Resolving Ethical Dilemmas: A Guide for Clinicians
3rd Edition

Patients usually agree with physicians’ recommendations. Such an agreement is particularly common in an acute illness, when the goals of care are clear, one option is superior, the benefits are great, and the risks are small. For example, a patient who suffers a wrist fracture almost always agrees to a cast. In such situations informed consent seems tantamount to obtaining the patient’s agreement to the proposed intervention. Physicians often speak of “consenting the patient,” implying that it is a foregone conclusion that the patient will agree.

Another view of informed consent is that patients have an ethical and legal right to be free of unwanted medical interventions. Hence, competent patients have the power to reject their physicians’ recommendations about care.

A broader view of informed consent holds that patients should have the positive right to choose among feasible options in addition to the negative right to refuse unwanted interventions. For instance, Case 3.1 involves not only the right to refuse surgery but also a choice between surgery and radiation therapy.

A still more comprehensive view is that informed consent is a process of shared decision making by the physician and patient (1). Both parties need to discuss the issues and reach a mutually acceptable decision. Through repeated discussions, physicians can educate patients about their conditions and the alternatives for care, help them deliberate, make recommendations, and to try to persuade them to accept the recommendations (2).

People want to make decisions about their bodies and health care in accordance with their values and goals. Decision-making power in health care is important because the stakes can be high. One court expressed this idea in sweeping terms, declaring, “Every human being of adult years and sound mind has a right to determine what shall be done with his own body (6).”

Patient choice should be promoted because in most clinical settings, different goals and approaches are possible, outcomes are uncertain, and an intervention might cause both benefit and harm (7). Individuals place different values on health, medical care, and risk. Some patients are wary about the side effects of medication, while others want to try risky therapies that promise better outcomes. Lung cancer patients are generally concerned about the short-term morbidity and mortality of surgery as well as long-term survival and cure (8). Older patients often prefer radiation therapy, which has a lower 5-year survival rate but also a lower likelihood of death during treatment. Physicians cannot accurately predict patients’ preferences. For example, patients with newly diagnosed cancer are more likely than physicians, nurses, and the general public to prefer intensive chemotherapy with little chance of cure (9).

The goal of medical care is to enhance patient well-being, which can be judged only in terms of the patient’s goals and values. The patient’s values are particularly important when various treatment approaches have very different characteristics or complications and involve trade-offs between short-term and long-term outcomes, when one of the options carries a small chance of a grave outcome, when the patient has unusual aversions toward risk or certain outcomes, and when there is uncertainty and disagreement among physicians (10). The choice between surgery and radiation in Case 3.1 has many of these characteristics. In addition, participation in decisions might have other beneficial consequences for patients, such as increased sense of control, self-efficacy, and adherence to plans for care.

Physicians might consider informed consent “a nuisance, an alien imposition of the legal system that must be tolerated… but can be dealt with in relatively mechanical ways, such as making sure
patients sign consent forms before major procedures (11).” Similarly, many patients are cynical about informed consent. In one study nearly 80% of patients said that the purpose of informed consent was to protect the physician (12).

Physicians need to discuss with patients information that is relevant to the decision at hand (Table 3-1). Most court decisions and legal commentaries use the term “disclose,” and, when summarizing legal doctrine, this book also uses this term. In general, however, we prefer the term “discuss” to emphasize that a dialogue between the physician and patient is preferable to a monologue by the physician.

Patients need to know the nature of the intervention, the expected benefits, the risks, and the likely consequences. In general, risks that are common knowledge, already known to the patient, of trivial impact, or very infrequent do not need to be discussed. For instance, patients do not need to be told about the nature of venipuncture, the rare risk of infection, or the minor discomfort of hematomas. On the other hand, for invasive interventions, courts have ruled that physicians need to discuss serious but rare risks, such as death or stroke. In Case 3.1, the physician should discuss with Mr. T the risks of surgical mortality, prolonged hospitalization, and long-term shortness of breath following surgery.

The risks of an intervention might include psychosocial as well as biomedical risks. For human immunodeficiency virus (HIV) and genetic testing, the pertinent risks are not the risks of venipuncture, but the risks of stigma and discrimination in employment or health insurance. Many states have enacted special provisions, requiring written informed consent and pretest counseling for HIV testing (13).

