Law & Medicine

Duty vs. confidentiality


Question: During an office visit, a patient confided in his primary care provider his intention to go on a shooting spree. When pressed, he admitted he owned a firearm, but did not truly intend to carry out the act. He denied having a specific target location or victim in mind. The primary care provider (PCP) opted to counsel the patient, and continued him on his antidepressant medication. The PCP did not make a referral to a psychiatrist. Several weeks later, the patient put his threat into action, shooting and injuring several innocent shoppers at a mall.

In a lawsuit against the PCP, which of the following is best?

A. It is unethical for the PCP to report this threat, because it violates the principle of confidentiality.

B. Failure to warn under the circumstances definitely makes the physician liable.

C. The PCP is liable under the Tarasoff doctrine.

D. Without a named victim, there is no one to warn.

E. Whether the PCP is liable may turn on the issue of foreseeability, and an immediate referral to a psychiatrist would have been a more prudent course of action.

Answer: E. One of the physician’s sacred promises is to maintain patient confidentiality. It is more than an unspoken contractual agreement; it’s a central component of trust, without which the doctor-patient relationship cannot exist. Indeed, Hippocrates had cautioned, “whatever in connection with my professional practice, or not in connection with it, I see or hear, in the life of men, which ought not to be spoken of abroad, I will not divulge as reckoning that all should be kept secret.”

Dr. S.Y. Tan, emeritus professor of medicine and former adjunct professor of law at the University of Hawaii

Dr. S.Y. Tan

Yet in special circumstances, a doctor may be expected, even required, to violate patient confidences, as when mandated by state statute; for example, disclosure of communicable diseases and reporting of gunshot wounds and suspected child abuse. A contentious issue revolves around a physician’s duty to warn of potential harm to a third party. This has typically been in the setting where there is some control over the patient’s actions, and where there is a named or easily identifiable potential victim.

Thus, the recent expansive decision handed down by the Washington state Supreme Court stunned many mental health professionals.1 In Volk v. DeMeerleer, a clinic patient had confided in his psychiatrist over 9 years of therapy his suicidal and homicidal ideations. He eventually went on to kill his former girlfriend and one of her sons, and then committed suicide. The psychiatrist did not have full control of this clinic patient, nor had the patient named a specific victim. Nevertheless, the Volk court held that the doctor’s duty extended to any foreseeable harm by the patient, leaving it up to the jury to decide on the issue of foreseeability.

The court held that the psychiatrist had a “special relationship” with the victims, because he had a duty to exercise reasonable care to act, consistent with the standards of the mental health profession, in order to protect the foreseeable victims.

The court reasoned that some ability to control the patient’s conduct is sufficient for the “special relationship” and the consequent duty of care to exist, writing: “We find that absolute control is unnecessary, and the actions available to mental health professionals, even in the outpatient setting, weigh in favor of imposing a duty.” The court also noted that a psychiatrist’s obligation to protect third parties against patients’ violence would normally set aside confidentiality and yield to greater societal interests.

The vexing issue of compromising confidentiality in the name of protecting public safety had its genesis in the famous Tarasoff v. Regents of University of California case.2

In Tarasoff, the California court imposed a legal duty on a college psychologist to warn an intended victim of harm, even though that meant breaching confidentiality in a professional relationship. A jilted patient had confided in the university psychologist his intention to kill his ex-girlfriend. The information, though shared with campus security, was not released to the intended victim, whom the patient stabbed to death 2 months later.

The court reasoned that the protection of public safety was more important than the sanctity of the doctor-patient confidentiality relationship, pointing out: “We recognize the public interest in supporting effective treatment of mental illness and in protecting the rights of patients to privacy and the consequent public importance of safeguarding the confidential character of psychotherapeutic communication. Against this interest, however, we must weigh the public interest in safety from violent assault. ... In this risk-infested society, we can hardly tolerate the further exposure to danger that would result from a concealed knowledge of the therapist that his patient was lethal.”

Following Tarasoff, it is generally accepted that there is no affirmative duty to warn where there is no readily identifiable victim. In a subsequent case, the same court explained that “the duty to warn depends upon and arises from the existence of a prior threat to a specific identifiable victim. In those instances in which the released offender poses a predictable threat of harm to a named or readily identifiable victim or group of victims who can be effectively warned of the danger, a releasing agent may well be liable for failure to warn such persons. Despite the tragic events underlying the present complaint, plaintiffs’ decedent was not a known, identifiable victim, but rather a member of a large, amorphous public group of potential targets. Under these circumstances, we hold that County had no affirmative duty to warn plaintiffs, the police, the mother of the juvenile offender, or other local parents.”3

Commentators have suggested that such post-Tarasoff cases, coupled with the advent of state statutes that codify Tarasoff, have witnessed a retreat from imposing on clinicians a duty to warn.4 By dispensing with the control and “readily identifiable victim” elements, Volk appears to contradict this perceived trend. Foreseeability may turn out to be the dispositive issue.

Some have speculated whether Volk’s expanded view will apply to other professionals such as a family practitioner or even a school counselor. Physicians who are not experienced in treating mental health patients will do well to promptly refer their unstable patients to a specialist psychiatrist.

Volk can also be read as offering judicial insight into patients who pose a health hazard by insisting on operating a vehicle against medical advice. In its 2012 Code of Medical Ethics, the American Medical Association had this to say: “In situations where clear evidence of substantial driving impairment implies a strong threat to patient and public safety, and where the physician’s advice to discontinue driving privileges is ignored, it is desirable and ethical to notify the department of motor vehicles.”5

That is now replaced in the AMA code with a broader version covering the duty to disclose: “Physicians may disclose personal health information without the specific consent of the patient ... to other third parties situated to mitigate the threat when in the physician’s judgment there is a reasonable probability that: (i) the patient will seriously harm him/herself; or (ii) the patient will inflict serious physical harm on an identifiable individual or individuals.”6

Under Volk, physicians are now more likely to forsake confidentiality in the name of public safety – even in the absence of a readily identifiable potential victim.


Next Article: