On January 10, 2020, less than a month after the first COVID-19 patient was admitted to a Wuhan hospital, researchers shared the full genome of the novel coronavirus in open access (ahead of the full publication in The Lancet), thus providing the basis for all research towards testing, treatment and vaccines. This was a remarkable achievement, considering that it took five months to release the sequence the genome of the deadly SARS outbreak in 2002-03, due to issues of data retention.
Indeed, the COVID-19 crisis has shed light on how policies that enable the sharing of research data can help accelerate the fight against pandemics and other global emergencies. Although the global sharing of research data has spurred collaboration and boosted discovery during the current crisis, significant challenges remain.
High-profile data breaches have encouraged risk-averse behaviour among those in charge of data management, particularly in cases involving sensitive data, such as the personal health records that are crucial to the search for new medical treatments and vaccines. Agents in charge of data management often aren’t properly recognised and rewarded, yet their personal responsibility is involved whenever incidents occur, contributing to more risk averseness. Data managers may also lack the skills necessary to apply proper data management processes – including storing, curating, providing access and re-using data – and scientific infrastructures are not always sustainably financed to handle exponentially increasing volumes of data. Issues around data ownership and stewardship can pose a further barrier. Data flows occur globally, yet national-level policies inhibit the free flow of some data, in particular the ‘sensitive’ category which can be provided for research purposes, but often only within a single country. All of this results in data that are insufficiently FAIR (Findable, Accessible, Interoperable, and Re-usable).
Technological and policy advances required new policy guidance
The OECD’s Recommendation concerning Access to Research Data from Public Funding aims to overcome these barriers by establishing access and global sharing of research data as a major policy priority, with the ultimate goal of making the global science system more efficient and effective. Initially adopted by the OECD Council in 2006, this pioneering initiative has contributed to establish access to data as a major policy priority and inspired a number of other multilateral and national policy instruments, including the European Commission’s Recommendation on access to and preservation of scientific information and UNESCO’s Policy Guidelines for the Development and Promotion of Open Access, both issued in 2012. A 2017 survey carried out by the OECD Committee for Scientific and Technological Policy found that the Recommendation had an “important” or “highly important” influence on 47% of the national initiatives concerning open access to data that were adopted between 2015 and 2017.
The scientific landscape has changed dramatically in the 15 years since the original OECD Recommendation was adopted in 2006. Data-driven innovation and data-intensive science are transforming society, and access to data has had far-reaching effects on the reproducibility of scientific results, diffusion of knowledge across society, cross-disciplinary co-operation, resource efficiency, productivity and competitiveness. Open science and open data have become mainstream trends, with at least 58 countries adopting dedicated national strategies and policies for open data and publications.
Such technological and policy advances required new policy guidance, which is why the OECD Recommendation on research data was recently updated. The revised Recommendation, released today, reaffirms the relevance and importance of several key principles set out in 2006: openness, flexibility, transparency, legal conformity, protection of intellectual property, formal responsibility, professionalism, interoperability, quality, security, efficiency, accountability, and sustainability. In addition, it provides updated policy guidance on seven areas that have emerged as crucial to enhancing access to research data in recent years (see figure below).
The revised Recommendation also expands its scope to cover not only research data, but also related metadata (data about data, specifying their sources, methodology and limitations), as well as the bespoke algorithms, workflows, models and software (including code) that are essential for their interpretation.
In committing to this Recommendation, the 37 OECD member countries as well as Argentina, Brazil and Kazakhstan have engaged to ensure that research data are “as open as possible, and as closed as necessary”. The term “as closed as necessary” refers to the fact that not all data can be made open, in order to protect privacy, intellectual property, national security and other legitimate public and private interests. In those cases, special arrangements can be made to provide selective access to a certified group of researchers within safe environments.
The OECD will continue working with governments on specific implementation guidelines in the seven policy areas that the Recommendation outlines. As COVID-19 has made clear, open data for science is key to address global challenges – including climate change, biodiversity, aging populations or the next pandemic.