Managing Incidental Findings in Human Subjects Research

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): This 2-year project tackles a profoundly important but widely ignored problem: how should researchers handle incidental findings that have potential clinical significance, what should consent forms say about this, and how should IRBs consider the potential for incidental findings in their review of protocols. By "incidental findings" we do not mean results on those variables of key interest in the research, but unexpected findings beyond the domain of focal research interest, such as a suspicious mass revealed in a functional MRI (fMRI) study, or an incidental finding of nonpaternity in genetic research on families. We will anchor our analysis by focusing on incidental findings in 4 research areas: (1) fMRI research, (2) CT colonography research, (3) genetic research on families, and (4) genomic microarray research. This project will have a normative and an empirical component. The normative work will focus on developing a consensus paper on how incidental findings should be managed by investigators and IRBs and model consent form language. This will be informed by collection and analysis of model consent forms and consent forms used by U.S. researchers in the 4 areas of study as posted on university, professional society, and governmental websites and on the World Wide Web. This project will be led by researchers in the University of Minnesota's Consortium on Law and Values in Health, Environment & the Life Sciences, collaborating with a senior empiricist expert in content analysis and a Working Group of prominent national scholars. Project outcomes will include: publication of the consensus recommendations and model consent form language, a conference presenting our conclusions plus papers on incidental findings management in the 4 targeted research areas and analysis of the legal implications of incidental findings, a published symposium in a prominent journal collecting these papers, and a website offering tools such as model consent form language and an annotated bibliography with web links.