Improving the reporting of surveys of clinicians

We congratulate Karen Burns and colleagues for their article on how to design and conduct self-administered surveys of clinicians.¹ We agree that there is a need to advance the idea of standardizing the reporting and the quality assessment of surveys of clinicians, and we believe that this extends to all studies targeting the practices of health professionals. Existing guidelines on how to report studies pertaining to the general population need to be adapted for studies of health professionals. For example, we systematically reviewed instruments to assess the perceptions that physicians have of the decision-making process in specific clinical encounters and adapted the STARD (Standards for Reporting of Diagnostic Accuracy) guidelines for evaluating the quality of study reporting.^2,3

As reviewers for the Effective Practice and Organisation of Care Group of the Cochrane Collaboration, we agree that the synthesis of studies examining the practices of health professionals suffers from incomplete review of the existing literature, lack of standardization of measurements and improper analytic methods.^2,4 However, we have also observed that such studies suffer greatly from the lack of a theoretical basis, which in turn hampers the development of effective interventions to improve clinical practices.⁵ Therefore, we suggest adding an item to the list of questions to consider when preparing a report of surveys in Table 4¹: “In the Introduction, is the model (or theory) or the conceptual framework clearly stated?”