BMJ  2004;328 (10 April), doi:10.1136/bmj.328.7444.0-g

Editor's choice

The teaching of communication skills may be misguided

Each day in Britain there are some two million consultations between doctors and patients and around 100 million decisions about patient care. Any fundamental reform of health care will depend on influencing those consultations and decisions, and they are, I suggest, so far largely untouched. "Thank God," I hear many doctors say, but who could disagree that much of what goes on in those consultations is capable of considerable improvement?

One method of improving what happens in consultations is through teaching doctors communication skills, but a paper from Liverpool suggests that the teaching has been driven too much by what doctors think is good communication rather than by what patients want (p 864). The authors studied consultations between 12 doctors and 39 patients with breast cancer and asked patients "to describe aspects of communication they valued or deprecated." The results suggest that some teaching on communication may be misguided.

Patients didn't think about their doctors in terms of how well they communicated. Instead, they cared about whether their doctors had expertise, had a personal relationship with them, and respected their autonomy. Teaching has emphasised shared decision making, but "no patient described a process of decision making." They liked instead being given a recommendation with the option to say no, and "they had concurred with clinicians' recommendations."

The paper offers practical examples of how the doctors communicated expertise, built relationships, and showed respect for the patients. Interestingly, "being a doctor was often sufficient" to convince patients of expertise. In other words, patients start by trusting doctors. Trust doesn't have to be won—but can be lost: "Trust in doctors' expertise was irretrievable if patients thought they had been misled." Doctors could also encourage belief in their expertise by displaying confidence and efficiency, making things happen, and by answering all questions without hesitation.

Non-verbal cues—eye contact, smiling, touching—were important in building a relationship. "The simplest verbal strategy was for the patient to be told she was special." Humour and idiosyncrasies also helped. Patients felt respected when doctors communicated with them at eye level and matched their language to the patients' expectations.

The authors argue that their results can be generalised, but I wonder. Patients with breast cancer may be radically different from patients with conditions that are not life threatening, and it's not clear if patients were offered the choice of sharing decisions. It might be that they were content with what they were offered but could have found more empowerment even more attractive. And—a question that interests me—how do you share the uncertainty that is ubiquitous in medicine and still seem to have expertise?