The 23andMe Research cohort includes over eleven million participants, most of whom opted into participating in research after becoming 23andMe customers; the cohort is therefore unselected in terms of participants’ health status and demographic characteristics. Customers who consent to 23andMe Research allow 23andMe researchers to use their genetic information, online survey responses, and certain other categories of data to study multiple topics including: predisposition to health conditions and traits, disease diagnosis and treatment, population genetics, and outcomes of learning about genetics and health. The 23andMe Research Consent document is open-ended, allowing recontact for unspecified future studies, and it allows 23andMe to share aggregate data with research collaborators and in publication. There are additional consents for individual-level data sharing and for recontact based on genetic information not included in 23andMe reports, as well as for specific research activities incurring additional risk. Here we describe strategies for maintaining a participant-centric research consent, and corresponding research practices. We describe our approach to providing appropriate levels of choice, while avoiding undue participant burden. We discuss data-sharing practices that balance vigorous privacy protections with maximizing the value of data participants contribute for research. Finally, we describe best practices for making changes to the 23andMe Research Consent document as external standards, research capabilities, and research aims evolve over time.
Authors: Sarah L Elson, 23andMe, Inc.; Emily DelloRusso, 23andMe, Inc.; Bianca A Llamas 23andMe, Inc.; Alex Moran, 23andMe, Inc.; Dominique Nguyen, 23andMe, Inc.; Leah Selcer, 23andMe, Inc.; Joyce Tung, 23andMe, Inc.
