the medical directive developed by Linda and Ezekiel Emanuel.¹⁰ These documents may present a series of increasingly dire scenarios and ask about overall preferences (“do everything possible to prolong life,” “continue aggressive care but reevaluate often,” “keep me comfortable, but do not provide care that prolongs my life”), or they may ask more general questions about what makes the person’s life “worth living.” It is hoped that this information will be helpful to a surrogate who must decide whether to continue supportive care in the case of irreversible injury or damage or even to continue disease-oriented care in the case of critical illness and impaired decision-making capacity.

For a variety of reasons, advance directives have not achieved wide popularity. When they exist, they are often not specific enough to provide meaningful guidance.¹¹ Even when a detailed directive exists, a question often remains about whether the individual was adequately informed. For example, a patient’s advance directive says that she would never want to be on “life support,” but when she is asked about mechanical ventilation in the case of reversible respiratory failure from pneumonia, she says of course she would want that. Thus, following a legally executed advance directive without verifying what was meant by the patient and whether the written wishes apply to the current illness is often quite problematic. It could in fact result in a preventable death in a patient who, with proper education, would wish to be treated.

A more limited form of advance directive, known as a code status, is sometimes sought on admission to the hospital, and especially on admission to the ICU. A code status is an advance directive that is specifically limited to a patient’s (or surrogate’s) preferences regarding cardiopulmonary resuscitation (CPR) and other measures in the event of cardiopulmonary arrest. In many hospitals and other healthcare institutions, as a matter of policy, any patient who suffers cardiac arrest is treated with interventions designed to attempt to reverse the life-threatening derangement, including CPR, electrical defibrillation, and intubation and mechanical ventilatory support. Because a patient who suffers a cardiopulmonary arrest will die in a very short time without interventions, the discussion about code status is as much about how a patient wishes to die as it is about whether he or she wishes to live. Tomlinson and Brody distinguish three distinct rationales for a do-not-resuscitate (DNR) status¹²: (1) CPR has such a low likelihood of producing the desired outcome that it is effectively “futile,” (2) there would be an unacceptable quality of life after CPR, and (3) there is already an unacceptable quality of life, and cardiopulmonary arrest would be a welcome deliverance. A decision about CPR may not give much useful information about a patient’s preferences regarding other aspects of his or her illness. A patient may choose aggressive disease-oriented measures well into a severe illness but still choose to forgo resuscitation in the event of an arrest. This approach may be voiced in a statement such as, “I want to fight this thing with all I have, but when it is my time, I want to go quickly without suffering.” Such a statement would be an opportunity to address resuscitation status, in addition to addressing overall goals of care (see later).

Many ICU patients who are actively receiving intensive disease-oriented care have a DNR code status. Such a directive may save surrogates and family members from the emotionally difficult task of removing life-supporting care. A patient’s acceptance of DNR status may signify acceptance of the limits of medical science; refusal of DNR status in the setting of progressive irreversible illness may be an indication that the patient has an incomplete and perhaps unrealistic understanding of the illness. Further discussion, addressing knowledge deficits or unspoken fears, may increase the likelihood that the patient’s true wishes will be followed.

A common error when discussing code status is the failure to address post-resuscitation issues. Patients who undergo CPR will most likely be incapacitated for at least a period of time after the resuscitation, even in the best scenarios. There is also a significant risk of permanent brain injury after cardiopulmonary arrest and resuscitation. Thus, it would be prudent for the patient to name a preferred surrogate as well.

Any discussion of advance directives should attempt to answer at least three questions: (1) In the event of cardiac arrest, do you want the healthcare team to attempt resuscitation? (2) If you become incapacitated, who do you want to make decisions for you? (3) If you were left significantly impaired after an attempt at resuscitation, would you want us to discontinue life-sustaining care? Preferences for resuscitation are best understood in the context of an individual’s values, beliefs, relationships, and culture.⁷

Many problematic end-of-life issues can be traced to a focus on interventions (“Would you wish to be intubated?”) without an adequate exploration of values (“What do you value about your life? What are the things that make your life worth living?”). It is also a mistake to think about advance directives as an issue limited to end-of-life situations. Advance directives are really just part of informed consent for any treatment, and discussion of advance directives is an important aspect of good medical care.

Informed Decision Making (Informed Consent)

In the United States, autonomy is one of the core principles that define the relationship between doctor and patient. Autonomy requires respect for the values and wishes of the individual. An individual patient’s autonomy is best respected when decision making takes place through informed consent. Without adequate information, the power of reason, and freedom of choice, patients’ decisions cannot be said to be autonomous. Informed consent (or more accurately, informed decision making) is a decision-making model designed to safeguard the autonomy of vulnerable patients.

Brock discussed the basic requirements for informed consent and identified three critical elements: the person giving consent must be competent, informed, and able to make a decision free from coercion.¹³ Competence has several critical elements.¹⁴ First, the decision maker must be capable of understanding relevant information, which involves both memory and mental processing. Second, it requires the ability to attend to and retain information, the ability to manipulate information, and the ability to foresee consequences. A third element is the ability to formulate and communicate choice. Some standards of competency strengthen this requirement by demanding the ability to communicate a stable choice (in this case, ambivalence may be a sign of incompetence).

To adequately participate in medical decision making, patients must have enough information to weigh the risks and benefits of various medical interventions. In the past, the standard for being “informed” was the standard practice of other physicians in the community.¹⁵ Subsequently, “informed” came to mean what a “reasonable person would want to know.” Because the main point of informed consent is to respect the rights and values of individuals, it is most appropriate to address this issue in terms of what a particular patient needs to know.¹⁶ In general, patients need to know about the illness and its natural history to make informed decisions about medical care. They need information about the effectiveness of treatment, the risks of treatment, and the likelihood of success with treatment. This information must be presented in a way that is understandable to the patient, at an appropriate educational level, and in the patient’s language. Whether enough information has been transmitted can be assessed at the most basic level by simply asking a patient whether he or she has any questions. Brock writes of “informed understanding” and notes that this “permits an informed exercise in self-determination and promotes a decision most in accord with the patient’s well-being.”¹³ In addition, this approach values autonomy.

The decision must also be voluntary—that is, free of coercion. The decision maker must have the freedom to accept or refuse the intervention or test being proposed. Consent given as a result of undue coercion is generally not valid.

Informed consent in the ICU raises some special issues. First, as mentioned earlier, the decision maker is often a surrogate rather than the patient. The surrogate decision maker should have access to all relevant information the patient would need to make informed decisions; however, the surrogate should not routinely be given confidential information simply because the patient is no longer competent. An example may be helpful in illustrating this point. An HIV-positive patient in the ICU has designated a family member as his surrogate; however, the family is unaware of his HIV status. The ICU physician believes a central line is indicated for continued care and seeks informed consent from the family member. In this case, it may be possible to obtain true informed consent for the procedure without divulging the patient’s HIV status. Alternatively, a decision about a test or treatment specifically related to the patient’s HIV status may require that this information be divulged to the surrogate for her to make an informed decision.

The adequacy of a properly designated surrogate is usually assumed but should be questioned in two situations. The first is when the surrogate acts in contrast to the patient’s known wishes. Anyone who knows that the surrogate’s directions conflict with the patient’s expressed wishes has an obligation to work with the surrogate to come to a treatment decision more in keeping with the patient’s wishes or to seek outside assistance from the hospital ethics committee or the hospital’s legal department. The second situation occurs when there is doubt about the surrogate’s competence, specifically his or her ability to retain and process information. Again, the ethics committee or the risk management department can be of help in this situation. An important study by Schneiderman et al.¹⁷ demonstrated the value of ethics consultation for ICU patients. In that randomized controlled trial, patients receiving an ethics consultation had shorter ICU and hospital stays as well as a decrease in the use of “non-beneficial treatments.” This study has led to a call in the literature for more frequent utilization of ethics consultations in the ICU.^18–20

In summary, ethics in critical care are founded on the same four primary directives common to all disciplines of medicine. Critical care decision making presents special challenges because these decisions often involve the life or death of patients who are unable to participate in the decision-making process. Although the balance between physician and patient responsibility for decision making may vary across cultures, the primary directive for physicians to act in the best interest of their patients is universal.

Key Points