Introduction UK doctors are now required to participate in revalidation to maintain their licence to practise. Appraisal is a fundamental component of revalidation. However, objective evidence of appraisal changing doctors’ behaviour and directly resulting in improved patient care is limited. In particular, it is not clear how the process of appraisal is supposed to change doctors’ behaviour and improve clinical performance. The aim of this research is to understand how and why appraisal of doctors is supposed to produce its effect.

Methods and analysis Realist review is a theory-driven interpretive approach to evidence synthesis. It applies realist logic of inquiry to produce an explanatory analysis of an intervention that is, what works, for whom, in what circumstances, in what respects. Using a realist review approach, an initial programme theory of appraisal will be developed by consulting with key stakeholders in doctors’ appraisal in expert panels (ethical approval is not required), and by searching the literature to identify relevant existing theories. The search strategy will have a number of phases including a combination of: (1) electronic database searching, for example, EMBASE, MEDLINE, the Cochrane Library, ASSIA, (2) ‘cited by’ articles search, (3) citation searching, (4) contacting authors and (5) grey literature searching. The search for evidence will be iteratively extended and refocused as the review progresses. Studies will be included based on their ability to provide data that enable testing of the programme theory. Data extraction will be conducted, for example, by note taking and annotation at different review stages as is consistent with the realist approach. The evidence will be synthesised using realist logic to interrogate the final programme theory of the impact of appraisal on doctors’ performance. The synthesis results will be written up according to RAMESES guidelines and disseminated through peer-reviewed publication and presentations.

Trial registration number The protocol is registered with PROSPERO 2014:CRD42014007092.