The researchers' role in knowledge translation: a realist evaluation of the development and implementation of diagnostic pathways for cancer in two United Kingdom localities.

<bold>Background: </bold>In examining an initiative to develop and implement new cancer diagnostic pathways in two English localities, this paper evaluates 'what works' and examines the role of researchers in facilitating knowledge translation amongst teams of local clinicians and policy-makers.<bold>Methods: </bold>Using realist evaluation with a mixed methods case study approach, we conducted documentary analysis of meeting minutes and pathway iterations to map pathway development. We interviewed 14 participants to identify the contexts, mechanisms and outcomes (CMOs) that led to successful pathway development and implementation. Interviews were analysed thematically and four CMO configurations were developed.<bold>Results: </bold>One site produced three fully implemented pathways, while the other produced two that were partly implemented. In explaining the differences, we found that a respected, independent, well-connected leader modelling partnership working and who facilitates a local, stable group that agree about the legitimacy of the data and project (context) can empower local teams to become sufficiently autonomous (mechanism) to develop and implement research-based pathways (outcome). Although both teams designed relevant, research-based cancer pathways, in the site where the pathways were successfully implemented the research team merely assisted, while, in the other, the research team drove the initiative.<bold>Conclusion: </bold>Based on our study findings, local stakeholders can apply local and research knowledge to develop and implement research-based pathways. However, success will depend on how academics empower local teams to create autonomy. Crucially, after re-packaging and translating research for local circumstances, identifying fertile environments with the right elements for implementation and developing collaborative relationships with local leaders, academics must step back. [ABSTRACT FROM AUTHOR]