Martin L Smith, STD, is an Associated in the Department of Bioethics, Cleveland Clinic Foundation, Cleveland, Ohio.
Kevin P Martin, MD is a Child and Adolescent Psychiatrist in the Department of Mental Heath, Kaiser Permanente, Cleveland.
AIDS (acquired immunodeficiency syndrome), now in pandemic proportions, presents formidable challenges to health-care professionals. The human immunodeficiency virus (HIV) infection and its related diseases have also raised a number of thorny ethical questions about government and social policy, health-care delivery systems, and the very nature of the physician-patient relationship. This article presents the case of an HIV-positive patient who presented the treating physician, a psychiatrist, with an ethical dilemma. We provide the details of the case, identify the ethical issues it raises, and examine the ethical principles involved. Finally, we present a case analysis that supports the physician’s decision. Our process of ethical analysis and decision making is a type of casuistry,1 which involves examining the circumstances and details of the case, considering analogous cases, determining which maxim(s) should rule the case and to what extent, and weighing accumulated arguments and considerations for the options that have been identified. The goal of this method is to arrive at a reasonable, prudent moral judgement leading to action.
The patient, Seth, is a 32-year-old, HIV-positive, gay, white male whose psychiatric social worker had referred him to a community-mental-health-center psychiatrist for evaluation. He had a history of paranoid schizophrenia that went back several years. He had been functioning well for the last two years as an outpatient on antipsychotic medications and was working full time, socializing actively, and sharing an apartment with a female roommate.
The social worker described a gradual deterioration over several months. Seth had become less compliant with his medication and with his appointments at the mental-health center, had lost his job, had been asked to leave his apartment, and was living on the streets. He was described as increasingly disorganized and paranoid. His behavior was increasingly inappropriate, and he had only limited insight into his condition.
On examination, Seth was thin, casually dressed, slightly disheveled, and with poor hygiene. His speech was spontaneous, not pressured, and loose with occasional blocking. [That is, he spoke spontaneously, he could be interrupted, and his speech was unfocused with occasional interruption of thought sequence.] His psychomotor activity was labile [unstable]. His affect was cheerful and inappropriately seductive, and he described his mood as ''mellow." He denied having hallucinations, systematized delusions and suicidal or homicidal ideation. He admitted having ideas of reference [incorrect interpretation of casual incidents and external events as having direct reference to himself], was clearly paranoid, and at times appeared to be internally stimulated. He made statements such as: “They’re blaming me for everything,” and “I’m scared all the time,” although he was too guarded or disorganized to provide more detail. His cognitive functioning was impaired, and testing was difficult given his distracted, disorganized state. His judgement was significantly impaired, and his insight was quite limited.
At the time of the evaluation, Seth indiscriminately revealed his HIV-positive status to the staff and other patients. He claimed he had been HIV positive for five years, and he denied that he had developed any symptoms of disease or taken any HIV-related medications. He was not considered reliable, and the staff sought confirmation. After he provided the location and approximate date of his most recent HIV test, the physician confirmed that the patient had been HIV positive at least since the test, about a year earlier.
When asked, Seth stated that the was not currently in a relationship. He appeared to be disorganized and could not name his most recent sexual partner(s). He could not remember whether he had been practicing safer sex and whether he had informed his partners of his HIV positive status.
In addition to the information he obtained during the evaluation, the psychiatrist, by chance, had limited personal knowledge of the patient. Through his own involvement as a member of the local gay community, the psychiatrist had briefly met the patient twice – once while attending an open discussion at the lesbian-gay community center, and later, at a worship service in a predominantly lesbian-gay church. The physician recalled that Seth had seemed to be functioning quite adequately, at least superficially. He was somewhat indiscriminately flirtatious, his behavior was otherwise appropriate, and he did not appear to be psychotic or disorganized in his thinking. He was not overtly paranoid and did not publicly reveal his HIV-positive status.
Through the church, the psychiatrist had also become acquainted with Maxwell and Philip, who were partners in a primary sexual relationship. Before Seth’s decompensation [deterioration of existing defenses, leading to an exacerbation of pathologic behavior], but after he was known to have tested HIV positive, Seth and Maxwell had been lovers. Maxwell left Philip and moved in with Seth for about two months, but then left Seth and returned to Philip around the time of Seth’s decompensation.
