Four Models of the Physician-Patient Relationship

Ezekiel J. Emanuel, MD, PhD, Linda L. Emanuel, MD, PhD

Journal of the American Medical Association, April 22/29, 1962; Volume 267, #16 

(References have been deleted)

DURING the last two decades or so, there has been a struggle over the patient's role in medical decision making that is often characterized as a conflict between autonomy and health, between the values of the patient and the values of the physician. Seeking to curtail physician dominance, many have advocated an ideal of greater patient control. Others question this ideal because it fails to acknowledge the potentially imbalanced nature of this interaction when one party is sick and searching for security, and when judgments entail the interpretation of technical information. Still others are trying to delineate a more mutual relationship. This struggle shapes the expectations of physicians and patients as well as the ethical and legal standards for the physician's duties, informed consent,and medical malpractice. This struggle forces us to ask, What should he the ideal physician-patient relationship?

We shall outline four models of the physician-patient interaction, emphasizing the different understandings of (1) the goals of the physician-patient interaction, (2) the physician's obligations, (3) the role of patient values, and (4) the conception of patient autonomy. To elaborate the abstract description of these four models, we shall indicate the types of response the models might suggest in a clinical situation. Third, we shall also indicate how these models inform the current debate about the ideal physician-patient relationship. Finally, we shall evaluate these models and recommend one as the preferred model.

As outlined, the models are Weberian ideal types. They may not describe any particular physician-patient interactions but highlight, free from complicating details, different visions of the essential characteristics of the physician-patient interaction. Consequently, they do not embody minimum ethical or legal standards, but rather constitute regulative ideals that are "higher than the law, but not above the law."

THE PATERNALISTIC MODEL

First is the paternalistic model, sometimes called the parental or priestly model. In this model, the physician-patient interaction ensures that patients, receive the interventions that best promote their health and well-being. To this end, physicians use their skills to determine the patient's medical condition and his stage in the disease process and to identify the medical tests and treatments most likely to restore the patient's health or ameliorate pain. Then the physician presents the patient with selected information that will encourage the patient to consent to the intervention the physician considers best. At the extreme, the physician authoritatively informs the patient when the intervention will be initiated.

The paternalistic model assumes that there are shared objective criteria for determining what is best. Hence the physician can discern what is in the patient's best interest with limited patient participation. Ultimately, it is assumed that the patient will be thankful for decisions made by the physician even if he or she would not agree to them at the time. In the tension between the patient's autonomy and well-being, between choice and health, the paternalistic physician's main emphasis is toward the latter.

In the paternalistic model, the physician acts as the patient's guardian, articulating and implementing what is best for the patient. As such, the physician has obligations, including that of placing the patient's interest above his or her own
and soliciting the views of others when lacking adequate knowledge. The conception of patient autonomy is patient assent, either at the time or later, to the physician's determinations of what is best

 THE INFORMATIVE MODEL

 Second is the informative model, sometimes called the scientific, engineering, or consumer model. In this model, the objective of the physician-patient interaction is for the physician to provide the patient with all relevant information, for the patient to select the medical interventions he or she wants, and for the physician to execute the selected interventions. To this end, the physician informs the patient of his or her disease state, the nature of possible diagnostic and therapeutic interventions, the nature and probability of risks and benefits associated with the interventions, and any uncertainties of knowledge. At the extreme, patients could come to know all medical information relevant to their disease and available interventions and select the interventions that best realize their values. 

The informative model assumes a fairly clear distinction between facts and values. The patient's values are well defined and known; what the patient lacks is facts. It is the physician's obligation to provide all the available facts, and the patient's values then determine what treatments are to be given. There is no role for the physician's values, the physician's understanding of the patient's values,  or her judgment of the worth of the patient's values. In the informative model, the physician is a purveyor of technical expertise, providing the patient with the means to exercise control. As technical experts, physicians have important obligations to provide truthful information, to maintain competence in their area of expertise, and to consult others when their knowledge or skills are lacking. The conception of patient autonomy is patient control over medical decision making.


