Exploring Data Sharing Practices in Ophthalmology Journals: A Cross-sectional Analysis

Background: Open data practices in medical research not only uphold the principles of reproducibility and integrity in scientific endeavors but also pave the way for collaborative advances by allowing researchers to build upon each other's findings. Despite the International Committee of Medical Journal Editors (ICMJE) making data sharing mandatory for clinical trials since 2017, barriers persist in research data sharing. Even decade-old ophthalmological data from the US National Institutes of Health remains unreleased as of 2023. The extent of ophthalmology's adoption of data sharing remains uncertain, influenced by journal and institutional policies, clarity of statements, and dataset prevalence and accessibility. This study aims to investigate the current state of data sharing in ophthalmology, identifying strengths, barriers, and improvement avenues. Understanding these practices is pivotal, not just for the sake of transparency, but also for ensuring the rapid advancement of ophthalmology as a discipline.

Methods: In this cross-sectional analysis, we identified the top ten Ophthalmology journals that published original research articles, ranking them based on the Clarivate Journal Impact Factor scores. We searched and compiled a list of all original research articles published in these 10 journals over the 5-year timeframe from 2018 to 2023. To be included, articles had to be original research studies focused on ophthalmology, providing new primary data (either quantitative or qualitative) and published between 2018 and 2023 in one of the top 10 selected journals. Exclusions comprised review articles, commentaries, editorials, letters, studies unrelated to ophthalmology, and those lacking novel primary data.

Using a standardized extraction form, two researchers independently extracted data on bibliographic details, presence and details of data sharing statements, repository details, study design, journal data policies mentioned, ethical approval obtained, and consent for data sharing mentioned. Two investigators independently conducted data extraction within the sample, resolving discrepancies through discussion. Upon completing data extraction, we computed the number and percentage of articles containing data sharing statements, categorized by journal and publication year.

Results: In the analysis of 1406 screened and evaluated articles, approximately one-fourth of articles included a data sharing statement. Among these, the majority indicated data availability upon request, while one-fifth had data accessible on online platforms. A smaller proportion explicitly mentioned that their data was not available for sharing. Notably, less than one-tenth of articles with data sharing statements had obtained consent for data sharing.

Conclusion: Our cross-sectional analysis of ophthalmology research articles reveals a landscape where data sharing practices are not yet pervasive, despite efforts to promote them. Approximately one-fourth of the examined articles included data sharing statements, with the majority offering data availability upon request and a smaller fraction making data accessible on online platforms. The limited mention of obtained consent for data sharing highlights a significant gap in ethical considerations. This study underscores the existing barriers in ophthalmological data sharing, influenced by journal and institutional policies, clarity of statements, and dataset accessibility. Addressing these issues is crucial not only for transparency but also for fostering collaborative advancements and ensuring the rapid progress of ophthalmology as a discipline.