Moving beyond the traditional teaching of interview skills by psychiatrists in the "see one, do one, teach one" mode, we report the development of an innovative approach to the teaching of psychiatric interview skills. This approach began four years ago as a partnership between mental health consumers and academic psychiatry to examine the ongoing feasibility of training mental health consumers as tutors for 4^th year medical students in psychiatry. The idea is based on the belief that consumers have a legitimate experience to share and a rich skill base on which to draw. The purpose of this ongoing project is to contribute to the ultimate development of a workforce of medical practitioners with clinical assessment skills that are better tailored to the needs of mental health consumers. More immediate potential benefits are the promotion of an engaging curriculum in psychiatry for students, offering direct meaningful contact with mental health consumers.

The attempt to measure attitudes deserves discussion. Attitudes are recognised as an important component of curriculum development yet remain the personal business of each of us. It would be reasonable to see education as a means of working past one's own attitudes rather than seeking to refine or replace student attitudes. Guidelines for working with consumers in health care assume that "for consumer participation to be effective, all participants in the process need to respect the different skills and expertise of the other participants" [12]. In this study, student attitudes to consumers had a tendency to improve across all dimensions measured. On average the medical students began the program with largely positive attitudes to working with and learning from consumers which may explain the lack of statistically significant difference in their attitudes pre and post the program. In addition, a finding of lack of significance using a pre and post test design with a small sample of subjects is not usual. The one attitude measured that did improve throughout the program and reached statistical significance was towards mental health clients in psychiatric wards. This finding is understandable in light of the fact that the training took place within a psychiatric unit and the program incorporated practice at live interviews with clients from the unit. In addition, the study found a change in the primary focus of medical student concerns regarding interviewing which moved from issues focused on the consumer (such as violence or unpredictability) to those focused on improving their skills in interviewing and seeing this as a worthwhile activity. In terms of their satisfaction with the training program, based on their open comments, the few students who objected at least had the challenge of working in an educational model they did not admire. This was thought provoking and engaging even if the response produced was negative.

Benefits for consumer tutors (as measured by their open evaluations) included enhanced self esteem and financial reward for work done. Consumer tutor curriculum development was novel such as utilising an art therapy vehicle to experience a non-pharmaceutical therapeutic device. Most consumer tutors have continued to teach, with appropriate breaks, and have mentored new tutors. Some have used this experience to step further into paid employment and to rehabilitate previous work skills. Consumer tutors have remained resilient and episodes of relapse appear to be multifactorial in origin (with teaching perhaps one of the factors). This robustness was also found in a study of psychological impact on consumers working in a peer support role in an acute care setting [13]. Consumer presentations have centred on the powerful personal effects of participation in learning new skills and gaining confidence. The largely positive ratings of tutors about the program was not as positive as the medical students, highlighting the need to evaluate both groups to adequately measure the effectiveness of the program.

Benefits for the health system included the placement of consumers in a 'professional' light. Consumer tutors shared the staff tearoom, were paid as other casual tutors and were seen as well contributors rather than being in the sick role. Professional tutors were aware of the consumer tutor teaching and perhaps viewed it as 'politically correct' rather than educationally effective. Dissemination of findings via service and conference presentations has helped address this common view.

Tutoring medical students is a skilful and potentially stressful role, and is not suitable for all mental health consumers. Following the training program some trained tutors realised that teaching was not their interest or strength (a proportion of whom did not teach at all). This was anticipated and should be factored into training plans. Some consumer tutors realised their tolerance level was insufficient to manage student junior skill level and found it hard to resist retelling their 'war stories' of difficult clinical encounters. This was a common theme in debriefing and required active refocusing on the curriculum of effective and ineffective interview techniques. The occasional protesting student required gentle persuasion to see that ongoing participation was a way of exploring contact with consumers. Like most education programs this approach did not run itself. Tutors required sustenance; feedback needed action and materials needed review. New tutors needed to be trained to add to the growing pool of available people. Despite these issues, our experience was that this was manageable and in keeping with maintenance of quality teaching by professional tutors.