THE INTERPRETIVE MODEL

The third model is the interpretive model. The aim of the physician-patient interaction is to elucidate the patient's values and what he or she actually wants, and to help the patient select the available medical interventions that realize these values. Like the informative physician, the interpretive physician provides the patient with information on the nature of the condition and the risks and benefits of possible interventions. Beyond this, however, the interpretive physician assists the patient in elucidating and articulating his or her values and in determining what medical interventions best realize the specified values, thus helping to interpret the patient's values for the patient.

According to the interpretive model, the patient's values are not necessarily fixed and known to the patient. They are often inchoate, and the patient may only partially understand them; they may conflict when applied to specific situations. Consequently, the physician working with the patient must elucidate and make coherent these values. To do this, the physician works with the patient to reconstruct the patient's goals and aspirations, commitments and character. At the extreme, the physician must conceive the patient's life as a narrative whole, and from this specify the patient's values and their priority. Then the physician determines which tests and treatments best realize these values. Importantly, the physician does not dictate to the patient; it is the patient who ultimately decides which values and course of action best fit who he or she is. Neither is the physician judging the patient's values; he or she helps the patient to understand and use them in the medical situation.

 In the interpretive model, the physician is a counselor, analogous to a cabinet minister's advisory role to a head of state, supplying relevant information, helping to elucidate values and suggesting what medical interventions realize these values. Thus the physician's obligations include those enumerated in the informative model but also require engaging the patient in a joint process of understanding. Accordingly, the conception of patient autonomy is self-understanding; the patient comes to know more clearly who he or she is and how the various medical options bear on his or her identity.

THE DELIBERATIVE MODEL

Fourth is the deliberative model. The aim of the physician-patient interaction is to help the patient determine and choose the best health-related values that can be realized in the clinical situation. To this end, the physician must delineate information on the patient's clinical situation and then help elucidate the types of values embodied in the available options. The physician's objectives include suggesting why certain health-related values are more worthy and should be aspired to. At the extreme, the physician and patient engage in deliberation about what kind of health-related values the patient could and ultimately should pursue. The physician discusses only health-related values, that is, values that affect or are affected by the patient's disease and treatments; he or she recognizes that many elements of morality are unrelated to the patient's disease or treatment and beyond the scope of their professional relationship. Further, the physician aims at no more than moral persuasion; ultimately, coercion is avoided, and the patient must define his or her life and select the ordering of values to be espoused. By engaging in moral deliberation, the physician and patient judge the worthiness and importance of the health-related values.

In the deliberative model, the physician acts as a teacher or friend, engaging the patient in dialogue on what course of action would be best. Not only does the physician indicate what the patient could do, but, knowing the patient and wishing what is best, the physician indicates what the patient should do, what decision regarding medical therapy would be admirable. The conception of patient autonomy is moral self-development; the patient is empowered not simply to follow unexamined preferences or examined values, but to consider, through dialogue, alternative health-related values, their worthiness, and their implications for treatment.

COMPARING THE FOUR MODELS

 Importantly, all models have a role for patient autonomy; a main factor that differentiates the models is their particular conceptions of patient autonomy. Therefore, no single model can be endorsed because it alone promotes patient autonomy- Instead the models must be compared and evaluated, at least in part, by evaluating the adequacy of their particular conceptions of patient autonomy.

 The four models are not exhaustive. At a minimum there might be added a fifth: the instrumental model. In this model, the patient's values are irrelevant; the physician aims for some goal independent of the patient, such as the good of society or furtherance of scientific knowledge. The Tuskegee syphilis experiment and the Willowbrook hepatitis study are examples of this model. As the moral condemnation of these cases reveals, this model is not an ideal but an aberration. Thus we have not elaborated it herein.

A CLINICAL CASE

To make tangible these abstract descriptions and to crystallize essential differences among the models, we will illustrate the responses they suggest in a clinical situation, that of a 43-year-old premenopausal woman who has recently discovered a breast mass. Surgery reveals a 3.5-cm ductal carcinoma with no lymph node involvement that is estrogen receptor positive. Chest roentgenogram, bone scan, and liver function tests reveal no evidence of metastatic disease. The patient was recently divorced and has gone back to work as a legal aide to support herself. What should the physician say to this patient?

