Perspective
200th Anniversary Article
Patients and Doctors — The Evolution of a Relationship
N Engl J Med 2012; 366:581-585February 16, 2012
Comments
open through December 31, 2012
- Article
The relationship between patients and doctors is at the core of medical ethics, serving as an anchor for many of the most important debates in the field. Over the past several decades, this relationship has evolved along three interrelated axes — as it is defined in clinical care, research, and society. Many of the pivotal discussions of these issues have appeared in the pages of the Journal (see Key NEJM Articles on Medical Ethics).
Clinical Care
The relationship between patients and doctors in the clinical realm has historically been framed in terms of benevolent paternalism. Until about 1960, most codes of medical ethics relied heavily on the Hippocratic tradition, framing the obligations of physicians solely in terms of promoting the welfare of the patient, while remaining silent about patients' rights. The past several decades have seen tectonic societal shifts that have resulted in increasing empowerment of individuals against the authority of government and institutions, creating a surge of rights-based movements, with patients' rights emerging as a societal demand alongside women's rights, minority groups' rights, consumers' rights, and others.
This dramatic shift appeared to move the locus of authority in decision making from the physician to the patient. And indeed the emergence of the Internet, with its myriad health-related websites and other sources of medical information, has given many patients the impression that they can largely manage their own medical affairs, with physicians serving primarily as consultants. But the reality is more complex: the wealth of information available to patients has proved to be as dangerous as it is helpful, and today patients and physicians are beginning to find a healthier balance of power through a process of shared decision making. With this approach, physicians are seen as having expertise and authority over matters of medical science, whereas patients hold sway over questions of values or preferences.
This division of labor reflects a recognition of the naturalistic fallacy, the erroneous notion that one can derive ethical conclusions from scientific facts; in truth, an “ought” cannot be deduced from an “is.” Although physicians may be experts on the medical facts of a patient's condition, those facts are never sufficient to specify a course of treatment; clinical decisions must always include consideration of the values and preferences of the patient. This approach has many implications — for example, in recognizing the right of a competent adult to refuse a lifesaving blood transfusion on the basis of his or her religious beliefs, or the right of a patient to refuse mechanical ventilation for a treatable and reversible cause of respiratory failure. Stated succinctly, today we acknowledge that competent patients have a virtually unlimited right to refuse unwanted medical care, even when physicians correctly claim that it would be medically effective and indeed even lifesaving. More important, however, is the way that the concept of shared decision making guides the many more mundane decisions that are made in clinics every day, when physicians present patients with what they see as reasonable medical options and then help them to incorporate personal values and preferences to arrive at decisions that make the most sense for them in terms of both the medical facts and their unique personal perspective. This approach to engaging patients has other benefits as well, such as promoting their sense of self-efficacy and improving their adherence to treatment recommendations.
Despite this vigorous consensus about the rights of patients to refuse unwanted care, equally strong disagreement persists about patients' rights to demand care that physicians regard as medically inappropriate. Consider situations in which the data clearly show that the likelihood of successful resuscitation would be less than 1%. Does this fact allow us to conclude that resuscitation should not be performed, regardless of the patient's preference? Or is the threshold for what should count as a reasonable likelihood of success fundamentally a question of value that should be defined by the patient? When the medical evidence overwhelmingly suggests that a patient is permanently unconscious, does this imply that further medical treatment is inappropriate and should not be provided (as stated in the guidelines of many professional societies), or is what counts as a “life worth living” a personal choice that should be respected?
In recent years, the Texas Advance Directives Act has defined one very concrete approach to addressing these dilemmas. When families demand treatments that have an exceedingly low likelihood of success or that sustain life of such low quality that one might reasonably say it is of no benefit to the patient, Texas law allows physicians to refuse to provide such treatments. Under the Texas legislation, demands by families for treatments that appear to meet these criteria are adjudicated by a hospital-based committee, and if the committee agrees with the clinicians, and if other providers cannot be located who are willing to provide such care, then treatment may be withdrawn without the permission of the patient's surrogate. Although Texas has the most experience with this approach, other states are showing interest in similar proposals that address both the financial implications of providing allegedly inappropriate care and the concerns of clinicians who must endure the moral burdens and burnout associated with being compelled to provide treatments they believe are ethically wrong.
