Medicine constantly holds the tensions of contrasts. Of life and of death. Of fighting and of folding. Of truth and of deceit. Herein lies the art of medicine. Holding the balance of the rational advance of science against the relative world of values and attitudes. This tension, in turn, reflects the human condition, wherein man is both symbolic and physical.

Some time ago I had a case that brought these issues into sharp relief. I had been asked to consult on an 86-year-old Russian lady who spoke only a few words of English. She had a slowly advancing head and neck tumor for which no further antitumour treatment was possible. She had neuropathic pain, which settled with a tincture of anticonvulsant. At the first consultation her son pushed urgently in front of her and whispered emphatically to me, “Do not tell her that she has cancer. If she knows, it will kill her.” I nodded neutrally. After 15 minutes of preliminaries, she blurted out in broken English, “Do I have cancer?” There was no time to reflect as in an editorial—there was no interpreter to impartially dissect the question—and the son's presence loomed large. I simply did not know her. I answered her in a reflex, as these thoughts flashed through my mind.

Robert Twycross, the British doyen of palliative medicine, recently wrote about the challenge of “truthful communication between health professionals and patients.” (Twycross, 2006). He cites the literature and 30 years of clinical experience to conclude that “although truth may hurt, deceit hurts more.” And that “a conspiracy of silence or of words” leads to “desolation, depersonalization … and suffering” of the patient.

Telling the truth (or not) has a particular ethical framework. A broader paradigm that incorporates truth telling would be of autonomy and paternalism. Should I, the doctor, paternalistically decide what information to reveal or to conceal?

A common defense for not telling the truth to the patient and deferring to the family is that the ethical and social unit is the family, not the individual—and the family knows the patient best. Therefore, although one may appear paternalistic to the individual, one is communicating truthfully with the relevant entity, the family. This is a not uncommon model in many parts of the world, including Eastern Europe.

An experienced psychiatrist recently noted that a family's psychological dynamics—coping, communicating, and caring—tends to override the influences of culture and country in clinical and therapeutic situations (S. Bloch, pers. comm., September 5, 2006). By analogy one might extend that to individuals. Around the world are we more similar to one another than not, in terms of the existential challenge of the fear of death (of nonexistence)?

Passing sensitive clinical information should be delicately imparted, based on timing and content, that is, how much to reveal and when to discuss it. At what point in the relationship between doctor and patient is it appropriate? At what point in the patient's own intellectual or spiritual development? This approach is, however, by implication paternalistic. I, the doctor, know better than the patient when to divulge information. In the West the culture is heavily invested in the paramount importance of the ethic of autonomy. In a landmark case in Canada's Supreme Court the judge intoned: “The patient has the right to be wrong.” (Hopp v. Lepp, 1980).

From a legal–ethical perspective, this statement appears reasoned, even reasonable. However how can a doctor stand aside—keeping in mind his duty of care to heal—and allow the patient to suffer as a consequence? A thoughtful case, well worth reading, analyzes a preventible death following complications of childbirth of a Jehovah's Witness. It highlights issues of autonomy, the ambient culture, and a doctor's duty of care. An addendum some years later described five orphaned children being swept up into the bosom of the community, seemingly not missing a beat. The doctors tried to persuade the patient and thereby fulfill their duty of care—to heal and to alleviate suffering (Stotland, 1999). It was the ethic of the community in one sense that loomed larger than the right of the individual. It is not clear that the patient's decision was wrong.

When the merits and risks of a medical decision are evenly balanced but filled with uncertainty, who is best placed to decide: the patient (autonomy) or the doctor (paternalism)? Surely this life or death decision rests with the patient. And whatever informed decision the patient makes the doctor will support. On the other hand, who could be more expert than an experienced specialist physician? Ultimately—short of forcing a person—the decision is the patient's. Paradoxically, though, the patient can nullify his or her autonomy in favor of the doctor.

Can a medical decision with its attendant risks and responsibilities be truly shared?

I have noticed a disturbing trend among medical staff at the “clinical coal-face” of cancer care. At times it appears as if doctors “hide” behind the shield of patient autonomy to avoid making difficult medical decisions. Were the paradigm of autonomy taken to its logical extreme, would doctors ever need to make a decision? We cannot presume to know what patients think or feel without asking them. We cannot decide for them. Nevertheless our duty of care demands that we assure ourselves that the patient is competent and decision making is not distorted by psychological burdens.

The ever-hovering cloud of medico-legal action also influences decision making. Doctors are more inclined to behave defensively and “enforce” autonomy to reduce their own risk taking and responsibility. Thus in a sense the patient–doctor relationship is diminished.

Whereto the physician? What was the correct stance to adopt with my 86-year-old Russian?

Martin Buber wrote his classic I and Thou in 1923 (Buber, 1923). In it, he develops a simple but profound thesis. One (“I”) can relate to the world either as an “It” or a “Thou.” For I–Thou the two have a shared reality where the essence of I belongs to both. Whereas in I–It the I is separated from It and the essence of I belongs to I only, not to It. I find this model informative in my daily practice. It helps describe the relationship between the doctor and the patient. If I–Thou, the relationship is dynamic, the doctor is immersed in the patient's circumstance, in their autonomy. If I–It, then the doctor is referring to the patient as an object—just as one might to a hammer knocking a nail into the wall. Similarly, it is preferable not to relate to the patient through formulaic dicta or bureaucratic procedures. In other words, using a blanket rule that one must at all times keep the patient fully informed of all findings and interpretations is, in my view, treating the patient as an It.

A relationship becomes dynamic and time bound when the doctor is invested in the essence of the patient. The truth becomes embedded in the relationship. Concern is raised that if one is paternalistic, false assumptions will be made about the patient, or if key information is withheld, then trust will be forever lost—and not be available for the next time. These are important considerations. In my experience, when the relationship is shared (in the sense that Buber proposed) and trust evolves—only then can decisions and responsibility be shared and the somewhat artificial polarization of autonomy versus paternalism becomes less relevant.

It seems, as Ecclesiastes indicated, that timing is critical. There is a time to talk, but there is also a time to judiciously keep quiet. The timing and content of the silence depends on a relationship based upon a shared vision—between me and you, not me and it. An effective clinical approach and ethical stance is a relationship underpinned by Buber's concept of “I and Thou,” with the doctor's duty of care being paramount.