In the paternalistic model a physician might say, "There are two alternative therapies to protect against recurrence of cancer in your breast: mastectomy or radiation. We now know that the survival with lumpectomy combined with radiation therapy is equal to that with mastectomy. Because lumpectomy and radiation offers the best survival and the best cosmetic result, it is to be preferred. I have asked the radiation therapist to come and discuss radiation treatment with you. We also need to protect you against the spread of the cancer to other parts of your body. Even though the chance of recurrence is low, you are young, and we should not leave any therapeutic possibilities untried. Recent studies involving chemotherapy suggest improvements in survival without recurrence of breast cancer. Indeed, the National Cancer Institute recommends chemotherapy for women with your type of breast cancer. Chemotherapy has side effects. Nevertheless, a few months of hardship now are worth the potential added years of life without cancer."

In the informative model a physician might say, "With node-negative breast cancer there are two issues before you: local control and systemic control. For local control, the options are mastectomy or lumpectomy with or without radiation. From many studies we know that mastectomy and lumpectomy with radiation result in identical overall survival about 80% 1O year survival. Lumpectomy without radiation results in a 30% to 40% chance of tumor recurrence in the breast. The second issue relates to systemic control. We know that chemotherapy prolongs survival for premenopausal women who have axillary nodes involved with tumor. The role for women with node-negative breast cancer is less clear. Individual studies suggest that chemotherapy is of no benefit in terms of improving overall survival but a comprehensive review of all studies suggests that there is a survival benefit. Several years ago, the NCI suggested that for women like yourself, chemotherapy can have a positive therapeutic impact. Finally, let me inform you that there are clinical trials, for which you are eligible, to evaluate the benefits of chemotherapy for patients with node-negative breast cancer. I can enroll you in a study if you want. I will be happy to give you any further information you feel you need."

The interpretive physician might outline much of the same information as the informative physician, then engage in discussion to elucidate the patient's wishes, and conclude, "It sounds to me as if you have conflicting wishes. Understandably, you seem uncertain how to balance the demands required for receiving additional treatment, rejuvenating your personal affairs, and maintaining your psychological equilibrium. Let me try to express a perspective that fits your position. Fighting your cancer is important, but it must leave you with a healthy self-image and quality time outside the hospital. This view seems compatible with undergoing radiation therapy but not chemotherapy. A lumpectomy with radiation maximizes your chance of surviving while preserving your breast. Radiotherapy fights your breast cancer without disfigurement. Conversely, chemotherapy would prolong the duration of therapy by many months. Further, the benefits of chemotherapy in terms of survival are smaller and more controversial. Given the recent changes in your life, you have too many new preoccupations to undergo months of chemotherapy for a questionable benefit. Do I understand you? We can talk again in a few days."

The deliberative physician might begin by outlining the same factual information, engage in a conversation to elucidate the patient's values, but continue, "It seems clear that you should undergo radiation therapy. It offers maximal survival with minimal risk, disfigurement, and disruption of your life. The issue of chemotherapy is different, fraught with conflicting data. Balancing all the options, I think the best one for you is to enter a trial that is investigating the potential benefit of chemotherapy for women with node-negative breast cancer. First, it ensures that you receive excellent medical care. At this point, we do not follow-up visits, tests, and decisions is specified by leading breast cancer experts to ensure that all the women receive care that is the best available anywhere. A second reason to participate in a trial is altruistic; it allows you to contribute something to women with breast cancer in the future who will face difficult choices. Over decades, thousands of women have participated in studies that inform our current treatment practices. Without those women, and the knowledge they made possible, we would probably still be giving you and all other women with breast cancer mastectomies. By enrolling in a trial you participate in a available but also enhance the care of women in future generations because medicine has learned something about which interventions are better. I must tell you that I am not involved in the study; if you elect to enroll in this trial, you will initially see another breast cancer expert to plan your therapy. I have sought to explain our current knowledge and offer my recommendation so you can make the best possible decision."

