Need for ethics approval and patient consent in clinical research.

A key principle in clinical research is the need to obtain approval from a Research Ethics Committee before commencing a study, and then to obtain consent from each subject enrolled into the study. This tenet is highlighted in the Helsinki Declaration and reinforced in subsequent principles such as Good Clinical Practice/International Committee on Harmonisation. However, not all clinical studies need ethics approval and consent—the most obvious example is clinical audit, which aims to review patient care against predefined criteria and then implement change to improve care and outcomes. Hence, audit is about healthcare processes, which contrasts with clinical research where the aim is to understand human disease, its presentation, investigation, management, and outcome. So, why this reminder about the need for ethics approval and consent in clinical research? I (P.B.) was one of three reviewers asked to assess the article of Steinhagen et al for the journal (and published in this issue) and consider its suitabil-ity for publication. One of the other reviewers and myself questioned the absence of any mention of ethics approval and patient consent in the manuscript. The authors modified the manuscript in light of the first set of comments and, in their response letter to the editor, responded that " Ethics approval was not needed because all investigations performed were part of standardized routine diagnostic workup. " This statement made no sense, because the study was clearly clinical research; further, the authors did not claim the study was audit and it would not have fulfilled the criteria for this if they had. Although the authors had addressed most of the other comments relating to the first version and the research findings were interesting and informative, I recommended rejection on the grounds that the study was unethical because it had no approval by an ethics committee and patients had not given consent. It is important to justify why the study must be considered a piece of clinical research. First, the work is clearly not audit because it does not examine care delivery. Second, and equally clearly, it is a piece of clinical research as evidenced by statements throughout the publication, such as: " … clinical signs of dysphagia were studied " (Abstract); the study " explored whether brain lesion topology relates to specific dysphagia patterns " (Introduction); patients were " prospectively studied " (Patients and Methods); the listing of inclusion and exclusion criteria (Patients and Methods); and …