Some patients retain a sufficient degree of capacity to make and express their own medical decisions even after admission to an ICU. Legal precedent is very clear that, except in cases in which the welfare of a third party such as a minor dependent is jeopardized, a properly informed, mentally capable patient has the right to make personal medical decisions, including a decision to refuse even life-prolonging treatment.
If the patient is decisionally incapacitated, a more difficult scenario may confront the physician.9 A surrogate decision maker may be identified (see the previous section) by either a state family consent statute, the patient's prior execution of a durable power of attorney naming a health care agent, court appointment of a guardian/conservator, or informally relying on available, willing family members. The surrogate is expected to make decisions consistent with what the patient would choose if he or she were presently able to make and express choices personally—the substituted judgment standard, or “donning the mental mantle” of the incapacitated person. If there is no reliable indication of the patient's preference under the circumstances (a number of states require proof of this fact by clear and convincing evidence),10 the surrogate must act benevolently in the patient's best interests, considering—from the patient's perspective—the proportionality or comparison of likely benefits and burdens associated with available medical alternatives. The surrogate frequently is guided in fulfilling his or her decision-making role by treatment recommendations offered by the attending physician.11
A serious problem arises when another relative or friend of the patient accuses the surrogate of making choices that are contrary to both the patient's substituted judgment and best interests and that relative or friend demands a contrary course of treatment. Such unfortunate circumstances, such as those surrounding the infamous Florida case of persistent vegetative state patient Terri Schiavo, often end up being resolved in a courtroom.12
Surrogate decision making for others is a difficult, stressful endeavor.13 Especially as the population ages, an increasing percentage of patients lack available and willing relatives or friends to act as surrogate decision makers for them. In such cases, there are several potential sources of guidance for the physician.
The patient may have executed a living will document while still decisionally capable. This type of advance directive, which is authorized by state statute (frequently called a natural death act), permits a capable adult to make a record of personal preferences regarding future medical treatment in the event of subsequent incapacity and critical illness. Although usually this directive is used to indicate a preference for the limitation of future medical intervention, the directive could be employed to request maximum medical intervention if there is any perceived likelihood of benefit to the patient.
A physician who complies in good faith with the patient's expressed voluntary, informed wishes to limit life-prolonging treatment is on firm legal ground. For the physician who, for ethical or other reasons, chooses not to follow a living will's directive to limit treatment, there is an ethical and legal obligation to notify the patient or surrogate (if one is present) of the physician's objection and to make a reasonable attempt to transfer the patient to another physician who is willing to comply; at the very least, the attending physician is obliged not to impede such a transfer.
Advance care planning often improves end-of-life care and patient and family satisfaction and reduces stress, anxiety, and depression in surviving relatives.14 However, “[a]lthough advance directives may stimulate discussions and reduce the stress of surrogate decision making, well-documented controversy exists over their clinical effectiveness, including their inability to affect clinicians' and families' understanding of patients' preferences and the type of care received.”15 Thus, there is a higher probability that the patient's wishes will be effectively honored by health care providers when those wishes have been ascertained in a timely fashion and incorporated into a written physician's order for life-sustaining treatment (POLST).16,17 The health care provider community is at different stages of adopting and implementing the POLST paradigm in various jurisdictions.
In many states, a public guardianship system has been created to make surrogate decision makers available, either through a government agency or a private agency under government contract, for incapacitated patients who lack available, willing relatives or friends.18 There are also projects in some locales that use charitably funded agencies and their volunteers to act as surrogate medical decision makers for incapacitated patients. The physician should consult hospital legal counsel to determine acceptable local sources of surrogate decision making for the incapacitated patient without relatives or friends.
The most controversial and complicated issue in the treatment limitation arena is still the status of artificial feeding and hydration.19 The courts have been unanimous in holding that feeding tubes (of all kinds) are merely another form of medical intervention that could be withheld or withdrawn under the same circumstances applicable to the withholding or withdrawal of any other type of medical intervention. Major medical groups endorse this position.20 However, some people argue that feeding and hydration, even when accomplished only through tubes surgically or forcibly inserted into the patient's body, are fundamentally different and more elemental than medical treatment, and therefore ought to be continued as long as they might keep the patient alive. A number of state legislatures have embodied this argument in living will or durable power of attorney statutes that are intended to severely restrict the prerogative of patients and surrogate decision makers to authorize the removal of feeding tubes.21 Both the wisdom and the constitutionality of these purported restrictions are extremely questionable.
