5
Informed Consent

Medicine has changed, in many respects radically, over the last twenty-five years. These changes have occurred not only in health policy but also in the conceptual underpinnings of the relations of physicians to their patients and to the institutions of medical practice. These changes in health law and policy have created a demand for a new theory of ethics in medicine that I have called just doctoring. Just doctoring properly should concern not only ideal relations between two individuals hut also determinations of the best possible shape for new health policy initiatives. From the physician's ethical viewpoint, new policies should conform to a liberal public morality and yet reaffirm the cooperation and egalitarianism central to the best possible liberal state and essential to the practice of medicine, Physicians must abandon their distrust of (indeed hostility toward) matters of law and politics—a posture that has belied their continuous efforts to maintain control over the practice of medicine—and now become activists. They must speak on behalf of patients when legal or policy reforms arc introduced.

In this chapter, we will examine the issue of informed consent in medical care, a subject of great concern for medical ethics. Reviewing informed consent will allow us to trace the evolution of a legal initiative that has helped to reintegrate medicine within the liberal state, laying the foundation for just doctoring.

The history of informed consent in this country is one of patients attempting to overcome the paternalistic attitudes of doctors generated by the inequality of knowledge that exists between them. The demand for informed consent has been justified in terms of rights. Patients

want to maintain their right to refuse treatment, and they do not want to have their liberty usurped by physicians.

Consider again the case of Mr. Z., the patient who was suffering from a bleeding duodenal ulcer. When he was first admitted to the hospital, Dr. A. was uncertain about the source of his bleeding. She recommended that he undergo endoscopic examination. This procedure entails passing a long tube of fiberoptic equipment down the esophagus and into the stomach. Its risks are generally minor, but it could potentially cause perforation of the stomach, necessitating emergency surgical intervention.

Mr. Z. was hesitant; he did not want to stay in the hospital. He did not like the idea of seeing doctors, and he felt he would forfeit personal autonomy once he entered the hospital. Dr. A. did not want to cause him further anxiety, but she did want the endoscopy performed. Therefore Dr. A. was reluctant to detail all the potential risks of the procedure.

One route Dr. A. might have chosen would have been to withhold some information concerning the risks of endoscopy, her justification being that sick people often act irrationally and that she had to take this irrationality into account. This is called "therapeutic privilege," an expression of physicians' paternalism toward sick patients. Dr. A. could have invoked therapeutic privilege and withheld information.

I would argue that physicians are less likely to invoke therapeutic privilege now than they were twenty-five years ago. Patients have defended their rights by suing physicians who have failed to inform them fully of the risks and benefits of a treatment. A review of the legal history of informed consent reveals the manner in which the law has reasserted control over medicine, overcoming physicians' resistance to supervision. It also reveals that physicians have an ethical duty to pursue further efforts to inform patients fully.

While there are many functions of informed consent suits, my attention is restricted to the patient's assertion that her negative freedom he respected, in keeping with the values of the liberal state.1 In addition, I will continue to assert that the law reflects the public morality that underlies the modern liberal state.2 Following Alan Donagan, I concentrate on suits that ensure that the "physician proposes and the patient decides" in a manner that is consistent with the liberal principles of society and health care institutions.3 ,4

Typically, a suit by a patient to whom a physician has failed to give adequate information when seeking consent to an operation is part of torts, or personal injury law. More precisely, such suits are usually

tried as cases of negligence; the patient asserts that the physician failed to meet the standard of care expected of the profession. In such cases, courts expect to hear professionals' expert testimony on the quality of the information provided to the patient in order to evaluate the consent given by the patient.5

Not all informed consent suits are framed this way. Some are heard as cases of battery. In these suits, the plaintiffs claim that they were victims of unconsented touching. There is no need in such suits for expert testimony, as there is no determination of a standard of care. Battery suits in effect state that the physician is liable even if she can find other physicians who agree that the amount of information she provided was appropriate. (In the real world, most plaintiffs will raise both points when they sue, and the court has some prerogatives over the grounds it thinks are best considered.) The tension between battery and negligence cases informs much of the recent history of informed consent suits. Analyzing cases from this perspective can help to reveal the manner in which courts have thought about informed consent and the rights of patients vis-à-vis doctors' ethical duties.

The Early Evolution of Legal Principles of Informed Consent

Anglo-American common law holds that no medical procedure can be performed on a person unless that person has given her informed consent. This prohibition is based on the respect that the law accords one's body, a respect that coincides with the concern for negative freedom, which lies at the heart of liberalism. The prohibition is an ancient one, as well.6 Until the early part of this century, it was usually expressed in terms of battery. As noted above, a battery is an intentional touching to which no consent has been given.7 As Fowler Harper and Fleming James have related, the battery doctrine is the patient's interest "in the physical integrity of the body, that it be free from physical contact," and "the purely dignitary interest in the body, that it be free from offensive contact."8 Katz has discussed the implications of the battery standard:

Battery, which evokes frightening visions of physical violence, is an uncompromising remedy, allowing few defenses. Only a very restricted question is asked: Did the patient know and agree to what was going to be done to him?

If the answer is no, law docs not require the patient be physically damaged by the intervention.... Nor is proof required that the patient's probable conduct in submitting to the touching would have been different, had the doctor fulfilled his duty to disclose the nature of the operation.9

Thus the battery doctrine provides unmitigated protection of the patient's exercise of free choice in line with the tenets of negative freedom.

Since the 1930s, courts have largely abandoned the battery standard in informed consent suits—and have adopted in its place the standard of negligence. This standard incorporates principles of medical professional negligence, treating failure to inform as negligent nondisclosure.10 The negligence standard differs from the battery standard in several important ways. First of all, it tics informed consent to professional norms. Often practitioners must testify as experts in order for the court to discern what negligent nondisclosure is. As Arnold Rosoff has noted, the important question is whether the physician has told the patient everything that an "average reasonable practitioner" would have revealed in a set of similar circumstances (as we shall see, the reasonable practitioner has been replaced, in some jurisdictions, by the reasonable patient).11 The second difference is that there is no purely dignitary damage in negligence: physical injury must be demonstrated. The informed consent suit that is tried as a case of negligence is thus much more dependent on professional standards and expertise than one that is tried as a case of battery.

The laws regarding battery and negligence are substantially different. The former directly protects the patient's person from interference; it guarantees negative freedom. The latter turns on the customs of a profession, with less of an emphasis on the patient's liberty. Battery concentrates on interference with a person's liberty without consent; negligence concentrates on a professional's transgression of professional norms. These differences should not be minimized by characterizing battery and negligence as two points on the same spectrum. Marcus Plante for example, has suggested that negligence applies to failure to warn about a collateral hazard of a procedure, while battery applies to a more central hazard.12 This analysis, however, misses the key tension within informed consent suits: the conflict between the patient's negative freedom and the prerogatives the physician seeks to maintain.

The decision of Pratt v. Davis in 1905 is representative of the judiciary's use of the battery standard when physicians fail to inform

patients.13 In this decision, Justice Brown of the Illinois Appellate Court affirmed that damages should be awarded to Mrs. Davis as a result of Dr. Pratt's failure to get her consent to a hysterectomy. Mrs. Davis had been afflicted by epilepsy and consulted Dr. Pratt, who told her that her lacerated womb would have to be repaired. He then proceeded to remove both ovaries and uterus, sterilizing the plaintiff and causing her great emotional pain.14

Pratt, the appellant, argued that "when a patient places herself in the care of a surgeon for treatment without instructions or limitations, upon his authority, she thereby in law consents that he may perform such operation as in his best judgment is proper and essential to her welfare."15 The court refused to accept this definition of the physician's prerogatives. Justice Brown's reasoning is worthy of lengthy quotation:

On the contrary, under a free government at least, the citizen's first and greatest right, which underlies all others—the fight to the inviolability of his person, in other words, his right to himself is the subject of universal acquiescence, and this fight necessarily forbids a physician or surgeon, however skillful or eminent, who has been asked to examine, diagnose, advise and prescribe (which are at least the necessary first steps in treatment and care) to violate without permission the bodily integrity of his patient.16

In this decision, then, we see the juxtaposition of opposing sets of values. Pratt's defense is the medical ethical principle of mutual partnership, reformulated such that the doctor's autonomy is emphasized. Mrs. Davis's suit is based on her right to her self, the negative freedom guaranteed to all in a liberal state. The suit itself was tried as a case of battery, and Mrs. Davis's rights were upheld accordingly.

The Pratt v. Davis decision appears to have assumed that the grounds for a battery suit were self-evident. In another case that year, a Minnesota court was forced to consider explicitly whether the battery. rubric applied to consent suits. In Mohr v. Williams, defendant had decided that the plaintiff's right ear was more diseased than his left and operated accordingly. The plaintiff had, before undergoing anesthesia, agreed only to an operation on the left ear. The court allowed that "the operation was in every way successfully and skillfully performed."17 The defendant argued that "under the facts, an action for assault and battery will not lie.... There is a total lack of evidence showing or tending to show malice or an evil intent on the part of defendant, or that the operation was negligently performed."18

The court addressed this contention directly: "Every person has a

right to complete immunity of his person from physical interference of others... and any unlawful or unauthorized touching of another... constitutes assault and battery."19 Judge Brown was not swayed by the argument that the right ear was diseased and the operation a success. The plaintiffs liberty to choose the operation had been infringed—a battery had occurred. Typical notions of negligence did not apply.

Benjamin Cardozo reemphasized this point nearly ten years later in the case of Schloendorff v. New York Hospital.20 The trial court had found that Mrs. Schloendorff was operated on without her consent by Drs. Stimson and Bartlett. Cardozo noted that "in the case at hand, the wrong complained of is not merely negligence. It is trespass. Every human being of adult years and sound mind has a right to determine what shall be done with his own body; and a surgeon who performs an operation without his partner's consent commits an assault, for which he is liable in damages."21 The court emphasized that the issue was battery, and that physicians must get a patient's full consent before operating.

The underlying principle in all these decisions was that the patient's negative freedom must be protected, in accordance with the principles of a liberal state. It is notable that all of these cases occurred just before the AMA began to consolidate the position of the medical profession in health care.22

This is not, however, to say that courts during this period were unanimous in their support of a battery standard and individual patient's rights. In Bennan v. Parsonnet, an appellate court reversed a jury verdict that awarded damages to the plaintiff after Dr. Parsonnet operated on his right groin.23 Dr. Parsonnet was to operate on the other groin, but decided after the patient was anesthetized that the right hernia was more serious. The court thought that the employment of anesthesia was critical and that

when a person has selected a surgeon to operate upon him and has appointed no other person to represent him during the period of unconsciousness that constitutes a part of an operation, the law will by implication constitute such surgeon the representative pro haec vice of his patient and will, within the scope to which such implication applies, cast upon him the responsibility of so acting in the interest of the patient that the latter will receive the flail benefit of that professional knowledge and skill.24

In this case, the situation was quite similar to that of Mohr v. Williams, yet the outcome was directly opposite. Here Judge Garrison appeared

to focus on the inequity of the knowledge and introduction of technology into the doctor-patient relationship. As a result, he asserted that the patient was to trust his physician, and the physician was to act on the patient's behalf, The reasonable expectation that the patient would so trust the surgeon was not addressed in the decision.

