An assessment of data and code sharing in research funded by the Nathan Shock Centers, 2017-2022: a meta-research study.

Abstract

Background and objectives: Data and code sharing facilitate reproducibility, trust, and efficient reuse of research resources. Growing recognition of their importance among researchers, journals, and funders led the Nathan Shock Centers (NSC) Coordinating Center to propose documenting and supporting data and code sharing in aging research. As a baseline, we characterized data and code sharing practices in NSC-funded research.

Research design and methods: We surveyed articles citing NSC grants published 2017-2022 indexed in PubMed. In full-text screening (n = 507), we excluded articles that did not generate or analyze data. For included articles, we classified data/code availability statements as available via repository, supplemental file, inclusion in paper, available on request, explicitly not available, no statement included, or other. We checked articles indicating open data/code to determine whether materials could be located per the provided statement.

Results: Of 400 articles included, 50% and 92% had no data or code availability statements, respectively. Data statements indicated availability via repository (30%), supplemental files (26%), inclusion in paper (10%), and on request (14%). For code, 6% indicated availability in repositories, and ≤1% each for other categories. Of those indicating open data or code, materials were located for 63% and 81%, respectively.

Discussion and implications: Availability statements were absent for about half of articles for data and most for code; when open data or code was claimed, materials were located in most, but not all. Future work will evaluate how these practices improve over time as the NSC Coordinating Center's support and journal and funder guidelines evolve.