The psychiatrist was not privy to details of Maxwell and Seth’s or Maxwell and Philip’s sexual practices. He did not know of the HIV status of Maxwell or Philip, or whether either had ever been tested. In addition, he was unaware of whether Maxwell or Philip know of Seth’s HIV-positive status at the time of Maxwell’s relationship with Seth, or at any time thereafter. During the evaluation, Seth did not recall having met the psychiatrist, nor did he mention his relationship with Maxwell.
Seth agreed to enter a crisis stabilization unit and to resume treatment with antipsychotic medications. Free to come and go at will during daylight hours, he left the unit on day two, failed to return, and was lost to follow-up. His mental status had not changed significantly before he left the crisis unit.
In this case, the physician’s duty to maintain physician-patient confidentiality conflicts with his duty as a psychiatrist to warn third parties at risk. Clearly, a patient’s status as HIV positive is a matter of confidentiality between doctor and patient. Just as clear is the risk for third parties to whom the patient may pass the virus via sexual intercourse. It is unknown whether everyone infected with HIV will develop AIDS, or how many months or years may intervene between infection and the appearance of full-blown AIDS. However, once AIDS develops it is always fatal.2 Therefore, there is a potentially lethal risk to a person having intercourse, particularly without employing safer sex-practices, with another infected with the HIV virus.
This ethical conflict raises two questions. Is it permissible to violate confidentiality to warn a third party at risk? Is there a duty to violate confidentiality to warn a third party at risk? The potential benefit to the third parties must be considered, as well as the strength of the principle of confidentiality in the patient-physician relationship. There is also wider societal consideration as to how breaches of confidentiality, even for good reasons, will affect voluntary testing and seeking of prophylactic treatment by HIV-positive persons. This societal consideration must be weighed against the benefit to the individual third party of knowing the risk and then choosing to be tested and treated and choosing to be tested and treated and choosing to take precautions against infecting others.
In this case, another issues arises from the fact that the physician’s of at least one third party who may have been placed at risk possibly without his knowledge, was obtained through personal knowledge, outside the professional relationship. Is it appropriate to bring this information into the clinical setting, particularly because it is so central to the primary ethical issue? Does the physician have an obligation to act on this information?
Finally, two additional sets of issues complicate this case. First, the patient’s decompensation and disappearance necessitate the physician’s choosing a course of action without patient consent or cooperation, and with patient-supplied information that is incomplete and probably unreliable. Second, a breach of confidentiality could greatly damage the physician’s position as a psychiatrist and a trusted member of the gay community, offering assistance directly to some and referral to many others. Given these issues, what should the physician do?
Some background information will be useful in analyzing the ethical issues of the case. This information includes basic ethical values and norms, and legal mandates and opinions about confidentiality, the duty to warn, and HIV/AIDS reporting.
Whether privacy is viewed as a derivative value from the principle of autonomy or as a fundamental universal need with its own nature and importance,3 privacy, and the associated issue of confidentiality, is generally accepted as essential to the relationship between physician and patient. The purpose of confidentiality is to prevent unauthorized persons from learning information shared in confidence.4 Stated more positively, confidentiality promotes the free flow of communication between doctor and patient, thereby encouraging patient disclosure, which in turn should lead to more accurate diagnosis, better patient education, and more effective treatment.
The Hippocratic Oath is evidence of the long-standing tradition of confidentiality in Western though: “What I may see or hear in the course of treatment… I will keep to myself, holding such things shameful to be spoken about.” More recently, the American Medical Association,6 the American Psychiatric Association,7 the American College of Physicians, and the Infectious Diseases Society of America8 have reaffirmed the right of privacy and confidentiality, specifically for HIV-positive patients. Without the informed consent of the patient, physicians should not disclose information about their patient. The Center for Disease Control also recommends that patient confidentiality be maintained, because the organization believes that a successful response to the HIV epidemic depend on research and on the voluntary cooperation of infected persons.9That is, the interests of society seem best served if the trust and cooperation of those at greatest risk can be obtained and maintained.10
Within the complexities of clinical care, should patient confidentiality be regarded as absolute, never to be breached under any circumstances (as claimed by the World Medical Association in its 1949 International Code of Medical Ethics11)? Or should confidentiality be regarded as a prima facie duty? (That is, should it be binding on all occasions unless it is in conflict with equal or higher duties?12)
Most commentators and codes conclude that patient confidentiality is not absolute and, therefore, it could – and even should – be overridden under come conditions.13 In other words, in a specific situation in which patient confidentiality is one value at stake, the health-care provider’s actual duty is determined by weighing the various competing prima facie duties and corresponding values, including confidentiality. (As might be expected, not all authors accept this conditional view of confidentiality and argue for its absolute quality.14) There is less unanimity about the circumstances under which patient confidentiality can be justifiably breached. More specifically for HIV-positive patients, the controversy revolves around the premise that some circumstances might create a duty to warn endangered third parties, even at the expense of confidentiality. The potential for harm to HIV-positive patients through breaches of confidentiality is great. Discrimination, isolation, hospitality, and stigmatization are all too real for these patients when their HIV-positive status has become know to others.15 Further, societal harm is possible if these patients – who might ordinarily seek medical attention voluntarily – refrain from doing so, knowing that professional breaches of confidentiality may ensue. Without ignoring this potential societal harm, the majority opinion of professional codes and of ethical and legal experts16 foresee the possibility of a duty to warn through discrete disclosure, especially if others are in clear and imminent danger if the patient cannot be persuaded to change hi behaviors or to notify those at risk of exposure.