Clinical Research
The Nuremberg trials in 1946 marked the beginning of modern discussion of the ethics of clinical research. Although Nuremberg showed how physicians could be led astray by a corrupt political regime, it was not until Henry Beecher's alarming exposé in 1966 that U.S. physicians had to confront the fact that blatantly unethical research — such as injecting patients with malignant cells without their knowledge or permission — was prevalent even on the wards of prestigious academic medical centers in the United States.1 These and other revelations led to the development of the principles outlined in the Belmont Report, federal regulations governing the conduct of clinical research, and the creation of institutional review boards charged with applying these guidelines to individual research protocols.
Despite the strong safeguards that are now in place to ensure that patients are fully informed and provide their consent before being enrolled in research trials, important tensions remain between the ethical obligations of the physician–patient relationship and those of the researcher–subject relationship. Both physicians and patients have a psychological tendency to minimize these tensions; neither wants to recognize the important ethical conflicts that may exist between clinical care and research. This phenomenon, known as the therapeutic misconception, has been shown by Paul Appelbaum and others to be ubiquitous among both researchers and research subjects, manifesting as the false and often implicit belief that the primary aim of research is to benefit the patient.2 Although patients often do benefit from their involvement in research, nearly all clinical research includes procedures that carry risks to subjects that are not compensated for by corresponding benefits. By definition, research protocols are designed to answer scientific questions, not to optimize the medical care of the patient.
The concept of clinical equipoise was developed with the promise of easing the ethical tension between clinical care and research. Clinical equipoise exists when there is genuine uncertainty in the medical community about which of two treatments is better in a given situation. When this condition is satisfied, a researcher enrolling his or her patients in a randomized trial can honestly say to the patient, “Although I may personally prefer treatment X, if you were being seen by another, equally competent physician, you could be given treatment Y. Hence, your medical care will not be compromised if you agree to enroll in this trial and have your treatment determined by chance.” Under clinical equipoise, there appears to be no conflict between the ethics of clinical care and those of research.
Over the past several years, however, clinical equipoise and other attempts to harmonize the ethics of clinical care and research have become less tenable, particularly in the context of placebo-controlled trials. Consider, for example, placebo-controlled trials of new antidepressants. Given that some antidepressants are known to be efficacious, clinical equipoise cannot be used to justify such trials, since a physician could never defend prescribing a placebo for a patient outside of the trial.
These problems extend beyond placebo-controlled trials, however. For example, oncologists have recently been engaged in an anguished debate about the ethics of a randomized, controlled trial of a new drug for metastatic melanoma in the face of impressive preliminary evidence supporting the efficacy of the new drug, in combination with the known dismal prognosis associated with the standard therapy given to patients in the control group.3 Trials like this one strain the concept of clinical equipoise beyond the breaking point: most oncologists would agree that these trials must be performed, but few would say that the physician researchers believe the treatments are in equipoise.
Given the scientific and ethical rationale for these trials and the failure of the clinical equipoise paradigm to provide justification for them, over the past several years a new way of looking at the ethics of clinical research has developed — one that regards the ethical principles governing clinical care and research as fundamentally distinct and indeed often in tension.4 If physicians forthrightly inform patients that the goal of clinical research is not primarily to optimize their clinical care but to advance knowledge for the benefit of future patients, both patients and physicians may be guided to a more transparent view of research that is not distorted by the therapeutic misconception.
Separating the ethics of research from clinical care has other advantages. With this approach, patients become more active participants in the research enterprise. They expect to be informed about research results and engaged in setting research priorities. Parallel to the evolution of the relationship between physicians and patients in clinical care, this new vision of the ethics of clinical research moves research subjects out from under the paternalistic umbrella of clinical investigators and empowers them to have a more equal and active role in the process of advancing medical knowledge.
