Eileen H. Kim
Michele A. Lorand
William Hayden
Cory M. Franklin
Informed consent is the process of providing patients with information about the risks, benefits, and potential alternatives to the care they are offered. Informed consent is an essential part of the therapeutic discussion and is central to the relationship created between patient and physician (1,2).
The roots of the informed consent doctrine can be traced as far back as the Magna Carta, but its practical basis was established in the early 20th century in the 1914 New York case Schloendorff v New York Hospital concerning a patient with a fibroid uterine tumor. The patient agreed to an abdominal evaluation under anesthesia but specifically refused any surgery (3). Despite this refusal, the surgeon removed the tumor while the patient was anesthetized. This resulted in the modern foundation for informed consent, authored by Judge Benjamin Cardozo (later to become a famous Justice of the U.S. Supreme Court),
“Every human being of adult years and sound mind has a right to determine what shall be done with his own body…” (3)
Ironically, today Schloendorff is nearly synonymous with patient autonomy and informed consent, but at the time there was no specific mention of informed consent as an actual principle. The case did not address issues such as what amount or type of information is necessary for a patient to make appropriate care decisions, nor did it result in damage recovery.
The process of informed consent did not become an established part of American medical practice until the late 20th century. Two historical tragedies proved instrumental in the creation of the informed consent doctrine as we know it today. The first event was the Nuremberg Code (1946–1949), which was developed as a result of the notorious Nazi medical experiments at Dachau during World War II (4). This code provided that “voluntary consent of the human subject is absolutely essential” and “the person involved…should have sufficient knowledge and comprehension of the elements of the subject matter as to enable him to make an understanding and enlightened decision” (5).
The Tuskegee Syphilis Study (1932–1972), conducted under the direction of the United States government, marked the second event. The study resulted in the deliberate withholding of syphilis treatment from several hundred rural African American males so that investigators could gain information regarding the serious complications of late-stage syphilis. When the facts surrounding this experiment finally became public, it raised the consciousness about the rights of patients and of research subjects regarding what information doctors must disclose (6).
Unfortunately, a wide gap still persists between the idealized elements of informed consent and what is commonly observed in clinical practice today. Too often, “informed consent” is simply another shopworn phrase of internal contradictions along the lines of “rush hour,” “United Nations,” or “reliable software.” Simply put, when a harried medical student, nurse, or ward clerk hurries into a patient's room with a boilerplate form, the patient is expected to sign immediately. Informed consent is thus often neither informed nor consent. Having a signed informed consent form is not the same as getting informed consent.
This chapter will discuss the current status of informed consent in the intensive care unit (ICU). It will stress the principles of sharing information, making good faith attempts to understand patient values and decision-making processes, and finally, avoiding manipulation and coercion of the vulnerable ICU patient.
The Ethical Foundations of Informed Consent
The ethical foundations of informed consent encompass the classic principles of autonomy, beneficence, and justice. These three virtues provide the moral framework for informed consent and present guidelines for appropriate clinical action (7).
Autonomy
Autonomy, from the Greek words for self (auto) and rule (nomos), refers to the capacity for self-governing and the patient's right to self-determination. This includes the right to select a course of medical therapy that best reflects individual values and preferences. A prerequisite of autonomy is that an individual maintains the right to hold certain beliefs and to exercise independent thought. From these principles arise the ability to choose a certain course of action, to act according to this preference, and to accept the subsequent consequences of that decision. For this to occur, an individual must have access to relevant information and must also possess freedom from both internal and external constraints.
Practically speaking, before patients can reasonably form an opinion regarding available therapeutic options, they must first appreciate the nature of their medical condition, recognize the range of possible interventions, and understand the possible risks, benefits, and consequences associated with each option. This is essentially the mental checklist the physician should perform when speaking with the patient. It is the physician's duty to ensure that the patient understands the medical diagnosis, the details of the proposed therapy, the available alternatives, and the consequences of refusal. Although the responsibility to provide this information lies with the physician, it is the patient who must ultimately integrate the facts and determine the most appropriate course of action.
Generally, autonomous action requires that individuals enter into the physician–patient relationship voluntarily and remain free to accept or refuse treatment without feeling coerced or intimidated. This is often not the case in the ICU. Patients frequently arrive in the ICU in a vulnerable condition, often admitted without their consent or knowledge. The additional stresses of critical illness leave them susceptible to fear, pain, or anxiety. With these factors in mind, the ICU physician must maintain a balance between talking to the patient and making prompt therapeutic decisions. Given the emergent nature of developments in the ICU, it may be impractical to engage in an extensive discussion regarding every procedure or therapy, but whenever possible, it is essential to provide patients with sufficient information to let them guide the overall course of their care. The balance between acting and letting the patient act characterizes the essence of informed consent in the ICU (8).
Beneficence
Beneficence (doing good), and its associated principle, nonmaleficence (not doing harm) affirm the physician's obligation to provide benefit while refraining from committing harm. Beneficence compels the physician to treat illness, provide other appropriate therapy, and relieve pain. Nonmaleficence compels the physician to avoid causing pain and to refrain from committing unnecessary harm. It would be unreasonable to state that physicians must avoid all risks in treating patients. Obviously, many ICU therapies and interventions pose considerable risk to the patient and may also cause pain. The therapeutic relationship in the ICU represents a working relationship between physician and patient, which balances potential benefits against potential harms whenever possible. Physicians are not neutral observers and, as long as they avoid coercive techniques, it is certainly acceptable—and some would argue mandatory—for them to provide their professional recommendation based on their clinical experience (9).
Justice
The third principle, justice, is generally not a source of conflict between the individual physician and patient in matters of informed consent. Ideally, the rules of informed consent serve to motivate the social virtue of justice; when conflicts do occur, they relate more commonly to societal versus individual claims and thus do not involve the physician–patient relationship. An exception is organ transplantation, a situation in which the transplant surgeon's primary duty is directed to the proper allocation of organs rather than to a specific patient (10). Implicit in the relationship of justice to informed consent is the specific involvement of society's instrument of justice: the court.
Legal Foundations of Informed Consent
The progression of legal opinions during the past century illustrates the evolution of the currently recommended standards of informed consent. The term informed consent was first used in 1957 by an unheralded attorney named Paul Gebhard, drawing on his experience in labor law negotiations (11). In Salgo v Leland Stanford Jr. University, Gebhard used the term in a friend-of-the-court brief on behalf of the American College of Surgeons to refer to the requirement that a physician must disclose to a patient the relevant risks and benefits of a procedure (12).
Determining the acceptable limits of appropriate treatment that a physician can offer is essential to understanding the legal considerations of informed consent. Schloendorff established the precedent that a physician performing nonemergent surgery without the patient's authorization constituted a form of battery, “a touching that is not consented to” (3). Judge Cardozo opined that “a surgeon who performs an operation without his patient's consent commits an assault.” Since Schloendorff addressed only issues of self-determination and autonomy, appropriate standards for adequate consent remained unaddressed for decades.
Canterbury v Spence (1972) later served to establish a minimum standard of information disclosure. The case concerned a patient who underwent a laminectomy for back pain (13). While recovering from surgery, the patient fell out of bed and suffered partial paralysis. After an initial trial failed to establish medical negligence, the argument shifted to the surgeon's failure to fully disclose all known risks associated with the surgical procedure. At the time, laminectomy carried an approximately 1% risk of paralysis, a risk that the surgeon openly admitted he did not disclose to the patient. The surgeon stated that he felt communication of that risk to the patient is not good medical practice because it might deter patients from undergoing needed surgery (13). The court rejected this position and found negligence in the failure to disclose the risk of paralysis. It further emphasized that a physician's obligations extended beyond merely diagnosing and offering treatment. Inherent in the duty to heal exists an obligation to communicate risks and benefits so that patients receive the information necessary to formulate an educated decision:
True consent to what happens to one's self is the informed exercise of a choice, and that entails an opportunity to evaluate knowledgeably the options available and the risks attendant upon each (13).
By establishing the physician's duty to disclose the relevant risks of surgery, Canterbury v Spence carved out an obligation on the part of the physician to present patients with adequate disclosure for any intervention. In Cobbs v Grant (1972), the California Supreme Court found a physician did not need to provide a “lengthy polysyllabic discourse on all possible complications” or a “minicourse of medical science.” The court required disclosure of “such information as a skilled professional would provide” (14). This introduced the concept of a medical community standard for informed consent disclosure, further codified in Truman v Thomas (15).
Whereas previous cases had merely emphasized obtaining patient permission and providing sufficient information prior to performing an intervention, Truman extended the obligations of the physician. A 30-year-old patient who ultimately died of cervical cancer had seen her physician regularly over numerous years, during which time she repeatedly refused to undergo a pap smear. When her cervical cancer was diagnosed, the mass was no longer operable, and she died shortly thereafter. The court, referencing principles previously established by Canterbury v Spence, ruled the physician had negligently failed in his duty to provide adequate disclosure because he did not specifically emphasize death as a potential consequence of refusing a pap smear. Citing Cobbs v Grant, the court concluded,
A patient must be apprised not only of the risks inherent in the procedure [prescribed, but also] the risk of a decision not to undergo the treatment, and the probability of a successful outcome of the treatment.
This case became the first to consider the consequences of refusing a medical therapy. It represents the transition from medical paternalism toward a more interactive discussion between physician and patient that recognizes the patient's right to self-determination. Our current models of informed consent have all developed as a result of that transition (16).
Current Ethical Models of the Physician–Patient Relationship
The models of informed consent, which propose strategies for presenting information to patients and discussing alternatives, arose originally from the paternalistic Hippocratic tradition. In the physician-centered model—alternatively known as the paternalist, parental, or priestly model—the physician is the authority figure and guardian (17). In prioritizing the principle of beneficence (or doing good), the physician engages the patient in decision making only to provide relevant information and encourage acceptance of the proposed therapy. Historically, Hippocrates advocated “concealing most things from the patient while you are attending to him.” Similarly, in 1871, Oliver Wendell Holmes asserted, “Your patient has no more right to all the truth you know than he has to all the medicine in your saddlebags…He should get only just so much as is good for him” (18).
In time, greater emphasis on patient self-determination emerged, along with a higher priority on patient autonomy. Consequently, the informative model—also known as the scientific, engineering, consumer, or independent choice model—emerged as an alternative patient-centered strategy. It minimized physician bias and value judgment while recognizing the physician as technician and source of information. This provided the patient with options regarding the range of medical choices, along with the risks and benefits of potential alternatives. In contrast to the physician-centered model, the informative model asserts the physician's duty to provide facts and medical knowledge without expressing bias toward any particular treatment strategy. Ultimately, it is only the patient who determines which course of action best suits his or her values and goals.
By minimizing physician input, this departure from paternalism represented an attempt to achieve complete patient autonomy. Nevertheless, this remained an unsatisfactory strategy for achieving informed consent. True informed consent requires an interactive process between physician and patient. In clinical practice, the physician–patient relationship is a collaborative process by which both sides take equal responsibility for participation with the shared goal of enhanced understanding. Clearly, the physician must be more than a technical adviser. The ICU is where the physician's training, knowledge, and experience are most important in providing interpretive guidance about diagnosis and treatment. This means that the patient may on occasion request and receive a great deal of information; other times this will be impossible, and the physician will be the primary decision maker.
Two current models of shared decision making propose strategies for mutual understanding through an interactive process. The first, the interpretive model, focuses on clarifying the patient's values and determining preferences regarding the goals of therapy. The physician may help the patient recognize and express his preferences by serving as a counselor who provides information and engages the patient in a joint process to achieve understanding. A discussion of treatment options allows the patient to recognize his or her own priorities and to determine which option may best realize these values. The physician's guidance allows the patient to demonstrate his or her autonomy and self-understanding.
The second model of shared decision making, the deliberative model, requires the physician to provide clinical information and then elicit information from the patient regarding his or her understanding and goals. In representing an idealized interaction between physician and patient, the physician integrates medical information with the patient's values. In this model, the physician should express opinions and preferences regarding appropriate therapy. Patient autonomy is preserved through the patient's moral understanding and action.
These idealized models of the physician–patient relationship recognize that informed consent is a process of shared decision making. Examining the values of both the patient and the physician contributes to decisions regarding treatment benefits or risks (19). The optimal model for the physician–patient relationship is one that achieves a level of interactive and shared decision making, thereby prioritizing patient autonomy while still engaging the participation of a concerned physician (20).
Current Legal Standards of Informed Consent
Considerable uncertainty and debate remain regarding how much information a physician should reasonably provide so that a patient can adequately appreciate the risks associated with any particular therapeutic intervention (21). The perpetual dilemma of informed consent in the ICU is that, in extreme situations, both of benefit and risk, a greater obligation lies on the physician to adhere strictly to the guiding principles of informed consent. At the same time, the ICU patients, because of their weakened condition, may be less able to comprehend and make decisions. In any discussion of possible risks, a physician should routinely disclose to the patient the complications that would most commonly occur; a reasonable figure would be a complication with a probability of at least 1% to 5%. If the potential risk is particularly serious or potentially fatal, it seems obvious that even rare complications with less than a 1% probability should be mentioned (e.g., the vascular complications of routine central venous catheter placement). However, some may argue that the occasional one-in-a-million fatal complication is not the appropriate standard for disclosure (to say nothing of the fact that some physicians may not be aware of these rare complications). Because opinions differ, there is no uniform legal standard that defines the level of information required to meet the standard of adequate disclosure (22). Consequently, three standards of disclosure have been developed and currently exist: the professional community standard, the reasonable patient standard, and the individual patient standard (23,24).
