Strategies needed to promote effective, ethical data sharing
The degree to which they share or withhold data is not solely a personal choice, as institutional and national factors greatly impact on data sharing. For instance, in the context of laws, regulations and policies, restrictions may apply to data sources that are copyrighted and may prohibit the publication of certain types of data.
Furthermore, data transfer agreements (DTAs) govern the transfer of identifiable human participant data, for which voluntary and informed consent have been obtained from participants. Both material transfer agreements (MTAs) and DTAs contractually govern biological material and data transfer between parties to safeguard the interests of stakeholders.
Complex agreements hinder sharing
These contractual agreements outline the specific purpose(s) for which the data may be used, as well as the related protections, rights and obligations of stakeholders and collaborators. Despite the important role that MTAs and DTAs play in the governance of bio-samples and data, they are occasionally perceived as an impediment to data sharing, given their complexity and associated bureaucracy.
As a result, it is important to develop strategies and policies to promote effective data sharing, while simultaneously maintaining privacy and confidentiality.
We, as the authors, firmly believe that it is a scientific and ethical imperative to seek potential solutions to improve data sharing, especially in Africa. For this reason, we conducted a study on the factors that promote or prevent the sharing of data in Sub-Saharan Africa (SSA). We surveyed researchers and scientists from 43 SSA countries on data use and reuse, data management, data protection, data sharing, and the use of others’ data.
Overall, our survey found that researchers and scientists were optimistic about data sharing, storage, data management support, and reuse. Most respondents were willing to share at least some of their data, provided that robust governance and regulation frameworks are in place. They also agreed to share their data if it resulted in proper citation credit, co-authorship, and an opportunity to collaborate.
However, respondents were split almost equally in terms of sharing their data for commercial gain.
While most respondents were satisfied with institutional data storage processes, some indicated that their organisations or institutions did not have a formally established process for storing data and failed to provide training, funding and long-term support for data management beyond the life cycle of the project. This could be related to a lack of resources, chronic institutional under-investment, and suboptimal research training and mentorship in SSA.
Scientists need tools to manage data
We also found that more than half of the respondents were satisfied with the available tools used for documentation preparation, while over a third were dissatisfied with the tools used for preparing their metadata. This could suggest that there is a need for adequate tools to assist SSA researchers and scientists in facilitating and enhancing their use and management of data.
Most respondents indicated that they do not use any data protection steps when sharing data other than using technologically based safety measures (eg, password protection or encryption methods). Whereas many researchers and scientists across SSA use various types of data agreements and security measures during data sharing, others lack such tools, approaches and data protection policies and frameworks that promote safe data sharing.
This is concerning as it suggests that researchers and scientists still use suboptimal or mediocre data practices, placing their data at risk for misuse or theft, among other concerns.
Going forward, researchers and scientists should be provided with the necessary resources, training and government support to reinforce their data-sharing processes and to help them harness big complex data sets to find solutions to scientific problems.
Considering the current advancements in digital technologies, SSA countries must implement data protection policies, legislation and frameworks that are a contextual fit, as this could provide assurances and confidence among researchers and scientists that measures are in place to secure their data sets during the sharing or transfer of data. Having policies or frameworks in place could encourage researchers and scientists in SSA to make their data electronically available.
Educate researchers about standards, criteria
It will also be crucial for organisations and institutions to invest and have sustainable funding for data-management services in SSA. In addition to funding, there is a need for institutional support of data management and capacity development in data governance for researchers and scientists. Relevant knowledge transfer between SSA countries should be facilitated.
Researchers and scientists in Africa must also be encouraged to prioritise good data practices by storing and sharing data in repositories. This can be accomplished by changing researchers’ negative perceptions around repositories by educating them on the standards and criteria of data repositories (increased security), as well as the benefits, such as adequately prepared metadata and the discoverability of the data.
Given that DTAs vary between projects and countries, having standardised templates for DTAs and data use agreements would expedite sharing agreements between research collaborators. This will enable researchers and scientists, their funders, journals, and institutions to collaborate and promote sustainable data sharing on the continent.
In this context, sustainable data sharing will include providing ethical incentive structures for researchers and scientists who are willing to share their data, as well as tangible infrastructure to facilitate such sharing.
Siti Kabanda, Nezerith Cengiz, Qunita Brown and Professor Keymanthri Moodley are affiliated with the Division of Medical Ethics and Law, and Professors Kanshukan Rajaratnam and Bruce Watson with the School for Data Science and Computational Thinking at Stellenbosch University (SU), South Africa. Professor Tonya Esterhuizen works in the Division of Epidemiology Biostatistics, also at SU. This article is an abridged version of their paper, ‘Data sharing and data governance in Sub-Saharan Africa: Perspectives from researchers and scientists engaged in data-intensive research’, published earlier in 2023 in the South African Journal of Science.