Public health regulations often reflect the same conclusion – that confidentiality can be compromised under certain circumstances – and therefore mandate reporting HIV-positive and AIDS patients to public health authorities. Patient confidentiality is not to be upheld so strictly that it obviates an ethically justified (and usually legally mandated) duty to report such cases to authorized health agencies. Those who support such public policies view society’s right to promote its health and safety, and the need for accurate epidemiological information, to be at least as important as an individual’s right to privacy and confidentiality.
In trying to balance patient confidentiality with other professional values, the California Supreme Court decision in Tarasoff v. Regents of the University of California17 has become a guideline for other courts and health-care professionals (although technically this decision applies to only one state and specifically addresses a unique set of circumstances). In this famous and controversial case heard before the California Supreme Court in 1976, the majority opinion held that the duty of confidentiality in psychotherapy is outweighed by the duty to protect an intended victim from a serious danger of violence. The court explained the legal obligation to protect and the potential duty to warn as follows:
When a therapist determines, or pursuant to the standards of his profession should determine that his patient presents a serious danger of violence to another, he incurs an obligation to use reasonable care to protect the intended victim against such danger. The discharge of this duty may require the therapist to take one or more various steps, depending upon the nature of the case. Thus it may call for him to warn the intended victim or others likely to apprise the victim of the danger, to notify the police, or to take whatever steps are reasonably necessary under the circumstances.18
Regarding the limits placed on confidentiality under these conditions, the court stated: “The protective privilege ends where the public peril begins.” This “Tarasoff Tightrope” identifies for the professional the dual duties of promoting the well-being and interests of the patient and protecting public and private safety.
Given the general jurisdictional autonomy of each state, the duty to protect and the potential duty to warn as adopted in California has been applied differently in different states.20 Although most commentators assume that Tarasoff is relevant for sorting out the issue of confidentiality relative to HIV-positive patients, this assumption is not universally accepted.21
Without a state statute or court case that specifically addresses the tension between patient confidentiality and the right of others to know whether they may have been exposed to HIV infection, and given the conundrum of legal principles relating to AIDS confidentiality, it is unclear as to who must be warned and under what circumstances.22 This lack of clarity is in indication that, in practice, the professional duty to warn is not absolute but always conditioned by the circumstances of the case (that is, the duty to warn is a prima facie value).
The above paragraphs describe an emerging consensus among health-care professionals who face confidentiality dilemmas, although universal agreement has not been fully achieved. Further, this emerging consensus and its contributing principles by no means provide easy answers to ethical quandaries. Each case, with its own specific set of relevant circumstances, must be analyzed and judged individually. Such an analysis of the presented case now follows.
Seth’s case, perceived as a dilemma by the psychiatrist, could be brushed aside easily if the information obtained outside the therapeutic relationship was simply ignored. But the lethality of HIV infection make it difficult to dismiss the information either as irrelevant or inadmissible for serious consideration. Had the information been obtained by unethical means (for example, by coercion or deception), a stronger justification for not using the information might be made. Such is not the situation. Without reason to ignore this information, the psychiatrist must incorporate this “data of happenstance” into his decision. To do so, of course, places him precisely at the crossroads of the dilemma: to uphold confidentiality, to warn the third party, or to create an option that supports the values behind these apparently conflicting duties.
