in the RDA COVID-19 working group. Photo: @weiwei_shoots.
During the current pandemic, many have expressed a greater need for transparency in research, especially in terms of faster sharing of data and research results. The volunteer and member-driven organisation Research Data Alliance (RDA) has addressed this need in their RDA COVID-19 Working Group. Gustav Nilsonne has been part of the working group and continues to work with some of these questions.
The RDA COVID-19 working group was assembled quickly in spring 2020. The group’s main task was to rapidly define guidelines for how to more efficiently share data and to address the related interests of researchers, policy makers, funders, publishers, and infrastructures. Their work resulted in an extensive report, which was published in summer 2020. This report is intended as guidance and a point of reference for anyone who has questions about research data concerning COVID-19.
Gustav Nilsonne is a researcher in neuroscience and meta-science at Karolinska Institutet. He is also the coordinator of the SND domain specialists, as well as a domain specialist in medicine. Gustav has been a member of the RDA working group since it started, both in a sub-group for clinical data and in a sub-group for legal and ethical aspects.
—The report is based on well-established principles. If you were to summarise the message, I’d say that it’s simply “as open as possible, as closed as necessary”. The report contains specific recommendations for, for example, data structures and file formats for different data types, such as immunological data and clinical imaging data, and so on. It has also compiled a large selection of links and resources for further reading. I don’t imagine it’s something you'd read from beginning to end, but if you have questions about something in this field, it can be an excellent resource.
If you want to read an overview of the report, it has been summarised in a article, which Gustav Nilsonne has also co-authored.
—The article is a way of highlighting the report from the RDA working group, and it also summarises some of the most important parts of the report. We have published it on Wellcome Open Research, which has a transparent review process. This is the first version of the article, and it has received some reviewer comments. We have gone through the comments and will soon submit a second version.
Even though the COVID-19 working group was created at short notice and worked under a tight deadline, Gustav nevertheless thinks that the work has gone very well.
—There’s a lot of competence in RDA, as well as a wide range of experts on various aspects of research data management, something that this working group demonstrated that you can use in a very efficient and successful way. It was fascinating to see that you can actually produce a material quite rapidly, and that you don’t need long-winding processes to reach consensus. And we could even do it online.
A ”Spin-Off” Group Continued to Work on Ethical and Legal Aspects
When the report was published, the RDA working group had achieved its objective, but one of the sub-groups continued to work on their own accord. Gustav Nilsonne says that the ethical and legal group has recently published an article (preprint) about international transfers of health data.
—We have continued to work simply because there were ideas and a willingness to keep working in the group. One of the members took the initiative to an article that studies international transfers of health research data in the light of the Schrems II judgment from this summer.
Previous regulations made it possible to transfer personal data from the EU to the USA if the recipient was connected to the Privacy Shield data protection agreement. The Schrems II decision ruled that the agreement doesn’t provide adequate protection for personal data. It subsequently means that it is no longer possible for data protection officers in the EU to transfer personal data to the USA, using Privacy Shield as grounds to do so.
—The article contains a legal analysis of the decision and the legal situation. It also includes a discussion about how to move forward, possible solutions, and what problems and challenges that may arise from these solutions.
As an example of a solution you could use within current legislation, Gustav Nilsonne mentions federated analytics. It’s a method that means that research data remain in the EU, and that third country parties send code to analyse the data. They will then get results that are no longer connected to the personal data. But the article authors would still like to see a comprehensive approach on an EU level.
—We think that they have caused a great deal of problems for the processing of research data when they regulated the transfer of data that contain personal information. The article is a call to the European Data Protection Supervisor and European legislators to update the current guidelines and adapt them, in order to address the needs of research and research data. This is a problem that needs to be solved.
You can read the first version of the article here.