There are several unargued assumptions within the decision. Judge Garrison also seemed to believe that the doctor's actions would be guided by an ethical code that prohibited abuse of the patient's trust. Moreover, this code was protected and endorsed by the court, which found that a doctor's ethical code was not to be threatened by suits asserting a patient's negative freedom, especially when no physical damages occurred. This proposition was formulated somewhat un-clearly in the court's decision:

This concluding suggestion may perhaps be ethical rather than legal, but it does not seem in good morals a patient ought not, in his efforts to obtain a money verdict, be permitted to repudiate the sound judgment exercised in his behalf by the surgeon of his choice in whose judgment, had he been capable of consulting, he would have unquestionably concurred.25

In other words, instead of protecting the patient's liberty to choose, as in Mohr, the court here exhorts the patient not to abuse the trust relationship with the physician.26 The influence of traditional medical ethics on the judge's interpretation of the tortious act is apparent.

Although there are few cases to support any specific propositions, it seems that battery claims were slowly abandoned over the period from 1920 to 1960. A typical case is that of Hunt v. Bradshaw, decided in 1955.27 Hunt was the victim of an occupational accident in which a piece of steel lodged in his brachial plexus, a large set of nerves in the shoulder and chest. He was told by Dr. Bradshaw that a simple operation could remove the metal and prevent its migration toward the heart. Hunt agreed, but as a result of the operation, he was partially paralyzed for life. Hunt charged that he had not been warned of the risks involved in the operation. The appeals court's decision does not mention battery, but drily notes that "the plaintiff's case fails because of lack of expert testimony that the defendant failed, either to exercise due care in the operation, or to use his best judgment in advising it."28 The decision is based on the fact that there was no expert testimony that labeled Dr. Bradshaw as negligent. The judicial emphasis on patient liberty and serf-determination witnessed in the earlier cases that were tried as battery was absent entirely in this decision.29

The trends noted in the cases above were consolidated in Salgo v. Leland Stanford University Board of Trustees.30 In this case, the California appellate court considered the relationship of Salgo's paraplegia to an abdominal aortography performed by Dr. Gerbode. Salgo sued for malpractice and also claimed that he was not informed of the serious risks of aortography. Judge Bray's decision consisted in large part of a discussion of the trial court's mishandling of the res ipsa locquitar instruction. (Res ipsa locquitar means literally "the thing speaks for itself"—no testimony is needed as to how negligence occurred if the outcome is in fact the product of negligence.)

The court then turned to the informed consent issue. While criticizing the trial court's broad instruction on the duty to disclose, Judge Bray nonetheless suggested that "a physician violates his duty to his patient and subjects himself to liability if he withholds any facts which are necessary to form the basis of an intelligent consent by the patient to the proposed treatment."31 This language would suggest that the court planned to embrace again the battery standard rather than negligence. Yet Bray tied the mandate to inform to the physician's sense of duty and self-imposed ethics: "At the same time, the physician must place the welfare of his patient above all else and this very fact places him in a position in which he must choose between two alternative courses of action."32 As the court outlined, one course of action was to explain every risk that may attend a procedure; the other was to recognize that each patient is different, and "that in discussing the element of risk a certain amount of discretion must be employed consistent with the full disclosure of facts necessary to an informed consent.33

Judge Bray had hopelessly intermixed opposing positions. First of all, he stated what sounded like a strict battery standard, one that would respect the negative freedom of the patient. As Katz has noted, "The court appeared to recognize for the first time that physicians might be held liable for failure to disclose important information beyond the ancient requirement of revealing the nature of the operation."34 However, Bray went on to pay tribute to the "patient comes first" core of medical ethics and appears to have made the bottom line that patients have to trust doctors' discretion. Thus, instead of advocating the patient's self-determination, he ceded authority to the physician, largely because of the physician's duty to place the patient's welfare above all else. In its initial use, as the expression was coined by the Salgo court, the term "informed consent" was thus tied to the

physician's discretion at the core of medical ethics, rather than to a battery standard that would better protect the patient's liberty. But this case also signaled to many lawyers that courts were willing to hear the patient's claims so long as there was expert testimony available to support the plaintiff's position. The Salgo decision, with its willingness to hear rights claims and yet accept the role of medical ethics, revolutionized informed consent law.

The notion of informed consent was further transformed by the decision of the Kansas Supreme Court in Natanson v. Kline.35 Justice Schroeder, writing in 1960, considered the plight of Mrs. Natanson, who had received a large amount of cobalt radiation while being treated for breast cancer by Dr. Kline. Among Mrs. Natanson's claims was Dr. Kline's alleged failure to warn her about the treatment. Schroeder's survey of the law led him to conclude that the difference between battery and negligence was that the former is intentional, the latter unintentional. As a result, he concluded that negligence was the proper standard for an informed consent case.36

Schroeder described the dilemma he faced. On the one hand he allowed that "there is probably a privilege, on therapeutic grounds, to withhold the specific diagnosis where the disclosure of cancer or some dread disease would seriously jeopardize the recovery of an unstable, temperamental or severely depressed patient."37 In other words, the doctors must have some discretion to employ their ethical reasoning regarding the good of the patient. On the other hand, he paid tribute to the negative freedom of the citizen in the liberal state:

Anglo-American law starts with the premise of thorough-going self-determination. It follows that each man is considered to be master of his own body and he may, if he be of sound mind, expressly prohibit the performance of lifesaving surgery or other medical treatment. A doctor might well believe that an operation or form of treatment is desirable or necessary, but the law does not permit him to substitute his own judgment for that of the patient by any form of artifice or deception.38