Standards of Disclosure
The Professional Community Standard
The professional community standard was, for decades, the traditional standard for informed consent. According to this standard, a physician should provide the level of information that physicians in the community would communicate to patients in comparable situations. Courts would assess physician disclosure based on the standard practices of other physicians with similar training and experience working under similar circumstances. Because of the imprecise definition of “professional community,” the professional standard was used to justify a broad range of interpretations, albeit without solid grounding in clinical criteria. The community could range from very specific practice locations to a broad geographic region or otherwise could refer to a level of specialized training or experience. In some circumstances, even the opinions of a “respectable minority” of physicians would constitute an appropriate practice standard. As such, the expectation of what the physician would tell the patient was notoriously imprecise. It was difficult to define which specific surgical or procedure risks a physician should appropriately disclose to a patient. Furthermore, physicians often invoked the concept of therapeutic privilege, which permitted them to withhold all information if they thought it would be injurious to the patient. This doctrine has fallen out of favor both clinically and legally (13).
Critics cited not only the imprecision but also the paternalistic nature of the professional community standard. According to this standard, the physician ultimately determined the threshold of risk that should be disclosed to the patient. The obvious problem with this model was that if the community standard did not include the disclosure of a potential complication or other information that patients may reasonably want to know, the physician was not obligated to disclose it. For example, physicians might prescribe penicillin, and, while a potentially lethal anaphylactic reaction to the drug was possible, because it was rare it would not necessarily be mentioned as one of the complications. Although physicians could not be expected to divulge every possible complication of a procedure or adverse reaction to a drug, many still felt it unacceptable that the standard for providing information rested solely in the hands of the physician.
The Reasonable Patient Standard
In response to this paternalistic standard, and in concert with the trend toward greater emphasis on the patient's right to self-determination, American courts began recognizing an alternative reasonable patient standard to judge the adequacy of risk disclosure. Since it was unreasonable to expect a physician to disclose every potential risk associated with a particular treatment, the reasonable patient standard required the physician to disclose all information a reasonable person would need to make an informed decision. This new standard deemed that even rare complications should be explained to the patient if the consequences (death, severe injury) were such that a reasonable person would want to know them.
However, there are also problems with this model. For one, the physician must divine what a reasonable person would want to know. (Would a reasonable person want to know about anaphylactic responses to penicillin?) Second, the ICU, a setting in which life and death decisions are commonplace, may not lend itself to the enforced neutrality of a reasonable person standard. The physician would be performing a grave disservice by simply reciting the potential complications of endotracheal intubation to a patient in respiratory distress, and thus, a different standard was needed.
The Individual Patient Standard
In the ICU, the physician's input is critical to good decision making, which is why the optimal model for physician–patient relationships is one of interactive and shared decision making. The individual patient standard addresses this relationship. Based on the interaction with the patient and an understanding of the patient's beliefs, the physician should disclose specific information so the patient can reach a decision consistent with his or her principles. The distinction between the different standards is subtle but significant. Under the professional community standard, the physician asks, “What should I tell the patient?” Under the reasonable person standard, the physician asks, “What does a reasonable person want to know?” Under the individual person standard, the physician asks, “What does this patient want to know?” Obviously, the most idealized standard of disclosure, the individual person standard, is ultimately the most difficult to achieve. Courts may not require such an idealized standard in all cases, but when questions of informed consent arise, this is the standard that courts are most likely to favor.
Adjustments of Standards
In discussing these legal standards, a note of caution is in order: These models represent guidelines for medical encounters where both parties—patient and physician—can interact. In the ICU, patient situations are constantly changing, life and death decisions are commonplace, and emergencies sometimes make the search for an ideal physician–patient relationship impractical. Unlike the long-term relationship between the patient and the primary physician, the intensivist is often meeting the patient for the first time under conditions of extreme duress (25). For the patient, admission to the ICU is almost always a stressful and potentially overwhelming situation in which critical illness creates an unusual dependence and power imbalance. The patient may be unable to comprehend or express his or her wishes (see Competence and Decision-Making Capacity, later). Other times, an autonomous patient may choose to relinquish medical decision making at the physician's discretion when acute care is required (26).
It is also important that physicians use language the patient can understand when explaining the risks and benefits of any intervention. This means not only adequate translation for patients who do not speak English but also making the explanation as nontechnical as possible. Even in the best of situations, patients may have difficulty extracting important information from discussions with physicians. When physicians lapse into technical jargon, the anxious, frightened patient may have little or no opportunity to process what is being said. It is important that in the appropriate situations, physicians make use of translators, family members, and other intermediaries.
An extensive discussion regarding the risks and benefits of care are the desired standard, but in the ICU, less is sometimes more. In an emergency, the necessity of keeping the patient informed sometimes becomes a luxury that time and circumstance may not permit. Emergency circumstances, in which a patient lacks decisional capacity and no proxy decision maker is identifiable, do not realistically allow for voluntary consent from a medically incompetent patient. In truly emergent situations, if the patient lacks capacity, no proxy decision maker is available, and the potentially lifesaving intervention must be administered immediately, the “emergency exception” to informed consent permits the physician to intervene without obtaining formal informed consent. In these situations, the intensivist should document the emergent nature of the situation and the difficulty in obtaining informed consent. (See When Consent Cannot Be Obtained.)
Competence and Decision-Making Capacity
Hospitalized patients, especially the critically ill, often suffer from impairments in their ability to comprehend, process, or analyze information. Under the influence of pain medication, sedation, or the physical and mental stresses of illness, even the healthiest ICU patient may not fully appreciate or be able to actively participate in health care decisions, as the emotional stresses of critical illness may temporarily compromise their decisional capacity (27). As one study noted, for very sick patients, the ability to perform simple cognitive tasks is impaired to the point that an adult patient may temporarily function at the level of a 10-year-old child (28). This presents a unique challenge for critical care practitioners when discussing medically complex issues.
Definition of Terms: Competence versus Capacity
“Competence” and “capacity” both refer to the patient's ability to make decisions. Although the terms are often used interchangeably, their legal and medical definitions differ. Strictly speaking, competence refers to a legal determination and does not refer specifically to the patient's ability to make appropriate health care decisions (29). A court decides whether or not a person is legally competent, and generally, when “competence” is used as a legal term, it refers to patients' ability or inability to conduct their personal affairs, not necessarily to make health care decisions. It is unusual (but not unheard of) for petitioners to go to court specifically to ask that the court declare a patient incompetent to make medical care decisions. More often than not, this legal determination regarding who decides care for a patient remains in limbo and is left to the patient's family and doctors.
If the courts are not involved with a patient's ability to make decisions, health care providers commonly invoke the term “medical competence,” but they are really referring to the patient's capacity. Decision-making capacity is a clinical judgment in contrast to legal competence, which specifically describes the patient's ability to make health care decisions. The need to assess capacity by the physician arises when there is reason to question whether the patient can make decisions about care (30). When assessing the patient's capacity, i.e., what most observers refer to imprecisely as whether the patient is competent to consent to care, the examining physician must determine whether the patient understands the five basic elements of capacity (31) (see Table 5.1). During an interview, if a patient demonstrates satisfactory understanding of these five facts, it can reasonably be inferred that the patient possesses adequate decision-making capacity.
|
Table 5.1 The Five Elements Patients Must Understand to Determine Their Capacity |
||
|
Informal Assessments
Informal assessments of a patient's cognitive abilities occur regularly throughout physician–patient interactions; in the critically ill patient, mental status may fluctuate during the course of hospitalization or even during the course of the day. Unless presented with evidence to suspect otherwise, the treating physician should assume that the patient remains capable of independent choice; this is the default position, unless the patient's decision-making capacity is questioned. If this is the case, the health care provider is obliged to demonstrate that the patient cannot make medical decisions. If the clinical situation suggests the patient is not capable of independent choice, a more formal evaluation may be initiated (32).