Patients also need to understand the alternatives to the proposed test or treatment and their risks, benefits, and consequences. In particular, the alternative of no intervention needs to be discussed. If a patient declines the recommended intervention, the physician needs to explain the adverse consequences of the refusal. In one case a court ruled that when a woman refuses a Pap smear, the physician needs to discuss how the test could diagnose cancer at an early stage and avert death through early treatment (14).

It is controversial whether physicians need to disclose information about themselves that affect patient outcomes (see Chapter 38). For example, patients might find it pertinent to know the outcomes of a surgical procedure at a given institution or by a particular surgeon, as contrasted to outcomes reported in the literature. Some states make such individualized outcome data for cardiac surgeons available to the public. In addition, the surgeon’s experience might be pertinent, because increased volume is associated with better outcomes for some operations and surgeons have a “learning curve” for new procedures. Another issue that patients might find pertinent is the role of trainees in their care, particularly with invasive or surgical procedures. As Chapter 36 discusses in detail, most patients want to know about the role of trainees and agree to their participation. The physician’s own health might also be pertinent to patient decisions. For instance, a patient might wish to seek a different provider if a physician is impaired because of alcoholism, dementia, or other medical conditions.

Although some courts have ruled that physician-specific experience needs to be disclosed for some operations, other courts have not (15). If disclosure is guided by what a reasonable physician would disclose, disclosure of such information would not be legally required. However, from the ethical perspective of providing patients pertinent information to make informed decisions, such disclosure is desirable.

Physicians must take the initiative in discussing information rather than wait for patients to ask questions. Patients, who have far less medical knowledge than physicians, might not even know what questions to pose. Empirical studies show that physicians often fail to provide sufficient information for patients to make informed decisions (16).

The extent of the disclosure will depend on the clinical context. For conditions such as appendicitis or pneumonia, or if there is only one realistic option and it is highly effective, relatively safe, and strongly recommended, a detailed consent process might be of little benefit to patients (17).

It is controversial whether physicians need to inform patients of alternatives for care that they do not believe are medically indicated. Obviously, physicians do not need to mention treatments that have no scientific rationale, would provide no medical benefit, or are known to be ineffective or harmful, such as laetrile for cancer; nor do physicians need to discuss complementary or alternative medicines that they do not accept as valid. However, physicians should inform patients of alternatives that other reasonable physicians would recommend. Thus, a physician who believed that surgery was the best approach to Mr. T’s lung cancer still ought to inform him about the option of radiation therapy.

Discussions about the proposed test or treatment and the alternatives should be conducted by the attending physician or by the physician performing the intervention, the proposed test or treatment, and the alternatives (18). Such discussions should not be delegated to nurses, medical students, or house officers. Some busy physicians who have already discussed an intervention with the patient during an office visit will ask a nurse or house officer to obtain the patient’s signature on a consent form in the hospital. Although this approach is understandable because it saves time, it might be problematic if the patient has questions that an inexperienced physician or a nurse cannot answer.

Patients must agree with the intended plan of care. For major interventions, such as surgery, obtaining explicit written authorization is standard. Written consent signals the patient that the decision is important. In ambulatory care, oral agreement to the plan of care is usual because the risks are lower and because patients can choose to discontinue medications (19,20).

Coercion and manipulation invalidate consent because they preclude free choices by patients. Coercion involves threats that are intended to control patients’ behavior and that patients find irresistible (21). An example is a threat to discharge a patient from the hospital if he does not agree with the recommended care. Manipulation of information might also thwart informed decisions. For example, physicians might misrepresent the patient’s condition or the nature of the proposed intervention. Coercion and manipulation contrast with persuasion, which is an attempt to convince the patient to act in a certain way by providing rational arguments and accurate data (21). Persuasion respects patient autonomy and, indeed, enhances it by improving the patient’s understanding of the situation and the options.

Certain constraints on patients’ choices are not coercive (4). The patient’s prognosis might be so grim that all alternatives are undesirable and the patient has no “real choice.” Warnings by the physician about the outcomes of choices or about the natural history of the illness are also not coercive because the physician makes no threat to bring about undesirable outcomes. Indeed, physicians would be remiss if they did not point out to patients the consequences of unwise choices.

Patients might lack the capacity to make informed decisions, as discussed in Chapter 10. For such patients, advance directives or appropriate surrogates should guide decisions (see Chapters 12 and 13).

Patients often do not recall information they have discussed with physicians, even basic information about the proposed treatment. In a study of cancer patients who had just consented to treatment, only 60% understood the purpose and nature of the treatment, 55% could name any complication, and 27% could name an alternative treatment. Furthermore, just 40% had read the consent form carefully (22). Many physicians have had similar experiences with patients and conclude that patients are unable to make decisions in an informed way.

Physicians, however, are partly to blame for patients’ poor comprehension. Doctors often use technical jargon that is incomprehensible to laypeople. Informed consent forms are usually difficult to read and understand (23). More importantly, physicians often fail to provide patients with basic information about interventions (19,24,25).

As shown in Case 3.1, some patients might not want to make decisions but instead defer to physicians or family members. This might be particularly true of patients from cultures where informed consent and autonomy are not as important as in the United States. Several older empirical studies indicate that although patients generally want information about their condition and treatment, many do not want to make decisions. In one study almost all hospitalized patients regarded the physician as the decision maker (24). Patients wanted information about plans for care in order to know what was going to happen and how to carry out those plans effectively. However, the only patients playing an active role in decision making were those with chronic diseases who needed to take medications regularly and to report changes in their condition. In ambulatory settings, although most patients want information about their illness, approximately one-half of the patients want to defer decisions to physicians (12,26). Many patients, however, want to play a more active role in decisions later, after they have experience with the prescribed medications (26). More recent studies suggest that about two-thirds of patients want to share decision making with physicians (27,28).

A common criticism of informed consent is that patients might make unwise or harmful choices. Some physicians fear that information about risks might cause patients to refuse medically beneficial interventions. Empirical studies, however, do not support these concerns. In one study of 104 refusals of inpatient treatment, none were attributed to disclosure of information (29). Fourteen patients, however, refused care because of inadequate information about tests or treatments.

Informed consent assumes rational, deliberate patient decision making. However, patients might make important decisions without such deliberation. For example, people who donate a kidney or part of their liver for transplantation often decide to do so as soon as they learn of the opportunity. For example, donors might say that they had to do it or that they had no choice. Moreover, they commonly decide to donate before learning about the risks of the procedure. Instead, their decision is driven by a commitment to specific people and to helping others.

Although patients might not use all disclosed information, it is nonetheless important that physicians give patients pertinent information. Even if a patient decided to pursue a course of medical care upon first hearing about it, that patient may still reconsider upon learning more information.

In some cultures patients might be expected to involve their families in medical decisions rather than make decisions as individuals (5). In some cultures women might traditionally be expected to defer decisions to their husbands or fathers. Clearly, physicians need to allow patients to involve
others in their medical decisions if they choose to do so. However, physicians must avoid creating an expectation that patients must involve others in decisions, because not all patients from a given culture might agree with traditional decision-making practices.

Physicians who do not obtain informed consent might be found liable in civil suits for battery or negligence (5). Battery is the harmful or offensive touching of another person. Physicians might commit battery if they carry out surgery without the patient’s consent or if the surgery exceeds the scope of patient consent (30). For instance, a physician might be liable for performing a mastectomy on a patient who had consented to only a biopsy, even if the intervention was medically appropriate, skillfully performed, and beneficial. This battery model, however, fits medicine poorly. Many cases do not involve physical touching of the patient, as when physicians prescribe drugs, fail to consider alternative approaches, or do not disclose information to the patient. In addition, battery requires that the physician intended to provide care without the patient’s consent. Most cases of malpractice, however, involve unintentional actions.

The modern approach to malpractice, which has supplanted the battery model, is to hold physicians liable for negligence. To be found negligent, the physician must breach a duty to the patient, the patient must suffer a harm, and the breach of duty must cause the harm. With regard to informed consent, the patient needs to prove that the physician failed to disclose a risk that should have been disclosed, that the patient would not have consented had the risk been discussed, and that the risk occurred and caused harm. A crucial issue in malpractice law, therefore, is what risks should be discussed.

Full or complete disclosure of all information that physicians know about a particular condition is impossible. Thus, the issue is not whether physicians should limit the amount and types of information they discuss with patients, but rather what information should be discussed or omitted.

Courts have used several standards to determine what information to disclose to the patient (5). A slight majority of states have adopted a professional standard: The physician must disclose what a reasonable physician of ordinary skill would disclose in the same or similar circumstances. This is equivalent to providing the information that colleagues customarily provide. The professional standard has been criticized because patients generally want more information than physicians customarily discuss.

