February 26, 1997

A Model of Empathic Communication in the Medical Interview

Author Affiliations

From the Primary Care Institute and the Department of Medicine, Highland Hospital (Drs Suchman, Markakis, Beckman, and Frankel), and The Program for Biopsychosocial Studies (Drs Suchman, Markakis, Beckman, and Frankel), Departments of Medicine (Drs Suchman, Markakis, Beckman, and Frankel) and Psychiatry (Dr Suchman), University of Rochester, Rochester, NY.

JAMA. 1997;277(8):678-682. doi:10.1001/jama.1997.03540320082047

Objective.  —To formulate an empirically derived model of empathic communication in medical interviews by describing the specific behaviors and patterns of interaction associated with verbal expressions of emotion.

Design.  —A descriptive, qualitative study of verbal exchanges using 11 transcripts and 12 videotapes of primary care office visits to a total of 21 physicians.

Setting.  —An urban health maintenence organization (HMO), an urban university-based general medicine clinic, and an urban community hospital general medicine clinic.

Analytic Method.  —Individual review of transcripts by each research team member to identify instances of expressed or implied emotional themes and to observe the physicians' responses. Individual ratings were compared in group discussions to achieve consensus about the classifications. Similar consensus-based classification was used for review of videotapes.

Results.  —We observed that patients seldom verbalize their emotions directly and spontaneously, tending to offer clues instead. If invited to elaborate, patients may then express the emotional concern directly, and the physician may respond with an accurate and explicit acknowledgment. In most of the interviews, the physicians allowed both clues and direct expressions of affect to pass without acknowledgment, returning instead to the preceding topic, usually the diagnostic exploration of symptoms. With emotional expression so terminated, some patients attempted to raise the topic again, sometimes repeatedly and with escalating intensity. We noted a parallel dynamic for encounters in which patients sought praise. We summarized the full interactional sequence in a simple descriptive model.

Conclusions.  —This empirically derived model of empathic communication has practical implications for clinicians and students who want to improve their communication and relationship skills. Based on our observations, the basic empathic skills seem to be recognizing when emotions may be present but not directly expressed, inviting exploration of these unexpressed feelings, and effectively acknowledging these feelings so the patient feels understood. The frequent lack of acknowledgment by physicians of both direct and indirect expressions of affect poses a threat to the patientphysician relationship and warrants further study.