Physicians may be more likely to doubt a patient's mental capacity if the patient's choices appear unreasonable or contradict the physician's personal values. The patient who refuses a relatively low-risk, high-benefit intervention, or a terminally ill patient who insists on pursuing a painful intervention with little proven benefit, are both scenarios that may prompt a physician to question the patient's decision-making capacity. In these situations, the physician must first attempt to decipher whether the patient's seemingly illogical behavior actually follows a rational thought process. A critically hemorrhaging patient who refuses a blood transfusion may be medically frustrating to care for, but this refusal becomes understandable once it is revealed that the patient is a Jehovah's Witness. Similarly, the patient with end-stage metastatic cancer who has failed multiple rounds of chemotherapy may seem irrational for insisting on pursuing invasive experimental procedures. This seemingly irrational insistence may become more understandable in the context of an upcoming family event, anniversary, or graduation.
External and Emotional Factors
In the ICU, external factors, including sundowning, sedation, pain medication, or altered sleep patterns, may contribute to transient, reversible episodes of incapacity. Whenever possible, attempts should be made to minimize the impact of these influences and optimize the patient's cognitive status prior to making a capacity assessment. The patient's judgment is often compromised by emotional factors, e.g., anger, fear, denial, depression, or pain; this is especially true in the ICU (and in the emergency room). The common scenario of the 50-year-old executive with crushing substernal chest pain who denies he is having a heart attack and wants to sign out of the hospital is an example of how denial may compromise a patient's judgment.
Health care providers should recognize that the patient's decisions under those conditions may not be those they would choose in a less stressful environment. The physician must attempt to ensure that external factors do not unduly influence the patient. True informed consent requires the patient's unhindered judgment. When the patient's judgment appears to be unduly compromised, the physicians should use appropriate measures such as family intervention, psychiatric consultation, or medication aimed at treating the specific problem. The frightened patient who refuses necessary medical care is often grateful if, after appropriate intervention, proper care is provided.
For particularly high-risk interventions or close calls, a second physician, generally someone with expertise in this area, such as a psychiatrist or neurologist, may be called to evaluate the patient. In such cases, the physician should inform the consultant in advance about the situation so the consultant can conduct a focused interview and provide the necessary information.
Legal Interventions
In rare circumstances, questioning a patient's capacity means seeking a court determination of legal incompetence. In practice, resorting to the court is rarely necessary. Rather than deferring to the court system, most states recognize the authority of a spouse, family member, or friend to make decisions in the best interest of the patient. However, when family members and health care providers cannot agree about the most appropriate course of action after attempts at resolution, a legal opinion may be the only option. Courts are generally reluctant to get involved in health care decisions, so this should generally be the last option.
Surrogate Decision Making
In the United States in 1990, a landmark piece of legislation, The Patient Self-Determination Act, established a patient's right to a name a legal, durable power of attorney for health care if that person should lose decision-making capacity. Most states have their own statutes governing advanced directives. These statutes regulate the authority of a health care proxy decision maker and establish rules that balance the authority of a surrogate decision maker with the patient's preferences. In the absence of a durable power of attorney for health care decisions, an increasing number of states currently recognize the authority of family and friends to act as legal surrogate decision makers, with the limits and scope of this authority varying by state (33). A detailed discussion of the role of surrogate decision makers, specifically concerning end-of-life decisions, is discussed elsewhere in this textbook. (See Chapter 2: Life and Death in the ICU: Ethical Considerations.)
The authority of potential decision makers generally gives priority to family members with closer blood ties. A typical ranking of default surrogate decision makers might be as follows:
1. spouse
2. adult children
3. parents
4. adult siblings
5. other relatives.
If multiple surrogate decision makers are identified (e.g., more than one adult child or adult sibling), some statutes require consensus among all family members, whereas others accept a majority decision. If no blood relative can be identified, several states also include “close friend” in the list of acceptable surrogates, generally giving them the lowest priority. In most states, however, any interested party can challenge the authority of the presumed decision maker, with court appointed guardians generally having top priority.
It has been shown that sometime during their ICU stay, approximately 50% to 75% of patients lack decision-making capacity (34). At least one study found that advanced directives are infrequently used and seldom effective. This suggests a very real need for the participation of family and friends in critical care decision making. At the same time, when determining the most appropriate course of action, the decision maker should select the options that are consistent with the patient's previously demonstrated behavior and values. Guidelines for proxy decision making have recommended substituted judgment, by which proxy decision makers refer to past interactions, experiences, or conversations with the patient and draw on these experiences to select the course of action the patient would have preferred. Physicians should present the same level of information to health care proxies that would have been provided to the patient to help them arrive at the best decision.
The shared, interactive, decision-making process between physician and patient extends to the proxy decision maker. Surrogate decision makers, in turn, serve as an extension of the patient, attempting to communicate the choices and concerns the patient would have expressed. If no advanced directive exists, and family or friends cannot comfortably infer the incapacitated patient's likely preferences, the most appropriate guide for proxy decision making is what they believe to be in the patient's best interests. When substituted judgment is not possible, the presumption remains that family members, sharing similar cultural background and upbringing, are most likely to reflect the patient's preferences (35).
Even though most patients trust that designated family members and physicians would accurately anticipate their preferences, several studies have questioned whether proxy decision makers misjudge patient preferences (36). Evidence suggests proxy decision makers correctly predict patient preferences approximately two thirds of the time when presented with hypothetical scenarios regarding end-of-life decisions (37). In general, proxy decision makers tended to pursue treatment more aggressively than patients would have chosen for themselves (38). This discrepancy possibly reflects a tendency of surrogates to select interventions that they would choose for themselves rather than accurately representing the preferences of the incapacitated patients (39). Current data remain inconclusive regarding whether explicit patient–surrogate discussion improves the accuracy of proxy decision making.
The inadequacies inherent in surrogate decision making highlight the problem of shared decision making in critical care medicine. Despite the questionable accuracy of substituted judgment, however, family members are the most appropriate surrogate decision makers. In spite of the demonstrated inaccuracy of proxy decision makers, most patients nevertheless indicate that they would prefer that a family member make medical decisions regarding their care (35). The participation of interested participants—including patient, family, friends, clinicians, attending physician, consulting physicians, and nurses—is a necessary but complex process in determining the most appropriate treatment for a patient who cannot decide.
