By Samia Rizk, MD, professor of clinical pathology at Cairo University
Open data refers to “data that can be freely used, shared, and built-on by anyone, anywhere.” Data which is considered “open” is further required to be legally acceptable, technically readable, and available to others. The many potential benefits attributed to data sharing are combined with its many challenges: assuring privacy and security, the autonomy of participants as it relates to their ability to give valid consent, population and cultural issues, measures for appropriate governance, commercialization, and sustainability of databases.
With no binding requirement, it is a scientist’s decision whether or not to share data. Since transparency, openness, and reproducibility are considered part of the scientific method in research, data sharing aligns with those tenets.
n response to the need for a universal framework as to how, when, and what types of data should be shared, the Institute of Medicine’s Committee on Strategies for Responsible Sharing of Clinical Trial Data, recently reported guiding principles and a practical framework to implement data sharing across the research enterprise. To advance understanding of this issue, PRIM&R held a webinar titled, Maximizing Benefits to Research with Human Subjects Through Data Sharing, during which speakers discussed the Committee’s recommendations as they apply to those who review research with human subjects, and addressed data sharing in a research environment and the culture of data sharing at academic institutions.
Many initiatives for data sharing have been driven by funders and research sponsors in order to maximize the usefulness of data gathered. GlaxoSmithKline made a noteworthy announcement in October, 2012, when it stated that it plans to openly share its trial data. In response to mounting pressure from transparency advocates, the large pharmaceutical companies in Europe and the United States adopted a common initiative of sharing data with qualified researchers. They released a joint set of principles that stipulates any released data is subject to terms of protection for patient privacy and confidential commercial information. Such a step by pharmaceutical companies represents a reasonable alternative to proposals put forth by the European Medicines Agency, who requested the release of all data relevant to newly approved drugs, which could undermine the competitiveness of the pharmaceutical industry.
Certain research contexts may require special considerations. Genome-wide association studies are one area in which data sharing is of utmost importance. However, these studies also pose unique concerns related to the big data produced, and the related sensitive and cultural issues. The Global Alliance for Genomics & Health promotes international cooperation and interoperability of data with the aim of harmonizing ethical and regulatory approaches, and promoting autonomy. However, in the context of research involving developing countries special challenges are faced. As an example, the Malaria Genomic Epidemiology Network, a partnership of malaria researchers in over 20 countries, has developed policies and procedures for data release with a key principle of retaining ownership by the contributing investigators through the use of an independent data-access committee and a legally binding data-access agreement.
Implementation of international guidelines for data sharing in developing countries may also face many roadblocks, especially notable is the absence of clear national laws and regulations for data protection and use.
Data sharing and openness is clearly the future of research, yet more discussion and infrastructure are needed to ensure strong adherence to ethical principles is maintained and the protection of human subjects is preserved.