Data Sharing of Research: Implications for Otolaryngology

The “ethical and scientific imperative” ¹ for data sharing has led to regulatory requirements. Since 2003, the National Institutes of Health (NIH) has required certain federally funded researchers to propose a data sharing plan, ² and this has been extended to include the release of all NIH-sponsored data that would allow for the validation and replication of research findings. ³ In 2022, the White House Office of Science and Technology Policy (OSTP) updated its support of data sharing, requiring all federally funded articles be freely available immediately and public access plans be updated to facilitate data accessibility. ⁴ The International Committee of Medical Journals (ICMJE) also requires clinical trial manuscripts to contain a statement and plan regarding data sharing to be considered for publication. ⁵ These directives promote data sharing to enhance transparency, reproducibility, and validation of biomedical research.

In “Data-Sharing across Otolaryngology: Comparing Journal Policies and their Adherence to the FAIR Principles” by Johnson et al, ⁶ the authors examined data sharing within otolaryngology journals. Using the FAIR principles, which offers a framework to guide the management and stewardship of data, they found that while most otolaryngology journals included a data sharing initiative, they were not standardized in reporting the FAIR principles. Notably, few required data sharing for publication.

This study highlights the tension between support for data sharing and the practical challenges of enforcing a system to support it. The variability in adherence to the FAIR principles is due to fragmented approaches to data sharing policies. Amongst otolaryngology journals, there is hesitation to mandate data sharing, perhaps due to regulatory requirements that only apply to federally funded researchers, a population that reflects only a small portion of researchers submitting manuscripts. While data sharing can enhance the rigor and quality of research, the feasibility of data sharing poses complexities that are highlighted below.

Data sharing requires dedicated data storage. Since the NIH 2003 Data Sharing policy, ² several repositories have been established through support from industry ⁷ and the NIH. ⁸ While repositories primarily support data obtained from clinical trials, a unified space is needed to encompass all types of data sets, particularly in otolaryngology where clinical trials are modest. Common data elements can help with data standardization ⁹ but administrative and financial support is needed to maintain the repository and steward the data upon request. ⁷

Ensuring data remains de-identified is also important, a concern echoed amongst otolaryngology journals. ⁶ If data sharing compromises data privacy, consent for research participation may decrease. ¹⁰ Protecting against privacy breaches is essential to honor the ethical obligation of respecting participant privacy and earn the public’s trust in open science. ¹¹ If data may be pooled or aggregated, data sharing may offer an avenue to promote research in marginalized communities by improving the generalizability of research findings to communities who have been underrepresented in research.^{11 -14}

Intellectual property is also a concern for original investigators. ¹⁰ Some have advocated for new investigators to work collaboratively with original investigators, culminating in co-authorship and acknowledgments of original investigators in subsequent manuscripts. ¹⁵ Others have supported offering original investigators an opportunity to appraise and comment on manuscripts that serve to verify original research findings. ¹⁶ Particularly for clinical trials, the purpose for the use of original data may be grouped into (a) verifying original results and (b) answering a new hypothesis. ¹ Delineating a framework that credits individuals appropriately will be necessary to empower scientific inquiry in a collaborative and fair environment.

Data sharing poses a promising solution to the reproducibility crisis in otolaryngology and beyond but is not without complexity. Regulatory efforts have made progress in promoting data sharing as a research norm, but medical institutions, professional societies, and journal editors can make headways by instituting and enforcing their own data-sharing standards beyond what is federally mandated.^5,17 Collaborative discussions between key stakeholders will be needed to build a unified framework that is just, enforceable, and practical. ⁵