Lacking the normal interchange with patients, these statements may seem contrived, even caricatures. Nevertheless, they highlight the essence of each model and suggest how the objectives and assumptions of each inform a physician's approach to his or her patients. Similar statements can be imagined for other clinical situations such as an obstetrician discussing prenatal testing or a cardiologist discussing cholesterol-reducing interventions.

THE CURRENT DEBATE AND THE FOUR MODELS

In recent decades there has been a call for greater patient autonomy or, as some have called it, "patient sovereignty" conceived as patient choice and control over medical decisions. This shift toward the informative model is embodied in the adoption of business terms for medicine, as when physicians are described as health care providers and patients as consumers. It can also be found in the propagation of patient rights statements, in the promotion of living will laws, and in rules regarding human experimentation. For instance, the opening sentences of one law state: "The Rights of the Terminally Ill Act authorizes an adult person to control decisions regarding administration of life-sustaining treatment. . . . The Act merely provides one way by which a terminally-ill patient's desires regarding the use of life-sustaining procedures can be legally implemented" (emphasis added).

Indeed, living will laws do not require or encourage patients to discuss the issue of terminating care with their physicians before signing such documents. Similarly, decisions in "right-to-die" cases emphasize patient control over medical decisions. As one court put it:

 The right to refuse medical treatment is basic and fundamental. Its exercise requires no one's approval. The controlling decision belongs to a competent informed patient. . . . It is not a medical decision for her physicians to make. . . . It is a moral and philosophical decision that, being a competent adult, is the patient's alone. (emphasis added)

 Probably the most forceful endorsement of the informative model as the ideal inheres in informed consent standards. Prior to the 197Os, the standard for informed consent was 'physician based.’ Since 1972 and the Cantebury case, however, the emphasis has been on a 'patient-oriented’ standard of informed consent in which the physician has a "duty" to provide appropriate medical facts to empower the patient to use his or her values to determine what interventions should be implemented.

 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. It is the prerogative of the patient, not the physician. to determine for himself the direction in which his interests seem to lie. To enable the patient to chart his course understandably, some familiarity with the therapeutic alternatives and their hazards becomes essential.
 

SHARED DECISION MAKING

Despite its dominance, many have found the informative model "arid." The President’s Commission and others contend that the ideal relationship does not vest moral authority and medical decision-making power exclusively in the patient but must be a process of shared decision making constructed around "mutual participation and respect." The President's Commission argues that the physician's role is "to help the patient understand the medical situation and available courses of action, and the patient conveys his or her concerns and wishes. Brock and Wartman stress this fact-value "division of labor' having the physician provide information while the patient makes value decisions-by describing "shared decision making" as a collaborative process in which both physicians and patients make active and essential contributions. Physicians bring their medical training, knowledge, and expertise; including an understanding of the available treatment alternatives to the diagnosis and management of patients' condition, Patients bring knowledge of their own subjective aims and values, through which risks and benefits of various treatment options can be evaluated. With this approach, selecting the best treatment for a particular patient requires the contribution of both parties.

Similarly, in discussing ideal medical decision making, Eddy argues for this fact-value division of labor between the physician and patient as the ideal:

 It is important to separate the decision process into these two steps. . . . The first step is a question of facts. The anchor is empirical evidence. The second step is a question not of facts but of personal values or preferences. The thought process is not analytic but personal and subjective. It is the patient's preferences that should determine the decision. . . . Ideally, you and I [the physicians] are not in the picture. What matters is what Mrs. Smith thinks. This view of shared decision making seems to vest the medical decision-making authority with the patient while relegating physicians to technicians "transmitting medical information and using their technical skills as the patient directs." Thus, while the advocates of "shared decision making" may aspire toward a mutual dialogue between physician and patient, the substantive view informing their ideal reembodies the informative model under a different label.Other commentators have articulated more mutual models of the physician-patient interaction. Prominent among these efforts is Katz in The Silent World of the Doctor and Patient. Relying on a Freudian view in which self-knowledge and self-determination are inherently limited because of unconscious influences, Katz views dialogue as a mechanism for greater self-understanding of one's values and objectives. According to Katz, this view places a duty on physicians and patients to reflect and communicate so that patients can gain a greater self-understanding and self-determination. Katz' insight is also available on grounds other than Freudian psychological theory and is consistent with the interpretive model.