Justice Schroeder's solution was to quote extensively from Salgo, and to accept the deference to the physician's ethical code that Justice Bray had set forth: "In our opinion the proper rule of law to determine whether a patient has given an intelligent consent to a proposed form of treatment by a physician was stated and applied in Salgo v. Leland Stanford University Board of Trustees. "39 Yet Justice Schroeder went a step further and stated that the extent of an informed consent was to

be determined by expert testimony. The duty to disclose "is limited to those disclosures which a reasonable medical practitioner would make under the same or similar circumstances. How the physician may best discharge his obligation in this difficult situation involves primarily a question of medical judgment."40 In effect, this meant that physicians were to tell courts of the extent of their "patient comes first" (doctor decides) ethics, and that courts were to accept this as expert testimony. Justice Schroeder thus sanctioned a more complete deference to medical ethics than did Justice Bray, in that he defined revelation of ethics as a matter of expert testimony.

Natamon v. Kline rationalized informed consent to a certain extent. Informed consent suits were, according to Schroeder, to be strictly analogous to malpractice suits. The difference between the two types of suits was that malpractice involved expert testimony on what was medically prudent and reasonable, whereas informed consent involved expert testimony on what was ethically prudent and reasonable. The profession was to decide what it would countenance in terms of the amount of information a patient needed to make a decision about a procedure. Thus Justice Schroeder significantly consolidated the power of the physician in the doctor-patient relationship by ruling that the ethical code that legitimated physician authority in the relationship was to be interpreted only by physicians.

It is ironic that two suits, Salgo and Natanson, that attempted to frame the doctor-patient relationship in terms of an encounter between right-bearing individuals resulted in an exaggeration of the physician's authority within that relationship. Following Natanson, doctors alone were to determine the extent of the duties shared by the doctor and patient. In addition, however, these cases enabled courts to hear patients' complaints about lack of informed consent without challenging the profession too directly. Cases tried as battery tended to make judges nervous, in that any battery claim could be countered by the physicians' assertion that lay people (and judges) should not tell doctors how to practice medicine. Use of the battery standard would have forced judges to challenge the contentions that if physicians told the whole truth, patients would act irrationally. Thus Salgo and Natanson allowed patients' rights claims but subordinated those claims to the powerful defense of medical ethics and doctor's therapeutic privilege. Paternalism was alive and well.

Informed Consent Suits in the Liberal State

According to Katz, Natanson v. Kline defined the law in almost all jurisdictions for twelve years.41 There were, however, some decisions that provided new refinements. In Shetter v. Rochelle, for instance, Judge Molloy more firmly tied the notion of informed consent to a physician's ethical duty: "[Courts] have reasoned from the fact that the relationship between a physician and his patient is one of trust and confidence [to] the conclusion that there is a duty to disclose all facts relative to the treatment prescribed.... This seems to this court to be a non sequitur."42 The patient's informed consent suit, demanding treatment as a free person with personal liberties, was thus turned away with a reminder that informed consent was a matter of the physician's duty. The patient's claim to negative freedom, to do as he wishes with his own body in accordance with the tenets of liberalism, is turned on its head: "The doctor's loyalty to the best interests of the patient must be given consideration in the formulation of a rule of law pertaining to disclosures and ... a breach of duty on the part of the physician is not made out by merely establishing a nondisclosure of possible failures inherent in a contemplated surgical procedure."43

Not all jurisdictions minimized the critical role of patients' tights. In the case of Mason v. Ellsworth, the court argued that "the basis of the theory of informed consent is both the patient's right to be reasonably informed of all material elements, including risks, which may affect his consent to the procedure, and the physician's duty to inform his patient in a medically sound fashion."44 The problem was that expert testimony determined the extent of the physician's duty, and this testimony usually sealed the decision. As Katz put it, "Since physicians ... are generally not committed to patients' participation in the medical decision-making process, recourse to a professional standard of disclosure had to stifle the ... call for self-determination."45 But Mason was the first major indication that courts might be willing to overturn physicians' therapeutic privilege and to provide greater respect for patient autonomy.

Indeed, Canterbury v. Spence explicitly reasserted patient rights. The decision by Judge Spottswood Robinson can be seen as a direct challenge to physicians' desire to maintain their positive freedom and pro-

fessional prerogatives.46 The case concerned a patient who was not warned of the possibility of paraplegia following back surgery. Judge Robinson began by citing Cardozo's self-determination quotation in Schloendorff as his root premise.47 From this premise, he argued that a patient's cause of action should not be dependent on a professional standard. "Respect for a patient's right of self-determination on particular therapy demands a standard set by law for physicians rather than one which physicians may or may not impose upon themselves."48 He recognized, moreover, that the ethical standard upon which the physician's duties were based was not inherently medical. Evaluation of behavior in line with an ethical code does not require special knowledge. "The decision to unveil the patient's condition and the chances to use mediation, as we shall see, is oft times a nonmedical judgment and, if so, is a decision outside the ambit of the special standard."49 Judge Robinson's point was that lay people can decide whether or not the "patient comes first," or whether the doctor treated the patient with due respect. Since all citizens are right-bearers in the liberal state, each should be able to recognize respect for autonomy. It follows that special medical knowledge has nothing to do with ethical duty. This was, of course, a radical departure from the Salgo-Natanson line.