The 1990 case of Nancy Cruzan is still the only US Supreme Court decision that deals directly with the issue of discontinuing life-prolonging medical treatment.22 Cruzan was an automobile accident victim who was kept alive in a permanent vegetative state within a government (Missouri) long-term care facility, through the use of feeding and hydration tubes. Her parents asked that this intervention be discontinued, a request they claimed was consistent with the patient's previously expressed (although not documented) wishes. The attending physicians refused to honor this request, and the Missouri Supreme Court upheld the trial court decision and denied the parents' request to discontinue treatment.
On appeal, the US Supreme Court held that a mentally capable adult has a fundamental constitutional right, under the liberty provision of the Fourteenth Amendment's due process clause, to make personal medical decisions, even regarding life-prolonging treatments including artificial feeding and hydration. For decisionally incapacitated patients, though, the court ruled that the public interest in preserving life is strong enough to permit a state, if the state so chooses, to require—before the state must comply with a surrogate's instructions to withdraw life-prolonging medical treatment—“clear and convincing” evidence that the patient would want that treatment withdrawn if the patient were currently able to make and express an autonomous choice. Presumably, a written declaration made by the patient while the patient was decisionally capable would suffice as evidence of treatment preference in the event of subsequent incapacity. Under the Cruzan decision, states are also free to set lower standards of proof than “clear and convincing” evidence for incapacitated patients, namely proof by a preponderance of the evidence (in other words, greater than a 51% likelihood).
One form of treatment limitation around which there is a high degree of current consensus is the Do-Not-Resuscitate (DNR) or No Code order, which instructs caregivers to refrain from initiating cardiopulmonary resuscitation (CPR) for a patient who suffers an anticipated cardiac arrest. There have been very few legal cases in this arena, but the well-accepted rule is that a decisionally capable patient has the right to refuse CPR, and that surrogates may elect to forego CPR for a patient if the likely burdens of this intervention to the patient would be disproportionate to any benefits (eg, mere continued existence until the next arrest) that might be derived. As is true for all medical decisions, a DNR order should be created only after a thorough consultation with the patient or surrogate and should be clearly documented in the medical record.23 A DNR order may be included as part of a more comprehensive POLST (discussed earlier).
When the patient or surrogate declines aggressive, technologically oriented interventions, the physician still has the legal obligation to provide basic palliative (comfort, pain control, and emotional support) and hygiene measures.24 Failure to do so could constitute negligence or form the basis for professional disciplinary action. Good palliative care may sometimes include the practice of palliative sedation (also called total, terminal, or controlled sedation) for intractable distress or suffering during the dying process.25
In every American jurisdiction, it is a criminal offense (as a form of homicide) for a physician to engage in positive or affirmative actions that are intended to hasten a patient's death (such as administering a lethal injection), even if the patient requested such action.26 Similarly, in every state except Oregon, Washington, and Montana,27 it is illegal for a physician to comply with a patient's request that the physician supply the patient with the means to hasten his or her own death (such as writing a prescription for a lethal dose of a medication, knowing that the patient intends to commit suicide by ingesting that lethal dose).28 The US Supreme Court has soundly rejected the argument that individuals have any constitutional right to physician-assisted death (PAD).29,30
The other side of the coin on treatment decision making is presented when the patient, or more usually the family, insists on initiation or continuation of medical treatment (“doing everything possible”) that the clinician concludes is futile in terms of benefit to the patient. Neither a patient nor the family has a legal right to, nor does a physician owe an obligation to provide, medical treatment that would be nonbeneficial.31-33 On the very rare occasions that courts have been involved prospectively with the futility issue, their holdings have been confusing, inconsistent, and poorly reasoned. However, no court has ever imposed liability for failure to begin or perpetuate futile interventions for a critically ill patient, even in the face of family insistence on doing everything technologically possible. In practice, clinicians usually seem to take the path of least resistance in such circumstances and “treat the family,” often out of misapprehension about potential liability exposure. In the vast majority of cases, better physician-family communication, in which the realistic implications of “doing everything possible” are spelled out clearly, can obviate serious disagreement over how to proceed.34