Francis Johnson was the first anthropologist to be hired for a newly created position of Clinical Social Scientist in a medical education program at a community teaching hospital. As part of a team research project, Johnson conducted ethnographic observations of "Code 99" events occurring on the evening shift. When "Code 99" was announced over the hospital loudspeakers, this was the signal for a team of doctors, nurses, and technicians to assemble to make strenuous attempts to resuscitate a dying patient. Although hospital policy demanded "coding" all patients who expired without a "DNR" (do not resuscitate) order on their charts, Johnson soon realized that residents in internal medicine, in advance and among themselves, were selecting patients who would receive a "slow code" on the night shift (a "slow code" being one that was intentionally conducted too slowly for resuscitation to occur). The residents offered three justifications for designating a "slow code" (one or more might be cited by different house officers to explain a decision about a particular case): (a) the patient was being kept alive by technology alone and should, as a moral decision, be allowed to die; (b) the patient had a chronic disease, which the residents found uninteresting, and from which they felt they could learn little; (c) the chronic disease the patient suffered from was beyond the resources of internal medicine, and the use of technology to prolong the patient's life was a waste of time and effort.
The consequences of documenting this slow code would bear most strongly upon the house officers, who found themselves under considerable professional pressure and competition. Johnson had been hired as part of a new program to upgrade the hospital, which included an affiliation with a more prestigious hospital, and an attempt to rid the hospital's residency program of its longstanding reputation of being composed primarily of graduates of foreign medical schools. At least half the current residents were, in fact, graduates of foreign medical schools, several had been fired since the new program was instituted, and all were uneasy about the renewal of their yearly contracts. In addition, residents feared an interrupted relationship with the cardiology department, whose letters of recommendation were influential for obtaining fellowships for residents who wanted further subspecialty training. A new cardiologist and director of the Cardiac Care Unit was counting "successful codes," (those resulting in resuscitation) as an index of his success, and would surely consider slow codes inappropriate.
Johnson felt a conflict. A decision to formally document the occurrence of slow codes bore not only on the welfare of the residents, but also on professional relationships within the hospital. Would documentation further polarize and politicize the uneasy relationships between professionals within the hospital? Would Johnson be perceived as an informer, a management "fink"? Would documentation betray his research population? Who, indeed, was his research population? Who was his "client"? And to whom did he owe primary loyalty?
Mark Siegler, Director for Clinical Medical Ethics, University of Chicago, Pritzker School of Medicine:
In its recently published ethics manual, the American College of Physicians recommended that "do not resuscitate" (DNR) orders for mentally competent adults be handled according to informed consent guidelines and be based primarily on the clinical indications in a case and on the preferences of the patient. There is general agreement that quality of life factors including age, mental illness, mental retardation, chronic illness, "uninteresting" illness, and cost of care, should not be primary grounds for withholding cardiopulmonary resuscitation (CPR).
There is also general agreement on the issue of "slow codes" or partial codes. The President's Commission commented: "Success" at resuscitation is rare enough when all efforts are expended, so limited efforts are usually doomed from the start. Thus, "partial codes" become a kind of dishonest effort that needs to be justified by reasons stronger than merely the providers' discomfort in discussing DNR decisions" (Deciding to Forego Life-Sustaining Treatment, pp. 250-51).
My own view goes even further. The decision to refrain from CPR should be clear, definite, and communicated in writing to all relevant parties. "Slow codes" or "show codes" should be repudiated as dishonest. If there are sound medical reasons to select among the techniques of resuscitation (that is, partial codes), then they should be clearly noted in the record.
The residents who practiced nocturnal slow codes acted in ways that were detrimental both to their patients and to themselves. Their three justifications for their (in)action were based on a confusing mixture of a dash of medical indications with a large measure of quality of life judgments, which seem to ignore completely any input from the patients. Further, by guaranteeing the failure of all their resuscitation efforts, they were courting the displeasure of the new cardiologist and were jeopardizing their own careers. Simultaneously, their actions constituted bad medicine, bad ethics, and bad personal judgment, and warranted censure if not outright dismissal from the program.