There is some reason to believe that Judge Robinson feared his decision went too far. He was concerned that ethics might be confused with medical knowledge and that physicians would be abused by lay decisions. So he hedged his decision: "When medical judgment enters the picture and for that reason the special standard controls, prevailing medical practice must be given its due."50 Moreover, Judge Robinson refused to draw any sharp lines with regard to situations in which medical expert testimony would be required. "There is no bright line separating significant from insignificant; the answer in any case must abide a rule of reason."51 Thus, while recognizing that the ethical relationship that exists between doctor and patient can be grasped by the nonexpert, Robinson refused to prohibit medical testimony and, in effect, moved away from the Natanson disposition of informed consent suits less radically than he might have.

Nonetheless, Robinson's decision did significantly diminish the physician-centric approach to informed consent. First of all, as noted above, Justice Robinson attempted to differentiate between ethical and medical decisions to disclose, arguing that the former was not a matter over which physicians had the final say. Medical ethics, insofar as it is

ethics, could properly be considered by judges and jurors, reflecting on the public morality of the liberal state. Only when esoteric medical facts came to the fore was the decision on disclosure to be solely in the hands of medical experts. Second, Robinson dealt with the question of the extent of disclosure from the point of view of the patient. Rather than relying on a standard based on what a prudent physician would have told a patient, Judge Robinson's standard is based on the reasonable patient, and what that patient is likely to need to know and what significance the patient is likely to attach to certain risks.52

Both of these points approach the key problem of informed consent suits: the power of the physician vis-à-vis the patient and the role of professional medical ethics. As we have seen, the ends of traditional medical ethics were to generate patient trust in a physician. An informed consent suit asserts that this trust was breached and that a patient's rights must be respected. Judge Robinson's description was thus quite insightful. By separating ethical and technical expertise, he teased out the role of medical ethics in informed consent suits. His emphasis on the patient's viewpoint in deciding what is to be disclosed served the same purpose: it denied reliance on physicians' ethical expertise and on their habits of disclosure that were steeped in notions of paternalism, which are antithetical to the liberal state.

The medical profession, like all professions, has a monopoly over a critical body of knowledge. Physicians must thus explain some technical issues to patients, especially when they are proposing therapy alternatives. To the extent that these proposals are incomplete, the traditional notion of the "patient comes first" ethic held that physicians would always act in the best interest of the patient. Indeed, many physicians believed that it was better to act on behalf of the patient than to discuss matters with him.

When a patient determines that medical paternalism has led to therapy he would not have chosen, the plaintiff-patient must act as a liberal citizen acts, arid assert his rights, which protect his negative freedom. Canterbury, for example, argued that even though Dr. Spence may have believed he was acting ethically by withholding certain critical information from him, the doctor should not have failed to warn him of the risks of a laminectomy. Canterbury himself would have preferred to have made an informed decision about the laminectomy. He felt that Spence should have made an effort to get past the issue of authority in their relationship—an issue based on the inequity of knowledge between doctor and patient. When this did not occur, he

believed that his only course was to assert his rights and the negative freedom they guarantee. He, like the plaintiff in any informed consent suit, sought a redefinition of the doctor-patient relationship in terms of an encounter between two rights-bearing citizens, each with a sphere of negative freedom that must be respected. The court agreed.

In lieu of a battery standard, however, it is not possible to completely prohibit the paternalistic aspects of the doctor-patient relationship. This is clear in Canterbury v. Spence. Despite Judge Robinson's intentions, the medical community retains some control with regard to the extent of informed consent. As noted, the decision allowed that medical expert testimony was still necessary to determine the physician's duty. to disclose. In essence, this meant that a doctor may justify his actions by arguing he was duty bound to do the best for the patient, and that he determined that the best for this patient was less than informed consent. Moreover, Canterbury held that the physician was still to decide what a reasonable patient needed to know before consenting to a procedure. This allowed a doctor to continue to assume a paternalistic posture that is atypical in the liberal state—and to justify this in terms of a special relationship with the patient. The individual whose negative freedom has been violated is thus given less than total regard throughout a trial based on the Canterbury v. Spence paradigm, Judge Robinson's dicta notwithstanding.53 Nevertheless, decisions like Judge Robinson's do restrict physician prerogatives. Canterbury v. Spence greatly expanded the notion of informed consent and signaled to the medical profession that patient autonomy was to be respected and that physician paternalism was not to be tolerated. Informed consent decisions of this sort point to a new understanding of doctor-patient relationships—they demonstrate an evolution of patients' rights and a reintegration of medical practice into the liberal state.

Since Canterbury v. Spence, courts have been forced to consider the lay standard and judge its worth compared to the professional standard. The Canterbury decision thus pervades legal reasoning on informed consent, even if it is not wholly embraced by every court. Indeed, there is a great deal of inconsistency in rulings about informed consent. Consider, for example, the case of Haven v. Randolph. This case involved a malpractice suit to which was appended an informed consent claim.54 The plaintiffs alleged that the risks of an aortogram were not fully explained to the patient's parents. The District of Co-

lumbia Federal District Court decision, handed down one day prior to Judge Robinson's Canterbury v. Spence decision, emphasized the importance of expert medical testimony to determine the limits of disclosure.55 There was no mention of lay testimony with regard to the duty to disclose. The decision was thus fundamentally at odds with Canterbury v. Spence. Moreover, the district court decision was approved on appeal by the same federal circuit court of appeals that had ruled on Canterbury.56 In a later decision in the D.C. Circuit Court, Judge Robinson noted, "We think it clear that nothing in Haven undercuts Canterbury in the least."57 Thus the courts appeared to be willing to countenance some ambiguity in decisions regarding informed consent.