OBJECTIONS TO THE PATERNALISTIC MODEL

 It is widely recognized that the paternalistic model is justified during emergencies when the time taken to obtain informed consent might irreversibly harm the patient. Beyond such limited circumstances, however, it is no longer tenable to assume that the physician arid patient espouse similar values and views of what constitutes a benefit. Consequently, even physicians rarely advocate the paternalistic model as an ideal for routine physician/patient interactions.


OBJECTIONS TO THE INFORMATIVE MODEL

The informative model seems both descriptively and prescriptively inaccurate. First, this model seems to have no place for essential qualities of the ideal physician/patient relationship. The informative physician cares for the patient in the sense of competently implementing the patient's selected interventions. However, the informative physician lacks a caring approach that requires understanding what the patient values or should value and how his or her illness impinges on these values. Patients seem to expect their physician to have a caring approach; they deem a technically proficient but detached physician as deficient, and properly condemned. Further, the informative physician is proscribed from giving a recommendation for fear of imposing his or her will on the patient and thereby competing for the decision-making control that has been given to the patient. Yet, if one of the essential qualities of the ideal physician is the ability to assimilate medical facts,prior experience of similar situations, and intimate knowledge of the patient's view into a recommendation designed for the patient's specific medical and personal condition, then the informative physician cannot, be ideal. Second, in the informative model the ideal physician is a highly trained sub-specialist who provides detailed factual information and competently implements the patient's preferred medical intervention. Hence, the informative model perpetuates and accentuates the trend toward specialization and impersonalizaticn within the medical profession. Most importantly, the informative model's conception of patient autonomy seems philosophically untenable. The informative model presupposes that persons possess known and fixed values, but this is inaccurate. People are often uncertain about what they actually want. Further, unlike animals, people have what philosophers call "second order desires," that is, the capacity to reflect on their wishes and to revise their own desires and preferences. In fact, freedom of the will and autonomy inhere in having "second order desires" and being able to change our preferences and modify our identity. Self-reflection and the capacity to change what we want often require a "process" of moral deliberation in which we assess the value of what we want. And this is a process that occurs with other people who know. us well and can articulate a vision of who we ought to be that we can assent to. Even though changes in health or implementation of alternative interventions can have profound effects on what we desire and how we realize our desires, self-reflection and deliberation play no essential role in the informative physician-patient interaction. The informative model's conception of autonomy is incompatible with a vision of autonomy that incorporates second order desires.
 

OBJECTIONS TO THE INTERPRETIVE MODEL 

The interpretive model rectifies this deficiency by recognizing that persons have second-order desires and dynamic value structures and placing the elucidation of values in the context of the patient's medical condition at the center of the physician/patient interaction. Nevertheless, there are objections to the interpretive model. 

Technical specialization militates against physicians cultivating the skills necessary to the interpretive model. With limited interpretive talents and limited time, physicians may unwittingly impose their own values under the guise of articulating the patient's values. And patients, overwhelmed by their medical condition and uncertain of their own views, may too easily accept this imposition.Such circumstances may push the interpretive model toward the paternalistic model in actual practice. Further, autonomy viewed as self-understanding excludes evaluative judgment of the patient's values or attempts to persuade the patient to adopt other values. This constrains the guidance and recommendations the physician can offer. Yet in practice, especially in preventive medicine and risk-reduction interventions, physicians often attempt to persuade patients to adopt particular health-related values. Physicians frequently urge patients with high cholesterol levels who smoke to change their dietary habits, quit smoking, and begin exercise programs before initiating drug therapy. The justification given for these changes is that patients should value their health more than they do. Similarly, physicians are encouraged to persuade their human immunodeficiency virus (HIV)-infected patients who might be engaging in unsafe sexual practices either to abstain or, realistically, to adopt "safer sex" practices. Such appeals are not made to promote the HlV infected patient's own health, but are grounded on an appeal for the patient to assume responsibility for the good of others. Consequently, by excluding evaluative judgments, the interpretive model seems to characterize inaccurately ideal physician-patient interactions.


