Sharing our Solutions to Data Sharing Challenges
September 15, 2016
by Stephanie Dyke, Ph.D.
Genomic researchers often feel stuck between a rock and a hard place when it comes to sharing the datasets underpinning their scientific studies. On the one hand, journals, funders, and increasingly research institutions, are reinforcing their data sharing policies to ensure as much knowledge as possible is extracted from valuable research data. On the other, researchers are well aware that the data they generate and analyze can include some of the most private data research participants can contribute: their genetic and health information.
In Canada, as in many other countries, this sharing must respect a complex web of not only research rules aimed at protecting the public from untoward consequences of their participation in research, but also more general privacy laws and regulations that apply to all kinds of businesses. Most research data sharing policies such as those of the main Canadian research funders acknowledge the restrictions on sharing that responsible privacy protection may engender. However, the specifics of such protections are not obvious and tend to be left to researchers and research institutions to handle on a case-by-case basis.
With my co-authors of the recent FACETS publication Evolving Data Access Policy: the Canadian Context, we are confronting many of the challenges of international data sharing firsthand as collaborators at the McGill Epigenomics Mapping and Data Coordination Centre, which was established to support large-scale human epigenome mapping for a broad spectrum of cell types and diseases and to share its data with the entire research community. As an interdisciplinary group of researchers in genomics, bioinformatics, ethics, policy, and law, we wished to share with other Canadian researchers our consideration of the ethical–legal questions and policy options for sharing data broadly while maintaining a reasonable amount of oversight to protect the public.
These were the most challenging issues that came up, which I’ll briefly review:
- providing access to data to a growing number of researchers;
- maintaining Canadian privacy standards while sharing controlled access data internationally;
- freedom of information requests; and
- providing more incentives for researchers to share pre-publication data.
In coming years, more and more researchers will have a legitimate need to access research data. While we tried to lower the barrier to access data for more interdisciplinary as well as industry researchers compared with somewhat more conservative access policies, the research community as a whole must face up to the realities of expanding access requirements and changing competitive landscapes.
Maintaining Canadian privacy standards while sharing controlled access data internationally
Canadian privacy standards are strong and need not be dismissed to enable access to data from Canadians for worthy research purposes. Through analysis of Canadian and Quebec law, we clarified general expectations and proposed legal provisions to reinforce privacy protection when sharing data internationally.
Freedom of information requests
Freedom of information requests may be at odds with the protection of research data, yet we didn’t find guidance on responding to such requests in this context. We therefore considered provincial access to information laws and how they might affect privacy protections and proposed that the issue be explicitly addressed in data sharing agreements.
Providing more incentives for researchers to share pre-publication data
One of the big barriers to data sharing is not ethical or legal, but rather social. To many researchers, sharing data from their studies still involves much greater risks than it does benefits. We drew attention to some steps that can be taken to acknowledge those who do share, but more needs to be done to create an environment in which data sharing in research is truly in everyone’s interest and becomes the norm.
Throughout the manuscript, we aimed to provide transparent data access agreement clauses that could easily be reused by other researchers with similar data sharing objectives. My hope is that in contributing to clarifying various aspects of data sharing policy, over time, the research community can build greater confidence in data sharing and realize its great benefits.
Read the full study, Evolving Data Access Policy: the Canadian Context by Stephanie O. M. Dyke, Katie M. Saulnier, Tomi Pastinen, Guillaume Bourque, and Yann Joly as open access in FACETS.
Dr. Stephanie Dyke has a Ph.D. in biochemistry (University of Cambridge) and M.Sc. in science communication (DCU). She has worked for over 10 years in bioethics and science policy (Irish Council for Bioethics, Wellcome Trust Sanger Institute, and Francis Crick Institute) and is currently conducting ethics and policy research at the Centre of Genomics and Policy at McGill University where she also teaches ethics and science communication.