Knowing who to trust: exploring the role of ‘ethical metadata’ in mediating risk of harm in collaborative genomics research in Africa
1 Department of Medicine, University of Cape Town, Anzio Road Observatory, Cape Town 7925, South Africa
2 Kenya Medical Research Institute (KEMRI)/Wellcome Trust Programme, Centre for Geographic Medicine Research, Coast, Kilifi District Hospital, PO Box 230, Kilifi 80108, Kenya
3 Medical Research Council Unit, PO Box 273, Banjul, The Gambia
4 Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK
5 Wellcome Trust Sanger Institute, Hinxton, UK
6 Nuffield Department of Population Health, University of Oxford, Old Road Campus, Oxford OX3 7LF, UK
7 The Ethox Centre, Nuffield Department of Population Health, University of Oxford, Old Road Campus, Oxford OX3 7LF, UK
BMC Medical Ethics 2014, 15:62 doi:10.1186/1472-6939-15-62Published: 13 August 2014
The practice of making datasets publicly available for use by the wider scientific community has become firmly integrated in genomic science. One significant gap in literature around data sharing concerns how it impacts on scientists’ ability to preserve values and ethical standards that form an essential component of scientific collaborations. We conducted a qualitative sociological study examining the potential for harm to ethnic groups, and implications of such ethical concerns for data sharing. We focused our empirical work on the MalariaGEN Consortium, one of the first international collaborative genomics research projects in Africa.
We conducted a study in three MalariaGEN project sites in Kenya, the Gambia, and the United Kingdom. The study entailed analysis of project documents and 49 semi-structured interviews with fieldworkers, researchers and ethics committee members.
Concerns about how best to address the potential for harm to ethnic groups in MalariaGEN crystallised in discussions about the development of a data sharing policy. Particularly concerning for researchers was how best to manage the sharing of genomic data outside of the original collaboration. Within MalariaGEN, genomic data is accompanied by information about the locations of sample collection, the limitations of consent and ethics approval, and the values and relations that accompanied sample collection. For interviewees, this information and context were of important ethical value in safeguarding against harmful uses of data, but is not customarily shared with secondary data users. This challenged the ability of primary researchers to protect against harmful uses of ‘their’ data.
We identified three protective mechanisms – trust, the existence of a shared morality, and detailed contextual understanding – which together might play an important role in preventing the use of genomic data in ways that could harm the ethnic groups included in the study. We suggest that the current practice of sharing of datasets as isolated objects rather than as embedded within a particular scientific culture, without regard for the normative context within which samples were collected, may cause ethical tensions to emerge that could have been prevented or addressed had the ‘ethical metadata’ that accompanies genomic data also been shared.