A Physician’s Suicide

When Dr. A began therapy, he told Dr. P he was his “last hope,” a remark that the psychiatrist felt suggested magical faith but perhaps implied that the patient would make minimal effort himself or take little responsibility in working toward improvement. He saw Dr. A as having a positive transference toward him but also some feelings of competition. In this context, he mentioned that during one period of improvement, Dr. A spoke of switching from anesthesiology to psychiatry and began reading extensively in the field.

Even though Dr. P had not known Dr. A before treating him, he was aware that he regarded his patient as a colleague and that this had complicated their work together. In retrospect, he regretted having allowed Dr. A’s wife to attend the therapy sessions and wished he had insisted on seeing Dr. A alone. He knew he would not have acquiesced to the patient’s pressure to include his wife if Dr. A had not been a physician. He said he had rationalized doing so because Dr. A’s wife was a pharmacist.

In the therapy, Dr. A referred to his wife’s affair as “something that happens in life that could not be avoided.” He was glad his wife had not left him and did not wish to discuss the matter any further. Dr. P did not believe that this superficial, stoical resignation accurately reflected Dr. A’s feelings but did not challenge his patient’s presentation of the matter. He saw the patient’s insistence on his wife’s presence in the therapy as reflecting his desire to force her to take care of him as a way of punishing her, but this was not discussed either. After the patient’s death, Dr. P wished he had explored more fully the nature of the couple’s relationship before the affair, which might have shed light on what precipitated it. He suspected that his patient had been emotionally neglectful of his wife and child even before he became depressed, but he did not feel comfortable discussing this.

It likely would have been easier for Dr. A to express his anger and hurt over the affair without the presence of his wife. Seeing the couple together might have had value if the therapy had dealt with their relationship. As it was, the wife attended the sessions as a consultant/caregiver, which only compounded Dr. A’s dependent tendencies and did not permit him to express the intense feelings that the therapist suspected underlay his passive resignation. Dr. A left a suicide note for his wife, asking to be forgiven for what he was doing, yet he arranged his suicide to shock her with the discovery of his body, rigged to the infusion tubing. This suggests both his anger and an enduring lack of forgiveness over the affair.

Work-related stress has been seen as contributing to depression in physicians. Dr. A attributed his difficulties, in part, to the stress of working with a new and difficult supervisor. Vaillant’s classic study that followed a group of physicians throughout their careers (1) suggests that the reverse is more often true. Work stress becomes a significant factor in already vulnerable physicians, a fact that Dr. P recognized. Dr. A’s vulnerability was evident in his 7-year depression after treatment for tuberculosis. He reacted similarly to his wife’s affair, the second major trauma of his life, in developing a depression that had persisted for over 4 years before his suicide.

Many physicians consider their identity as intimately tied to their work (2), and this may aggravate their distress when depression makes working impossible. This was certainly evident in Dr. A’s case. His intense identification with his work as a physician was likely rooted in his life-long struggle to fulfill his mother’s career expectations for him and his fears about his sense of inadequacy.

In many cases we have seen, therapists who believed their patients were imminently suicidal were nevertheless reluctant to hospitalize their patients. Dr. A’s psychiatrist had no such difficulty, putting Dr. A in the hospital immediately after learning that he had bought a gun. At that point in the treatment, Dr. A had been fully compliant with taking his medications and was compliant as well with the hospital admission. He undoubtedly knew that once he told his wife, his psychiatrist would be informed of his gun purchase and would likely take action to prevent his suicide.

Following the hospitalization and the unsuccessful course of ECT, Dr. A appeared to become less forthcoming about his suicidal intentions, insisting over his psychiatrist’s objections that he was ready to return to work. He may have interpreted the attempt at dissuasion as indicating that his “last hope” was not so hopeful about his prognosis. It was at this point that Dr. A began using benzodiazepines to treat his intense anxiety, a symptom that is a frequent warning sign of an imminent suicidal crisis (3, 4). Although Dr. P expressed shock at the suicide, that he sensed some danger is suggested by his calling his patient the night of the suicide to see how he was doing.