When Consent Cannot be Obtained
In certain situations in the ICU, it is impossible to obtain consent from the patient. Family members may on occasion be available to provide surrogate consent, but often they are either not present or, for various reasons, are unable to give consent. Currently, no standardized guidelines exist to direct the most appropriate course of action in those cases. When an emergency exists—defined as any situation where delay in providing care would endanger the patient's life or substantially affect his or her health adversely—health care providers can then assume implied consent and proceed with essential treatment (22). The treating physician must determine situations of implied consent in an emergency. Most authorities believe there are three major areas of clinical incompetence for which implied consent permits treatment:
1. the patient who cannot communicate (e.g., unconsciousness);
2. the patient who is unable to understand the situation (e.g. delirium, mental retardation, or language barrier); and
3. the patient who has attempted suicide (e.g., the patient who takes an overdose or shoots himself or herself) but now refuses care.
In these cases, the physician should proceed immediately with any treatment considered lifesaving.
The intent of implied consent is to protect seriously ill impaired patients at risk of imminent death. The general rule is when life or limb is imminently threatened, disclosure can be more limited. The health care provider must balance the urgency of the situation with the time it takes to obtain consent from a surrogate.
Procedures not considered immediately lifesaving that would otherwise require consent, such as central venous access, arterial and pulmonary artery catheters, or diagnostic bronchoscopy, can provide significant benefit in emergencies but do carry risks. Performing these procedures without consent may theoretically violate a patient's autonomy, but electing to forego them because consent cannot be obtained denies patients the benefit of these interventions. In most emergencies, these interventions should be performed without consent, as they represent the best interests of the patient. Diagnostic tests and procedures presenting minimal risk, including blood draws or peripheral IV placement, generally do not require formal consent, nor do noninvasive diagnostic modalities, including radiographs or computerized tomographic scans.
Implied consent gives critical care practitioners a great deal of latitude. However, there are limitations to the discretion afforded. More careful disclosure is mandated, even in emergencies, if the therapy is elective or in borderline situations when reasonable alternatives exist. For example, the patient with a potential intra-abdominal infection may not understand the need for immediate exploratory laparotomy. If the patient is stable and surgery is merely one diagnostic option, computerized tomographic scans and frequent observation may provide information equivalent to a laparotomy. In selected situations such as described, the treating physician should remain flexible when possible.
Signing Out of the Hospital Against Medical Advice
One of the most difficult situations for the ICU staff is the extremely uncooperative, combative patient. In most cases, these patients are reacting to fear, pain, illicit drug ingestion, or alcohol withdrawal. Usually, care can be delivered after appropriate sedation or analgesia, although, occasionally, a patient must be physically restrained. The indication for physical restraint is when patients present a risk to themselves or to others. In rare circumstances, despite the best efforts of the staff, a patient may refuse all treatment and demand to leave the hospital. The staff is then forced to reconcile the conflict between respecting the patient's rights and their duty to care for and protect the patient from harm. There may be no easy resolution of this problem.
The right to leave the hospital against medical advice is the prerogative of the competent patient. If the patient meets the general test of medical competence (as described in the five basic elements of capacity [31] [Table 5.1]), the patient must be allowed to leave the hospital, even if the staff disagrees with the decision or the decision seems irrational. However, all decisions by patients to leave against medical advice should be scrutinized by senior staff to ensure the patient truly is competent. Those decisions that appear irrational should be scrutinized with even more care. A classic example is the aforementioned 50-year-old male with an acute myocardial infarction, otherwise competent but who, in a fit of denial, demands to sign out of the hospital. All avenues should be used to get the patient to stay, including a detailed discussion of the situation and an appeal to family or friends who accompany the patient to the hospital. Ultimately, however, if the patient refuses to listen, because he is competent, he must be permitted to leave. In such situations—very trying ones, indeed—health care providers must attempt to provide any appropriate care or workup before the patient leaves. Staff should avoid recriminations, and the patient should also be reassured he or she can return for care at any time.
Other situations are not so clear-cut. In many cases, the patient's competence is in question. Possible physiologic causes for the patient's condition, e.g., hypoxemia, electrolyte disturbances, sepsis, should be identified. If the patient cannot be deemed competent, staff may decide to institute treatment over the objection of the patient, which might even entail physical restraint. Failure to restrain when indicated carries a significant risk to both patient and health care providers. The classic counterpoint to the aforementioned myocardial infarction patient is the patient in a motor vehicle accident who appears intoxicated but wants to sign out of the hospital. If the physician suspects that the patient is intoxicated, based on clinical observation even before confirmation of blood alcohol concentration, the patient should not be allowed to leave until appropriate radiologic assessment of the head and neck have been performed.
Some health care providers are overly concerned with the liability they may incur by treating a patient against his or her wishes. When competence cannot be established with certainty, if the staff decides to restrain the patient, they may theoretically open themselves to charges of battery. Such an outcome is extremely unlikely, and almost certainly less likely than the alternative of being charged with negligent discharge. The consequences of being responsible for the negligent death of unrestrained patients are far more serious than the responsibility for holding patients against their will for several hours. When the staff's actions are medically reasonable, they are acting in good faith, and if they document their decision (see next section), the likelihood of successful litigation against them is remote.
|
Table 5.2 Documentation of Patients Who Leave the Hospital against Medical Advice |
|
|
Documentation
All situations of implied consent or decisions to leave the hospital against medical advice require scrupulous and detailed documentation. The information that should be included in the medical record is listed in Table 5.2.
Specialty consultations with neurology or psychiatry are not mandatory but may be useful in assessing the patient and documenting the situation. In difficult cases, it may be necessary to involve a representative of the hospital's administration or legal counsel.
Patient Comprehension of Consent Forms
Currently there is no medical or legal consensus regarding which procedures require formal consent. As a general guideline, there is a greater need to discuss the risks and benefits more formally for procedures that carry greater risk (16). This standard results in consent policies that vary significantly from hospital to hospital. One survey of informed consent practices found that, while over 90% of hospitals surveyed required formal consent for gastrointestinal endoscopy, fiberoptic bronchoscopy, or medical research, fewer than 10% required consent for nasogastric intubation or bladder catheterization (40). Of note, requirements for consent varied between medical and surgical services, even within the same institution.
Achieving satisfactory informed consent may not always be possible, even when physicians explain the procedures to patients. In one study of patients who consented to moderately invasive bedside procedures (thoracentesis, paracentesis, bone marrow aspirate, or lumbar puncture), 90% of the patients surveyed reported that the physician had explained the indication for the procedure, although only 70% could correctly recall the reason for the procedure. Although 86% of patients reported that the physician had informed them of the risks of the procedure, only 57% could later name any of the risks (41). This raises the question of how effectively information is communicated to patients, as well as how accurately patients understand and recall information presented to them.
To anticipate patient needs and to simplify the consent process, many institutions use standardized consent forms that include essential information regarding particular procedures or interventions. The use of a standardized ICU admission consent package, describing and requesting consent for the most commonly performed procedures, can enhance the informed consent process (42). Standardized forms, however, do not necessarily guarantee clear communication. As a response to defensive medicine concerns, standardized consent forms may describe all possible adverse consequences instead of actually trying to inform the patient (43). Moreover, standardized lists of complications may not communicate the risks most relevant to any particular patient, especially in the ICU, where a patient's changing medical condition can present a dynamic series of risks and benefits. The risk of an iatrogenic pneumothorax from central venous catheter placement during mechanical ventilation carries different implications than a similar complication when the catheter is simply placed for fluid replacement.
Despite these caveats, standardized consent forms for common procedures may be useful in initiating dialogue between patient and physician. Standardized consent forms may also be necessary for especially complex ICU surgical procedures such as organ transplantation or experimental surgery. In these situations, a detailed informational document provides patient and health care providers a ready reference. The language of such forms should be reviewed periodically to ensure simplicity and reader-friendly, understandable language. Even when standardized forms are used, health care providers should write a note in the patient's medical record detailing the conversation between the patient and physician.
Research Consent in The ICU
Like every specialty, critical care medicine has achieved progress through research involving the participation of volunteers (44). Critically ill patients represent a particularly vulnerable population, which raises concerns about their ability to give voluntary, autonomous consent to participate in clinical research (45). Ongoing critical care research recognizes a corresponding obligation to protect this vulnerable population. Requesting consent for voluntary participation in research differs fundamentally from discussing informed consent for therapeutic interventions. When discussing the risks, benefits, and alternatives of any therapeutic or diagnostic intervention, both the clinician and patient seek a course of action that would maximally benefit the patient. In contrast, the goal of research is to generate information that may benefit future patients but does not necessarily benefit the individual research participant (46). This creates a potential conflict of interest between researcher and patient, and thus, researchers must exercise particular caution in protecting the rights of patients. Federal regulations, known as “the common rule,” have been designed to protect this vulnerable population of research participants (47).
Patients participating in clinical research may misunderstand or overestimate the individual benefits of participation; alternatively, they may not fully recognize the potential risks. Although the possibility for personal benefit does exist, a patient might be randomized to a nontreatment arm of a trial or may alternatively receive experimental therapy with unexpected, hazardous side effects (48,49). Researchers are obligated to ensure that research participants recognize the additional risks and benefits of participation. Occasionally, this means a researcher may ask a research participant to accept a disproportionate share of risk, with no prospect of additional individual gain. The exact limits of risk a vulnerable patient may be asked to accept have not been specifically defined (50).
The research consent process should clearly delineate the nature of the research and structure of the trial, specifically including details on any randomization process (46). In contrast to therapeutic interventions, informed consent for research represents a process that continues throughout the course of a clinical study, and thus routine updates for the patient may be necessary. Standardized consent forms may be useful in communicating the relevant information, and the physician must ensure that information is clearly explained. Clinical research consent forms may be used as reference documents during the study. Because of concerns regarding literacy and language comprehension, the complexity of language should generally target comprehension for no higher than a sixth-grade reading level.
Since few patients are likely to have voiced their preferences regarding research participation, proxy decision makers for patients who lack decision-making capacity are left to infer the most appropriate actions in certain situations (51). Emergency situations in which patients cannot consent and surrogates cannot be located raise concerns about the ethics of conducting research in these cases. Although regulations have sought to protect potential research subjects, they acknowledge that denying research participation to patients who cannot give consent may also deny them potentially beneficial therapy. A 1996 amendment to the Code of Federal Regulations for the Protection of Human Subjects permits emergency research with certain provisions if consent cannot be obtained (52). Clinical trials describe waivers of consent based on implied consent (53). Delayed consent is another mechanism that has been used for clinical trials that compared two clinically acceptable therapies (54).
The Ethical Care of Children in an Intensive Care Unit
The same basic principles of ethical decision making outlined throughout this chapter apply when treating children, with the understanding that the surrogate decision makers in most pediatric cases are the child's parents. However, dilemmas can arise, and when they do, can prove to be distressing to all involved. These issues arise in the care of critically ill children as well as in the care of the general pediatric population (55,56,57).
A basic concept in caring for a critically ill child is the age of responsibility, i.e., when children can decide for themselves. Our society affords adults certain rights that it does not allow to children, such as voting, driving, and purchasing alcohol or tobacco. Included among these is the right to make decisions regarding one's own medical care. The age of responsibility, and therefore the age at which a young person can consent to medical care, is a complex dilemma in medical ethics. There is no standardized answer as to when minor patients should be able to make decisions regarding their treatment. The factors to be considered include not only the children's age, but also their reasoning skills, level of understanding, experience, severity of illness, and the type of procedure or treatment being offered. In most states, 18 years is the age when a person becomes an adult and can make his or her own health care decisions. This is known as the age of majority. However, there are numerous consent statutes or rules of common law that the critical care physician should be aware of that allow medical treatment of a minor younger than 18 years of age without parental consent (58).
The Emergency Exception
A minor can seek emergency medical care without parental consent. Emergencies are generally defined the same way they would be for adults (see When Consent Cannot Be Obtained).
The Emancipated Minor Exception
Emancipated minors are children younger than 18 years of age (or whatever the age of majority is in the state of residence) who can decide their own medical care. The specific criteria for emancipated minors vary from state to state, but generally pertain to minors who are either married, pregnant, a parent, in the military, financially independent and living apart from their parents, or those who have been legally declared emancipated by the court. Pediatric critical care practitioners should be familiar with laws concerning the age of majority and emancipated minors in the state where they practice.
The Mature Minor Exception
This concept involves minors generally 14 years of age or older deemed capable of providing informed consent. In certain cases, if minors are sufficiently mature and possess the intelligence to understand and appreciate the benefits, risks, and alternatives to the proposed treatment, and can make voluntary and rational choices, they can decide their own care. States differ in how much weight they give to the concept of mature minors and, again, practitioners should be familiar with laws concerning mature minors in the state in which they practice.
Exceptions Based on a Specific Medical Condition
The minor who seeks care for certain conditions can give consent without parental involvement. These conditions generally include mental health services, pregnancy and contraceptive services, testing or treatment for human immunodeficiency virus infection or AIDS, sexually transmitted or communicable disease testing and treatment, drug or alcohol dependency counseling and treatment, or care for a crime-related injury.
Exceptional Situations
There are several other situations involving consent for minors that the critical care specialist may encounter; these are described in the sections that follow.
When Parents Cannot Be Reached or Are Absent
Parents are generally the best surrogate decision makers for their children. Custom and practice dictate that surrogate decision makers are generally ranked in the following order: parents, health care professionals, courts, and finally, social/governmental agencies. Unfortunately, there are occasions when these principles do not apply, and disagreement may arise regarding who best represents the concerns of the minor patient. When parents are absent and cannot be reached in emergency situations, critical care practitioners may use their best medical judgment in instituting treatment. Physicians should seek the support of their decision from other practitioners, as well as clearly documenting that good faith attempts have been made to reach the parents. Documentation should ensure that medical criteria for the particular intervention have been met and should include the circumstances surrounding the inability to contact the parents.