The sometimes schizophrenic posture of the courts is well demonstrated by the decision of Cobb v. Grant.58 Justice Mosk, writing for the California Supreme Court, penned the first half of a classic informed consent decision—negligence based on and dominated by the testimony of medical experts. He noted that "battery theory should be reserved for those circumstances when a doctor performs an operation to which a patient has not consented."59 He then went on to consider the extent of the physician's duty to disclose. Justice Mosk allowed that "even if there can be said to be a medical community standard as to the disclosure requirement for any prescribed treatment, it appears so nebulous that doctors become, in effect, vested with absolute discretion."60 Be this as it may, Justice Mosk could not limit the physician's discretion, having accepted a negligence standard. Negligence claims require expert testimony. Justice Mosk could only advocate "full disclosure, tempered by the fact that 'a mini-course in medical science is not required,'"61 and note that there was no duty to disclose "when it is common knowledge that such risks inherent in the procedure are of very low incidence."62 Justice Mosk's decision thus further reflects judicial willingness to hear patients' rights claims, and to compromise the positive rights claimed by physicians. It also reflects courts' hesitance to force physicians into a position in which they must reveal every possible hazard of every therapy.

It would be an exaggeration to suggest that physicians' positive rights have been totally removed by recent court decisions on informed consent. Nonetheless, courts have straightforwardly asserted that doctors must inform patients of alternative treatments.63 In similar development, a federal circuit court has ruled that a hospital must

inform patients of the limits of the care offered by an institution.64 The momentum of patients' rights is slowed only by the courts' questions about how to deal with the intricacies of medical technology:

The matters involved in the disclosure syndrome, more often than not, are complicated and highly technical. To leave the establishment of such matters to lay witnesses, in our opinion, would pose dangers and disadvantages which far outweigh the benefits and advantages a "modern trend" rule would bestow upon patient-plaintiffs. In effect, the relaxed modern trend rule permits lay witnesses to express, when all is said and done, what amounts to medical opinion.65

Despite this caveat, the modern trend is that the medical profession's prerogatives no longer totally determine the parameters of the doctor-patient relationship. Courts have asserted that patients' autonomy must be respected.

The final rebuilding of patient autonomy must be left to the profession itself. Up to this point, the interest of many medical ethicists in an autonomy-based standard has no doubt been secondary to the influence courts have had. But courts cannot and will not usurp the professional role of physicians as healers. They are too afraid that they might get the technical aspects of medical practice wrong and cause great harm. They have initiated and pushed along the task of reintegrating the doctor-patient relationship into the liberal state by insisting on informed consent. Now physicians must complete the task by supporting a set of ethical principles that reiterate patient rights and that seek to maintain the autonomy of the patient (principles that I argue arc embodied in the theory of medical ethics as just doctoring). Physicians have slowly come to realize that there is no role for paternalism in medicine, even when it is employed in the name of the patient's best interest.

Over the last ten years, informed consent rulings by courts have become murkier.66 This is largely attributable to the fact that informed consent suits were perceived by the profession as a major contributor to what it felt was a malpractice crisis (an issue to which we will return in chapter 6). Organized medicine has put pressure on state legislatures to reinstate the professional standard. According to A. Meisel, most of the resulting statutes have been specifically designed to proscribe action by plaintiffs,67 and there is evidence that attorneys for medical societies have actually drafted laws in some states. In any case,

fourteen state legislatures have adopted a professional standard for evaluating disclosure, while only two allow lay evaluation.

At the same time, state courts have continued to endorse the patient perspective as the best legal standard to guide disclosure.68 Moreover, some courts may now be willing to grant remedies, including money damages, to patients when the only harm suffered is failure to receive full disclosure.69 In essence, this is equivalent to compensating for infringement on negative freedom. Perhaps most important, there is evidence that physicians are remarkably less likely to mislead or misinform patients now than they were twenty years ago.70 Informed consent is now the rule, rather than the exception, and is a central tenet of medical ethics as just doctoring (the one exception perhaps being resuscitation status,71 a matter we return to in chapter 8).

In summary, the judiciary's insistence that informed consent suits be considered as part of the doctrine of negligence is best seen as a tacit understanding of the role of medical ethics. Yielding to medical expertise meant not only technical expertise, but also ethical expertise. When courts asked doctors to testify as to the extent of a plaintiff’s informed consent, they were asking those doctors to comment on the defendant's fulfillment of his duty. Thus physicians were deemed to be experts on the depth of a mutual trust relationship, and suits were possible only when other physicians were willing to say that medical ethics had been violated.

The patient, of course, wanted his negative freedom preserved. Perhaps the most direct way to have accomplished this goal would have been to return to the battery standard. This standard, as applied in Pratt v. Davis and Schloendorff v. New York Hospital, effectively protected the patient's right to self-determination. As I have argued, the battery standard is uncompromising in its protection of the individual from unconsented touching. It also prohibits the use of professional testimony, the vehicle through which the patient's purest demands for negative freedom have been so effectively blunted. Thus, with a forceful simplicity, the battery standard enforces the patient's claim to autonomy, in keeping with the tenets of the liberal state.

Courts were, however, generally loath to embrace the battery standard, especially since physicians so often thought they were acting ethically in keeping frightening information from patients. The battery standard even strikes some rights advocates as inappropriate for determinations of doctor-patient relations. The reasons for this were

summarized in the case of Trogun v. Fruchtman.72 Justice Wilkie noted first that informed consent suits do not "fit comfortably within the traditional concepts of battery or the intent to unlawfully touch the person of another."73 Second he noted that "the failure to inform is probably not an intentional act.... Third, the act complained of is not within the traditional idea of contract or touching."74 Each of these propositions underscores the fact that the doctor-patient relationship is rarely considered a setting for a battery. Nondisclosure— intentional or not—does not deserve this charge.

