As medical research grows ever more complex, and as the public comes to expect more and more from medicine and science, the time has long gone when publication of research in a journal could be considered the end of a story. Increasingly and rightly, the research paper is becoming viewed as just a snapshot of the research, overlaid by the authors’ interpretation of their analysis and that the crucial output of the research is the raw data itself.
Journals are now therefore increasingly requiring that authors share data with anyone wishing to replicate the analyses. Though in some fields such as genetics this data sharing is the accepted norm, it is not always welcomed by researchers, requiring as it does extra work on the part of the authors to keep their data in a shareable state, to have mechanisms in place to share data, to allow others to gain academic credit from data collected by others, and of course it opens up the research to alternative explanations, which may conflict with the original researchers’ hypotheses. In addition, for medical data, there are often problems of confidentiality.
Nonetheless, and despite all these objections, the advantages of sharing data now far outweigh the disadvantages. Funders are crucial to such initiatives. They are in a unique position to change the data-sharing behaviour of researchers. Journals have in the past been criticised for not requiring data availability, or indeed for failing to require authors to uphold other standards, but ultimately editors have limited powers in the face of a culture that thus far has neither required or rewarded altruistic behaviours by authors.
The statement today by 17 research funders to commit “to increase the availability to the scientific community of the research data we fund that is collected from populations for the purpose of health research and to promote the efficient use of those data to accelerate improvements in public health.” is therefore to be welcomed.
The statement comes after several rounds of discussions spread over several years. Last January the H8 published a paper on data sharing in which they committed to develop a Code of Conduct in data sharing.
The full statement is available on the Wellcome Trust’s website but in essence the funders commit to promote greater access to and use of data in ways that are “Equitable…Ethical…and Efficient”. In the short term they commit to develop standards of data management and to recognize data sharing as a professional achievement, and encourage the responsible behaviour of secondary data users. Longer term aims include having documented data sets available for secondary analysis, strengthening the capacity to manage and analyse data, that published work and data are linked and archived, and data sharing is sustainably resourced for the long term.
These are all worthwhile goals and the backing of all these funders together give them a good chance of succeeding. The funders note that “Each funding institution will work within its own legal and operational framework” but that they “are committed to working towards these goals together.” What is needed now are specific dates for some of the immediate objectives and concrete indications of how the funders will enable their researchers to meet these. The statement of course is just the beginning; working out the details will require long term commitment – both of resources and will.