“You have to cover up the words of the doctor”

PURPOSE: This article explores issues of trust in narratives of interpreted consultations in primary health care. DESIGN/METHODOLOGY/APPROACH: The paper is based on empirical data from a qualitative study of accounts of interpreted consultations in U.K. primary care, undertaken in three north London boroughs. In a total of 69 individual interviews and two focus groups, narratives of interpreted consultations were sought from 18 service users, 17 professional interpreters, nine family member interpreters, 13 general practitioners, 15 nurses, eight receptionists, and three practice managers. The study collected and analysed these using a grounded theory approach and taking the story as the main unit of analysis. It applies a theoretical model that draws on three key concepts: Greener's taxonomy of trust based on the different "faces" of power in medical consultations; Weber's notion of bureaucratic vs traditional social roles; and Habermas' distinction between communicative and strategic action. FINDINGS: Trust was a prominent theme in almost all the narratives. The triadic nature of interpreted consultations creates six linked trust relationships (patient-interpreter, patient-clinician, interpreter-patient, interpreter-clinician, clinician-patient and clinician-interpreter). Three different types of trust are evident in these different relationships--voluntary trust (based on either kinship-like bonds and continuity of the interpersonal relationship over time, or on confidence in the institution and professional role that the individual represents), coercive trust (where one person effectively has no choice but to trust the other, as when a health problem requires expert knowledge that the patient does not have and cannot get) and hegemonic trust (where a person's propensity to trust, and awareness of alternatives, is shaped and constrained by the system so that people trust without knowing there is an alternative). These different types of trust had important implications for the nature of communication in the consultation and on patients' subsequent action. RESEARCH LIMITATIONS/IMPLICATIONS: The methodological and analytic approach, potentially, has wider applications in the study of other trust relationships in health and social care. Practical implications - Quality in the interpreted consultation cannot be judged purely in terms of accuracy of translation. The critical importance of voluntary trust for open and effective communication, and the dependence of the latter on a positive interpersonal relationship and continuity of care, should be acknowledged in the design and funding of interpreting services and in the training of both clinicians, interpreters and administrative staff. ORIGINALITY/VALUE: This is the first study in which interpreted consultations have been analysed from a perspective of critical sociology with a particular focus on trust and power relations.