Data sharing in research - cultural and technical barriers in Life Sciences and Public Health

As promised for the PDF-shy - the first section of our review of attitudes to data sharing in Life Sciences and Public Health - it's also well worth reading the Research Information Network's (RIN) publication Data centres: their use, value and impact which RIN neatly summarise here. This review was meant to serve as a 'primer' aggregating analyses from the last few years in this area and as such references other publications pretty heavily - so do check out the footnotes if you want to explore further. We'll post the next section on Funders' policies tomorrow.

Summary: Initiatives pursued by the funders and publishers of research to promote data-sharing have moved the agenda forward, however, cultural and technical barriers remain. Attitudes and abilities vary across specialities within biomedical research and though obliging data sharing has had some success, incentives and expertise are still lacking in many areas.

  The Current Situation

 The last ten years have seen consistent and targeted promotion of data sharing in research by organisations such as the Research Councils, the National Cancer Research Institute (NCRI), research charities and other funders and publishers of research. The aims are clear: “Ensuring data are made widely available to the research community accelerates the pace of discovery and enhances the efficiency of the research enterprise.”[1]

The assumption is that data sharing is a ‘good thing’ and that connectivity between data will enable greater research potential. There has been a lot of activity in areas such as access and governance, initiatives providing portals and developing standards and there is evidence that “in many research fields – from genetics and molecular biology to the social sciences –data sharing is ingrained in how researchers work”[2] and that, with regard to Systems Biology at least, “data sharing, despite some anomalies, is the prevailing ethic.”[3]

The picture is mixed across the life-sciences, however, as noted in this recent call for contributions to a thematic series on data standardisation, sharing and publication by the online Journal BM Research Notes: “different disciplines have embraced the possibilities of data sharing and open data to differing extents, and it can take the leadership of a small number of individuals to develop and promote their standard to secure widespread adoption, and enable interoperability of scientific data… In other cases a standard of data collection and preparation might be well known amongst circles of experts but perhaps unknown to researchers in different or even related fields. But with few journals considering data-driven articles and apparent inconsistencies in incentives and rewards for data publication, the availability of definitive and freely-available examples of re-usable, standardized data across the life sciences is patchy at best.”[4]

Funders and journals are addressing this issue, promoting data sharing by various means including policies obliging researchers to make their output publicly available. However, “Where funder policies do not reach, there is a mix of results. Some researchers make great efforts to share data while others may retain their findings or publish in a form that means that although data are available, they are not readily accessible”[5] or ‘protecting by pdf’ as the practice is known within the community.

The situation is least advanced in Social and Public Health Sciences: “There are many datasets produced by individual researchers or small project teams that could have long term viability if they were offered to an appropriate data centre, but this tends not to be the natural course of things. The sharing of datasets from small scale research projects appears to be relatively uncommon at present.”[6] Other analysts concur: “By contrast, this culture has yet to be widely embraced by the public health research community.”[7]

Whilst the battle for cultural change appears to be far advanced in some areas and at least engaged in others, attitudes are not the only barrier: “Problems of reuse centre around… technical issues… – the variety of formats, the non-standardisation of formats, the need for proprietary software and so forth.”[8]

The picture is not uniform across all fields within biomedical research - in some areas data sensitivity and cultural barriers remain more challenging to address than technical and ethical issues. Others, genomics for example, appear comparatively mature, with both cultural and technical issues well in hand.

---

[1] Walport M, Brest P. Sharing research data to improve public health. The Lancet, Early Online Publication, 10 January 2011

[2] Ibid.

[3] To Share or not to Share: Publication and Quality Assurance of Research Data Outputs - Report commissioned by the Research Information Network (RIN) in association with the Joint Information Systems Committee and the National Environment Research Council (NERC) – published June 2008. This report covered six discrete research areas, two of which were Social and Public Health Sciences and Genomics and two interdisciplinary areas, one of which was Systems Biology.

[4] A call for BMC Research Notes contributions promoting best practice in data standardization, sharing and publication; http://www.biomedcentral.com/1756-0500/3/235/

[5] Ibid.

[6] Ibid.

[7] Walport M, Brest P. Sharing research data to improve public health. The Lancet, Early Online Publication, 10 January 2011

[8] To Share or not to Share: Publication and Quality Assurance of Research Data Outputs