Browse by author
Lookup NU author(s): Professor Madeleine Murtagh, Dr Neil Jenkings, Sian Pointon, Emeritus Professor Paul BurtonORCiD
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
Contemporary bioscience is seeing the emergence of a new data economy: with data as its fundamental unit of exchange. While sharing data within this new ‘economy’ provides many potential advantages, the sharing of individual data raises important social and ethical concerns. We examine ongoing development of one technology, DataSHIELD, which appears to elide privacy concerns about sharing data by enabling shared analysis while not actually sharing any individual-level data. We combine presentation of the development of DataSHIELD with presentation of an ethnographic study of a workshop to test the technology. DataSHIELD produced an application of the norm of privacy that was practical, flexible and operationalizable in researchers’ everyday activities, and one which fulfilled the requirements of ethics committees. We demonstrated that an analysis run via DataSHIELD could precisely replicate results produced by a standard analysis where all data are physically pooled and analyzed together. In developing DataSHIELD, the ethical concept of privacy was transformed into an issue of security. Development of DataSHIELD was based on social practices as well as scientific and ethical motivations. Therefore, the ‘success’ of DataSHIELD would, likewise, be dependent on more than just the mathematics and the security of the technology.
Author(s): Murtagh MJ, Demir I, Jenkings KN, Wallace SE, Murtagh B, Boniol M, Bota M, Laflamme P, Boffetta P, Ferretti V, Burton PR
Publication type: Article
Publication status: Published
Journal: Public Health Genomics
Year: 2012
Volume: 15
Issue: 5
Pages: 243-253
Print publication date: 20/06/2012
ISSN (print): 1662-4246
ISSN (electronic): 1662-8063
Publisher: S. Karger AG
URL: http://dx.doi.org/10.1159/000336673
DOI: 10.1159/000336673
Altmetrics provided by Altmetric