OBJECTIONS TO THE DELIBERATIVE MODEL

The fundamental objections to the deliberative model focus on whether it is proper for physicians to judge patients' values and promote particular health-related values. First, physicians do not possess privileged knowledge of the priority of health-related values relative to other values. Indeed, since ours is a pluralistic society in which people espouse incommensurable values, it is likely that a physician's values and view of which values are higher will conflict with those of other physicians and those of his or her patients. Second, the nature of the moral deliberation between physician and patient, the physician's recommended interventions, and the actual treatments used will depend on the values of the particular physician treating the patient. However, recommendations and care provided to patients should not depend on the physician's judgment of the worthiness of the patient's values or on the physician's particular values. As one bioethicist put it:

The hand is broken; the physician can repair the hand; therefore the physician must repair the hand-as well as possible-without regard to personal values that might lead the physician to think ill of the patient or of the patient's values. At the level of clinical practice, medicine should be value-free in the sense that the personal values of the physician should not distort the making of medical decisions.

Third, it may be argued that the deliberative model misconstrues the purpose of the physician-patient interaction. Patients see their physicians to receive health care, not to engage in moral deliberation or to revise their values.

Finally, like the interpretive model, the deliberative model may easily metamorphose into unintended paternalism, the very practice that generated the public debate over the proper physician-patient interaction.


THE PREFERRED MODEL AND THE PRACTICAL IMPLICATIONS

Clearly, under different clinical circumstances different models may be appropriate. Indeed, at different times all four models may justifiably guide physicians and patients. Nevertheless, it is important to specify one model as the  shared, paradigmatic reference; exceptions to use other models would not be automatically condemned, but would require justification based on the circumstances of a particular situation. Thus, it is widely agreed that in an emergency where delays in treatment to obtain in-formed consent might irreversibly harm the patient, the paternalistic model correctly guides physician-patient interactions. Conversely, for patients who have clear but conflicting values, the interpretive model is probably justified. For instance, a 65 year old woman who has been treated for acute leukemia may have clearly decided against reinduction chemotherapy if she relapses. Several months before the anticipated birth of her first grandchild, the patient relapses.The patient becomes torn about whether to endure the risks of reinduction chemotherapy in order to live to see her first grandchild or whether to refuse therapy, resigning herself to not seeing her grandchild. In such cases, the physician may justifiably adopt the interpretive approach. In other circumstances, where there is only a one time physician-patient interaction without an ongoing relationship in which the patient's values can be elucidated and compared with ideals, such as in a walk-in center, the informative model may be justified.

Descriptively and prescriptively, we claim that the ideal physician-patient relationship is the deliberative model. We will adduce six points to justify this claim. First, the deliberative model more nearly embodies our ideal of autonomy. It is an oversimplification and distortion of the Western tradition to view respecting autonomy as simply permitting a person to select, unrestricted by coercion, ignorance, physical interference, and the like, his or her preferred course of action from a comprehensive list of available options. Freedom and control  over medical decisions alone do not constitute patient autonomy. Autonomy requires that individuals critically assess their own values and preferences; determine whether they are desirable; affirm, upon reflection, these values as ones that should justify their actions; and then be free to initiate action to realize the values. The process of deliberation integral to the deliberative model is essential for realizing patient autonomy understood in this way.

Second, our society's image of an ideal physician is not limited to one who knows and communicates to the patient relevant factual information and competently implements medical interventions. The ideal physician-often embodied in literature, art, and popular culture is a caring physician who integrates the information and relevant values to make a recommendation and, through discussion, attempts to persuade the patient to accept this recommendation as the  intervention that best promotes his or her overall well-being. Thus, we expect the best physicians to engage their patients in evaluative discussions of health issues and related values. The physician's discussion does not invoke values that are unrelated or tangentially related to the patient's illness and potential therapies. Importantly, these efforts are not restricted to situations in which patients might make "irrational and harmful" choices, but extend to all health care decisions.