We turn now to the more interesting problem: To whom and for what was Francis Johnson, the Clinical Social Scientist, responsible? Specifically, should he reveal the existence of nocturnal slow codes? The answer to this depends largely on Johnson's conditions of employment and on his role and its obligations. The responsibilities of his newly created position as Clinical Social Scientist are described in different ways including (1) responsibility for medical education, (2) responsibility for a research project, (3) responsibility for ethnographic observations, and (4) "Johnson had been hired as part of a new program to upgrade the hospital. . . ."
At times, these various roles of educator, researcher, ethnographer, and "hospital upgrader" may conflict and they do so here. But which is his primary obligation? The answer seems clear to me. Johnson has agreed to work for a salary primarily as an applied social scientist--a clinical anthropologist. The hospital has bought his expertise to help it improve its operation by developing applied research and education programs. In the course of his activities, he has uncovered a recurring pattern of bad practice--call it if you like "malpractice"--and it is his duty to his employers, not to patients, to "blow the whistle" so that hospital policy and function can be improved. In doing so, he will have earned his salary.
Note that Francis Johnson is not functioning primarily as an observer who is conducting basic ethnographic research on the sort reported by Charles Bosk in his excellent book, Forgive and Remember. Johnson is not employed as a patient advocate. He surely is not employed as a physician advocate. And, unlike physicians and other health professionals, he does not have a primary commitment or role responsibility to patients. In each of these roles, he would likely face a conflict of interest and would have multiple and divided loyalties. Johnson is a lucky man; his responsibilities are undivided; he should be "a management `fink' " and should be proud of it.
One final cautionary note. As this protocol indicates, clinical social scientists like Francis Johnson, should be very sensitive to the many ethical issues raised by their very presence in the clinical setting. In contrast to doctors and nurses, and even to social workers and chaplains, clinical social scientists do not provide services directly to patients, and thus do not have the legitimacy that these traditional health providers can claim. Clinical social scientists should begin asking the ethical question of how they gained access to the medical context, and whether this access is appropriate. It should be their responsibility to inform physicians and housestaff, other members of the health care team, and even patients of their plans to conduct ethnographic observation. Indeed, they should be expected to obtain informed consent when they are doing such research. Otherwise, they are violating patient and health professional privacy and confidentiality and they may be acting deceitfully and manipulatively with respect to both groups. I am also concerned that the presence of such clinical social scientists may interfere with medical activities and with the clinicians' intuitive judgments, which could cause harm to patients. Even more troubling is the possibility that some physicians might abdicate their medical responsibility by delegating difficult decisions to the social scientist expert. Naturally, the trained social scientist lacks technical medical knowledge, has not been prepared in the ethics of human caring, has no responsibility or accountability for decisions made, and has not been sanctioned by the patient to be making such decisions. Thus, the delegation of decision making responsibility to the social scientist would be detrimental to medicine and patients and clearly unethical for physicians. I merely wish to note that the presence of clinical social scientists or clinical anthropologists on the wards may generate some new ethical challenges that we must be aware of.
Jessica H. Muller, Division of Medical Ethics, and M. Margaret Clark, Medical Anthropology Program, University of California-San Francisco:
Let us begin by addressing the final question first: to whom is Francis Johnson's primary loyalty? It is a commonly held axiom that the anthropologist's paramount responsibility is to those he studies (see Principles of Professional Responsibility). However, as this case, among others, suggests, research often involves more than the anthropologist and his or her informants, thereby rendering the question of primary loyalty more complex. It appears in this situation that there are in fact three constituencies to whom Dr. Johnson has some measure of responsibility: the director and staff of the medical education program at the community teaching hospital; the residents participating in the internal medicine training program at the hospital; and those seriously ill patients who may be candidates for slow codes. These three constituencies determine the courses of action open to Dr. Johnson.
