Evaluating data sharing statements in leading emergency medicine journals: A mixed methods review

Faith Hendrickson; M D Hossain; Jacob Duncan; Brody Dennis; Alex Hagood; Taylor Gardner; Eli Oldham; Eli Paul; Alicia Ito Ford; Matt Vassar

doi:10.1016/j.ajem.2024.12.027

Evaluating data sharing statements in leading emergency medicine journals: A mixed methods review

Am J Emerg Med. 2024 Dec 16:89:159-168. doi: 10.1016/j.ajem.2024.12.027. Online ahead of print.

Authors

Affiliations

¹ Office of Medical Student Research, Oklahoma State University Center for Health Sciences, Tulsa, OK, United States. Electronic address: faith.hendrickson.research@gmail.com.
² Office of Medical Student Research, Oklahoma State University Center for Health Sciences, Tulsa, OK, United States.
³ Office of Medical Student Research, Oklahoma State University Center for Health Sciences, Tulsa, OK, United States; Department of Psychiatry and Behavioral Sciences, Oklahoma State University Center for Health Sciences, Tulsa, OK, United States.

PMID: 39729683
DOI: 10.1016/j.ajem.2024.12.027

Abstract

Background: Emergency medicine remains as the front line of healthcare, providing acute care to a diverse population during urgent and critical moments. Our objective was to systematically quantify the prevalence of data sharing statements (DSS) in high-impact emergency medicine journals and assess their implementation by contacting corresponding authors who indicated data available upon request.

Methods: We conducted a cross-sectional analysis to identify the prevalence and content of data sharing statements in clinical studies published in high impact emergency medicine journals between 2018 and 2023, followed by a hierarchical logistic regression analysis to identify variables impacting the incorporation of data sharing statements into emergency medicine studies. Lastly, we conducted a thematic analysis of the data sharing statements extracted from our included sample to identify prevailing themes.

Results: Of the 875 articles reviewed, only 84 (9.6 %) included DSS. DSS prevalence increased over time, from 2.12 % (4/189) in 2018 to 23.71 % (23/97) in 2023. Studies without listed funding sources were significantly less likely to include DSS (estimate = -2.027, p = 0.000). Articles published by BioMed Central (mean effect = 2.497) were more likely to include DSS compared to other publishers. Our review of the DSS content revealed prevailing themes, such as gatekeeper role and conditional data availability, concerning the oversight and regulation of information and the stipulations for data sharing based on certain criteria.

Conclusion: Our study revealed a lack of adherence to DSS, with a slight improvement recently. Listed funding sources and publishers influence DSS presence, highlighting their policy impacts on open science. Our recommendations include increased funding for data sharing, implementation of FAIR (Findability, Accessibility, Interoperability, and Reuse of digital assets) principles, and encouraging DSS inclusion as a publication requirement to enhance transparency and reproducibility, improving patient care. Further research needs to address barriers to data sharing and develop sustainable solutions in a dynamic research environment.

Keywords: Data sharing; Data sharing statements; Emergency medicine; Open science; Reproducibility.

Publication types

Review