When Abuse and/or Neglect Is Suspected
The law is clear that whenever child abuse or neglect is suspected, medical practitioners are mandated reporters. They are required to report their suspicions to state or local authorities in accordance with the child abuse reporting laws in the state where they practice (59,60). In most states, the practitioner is allowed to proceed with the initial diagnosis and stabilization of the suspected abused child, even in the absence of parental consent. It is obviously best to engage the parent in a straightforward manner concerning the medical decision-making process and obtain consent; however, there may be situations in which the parent is unavailable or unwilling to participate. In these cases, the practitioner should consult the local state child abuse laws and either take protective custody of the child or begin custody proceedings with the appropriate authorities. Most major children's hospitals have child abuse teams with experience in cases of suspected child abuse, along with knowledge of the procedures related to parental consent in these difficult situations. Critical care practitioners should consult these teams early in such cases. Generally, a physician or social worker with expertise in child abuse can work with a family to secure the best interests of the child.
When Parents' Religious Beliefs Interfere with Lifesaving Medical Care
The most common religious issue encountered in pediatric critical care is the child who needs a blood transfusion but cannot obtain it due to the refusal of the parents, who are perhaps Jehovah's Witnesses, to provide consent because it is forbidden by their faith. In some jurisdictions, these cases can be pursued as neglect, but in other jurisdictions, exemptions to child abuse and neglect laws prevent legal repercussions when the alleged abuse or neglect has occurred in the name of religion. However, there is ample precedent to support the decision to administer blood if a child's life is in danger, although this action generally requires legal intervention and hospital involvement (61). Most hospitals have ethics committees or consultants to help practitioners and surrogates work out the critical issues. Depending on the hospital model, ethics committees can be either informational, ensuring that all parties have the same facts and understand the issues and options, or act as decision makers to whom the clinicians can defer. It is both prudent and fair to the child and family to involve a medical ethicist or hospital ethics committee before proceeding with the medical treatment needed. Furthermore, these actions will demonstrate an effort to preserve the relationship between doctor, patient, and family so that all parties have access to the advocates available to discuss the issue. Surreptitious action to institute treatment is never indicated.
When the Parents or Surrogate and Health Care Team Disagree on the Care of a Critically Ill Child
This situation arises most commonly in end-of-life situations regarding the propriety and timing of providing care (62,63). These are obviously emotionally wrenching circumstances if health care professionals and parents or surrogates disagree. Health care providers may anticipate the termination of ventilator support in a severely brain-injured patient or a terminal cancer patient days, or even weeks, before parents or surrogates reach an understanding that this is the proper decision. Parents or surrogates may hold out unreasonable hope, however understandable, in the face of a child's impending death. Patience is usually the best approach. Time and reasoned discussion generally resolve these issues. This strategy requires the understanding that those involved may be at different stages of acceptance. Collegial communication eventually brings the concerned parties to an acceptable conclusion. It is imperative that critical care practitioners offer parents or surrogates sufficient opportunity to discuss their feelings and emotions. A distant, emotionally detached approach by the critical care provider is inappropriate and complicates the delivery of care.
When there is neither common ground nor hope of agreement between the critical care provider and parents or surrogates, the final resort is to take the issue to the courts. Practitioners should not undertake formal legal action lightly. Experience has shown that in most cases, attorneys, courts, and judges, rather than hearing such cases, prefer resolution outside the courtroom. Besides being expensive, the legal process requires time and energy, both physical and emotional, on the part of all involved. In this adversarial process between the critical care provider and parents or surrogates involving end-of-life decisions, medical professionals should keep in mind that, in some cases, courts have ruled against the medical team who originally instituted the proceedings (64). Whenever pediatric critical care practitioners consider going to court for resolution, they should consult the hospital's legal staff and ethics committee to explore other options and coordinate an optimal strategy for all parties involved. They should remember that the best interests of the pediatric patient are their paramount concern.
Summary
A traditional attitude in medicine has been to equate informed consent with obtaining a patient's signature on a boilerplate document. However, informed consent has evolved significantly from that approach so that today, it means much more than simply a signed form. It is now understood as a process intended to create trust between the critical care practitioner and patients. The physician has an obligation to explain the therapy, complications, alternatives, and the risks of alternatives to the patient. There is no simple answer to define exactly how or what the physician must tell the patient. The approach to disclosing information will vary, depending on the goals of care, the urgency of the situation, and the relationship that develops between the patient and critical care provider. There will never be an ideal single standard of informed consent.
Depending on the clinical situation, the requirements for informed consent disclosure lie along a continuum. In life-threatening emergencies where a patient may lack the capacity to decide, limited disclosure is acceptable and, in some cases, the only option. At the other end of the spectrum is the heightened level of disclosure required when asking patients to participate in critical care research. The sine qua non of medical research is the guarantee of the patient's complete understanding and voluntary consent. In those cases, instead of a 5-minute discussion with the patient, obtaining informed consent may require several sessions and considerable effort. Between these two extremes are the quotidian clinical situations in which mutual trust becomes the watchword. Regardless of the complication rate of a particular procedure, or which legal standard the physician seeks to meet when informing the patient, informed consent is optimal when the physician has earned the patient's trust and is comfortable with the patient's understanding of the clinical situation.
There is a belief among some learned but cynical observers that true informed consent represents an impossible goal, because patients will never have the same knowledge base or level of clinical understanding as their critical care physician, and hence, the conclusion that detailed explanations are merely an exercise in futility. Such cynicism misses the point. It is precisely because of that imbalance of knowledge and power between physician and patient that when health care providers seek to obtain informed consent, they redouble their efforts not only to inform patients, but also to earn their trust.
References
1. Beauchamp T, Faden R. A History and Theory of Informed Consent. New York, NY: Oxford University Press; 1986.
2. Katz J. The Silent World of Doctor and Patient. Baltimore, MD: Johns Hopkins University Press; 2002.
3. Schloendorff v Society of New York Hospital, 211 NY 125, 105 NE 92 (1914).
4. Meisel A. Legal and ethical myths about informed consent. Arch Intern Med. 1996;156:2521.
5. Trials of War Criminals before the Nuremberg Military Tribunals. Vols I and II. The Medical Case. Washington, DC: U.S. Government Printing Office; 1948.
6. Jones J. Bad Blood. New York, NY: The Free Press; 1981.
7. Menikoff J. Law and Bioethics. Washington, DC: Georgetown University Press; 2001.
8. Cook D. Patient autonomy versus paternalism. Crit Care Med. 2001;29(2 Suppl):N24.
9. Quill TE, Brody H. Physician recommendations and patient autonomy: finding a balance between physician power and patient choice. Ann Intern Med. 1996;125:763.