Third, the deliberative model is not a disguised form of paternalism. Previously there may have been category mistakes in which instances of the deliberative model have been erroneously identified as physician paternalism. And no doubt, in practice, the deliberative physician may occasionally lapse into paternalism. However, like the ideal teacher, the deliberative physician attempts to persuade the patient of the worthiness of certain values, not to impose those values paternalistically; the physician's aim is not to subject the patient to his or her will, but to persuade the patient of a course of action as desirable. In the Laws, Plato characterizes this fundamental distinction between persuasion and imposition for medical practice that distinguishes the deliberative from the paternalistic model: A physician to slaves never gives his patient any account of his illness . . . the physician offers some orders gleaned from experience with an air of infallible knowledge, in the brusque fashion of a dictator. . . . The free physician, who the patient and his friends his ailment. This way he learns something from the sufferer and simultaneously instructs him. Then the physician does not give his medications until he has persuaded the patient; the physician aims at complete restoration of health by persuading the patient to comply with his therapy.

Fourth, physician values are relevant to patients and do inform their choice of a physician. When a pregnant woman chooses an obstetrician who does not routinely perform a battery of prenatal tests or, alternatively, one who strongly favors them; when a patient seeks an aggressive cardiologist who favors procedural interventions or one who concentrates therapy on dietary changes, stress reduction, and life-style modifications, they are, consciously or not, selecting a physician based on the values that guide his or other medical decisions. And, when disagreements between physicians and patients arise, there are  discussions over which values are more important and should be realized in medical care. Occasionally, when such disagreements undermine the physician-patient relationship and a caring attitude, a patient's care is transferred to another physician. Indeed, in the informative model the grounds for transferring care to a new physician is either the physician's ignorance or incompetence. But patients seem to switch physicians because they do not like a particular physician or that physician's attitude or approach. 

Fifth, we seem to believe that physicians should not only help fit therapies to the patients' elucidated values, but should also promote health-related values. As noted, we expect physicians to promote certain values, such as "safer sex" for patients with HIV or abstaining from or limiting alcohol use. Similarly, patients are willing to adjust their values and actions to be more compatible with health-promoting values. This is in the nature of seeking a caring medical recommendation.

Finally, it may well be that many physicians currently lack the training and capacity to articulate the values underlying their recommendations and persuade patients that these values are worthy. But, in part, this deficiency is a consequence of the tendencies toward specialization and the avoidance of discussions of values by physicians that are perpetuated and justified by the dominant informative model. Therefore, if the deliberative model seems most appropriate, then we need to implement changes in medical care and education to encourage a more in keeping with the legal standards of informed consent and medical malpractice; we must educate physicians not just to spend more time in physician-patient communication but to elucidate and articulate the values underlying their medical  care decisions, including routine ones; we must shift the publicly assumed conception of patient autonomy that shapes both the physician's and the patient's expectations from patient control to moral development. Most important, we must recognize that developing a deliberative physician-patient relationship requires a considerable amount of time. We must develop a health care financing system that properly reimburses-rather than penalizes-physicians for taking the time to discuss values with their patients.


CONCLUSION

Over the last few decades, the discourse regarding the physician-patient relationship has focused on two extremes: autonomy and paternalism. Many have attacked physicians as paternalistic, urging the empowerment of patients to control their own care. This view, the informative model, has become dominant in bioethics and legal standards. This model embodies a defective conception of patient autonomy, and it reduces the physician's role to that of a technologist. The essence of doctoring is a fabric of knowledge, understanding, teaching, and action, in which the caring physician integrates the patient's medical condition and health-related values, makes a recommendation on the appropriate course of action, and tries to persuade the patient of the worthiness of this approach and the values it realizes.The physician with a caring attitude is the ideal embodied in the deliberative model, the ideal that should inform laws and policies that regulate the physician/patient interaction. .Finally, it maybe worth noting that the four models outlined herein are not limited to the medical realm; they may inform the public conception of other professional interactions as well. We suggest that the ideal relationships between lawyer and client, religious mentor and laity, and educator and student are well described by the deliberative model, at least in some of their essential aspects.