In medical research, discovery and innovation data are easy to share, transparency and scalability are possible, and confidence in science is increased. Spurred on by the COVID-19 pandemic, demands for data sharing have accelerated as calls grow for faster publication, evaluation, integration, and analysis of new medical research findings.
Contemporary recommendations for data sharing are based on policies developed 4 decades ago. For example, GenBank was created in 1982 as a public access repository of nucleotide sequences. Now maintained by the US National Center for Biotechnology Information, GenBank contains over 240 million publicly available sequences. In 1985, the US National Research Council (NRC) issued 3 reports on data sharing and it continues to be a useful guide for researchers, authors, editors, and journals. The following NRC recommendations are relevant to scientific journal publishing
Sharing data should be a regular practice.
Investigators must share data up to publication.
Data-sharing plans should be an integral part of the research plan.
Subsequent researchers who request data from others should bear the associated additional costs, should try to minimize the burden of data sharing, and should explicitly acknowledge the contribution of the first investigators.
Journal editors should require authors to provide access to data as part of the peer review process, give greater emphasis to reports of secondary analyzes and replication, require full credit and appropriate citations for original datasets in reports based on secondary analyses, and encourage authors to make detailed data accessible other researchers.
Despite these early recommendations, policies encouraging data sharing varied widely among research institutions, government agencies, funders, and journals, and it took four decades for policies to encourage data to become requirements. Journals have taken different policies to encourage data sharing across different data and study types. There has been debate about whether it is within the responsibilities of journals to enforce data sharing, particularly when data sharing requirements are seen as unfunded mandates and the method of data sharing is not universally accepted.
In 1989, PNAS was among the first journals to encourage the deposit of genetic data sets in public repositories, and many journals (including JAMA) have taken this approach and encouraged authors to deposit genetic data sets (such as a protein or DNA sequence) in a database. Publicly available data. Among medical journals, the Journal of Internal Medicine Annual and BMJ began requiring authors of research articles to provide data or share data in 2007 and 2009, respectively, and in 2014, the PLOS journals began asking authors to provide all data underlying the findings described. In their manuscripts available entirely without restriction, with rare exception.” Data to be published with clinical trial reports and journals that follow ICMJE policy (including JAMA and JAMA Network Journals) have made this a publication requirement for clinical trials. Although these policies encourage data sharing and require the publication of a data sharing statement, none of them requires or enforces the actual sharing of data.
In 2003, the US National Institutes of Health (NIH) issued a statement encouraging data sharing publicly and asking investigators to submit applications for grants of $500,000 or more per year to include a data-sharing plan or explain why the data is shared. On how to preserve and share scientific data by presenting a data management and sharing plan Expect to develop appropriate sharing of scientific data resulting from research funded by the National Institutes of Health, with justified limitations or exceptions.
To facilitate the reporting of data management and sharing plans, JAMA Journals and the JAMA Network have expanded data-sharing statement requirements for authors of all research articles, not just clinical trial reports. Starting in October 2022, at the time of manuscript submission for any research article, the authors were asked a series of questions regarding their plans to share the data. Authors are asked to identify the data that will be made available to others, including individual patient data, a data dictionary that defines each field in the data set, and supporting documentation (eg, statistical/analytic code); when, where, and how the data will be available (eg, a link to a data repository); types of analyses that are permitted; and if there will be any restrictions on the use of the data. Authors also have the option to explain why data may not be shared. While the JAMA Network journals endorse the principle of data sharing and have adopted this policy to encourage the sharing of data, sharing is not mandated by the JAMA Network. An author’s intention to share data or not will be considered in the editorial decision. The JAMA Network journals will continue to publish data-sharing statements in an online supplement linked to the published article.
JAMA Network Journals will continue to publish entry data in an online supplement linked to the published article.
Sharing data promotes the growth of medical knowledge. Keep in mind that some researchers may want to protect their data or may not have the resources to share their data effectively. However, as mandates from funders and others continue to proliferate and strengthen, JAMA Network’s data-sharing policy and process enables and encourages broader data sharing by researchers and thus enhances discovery, transparency, reproducibility, and trust in the medical sciences.