The compromise approach, as championed by Judge Robinson in the Canterbury case, used the negligence standard but tied it to the opinion of the layperson. This approach required integrated consideration of the patient's wishes and blunted physician paternalism. It acted, in effect, as a method for introducing liberal assumptions into the patient-physician relationship. Rights advocates like Judge Robinson have forcefully insisted that doctors are not the sole judges in that relationship. Patients' rights must be considered and their judgment respected by truly informed consent.

Decisions like Canterbury, handed down by a prestigious court like the District of Columbia Circuit Court of Appeals, cannot help but promote informed consent suits in other courts. The law, as a reflection of the public morality, of the state, thus has sought to reintegrate the doctor-patient relationship into the liberal state. Patients who sue for informed consent downplay the role of medical ethics in generating patient-doctor trust. Courts that rule in favor of plaintiffs acknowledge that professional ethics cannot replace public morality. When patients perceive that the political and economic setting of medical practice is at odds with a "patient comes first" attitude, they will be less inclined to give up autonomy in deference to the physician's self-assumed moral code.

This reintegration process has prompted the revisions in medical ethics outlined in previous chapters. Medical ethicists, and indeed many physicians, insist that paternalism cannot be acceptable to any theory of medical ethics, and that patient autonomy must be paramount. These values are enshrined in the principles of ethics of the American Medical Association and American College of Physicians. The patient as rights-bearer is a central concept in most recent theories of ethics. Indeed, physicians today are much more likely to try to gain truly informed consent from patients, discussing the risks of treatment and the course of disease. Medical ethics, prompted by the liberal state,

now encourages informed consent; certainly medical ethics as just doctoring does.

This is not, however, the final word on the subject. Much of the discussion of informed consent, and especially the case law, emphasizes the importance of informing the patient of the risks of a procedure. But cases of informed consent usually are tried not only on risks but on benefits.

I would argue that if we accept an ethical code that centers on maintaining patient autonomy and repudiating the restriction of ethical thinking to the personal encounter between doctors and patients, then we cannot restrict ourselves to a discussion of the risks of a procedure to the patient. In a theory of medical ethics as just doctoring, we must also try to share with patients the likelihood of the benefits of the procedure. In medical terminology, a test or procedure that holds the "promise of great benefit" for the patient is an appropriate test. Thus informed consent must include a focus on appropriateness.

Appropriateness Research as a Form of Informed Consent

Typically, informed consent has centered on the physician's description of the possible detrimental side effects of a therapy, and of the possible alternatives to that therapy. Far less attention has been paid to those cases in which the patient has undergone therapy that has not been rigorously proven to be beneficial. Such cases, which must be brought under an implied warranty rubric, or negligence, are notoriously difficult. When good data are lacking that suggest that a treatment is efficacious, each party to the litigation will be able to present experts who will testify to the benefits, or drawbacks, of the procedure. Thus a plaintiff can rarely win this kind of suit.

But plaintiffs' failure to win such suits does not mean that all forms of therapy are efficacious. In fact, there is good reason to believe that many procedures are done unnecessarily. Most of our insight into this problem is provided by research into variations in rates of procedures between geographic areas. The pioneer of this research is John E. Wennberg, and a review of some of his writings is informative.

Wennberg has been studying the variation in rates of procedures between geographic regions for the past fifteen years. Examining various regions that are well matched demographically, Wennberg has demonstrated that procedure rates for simple surgical techniques can vary by a factor of four to five.75 Operations with very high rates of variation include knee operations, transurethral prostate resections, tonsillectomies, hysterectomies, and dilation and curettage (DNC) operations. Wennberg and colleagues have produced data demonstrating that these variations are a universal phenomenon. The procedures with highest variation all appear to share one characteristic: they deal primarily with common problems of aging.

Wennberg has hypothesized that there are many reasons for practice variations. These include specialty training, clinical experience, and personal characteristics of the patient and physician. Variation is especially prevalent when there is little consensus about the best way to handle a problem. Presumably, one rate for a procedure is optimal in a given population. When some region has a very low rate, this suggests the procedure is not being done often enough. When the rate of the procedure is much higher than expected, it suggests inappropriate procedures are being done.

Wennberg and B. A. Barnes have argued that supplier induced demand may be playing a large role in creating inordinately high rates of procedures.76 As our analysis has made clear, physicians have traditionally controlled both supply and demand of medical care. This creates the temptation to provide more services than are strictly necessary. The patient is not a rational consumer, given his deficit of information. Given as well a lack of consensus among physicians about the appropriateness of many procedures, the pressure of economic reimbursement for doing the procedure will push rates up. Thus the decision-making process between doctor and patient is isolated from the values of the liberal market, which will presumably increase rates of unnecessary procedures.

How does one go about defining which procedures are necessary? Wennberg and his collaborators have begun to address this problem. They have recently published a comprehensive review of their attempt to define appropriateness for one procedure, the resection of prostates.77 Many elderly men develop benign enlargement of the prostate gland. This hypertrophy can become great enough that urinating becomes difficult. As these symptoms develop, patients are often referred

to urologists, who often recommend resection of the prostate. The preferred method of resection is the transurethral procedure (TURP) as opposed to an open procedure. Unfortunately, the procedure is often accompanied by bleeding, and it sometimes falls to relieve symptoms. The optimal timing of the operation as the disease advances is also difficult to discern. Procedure rates vary greatly for the TURP from region to region.

