Abstract
Prevalence and Characteristics of Data Sharing Policies Across the Health Research Life Cycle: Funders, Ethics Committees, Trial Registries, Journals, and Data Repositories
Aidan C. Tan,1 Sol Libesman,1 Weber Liu,1 Zijing Yang,1 Rani R. Chand,1 Kylie E. Hunter,1 Angela Webster,1 Anna Lene Seidler1
Objective
Most principal investigators support the concept of data sharing in principle, but few commit to sharing data in practice.1 One way shown to reduce this gap is if major stakeholders across the research life cycle implement policies to recommend or require data sharing. The objective of this study was to determine the prevalence and characteristics of data sharing policies.
Design
This was a cross-sectional study of data sharing policies of health research funders, research ethics committees, clinical trial registries, peer-reviewed scientific journals, and research data repositories. It included the 55 largest private and 55 largest public and philanthropic health research funders by annual health research expenditure, all national ethics committees, all clinical trial registries, the 5 highest-impact peer-reviewed scientific journals by journal impact factor for each of the 59 fields of clinical medicine, and all research data repositories in clinical medicine. Investigators reviewed all official websites, online reports, and gray literature information sources of stakeholders for the presence of a data sharing policy. If present, investigators assessed its magnitude of support for data sharing. If it recommended or required data sharing, investigators assessed its characteristics. All data were abstracted in duplicate by 2 independent investigators who compared the relevant information against structured criteria on a prepiloted data extraction form and resolved disagreements by discussion and a third investigator.
Results
Overall, 110 health research funders, 124 national ethics committees, 18 clinical trial registries, 273 peer-reviewed scientific journals, and 410 research data repositories were included. More than half of health research funders either recommended (15 [15%]) or required (45 [41%]) data sharing. These policies typically applied to all data from only interventional studies, with justified exceptions, and specified data to be shared before a predetermined period with independent committee–approved investigators for research proposal–approved purposes via third-party websites. Only 4 national ethics committees (3%) recommended data sharing. These policies typically applied to all studies, with justified exceptions, and specified data to be shared via third-party websites. Only 1 clinical trial registry (6%) required data sharing. This policy applied to only interventional studies, with justified exceptions, and specified data to be shared via third-party websites. Almost two-thirds of peer-reviewed scientific journals either recommended (120 [44%]) or required (52 [19%]) data sharing. These policies typically applied to only some data from all studies, with unjustified exceptions, and specified data to be shared with anyone for any purpose via third-party websites. Few research data repositories recommended (26 [6%]) or required (24 [6%]) data sharing. These policies typically applied to all data from all studies, with unjustified exceptions, and specified data to be shared with anyone for any purpose and via third-party websites.
Conclusions
Data sharing imperatives were not met by most stakeholders.
Reference
1. Tan AC, Askie LM, Hunter KE, Barba A, Simes RJ, Seidler AL. Data sharing—trialists’ plans at registration, attitudes, barriers and facilitators: a cohort study and cross-sectional survey. Res Synth Methods. 2021;12(5):641-657. doi:10.1002/jrsm.1500
1NHMRC Clinical Trials Centre, The University of Sydney, Sydney, Australia, aidan.tan@sydney.edu.au
CONFLICT OF INTEREST DISCLOSURES
None reported.