Populations and Health Care Systems
One of the most revered principles in medical ethics has been that physicians should be exclusively devoted to the best interests of their patients. As Norman Levinsky put it, “Physicians are required to do everything that they believe may benefit each patient without regard to costs or other societal considerations. In caring for an individual patient, the doctor must act solely as that patient's advocate, against the apparent interests of society as a whole, if necessary.”5 In reality, this has never been more than a lofty ideal, since physicians have always had competing pressures that prevent them from providing everything that might be of medical benefit to a patient, beginning with the fact that no physician can spend an unlimited amount of time with any one patient.
But this ideal is itself coming under increasing scrutiny, as both physicians and society come to recognize that the benefits of a solely patient-centered focus to care must be balanced against the value of offering entire populations of patients equitable access to necessary health care.
The issue first arose in U.S. politics in 1962, when dialysis became available for only a limited number of patients and the public was horrified to learn that a small committee of anonymous citizens was tasked with deciding who should live and who should die. Rather than face this difficult problem head on, Congress eventually passed legislation mandating that renal-replacement therapies be fully funded by the government, a decision that stands to this day. Clearly, however, taking this approach to solving every rationing decision would lead to financial disaster.
More recently, the patient-versus-population dilemma played out over new recommendations for screening mammography. Although analysis shows that current screening practices are exceptionally cost-inefficient, there is no doubt that they have saved the lives of many women. Yet given the seeming impossibility of having a public debate about when it may be permissible to forgo some beneficial care that is very expensive in favor of providing other benefits that are a much better value, those who supported the new screening recommendations were forced to justify them solely in terms of what would be best for an individual woman, regardless of cost. Not surprisingly, by acceding to the taboo that renders open discussion of cost-effectiveness off limits, proponents of the new recommendations appear to have lost the debate.
The United States already spends more on health care than any other country on earth yet does not attain the health benefits that are achieved in many countries with much more limited resources. By refusing to bring the issues of cost-effectiveness and rationing into the political discourse, we allow the myth to persist that there is some yet-to-be-discovered alternative to a thoughtful and systematic approach to the allocation of resources. Despite Levinsky's seductive view that the relationship between physicians and patients should be isolated from any external pressures, we must recognize that population-based factors such as justice, efficiency, and fairness are also ethically relevant. Overcoming our inability to muster the political will and courage to acknowledge the necessity of rationing and to grapple with the best way to use the tremendous resources currently being devoted to health care is likely to be the greatest challenge in the evolving relationship between physicians and patients in the decades to come.
Although the relationship between patients and doctors is often idealized in terms of universal and timeless principles, it has not been immune from the larger social and cultural forces surrounding it. This relationship has been profoundly shaped by the human rights movement of the past several decades, and clinical care today is guided by norms of shared decision making rather than benevolent paternalism. Clinical research is no longer regarded as a side benefit of providing patients with clinical care, but rather as a compatible but distinct activity that requires us to view patients as partners in the process of advancing medical knowledge. And finally, the greatest challenge still lies largely before us, as we will struggle in the years to come to balance the personal advocacy that all patients rightfully expect from their physicians with the equally compelling obligation of physicians to see that health care resources are used wisely in ways that are efficient and fair.
Key NEJM Articles on Medical Ethics.
Clinical Care
Clinical Research
Populations and Health Care Systems
Disclosure forms provided by the author are available with the full text of this article at NEJM.org.
Source Information
From the Departments of Anesthesia and of Global Health and Social Medicine, Harvard Medical School; and the Division of Critical Care Medicine, Children's Hospital Boston — both in Boston.
- References
References
1
Beecher HK . Ethics and clinical research. N Engl J Med 1966;274:1354-1360
Full Text | Web of Science | Medline2
Appelbaum PS ,Roth LH ,Lidz CW ,Benson P ,Winslade W . False hopes and best data: consent to research and the therapeutic misconception. Hastings Cent Rep 1987;17:20-24
Web of Science | Medline3
Miller FG ,Joffe S . Equipoise and the dilemma of randomized clinical trials. N Engl J Med 2011;364:476-480
Full Text | Web of Science | Medline4
Miller FG ,Brody H . A critique of clinical equipoise: therapeutic misconception in the ethics of clinical trials. Hastings Cent Rep 2003;33:19-28
Web of Science | Medline5
Levinsky NG . The doctor's master. N Engl J Med 1984;311:1573-1575
Full Text | Web of Science | Medline
- Comments (3)
-
3 Reader's Comments
Boris Veysman | Physician | Disclosure: NoneFebruary 18, 2012Clinical Analogy: a Physician's Unique Contribution.
Easily accessible wealth of medical information often leaves the patient overwhelmed and confused about what to do in his actual clinical scenario. Even when available medical knowledge is consistent among several sources and is credible (internet-based data often lacks on both fronts), using it for decision-making is best done in consultation with a thoughtful clinician.
Physicians can help uniquely: talking to the patient about his life and work in order to identify areas of expertise for the patient analogous to the medical issue at hand. Then, the doctor can use a concept the patient fully understands to explain physical, cognitive and emotional nuances of the medical decision in question. This use of analogy allows for informed dialogue and shared decision-making, which is otherwise rarely possible. This approach respects and empowers the patient as an intelligent participant in the process, and in turn the patient is appreciative and invested in the therapeutic effort.
The doctor-patient relationship continues to evolve. As medical information is becoming available to our patients in unprecedented quantity and complexity, they need us at their side more than ever.stephen liben, MD | Physician | Disclosure: NoneCanadaFebruary 16, 2012Removing "Therapeutic misconception" = Mission Impossible
The author states: “If physicians forthrightly inform patients that the goal of clinical research is not primarily to optimize their clinical care but to advance knowledge for the benefit of future patients, both patients and physicians may be guided to a more transparent view of research that is not distorted by the therapeutic misconception.” To which my response is “if only it were so!”.
Informing patients that the goal of enrolling in clinical research is not to help them personally but rather is for knowledge to benefit future patients, is a nice idea in principle but is rarely borne out in actual practice. There is strong evidence for example, to show that no matter how much phase 1 cancer patients are told they are not going to get benefit from enrolling as a subject, nonetheless they still believe they will.
Why this is so is not really a mystery and the answer emerges from the fact that patients are not consumers making a purchase decision (“do I want to “buy” into this altruistic research or not?”), but rather are fragile vulnerable sick persons. The sicker the patient the more that the fear of dying looms and from that the cognitive and emotional coping mechanisms that include wishful and non-rationale thinking (that may actually have survival value from an evolutionary perspective).
While I agree we need to be honest and “forthright” (and we have a history of being much less than that) we also need to be aware of the powerful psychological contexts in which both patient and doctor find themselves. Decision making for and about people facing the possibility of dying will always involve much more than a logic-information based approach. So what to do? One possibility is to continue to further the evolution of medical education along the continuum from mechanistic-reductionism towards emotionally intelligent congruent whole person care.
therapeutic misconception is itself sometimes a misconception
This is written from the perspective of a medical oncologist who refers patients for phase 1 trials. Oncologists have worked hard to have a viable clinical research enterprise that includes phase 1 trials. We have done this in the sincere belief that this would allow us to offer therapeutic options to our patients beyond those currently available.
These trials are written as the carefully done scientific experiments they are because anything less than the best science would be unethical and inhumane. However, to say that the only goal of a trial is to achieve the defined objectives is akin to saying that the only goal of a chemotherapy order is to get the drug into the vein.
The likelihood of a phase 1 trial helping an individual patient is currently about 5%. The fact that the odds are low does not change the fact that for the patient and for his/her doctor that is still the goal. These odds are for many patients comparable to the odds of benefiting from "standard" treatment, which is frequently given with considerably less scrutiny.
For many cancer patients, the treatment of choice is participation in a carefully designed and explained clinical trial.