HIV complicates question of who to tell what

Ethics Forum. Oct. 6, 2003.

Should you avoid telling a patient you have to report HIV status? - Can you warn one patient if it violates another's confidence?

Scenario: Should you avoid telling a patient you have to report HIV status?

A general internist strongly suspects from his patient's history and the findings on a physical exam that his patient is HIV-positive. In the state where he practices, people who test seropositive for HIV must be reported by name to the county health department. The physician knows if he tells the patient of his duty to report, the patient will not consent to the test. If he withholds the fact in order to get consent, and the seropositive patient is later contacted by the health department, the physician may lose the opportunity to treat him.

ETHICS FORUM

Answers to questions on ethical issues in medical practice

SEE ARCHIVES

Reply:

With the advent of named reporting for HIV in the United States, some physicians are having to weigh the risk of clients' refusal of needed HIV testing with the benefits of maintaining an honest relationship with them.

Yet, the concern that named reporting will deter patients from being tested has not been substantiated.

In surveying people at risk, colleagues and I found that, theoretically, named reporting does constitute a barrier to testing.

See related content

At least two other reported studies, however, concluded that named reporting had little or no effect on actual testing uptake.

Despite the paucity of data linking named reporting to HIV testing acceptance, the physician in this case believes a discussion of named reporting would deter the patient from being tested.

The question of whether to be honest with this patient is not difficult to answer -- "informed" consent requires that patients be given information about named reporting so that they can make a truly informed decision about whether to test.

Sharing information on the state's requirements for named reporting is both legally and ethically correct. In addition, the literature on patient-centered care suggests that frank discussions and shared patient-physician decision-making lead to better health outcomes, as well as greater patient and physician satisfaction.

In this case, what is most important is how the physician discusses the importance of and options for testing with his patient.

Early in the epidemic, before the advent of life-prolonging treatments, some people argued that the risks of discrimination outweighed the benefits of knowing one's status.

Now, however, it is clear that early detection of HIV infection is critically important both for early treatment as well as for prevention.

The benefits of testing far exceed the risks of named reporting, which to date has never resulted in a breach of confidentiality, despite the attempt (unsuccessfully) of one activist who was criminally prosecuted for trying to sell a list of names to local media.

In discussing HIV testing with this patient, it will be important to explore his concerns, and to make sure that he is aware of the benefits of knowing one's HIV status as well as the risks of delaying testing.

If his concerns over named reporting are not adequately assuaged, the physician might offer the option of anonymous testing, which is offered by most public health departments and which does not result in named reporting.

Some physicians may be able to offer anonymous testing by creating an anonymous code for the patient, keeping the test results separate from the patient's chart, and collecting clinic and testing fees without involving the patient's insurance source.

Although a practitioner is not obligated to report results of an anonymous test, once an HIV-positive patient seeks care, the physician would be obligated to report him or her.

Some health departments also have anonymous first physical exams so patients who are concerned about disclosure of their HIV status can get important clinical information on the need for antiretroviral treatments before having to seek treatment officially and have their names reported.

Although there is some risk that if this patient is referred elsewhere for anonymous testing he could be lost to follow-up, if the recommended discussions occur, this patient would likely be more trusting of his physician and either accept testing from the physician or follow up with him for treatment after receiving the anonymous test results.

Alternatively, if this physician chooses to withhold information about HIV reporting practices, this patient might feel uncomfortable with the physician when he discovers this breach of trust and may not value the physician's recommendations regarding treatment.

Honesty is still the best policy, especially when it comes to the physician-patient relationship.

--Freya Spielberg, MD, MPH, Assistant professor of family medicine, University of Washington School of Medicine, Seattle

--Robert W. Wood, MD, Director, HIV/AIDS Control Program For Public Health, Seattle & King County, Seattle; associate professor of medicine, University of Washington School of Medicine, Seattle

Note: Dr. Spielberg's time to prepare this manuscript was supported by a grant from the National Institute for Drug Abuse.

Scenario: Should you avoid telling a patient you have to report HIV status?

An internist has both a man and his longtime girlfriend as patients. The woman consented to be tested for HIV and had a positive result, which she feared but suspected. She refuses to tell her partner, or allow the physician to tell, saying, "It would ruin our relationship." The patient says she will continue to have sexual relations with her partner. She says they do not want children and her partner has had a vasectomy, so they do not need to worry about transmitting the virus to a fetus.

Reply:

Ethical decisions are not themselves difficult. Most of us have intuited social norms and can tell right from wrong. Context is the real difficulty, by creating nuance and generating conflict between competing ethical priorities. This scenario is a prime example. One ethical imperative is the duty to warn: the male partner should be informed. Another ethical imperative is to respect personal autonomy and patient confidentiality: the female partner's wishes should be followed. The physician, often under the duress of clinic practice, must make a decision.

First, the physician could fall back on the law. Jurisdictions vary in their requirements: The physician may be prevented from informing the male partner, or may be required to inform the male partner. Or the law may be silent. But for the question posed here, invoking the law is a cop-out. Laws cannot deal comprehensively with context and nuance and may not be in keeping with specific ethical concerns.

Second, the physician can delve into the context, and look for clues in his patients and their relationship that might provide an answer. The character of the people involved and the circumstances of their relationship, including its exclusivity, are critical. (It was said of Oliver Wendell Holmes that he would have given vapors to the Vestal Virgins; the physician must always be ready for the lurking presence of other sexual contacts in the most unsuspected places.)

The physician would probably want to explore that phrase "ruin our relationship" in some depth with the female partner. She might be encouraged to compare the ruin caused by informing her partner now to the future ruin when he does find out. (And he will find out, either through things that happen to her or the discovery of illness in himself.)

She might want to consider the consequences of continuing sexual exposure with knowledge of the potential to transmit a fatal (or at least life-limiting) disease.

The physician should probably understand better how she acquired the disease and determine if drug injection or crack cocaine were involved, particularly since the probability of penile-vaginal transmission is low. If drugs are a part of their relationship, the probabilities of his contracting the disease are strongly affected, and the complexity of the problem is significantly increased.

The female patient raises the issue that transmission to a fetus is not a possibility, but this changes little. It removes vertical transmission as an item for consideration but provides no help with the horizontal.

Ultimately, we must assume that in the worst case, the physician will be unable to convince the female patient that her long-term benefit lies in informing her partner, despite potential short-term consequences. In such a case, the physician must use judgment to decide whether the probability of adverse consequences is greater from violating one ethical imperative or another -- whether in this context the duty to warn is more important than respecting autonomy and patient confidentiality.

Given this worst case, I would make it clear to her that either she tell him or I will. But to do so generates the ethical imperative of taking responsibility for one's own action. At the very least, I must be prepared to provide counseling and guidance either directly or through referral.

This case thus reflects more than the phenomenon of competing ethical priorities. Ethical issues are often presented as stand-alone dilemmas. In fact, they are cascades that all concerned must confront.

Unfortunately, such long-term commitment, intrinsic to being a physician, too often involves Hobson's -- or even worse, Sophie's -- choice.

--Richard Rothenberg, MD Professor, Dept. of Family and Preventive Medicine, Emory University School of Medicine, Atlanta

Ethics Forum discusses questions on ethics and professionalism in medical practice. Readers are encouraged to submit questions and comments to philip.perry@ama-assn.org or to Ethics Group, AMA, 515 N. State St., Chicago, IL 60654; fax 312-464-4613. Opinions in Ethics Forum reflect the view of the author and do not constitute official policy of the AMA.