As we have seen, Francis Johnson has been hired as Clinical Social Scientist in a medical education program that has been developed to upgrade a community teaching hospital suffering from a poor reputation. Since Dr. Johnson was brought into the institution in this capacity he clearly has some obligation to his employer. We are not told who this is, but it was probably the director of the medical education program, possibly in conjunction with the hospital administration. As Clinical Social Scientist one of his duties apparently was participation on a team research project, one part of which was the study of "Code 99" decisions. Although we are not informed about the specific objectives of the research, it is possible that the research was being conducted, at least in part, to discover and document the strengths and weaknesses of the medical education at the hospital, specifically the residency training program. Any problems that existed could then be addressed by the staff of the medical education program or the hospital administration in the effort to upgrade the hospital. Presumably Dr. Johnson understood the duties associated with his position when he accepted it; therefore, he is ethically bound to his employer, and to the research team of which he was one member, to document what he has observed. If the practice of calling slow codes became apparent in the course of his research, he must respond to it in some fashion.
This could be accomplished in such a way that would preserve the confidentiality of individual residents and patients, as well as shift the "blame" away from the residents themselves. He could, for example, profitably employ his knowledge and skills as an anthropologist to elaborate the social context of "code" decisions. What are the social and structural features of the setting that might influence the use of slow codes? Are attending physicians--who have the legal responsibility for writing a "do not resuscitate" order in the patient's chart--available when it is necessary to discuss with the residents whether a particular patient will be resuscitated in case of cardiopulmonary arrest? Are there particular characteristics of the residents' work that might encourage them to resort to an unofficial slow code policy, such as fatigue or being overwhelmed by other, more "salvageable" patients? Is the issue of "terminal care," including code decisions, being pushed aside in discussions in favor of the challenges presented by obscure diseases or advances in medical technology? Such an analysis might include an examination of the roles of all of the actors who may be involved in these situations, in addition to the residents, Including physicians, nurses, the patient, and members of the patient's family.
Information of this nature could be analyzed and presented to the staff to illuminate how code decisions are made in general, as well as why slow codes in particular may be occurring. If this were done well, it could shift the focus of attention away from the residents onto the situation, with possible positive results for improving the ethical standards of the hospital. Needless to say, all documentation of this information must be done in such a way so the data are rendered anonymous, to preserve the confidentiality of both patients and residents.
In addition to this research function, Dr. Johnson's position in the medical education program also involves an education and training component. In this capacity he could, for example, initiate discussions on the acceptable criteria for code decisions. One way of doing this would be to organize a Grand Rounds (a weekly meeting where topics of interest to physicians are presented and discussed) where a physician and bioethicist could discuss the general issue of the acceptable criteria to be used in making code decisions for the benefit of medical students, residents, and attending physicians. Furthermore, if Dr. Johnson discovered that other personnel, such as nurses, were participating in calling slow codes, he might also invite them to participate in the general discussion on code decisions.
One issue that should be addressed in a general way in either Dr. Johnson's ethnographic account or in the public forum is the emphasis placed by the director of the Cardiac Care Unit on successful codes. If the residents are experiencing pressure to resuscitate patients, regardless of outcome--a poor quality of life or a chronic vegetative state, for example--then this must also be documented. The issue could be addressed in terms of the general policy implications of these actions, rather than the desires of a particular individual.
A final issue concerns Dr. Johnson's ethical obligations to the patients. It appears to us that he is ethically obligated to raise the question of the treatment of patients who are not being allowed the possibility of being resuscitated. Although in raising this question he may in fact influence the professional relationships in the hospital, an overriding issue here is the welfare of the patient. In situations of conflicting ethical principles, there may be certain circumstances under which the anthropologist has to give high priority to the population most at risk within his or her purview. In this case, the patients are more vulnerable than the residents.
In short, it appears to us that documentation goes with the job for which Dr. Johnson was hired. This can be done in a way, however, to protect individuals, to utilize materials for educational purposes, and to point out underlying structural problems in the institution that need to be corrected.