At the time of his death, Dr. A’s depression was described by Dr. P as moderate. Although the patient expressed little emotion in his sessions, he was clearly anxious about returning to work, humiliated over his inability to do so, and hopeless over his general lack of improvement. We suspect that he had desperately set returning to work as a condition for living—a condition that he could not possibly meet in his current state—and killed himself when this became evident to him. If he could not live as an anesthesiologist, he would at least die as one.

There are many cases in which medication makes psychotherapy possible. In our experience, in a significant number of cases, unless an ongoing psychosocial problem is addressed, medication does not work. This seemed to be such a case. Our experience does not support the notion that medication management and psychotherapy should be conducted by different persons, as Dr. P had initially recommended. Although we recognize there are situations where this may be necessary or advisable, we have seen far more cases in which the treatment by two different clinicians was clearly harmful. Poor communication between the two can result in problems going unnoticed. Too frequently, the patient can fall between the cracks.

Of interest, his experience with Dr. A appeared to cause Dr. P to rethink his approach of separate medication management and psychotherapy. In summarizing what he might have done differently, he stated that once he had agreed to perform psychotherapy, he should have done it more thoroughly, insisting on seeing his patient alone and on exploring the couple’s marital difficulties. He added that he no longer emphasizes psychopharmacology to the detriment of psychotherapy and now favors a combined approach.

The deference shown to Dr. A is not unusual. An experience recently related to us by a psychiatrist who knew of our interest in this problem illustrates this point. On duty in a psychiatric emergency room years before, the psychiatrist had examined a physician-patient brought in by his wife after a suicide attempt. Although there were no medical indications for admission, the psychiatrist at once recognized that his patient was seriously suicidal and recommended admission to the psychiatric unit of the hospital. Both the physician and his wife resisted the admission, insisting that it was not necessary. They persuaded the psychiatrist, against his better judgment, to allow the physician to go home. The psychiatrist learned later that the physician had killed himself that night. The psychiatrist was sure that had the patient not been a physician, he would have insisted on hospitalization. The case had troubled him for years, and he wanted not merely to unburden himself but to have others learn from his experience.

In another case of physician suicide that we have examined, the psychiatrist, who was performing psychotherapy as well as prescribing medication, referred his physician-patient to a gay therapist to deal with issues related to his emerging homosexuality. Although the psychiatrist considered himself to be generally comfortable working with homosexual patients, he was reluctant to explore the physician’s inability to become involved in a sexual relationship. There was virtually no communication between the two therapists, and only after the patient’s death did the psychiatrist learn the full extent of the difficulties that had become evident during the other treatment. One of his main regrets was his decision not to address them himself in his treatment of the patient.

Dr. A’s case was presented as part of a project, the Suicide Data Bank, which systematically collects and analyzes information from therapists about their patients who died by suicide while in treatment (3, 5, 6). Two of these cases had been physicians, an occupational group that has been shown to die by suicide more frequently than the general population (7–10), even though depression does not appear to be more common among physicians (11, 12).

Numerous reasons for the high rate of physician suicide have been put forth. These include the physicians’ tendency not to recognize depression in their patients or themselves (13–15), less frequent identification of depression among physicians because of their lower use of a regular source of health care (16), and institutional barriers to help-seeking in the form of restrictions regarding medical licensing, hospital privileges, and health and malpractice insurance placed on physicians after a psychiatric diagnosis (17, 18).

An additional factor that has been noted (19) is the deference psychiatrists too often show their medical colleagues who do seek help, which compromises the treatment process and may render it ineffective to prevent the patient’s suicide. We are learning that trying to be too “nice” to colleagues is sometimes not nice at all. Other doctors deserve the best of our skill and judgment, even when insisting on good treatment annoys them or causes them pain. Our surgical brothers and sisters know that a laparotomy is not always welcome, but they do not fail to insist on it when the indications are clear. The patient may choose whether or not to have the operation but does not decide how the procedure is conducted, and the family is not invited into the operating room.

Received March 5, 2003; revision received June 17, 2003; accepted June 20, 2003. From the American Foundation for Suicide Prevention; the Department of Psychiatry, New York Medical College, Valhalla, N.Y.; the Department of Psychiatry, Harvard Medical School, Cambridge, Mass.; and the Department of Health Services, Lehman College of the City University of New York, New York. Address reprint requests to Dr. Hendin, American Foundation for Suicide Prevention, 120 Wall St., New York, NY 10005; [email protected] (e-mail).