Many states have adopted a patient-oriented standard for disclosure: Physicians should disclose what a reasonable patient in the same or similar situations would find relevant to the medical decision. Generally, this standard requires more disclosure than the professional standard and is more consistent with the goal of promoting patient decision making and choices.

Some individuals, however, might desire more information than the standard “reasonable” patient. For example, a carpenter might be particularly concerned that a new medication might impair his or her dexterity or alertness. To accommodate individual patient needs fully, a few states have adopted a subjective standard for disclosure: The physician must provide information that the individual patient would find pertinent to the decision. This subjective standard for disclosure is problematic in malpractice litigation. If a rare, undisclosed complication occurs, the patient might claim that he would not have consented to the intervention if the physician had mentioned that particular risk. In hindsight, it might be difficult to decide whether this assertion is plausible.

In some states statutes specify that certain risks need to be disclosed—for example, “brain damage” or “loss of function of any organ or limb (5).” State courts have also ruled on issues of disclosure. A California court ruled that a physician did not have to give a quantitative estimate of life expectancy to a patient with pancreatic cancer. His widow claimed that if he had known such information, he would have declined chemotherapy and arranged his business affairs (31).

The consent form documents that the patient agreed to treatment. In some states a signed consent form provides a legal presumption of valid consent (32). However, a signed consent form is not tantamount to informed consent because the discussion of the risks, benefits, alternatives, and consequences might be inadequate (5).

When patients lack decision-making capacity, an appropriate surrogate should give consent or refusal, following the patient’s previously stated preferences or his or her best interests (see Chapter 4).

In an emergency, delaying treatment to obtain informed consent might jeopardize the patient’s health or life. Legally, the courts have recognized a doctrine of implied consent: Because reasonable persons would consent to treatment in such emergency circumstances, physicians may presume that the patient in question also would consent. Few people would object to treating life-threatening emergencies, such as an impending airway obstruction in anaphylaxis, without the patient’s explicit consent. It is often possible to abbreviate the process of disclosure and consent in an urgent situation, rather than dispense with it altogether. In addition, the process of informed consent can often be initiated while the treatment is being started.

The emergency exception should not be used when informed consent is feasible or if it is known that a particular patient does not want the treatment. For example, terminally ill patients might have indicated that they do not want cardiopulmonary resuscitation (CPR). If such patients seek emergency care, the usual presumption that CPR should be initiated in case of cardiac arrest would not be valid.

Some physicians claim that consent is implied when a patient seeks care from a hospital or signs a general consent form upon admission. The implication is that informed consent for specific tests or treatments is unnecessary. However, this use of “implied consent” is unacceptable, because it allows physicians to administer any type of care they choose. When patients come to a hospital, they do not give physicians carte blanche. Most patients would probably agree to certain interventions, such as diagnostic testing, but would want to base further decisions on new information.

Physicians may withhold information when disclosure would severely harm the patient or undermine informed decision making by the patient (1). For example, a patient might be depressed and have a history of previous suicide attempts in response to serious medical diagnoses. Telling such a patient he has cancer might provoke another suicide attempt. However, the concept of therapeutic privilege needs to be sharply circumscribed. The likelihood that the patient will feel sad does not justify withholding a serious diagnosis. Therapeutic privilege also does not allow the physician to “remain silent simply because divulgence might prompt the patient to forego therapy the physician feels the patient really needs (33).”

Patients like Mr. T in Case 3.1 might not want to participate in making decisions about their care. Ethically and legally, patients’ requests to waive the right of informed consent should be respected. Self-determination would be undermined if patients were forced to participate in decision making
against their wishes. Shared decision making entitles patients to participate actively in health care decisions but does not require them to do so (4). To be ethically valid, a waiver of informed consent must itself be informed. Patients must appreciate that they have the right to receive information and to make decisions about their care. Physicians can give patients the option to not receive information or make a decision without thereby suggesting that they should do so. Physicians must keep in mind that patients might later want to participate more actively in decisions.

Physicians can encourage patient involvement in decisions, even with patients like Mr. T in Case 3.1 who defer to their judgment. Mr. T’s doctor might say, “I’d be glad to tell you what I think is best for you. But first I need to understand what is important to you.”

The guideline of beneficence requires physicians to help patients make decisions that are in their best interests (see Chapter 4). In addition to providing information, physicians should help patients deliberate about their choices.