Academic General Practitioners as realist researchers: Lessons learned

Problem

Realist synthesis and evaluation are gaining traction in health services research and are increasingly recognised by funders as methods which can generate important learning by exploring complexity. Realist research may hold intuitive appeal for clinicians who are already familiar with the messy reality of clinical practice, having witnessed widely varying responses when using similar approaches with different patients. As realist research is a highly interpretative process academics with a clinical background need to be aware of the various ways in which their role may have an influence.

Approach

We use this presentation to discuss reflections from two studies in which academic general practitioners collected and analysed realist data. We will describe how being a clinician researcher influenced the interpretive processes of theory building and testing, and our suggestions on what clinical academics might need to consider when starting out in realist research.

Findings

Initial theorising was informed by our insider knowledge of the health system, and our understanding of what drives our own clinical decision-making. However, we needed to actively engage with other stakeholders to challenge our assumptions and ensure we did not prioritise theories which resonated most with our own experiences. Data collection involved observing and interviewing clinicians (from our own and other clinical backgrounds). We made our clinical roles clearly visible and believe this facilitated rapport building and encouraged engagement with the research. Participants perceived our roles differently, with some appearing to be more open, while others appeared concerned about being judged, feeling obliged to justify to us the decisions they made. The realist teacher-learner style of interviewing allowed us to demonstrate that we were not looking for ‘the right answer’ but that we recognised and sought to explore variations in practice. While our understanding of jargon and the way systems work often helped interviews to flow, we recognised that we sometimes assumed that we understood our colleagues’ experiences and could have missed opportunities to probe further. Reading our own interview transcripts, and discussing these with non-clinical colleagues helped us to identify these risks and adjust our interviewing style. Observations of clinical interactions proved challenging, as it was difficult not to focus on what we might have done in the same situation, raising possible ethical dilemmas. During the analysis stage, while our understanding of context often made us more confident about the interpretive process of retroduction, we also risked going too far beyond the data, relying too heavily on our own perspectives.

Consequences

Reflexivity is essential for all researchers. We present the lessons learned from our own reflections and are keen to hear from other clinicians to build a shared understanding of particular issues that ‘insider researchers’ using a realist approach may need to consider.