10. Franklin C. Organ transplants: a doctor's dilemma. New York Times. January 3, 2000:A18.
11. P.G. Gebhard, 69, Developer of the term ‘informed consent’ [obituary]. New York Times August 26, 1997:A13.
12. Salgo v Leland Stanford Jr. Univ. Bd. Of Trustees, 154 Cal App. 2d 560, 317 P.2d 170 (1957).
13. Canterbury v Spence, 464 F2d 772 (DC Cir 1972), cert denied, 409 US 1064 (1972).
14. Cobbs v Grant, 8 Cal. 3d 229, 104 Cal Rptr. 505, 502 P.2d 1 (1972).
15. Truman v Thomas, 611 P.2d902 (Cal. 1980).
16. Dagi TF. Changing the paradigm for informed consent. J Clin Ethics. 1994;5:246.
17. Emanuel EJ, Emanuel LL. Four models of the physician-patient relationship. JAMA. 1992;267:2221.
18. Laine C. Patient-centered medicine. A professional evolution. JAMA. 1996;275:152.
19. Brock DW. The ideal of shared decision making between physicians and patients. Kennedy Inst Ethics J. 1991;1:28.
20. Whitney SN, McGuire AL, McCullough LB. A typology of shared decision making, informed consent, and simple consent. Ann Intern Med. 2004;140:54.
21. Bernat JL, Peterson LM. Patient-centered informed consent in surgical practice. Arch Surg. 2006;141:86.
22. Sprung CL. Informed consent in theory and practice: legal and medical perspectives on the informed consent doctrine and a proposed reconceptualization. Crit Care Med. 1989;17:1346.
23. Karlawish JH. Shared decision making in critical care: a clinical reality and an ethical necessity. Am J Crit Care. 1996;5:391.
24. Piper A Jr. Truce on the battlefield: a proposal for a different approach to medical informed consent. J Law Med Ethics. 1994;22:301.
25. Schweickert W, Hall J. Informed consent in the intensive care unit: ensuring understanding in a complex environment. Curr Opin Crit Care. 2005;11:624.
26. Lidz CW, Meisel A, Osterweis M, et al. Barriers to informed consent. Ann Intern Med. 1983;99:539.
27. Marzuk PM. The right kind of paternalism. N Engl J Med. 1985;313:1474.
28. Cassell EJ, Leon AC, Kaufman SG. Preliminary evidence of impaired thinking in sick patients. Ann Intern Med. 2001;134:1120.
29. Berg JW, Appelbaum PS, Grisso T. Constructing competence: formulating standards of legal competence to make medical decisions. Rutgers Law Rev. 1996;48:345.
30. Brody H. Shared decision making and determining decision-making capacity. Prim Care. 2005;32:645.
31. Franklin C, Rosenbloom B. Proposing a new standard to establish medical competence for the purpose of critical care intervention. Crit Care Med. 2000;28:3035.
32. Appelbaum PS, Grisso T. Assessing patients' capacities to consent to treatment. N Engl J Med. 1988;319:1635.
33. Menikoff JA, Sachs GA, Siegler M. Beyond advance directives: health care surrogate laws. N Engl J Med. 1992;327:1165.
34. Burchardi H. A surrogate for decision-making in the ICU. Intensive Care Med. 2001;27:1243.
35. Arnold RM, Kellum J. Moral justifications for surrogate decision making in the intensive care unit: implications and limitations. Crit Care Med. 2003;31(5 Suppl):S347.
36. Seckler AB, Meier DE, Mulvihill M, et al. Substituted judgment: how accurate are proxy predictions? Ann Intern Med. 1991;115:92.
37. Shalowitz DI, Garrett-Mayer E, Wendler D. The accuracy of surrogate decision makers: a systematic review. Arch Intern Med. 2006;166:493.
38. Suhl J, Simons P, Reedy T, et al. Myth of substituted judgment: surrogate decision making regarding life support is unreliable. Arch Intern Med. 1994;154:90.
39. Sulmasy DP, Terry PB, Weisman CS, et al. The accuracy of substituted judgments in patients with terminal diagnoses. Ann Intern Med. 1998;128:621.
40. Manthous CA, DeGirolamo A, Haddad C, et al. Informed consent for medical procedures: local and national practices. Chest. 2003;124:1978.
41. Sulmasy DP. Patients' perceptions of the quality of informed consent for common medical procedures. J Clin Ethics. 1994;5:189.
42. Davis N, Pohlman A, Gehlbach B, et al. Improving the process of informed consent in the critically ill. JAMA. 2003;289:1963.
43. Hopper KD, TenHave TR, Tully DA, et al. The readability of currently used surgical/procedure consent forms in the United States. Surgery. 1998;123:496.
44. Drazen JM. Volunteers at risk. N Engl J Med. 2006;355:1060.
45. Luce JM, Cook DJ, Martin TR, et al.: American Thoracic Society. The ethical conduct of clinical research involving critically ill patients in the United States and Canada: principles and recommendations. Am J Respir Crit Care Med. 2004;170:1375.
46. Silverman HJ, Luce JM, Lanken PN, et al. NHLBI Acute Respiratory Distress Syndrome Clinical Trials Network (ARDSNet). Recommendations for informed consent forms for critical care clinical trials. Crit Care Med. 2005;33:867.
47. Karlawish JH. Research involving cognitively impaired adults. N Engl J Med. 2003;348:1389.
48. Dunn LB. Enhancing informed consent for research and treatment. Neuropsychopharmacology. 2001;24:595.
49. Suntharalingam G. Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med. 2006;355:1018.
50. Weijer C. The ethical analysis of risk in intensive care unit research. Crit Care. 2004;8:85.
51. Bigatello LM, George E, Hurford WE. Ethical considerations for research in critically ill patients. Crit Care Med. 2003;31(3 Suppl):S178.
52. Luce JM. Is the concept of informed consent applicable to clinical research involving critically ill patients? Crit Care Med. 2003;31(3 Suppl):S153.
53. Fisher M. Ethical issues in the intensive care unit. Curr Opin Crit Care. 2004;10:292.
54. Finfer S. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350:2247.
55. Zawistowski CA. Ethical problems in pediatric critical care: Consent. Crit Care Med. 2003;31(5 Suppl):S407.
56. Cooke RW. Good practice in consent. Semin Fetal Neonatal Med. 2005;10:63.
57. Nelson RM. Ethics in the intensive care unit. Creating an ethical environment. Crit Care Clin. 1997;13:691.
58. Committee on Pediatric Emergency Medicine, American Academy of Pediatrics. Consent for emergency medical services for children and adolescents. Pediatrics. 2003;111:703.
59. Child Abuse Prevention and Treatment Act (CAPTA), 1974. U.S. Department of Health and Human Services.
60. Keeping Children and Families Safe Act of 2003 (P.L. 108-36). U.S. Department of Health and Human Services, Administration for Children and Families.
61. American Academy of Pediatrics, Committee on Bioethics. Religious objections to medical care. Pediatrics. 1997;99:279.
62. In re Karen Quinlan. 70 N.J. 10 (1976)—Supreme Court of New Jersey.
63. Illinois Health Care Surrogate Act of 1998. 755 ILCS 40.
64. Matter of Hofbrauer, 65 App. Div. 2d 108, 411 N.Y.S. 2d 416 (App. Div. 1978); 47 NY 2nd 648,419 NYS 2d 936, 393 NE 2d 1009 (1979).