Several factors ethically support both a breach of confidentiality and the physician’s duty to warn and protect the third party: the emerging professional consensus that confidentiality is not absolute; the known identity of a third party who stands in harm’s way; the risk to unknown and unidentified sexual partners of the third party; and the deadliness of AIDS. Such a combination of factors is what the professional statements noted above23 have tried to address in their allowance for limits to patient confidentiality. In this case, the risk to the known third party has already been established, but other people may be at risk, including sexual partners of the patient and those of the third party. Individuals infected and unaware will not benefit from prophylactic therapies.
An additional reason for the psychiatrist to warn the third part is the patient’s mental status, which probably renders him incapable of informing his sexual partner(s) or of consenting to the physician’s informing them. On admission, he was not able to name his partner(s), and he was lost to follow-up without significant change in his mental status. Without the decision-making capacity of the likelihood of action on the part of the patient, any warning to the third party would have to come from the physician or through public health officials notified by the physician.
But the duty to warn, incidental information, and mental status are not the only factors that need to be considered here. Patients are subject to the risks of discrimination when their HIV status is disclosed. But for a patient who has indiscriminately revealed his own HIV-positive status, the physician’s contribution to this risk of discrimination through discreet disclosure to one person may be minimal.
Also to be considered is the societal risk that testing and prophylactic treatment of HIV-positive persons will decrease if confidentiality is not upheld. Members of the lesbian and gay community are often mistrusting of medical and mental health professionals,24 perhaps with valid reason. Mistrust, fear, and nonparticipation in voluntary programs may increase if confidentiality cannot be assured. Persons will be less likely to come forward voluntarily for education, testing, or other assistance if their well-being is threatened as a result. In Tarasoff, the court declared that protective privilege ends where social peril begins. In this case, overriding the protective privilege of the individual could lead to greater societal peril. Trust in this physician by members of the lesbian and gay community benefits individuals and the community as a whole, by improving access to medical and mental health services. A breach of confidentiality, if it became known, could damage this trust, as well as the physician’s reputation, reducing his professional contributions to the community. This professional loss would be significant.
Also to be considered is the general knowledge of the higher risk among gay and bisexual men for HIV infection, as well as the information in the gay community as to what constitutes high risk behavior and what precautions can decrease risk of infection. Thus, we can reasonably assume that a gay or bisexual male is already aware of his risk and that of his sexual partner(s) for carrying the HIV virus. Warning a probably knowledgeable third party about the HIV-positive status may be of little benefit to the third party, while it risks the greater individual, societal, and professional harms discussed above. Regarding the risk to unknown sexual partners of the patient, whatever their number, the physician is powerless to change their fate precisely because they are unknown to him.
The duty to maintain patient confidentiality and the duty to warn third parties at risk can both be viewed as prima facie duties. In clinical situations such as the one described here, when one duty must be weighed against another to arrive at an ethically supportable solution, the weighing should take place only in the context of the given case. In this case, we found no solution that upholds all the duties; thus a choice must be made between the two duties.
We submit that, although there is support for the physician to warn the third party, there is greater support for upholding confidentiality in this case. The individual risk of discriminatory harm from disclosure is possible, although admittedly small. Further, it is reasonable to presume the third party's awareness of his risk and of the risk to his sexual partner(s) of carrying the HIV virus, and thus, his awareness of the need for appropriate precautions.
Even more persuasive is the peril to the local gay community and the wider society if a breach of confidentiality increases mistrust of the healthcare system and decreases the effectiveness of this particular psychiatrist to provide quality professional care. In this case, the confidentiality of the physician-patient relationship should be maintained.
What has been presented here can serve as a model for ethical decision making within the complexities of clinical care. As cases and their accompanying ethical questions arise, the details of each case should be gathered. Any tendency to label the case prematurely as a particular type (for example, a duty-to-warn case) should be resisted. Such a label can divert attention from relevant details that make each case unique. In examining the facts of the case and judging their significance, the values and duties at stake can be identified. If necessary and practical, background material and analogous cases should be researched. Ethical dilemmas present persons with hard choices. While several solutions may have some ethical support, few can be labeled as perfect solutions. Often, choosing one solution over another leaves behind an ethically significant value and regrettably may even produce harm. The circumstances described here presented the psychiatrist with a hard choice and no easy answer. We have suggested an ethically supported solution, but we found no perfect solution for the dilemma.
Medical Association," in Encyclopedia of Bioethics.
. Archives of Path fo_gy, & . l,.!:z6oratory Medicine 114 (1990):304 8; D. Seiden, "HIV ·Seropositive Patients and Confidentiality," Clinical Ethics Report (1987): 1-8H.Zomina, "Warning Third