Wennberg and collaborators have attacked this problem with a fourfold research approach. First they reviewed and carefully evaluated all published reports on the natural history and treatment of prostatic hypertrophy. Second, they obtained large data bases of claims information from insurers. Since many patients have some form of insurance, and since their treatment leads to claims for payment, this is an excellent way to obtain information on the number and types of procedures. These data also provided information on complications, and on the efficacy of various forms of alternative treatment. Next, Wennberg's team interviewed patients and physicians to assess patients' symptoms as well as functional status. The interviews also provided data on complication rates. Finally, the researchers entered much of this information into a set of algorithms called decision analysis. Using data on outcomes and alternatives, the decision analysis allows the researchers to test a series of hypotheses.

Some of their conclusions were quite surprising. First, they found that the frequency of complications following prostatectomy was higher than that reported in the literature. The patient interviews proved to be most helpful in this assessment. They also found that surgery actually decreases the average life expectancy of those operated on, although it improves their quality of life somewhat.

The researchers concluded that in the case of prostatectomy, and presumably in a number of other high variability procedures, patient choice was the critical factor in determining the efficacy of the treatment. Some patients may wish to trade a few days of life, overall, for higher quality days while alive. Others may choose to avoid procedures that pose any risk, however small. Only full information for the patient will allow an optimal outcome in such procedures. Wennberg and his colleagues propose an interactive, computer-driven, video presentation to help patients choose between prostatectomy and watchful waiting. The video device presents information on the probabilities of outcomes associated with watchful waiting and operation. Videotaped

interviews made of actual patients who have opted for or against the operation help the patient understand the significance of the various outcomes.78

This seems to be a straightforward approach to the problems of informed consent when the advantages of a procedure are not clear. In essence, the video helps inform the patient about his options. It helps create a relationship between doctor and patient that is more akin to one between individuals in the liberal state than the one that was the model for traditional medical ethics. Allowing the patient to make an informed decision is a matter of maintaining his autonomy. Thus this innovation is one that ethical physicians should support.

The importance of Wennberg's research agenda for medical ethics does not, however, stop here. Wennberg has proposed that in order to develop rational patient choice in a significant manner, it will be necessary to monitor and distribute information on procedure rates to hospital markets.79 Using insurer records and information from hospitals, Wennberg would feed back to providers information on: (1) the status of health resources allocation to specific communities; (2) the utilization of surgical services and diagnostic procedures, as well as causes for admissions; and (3) reports on outcomes from insurance data and interviews.

To whom would this information be transmitted? Wennberg asserts that medical societies, specialty societies, and hospital staffs should be the target. He believes that educating providers can bring about changes in the indications for procedures and will also lead to the best possible education of patients.

Again, Wennberg's proposals have an ethical dimension, at least in my vision of medical ethics as just doctoring. Patients should be accorded the full respect due rights-bearers in the liberal state. To maintain their negative freedom, it is necessary that they be informed. They can only be informed about the appropriateness of procedures insofar as physicians undertake the kind of programs Wennberg has outlined. Thus there is an ethical imperative for physicians to participate in programs that help define appropriate procedures and allow, the most informed decisions by patients. Medical societies, hospital staffs, and individual physicians must take the lead in developing and responding to research initiatives on appropriateness.

This will strike those schooled in traditional medical ethics as odd. Medical ethics has historically focused on the relation between a doctor and her patient. What I propose is an ethics that involves research

and concerted action by many physicians. These efforts will not affect or involve the individual patient I am now treating. Physicians must act on their concern for, not only the individual patient, but the class of patients. This does not demean or belittle the commitment to the individual patient; it does add new ethical importance to the larger issues of health care. It is precisely what must occur once we conceive of medical ethics as part of the liberal state.

So what am I asking of physicians? Let me return to the case of Mr. Z. Suppose that instead of an obvious bleeding ulcer, he had had only peptic symptoms. Given these symptoms, Dr. A. might have referred him to a gastroenterologist, who in turn might have argued that such a young patient should undergo endoscopy to discern the cause of his symptoms. Some would have suggested that this endoscopy was inappropriate.80 Others might have thought it was indicated.81 As an ethical physician, what should Dr. A. do?

In this situation, Dr. A. should give Mr. Z. the best rendition of the facts she can muster, as well as her recommendation. With this information, he can decide. In addition, however, she and the other physicians in our hospital or community should take an active interest in the issue of appropriateness. In this sense, we have an ethical duty to determine whether procedures are being done appropriately and to regulate our practice if they are not. This will require careful record keeping and also significant amounts of time reviewing data and developing criteria. All physicians should participate in the gathering of data and the feedback process, as well as in the development of educational devices so that patients like Mr. Z. can make informed decisions. While this may sound bureaucratic and unexciting, it is as integral a part of medical ethics as just doctoring.

I have argued that the shape of medical ethics in the liberal state should be different from traditional medical ethics. Most important, I have argued that the maintenance of, and respect for, patient autonomy should be paramount. Others have agreed. I have also argued that medical ethics as just doctoring should have a public role and that physicians should reform medical institutions to fit with the sense of altruism that informs the attitude that the "patient comes first."

The law of informed consent has allowed us to specify the nature of the model I have suggested. First, the evolution of the law demonstrates the manner in which the state has sought to reintegrate medicine and require respect for the public morality of liberalism. The law has stated that patients are to be fully informed. The profession has

responded and informed consent is now enshrined in principles of ethics as well as in the mind-set of physicians. Second, evolution of informed consent indicates new ethical requirements for physicians. We must now begin to struggle to define the appropriateness of procedures and to develop new and better ways to inform patients about the utility of our therapies. Thus medical ethics as just doctoring sets forth a public role for physicians in the liberal state. To formulate a medical enterprise that respects patient autonomy, but also incorporates altruistic behavior by physicians and a concern for the social implications of our action, we must consider, as a profession, new policy initiatives. The development of information for patients on the appropriateness of our procedures is one such initiative.