Why code sharing matters in medicine

Modern medical science increasingly depends on computational approaches: from algorithms that process large genomic datasets, to statistical scripts underpinning observational studies, and code that implements predictive models of disease progression. These digital tools are not peripheral - they are central to how conclusions are reached, interventions are judged, and evidence influences practice. Yet, historically, much of this code has remained invisible to readers, reviewers, and secondary analysts.

By making author-generated code accessible at the time of publication, PLOS Medicine is strengthening the scientific record in three key ways:

1. Reproducibility—code sharing allows independent researchers to reconstruct analyses, verify findings, and explore alternative interpretations with confidence.
2. Usability—shared code becomes a usable resource for others to build upon, adapt, and extend to new questions that matter to patients and societies.
3. Trust—transparency fosters trust among clinicians, funders, and the public by making every step of a research finding accessible and open to scrutiny.

Code sharing is not simply an add-on to publication; it is part of robust scientific communication.

What the policy means for authors

Under the new policy, authors submitting articles to PLOS Medicine in which author-generated code supports the results of the manuscript are expected to make that code publicly available upon publication. Code should be accompanied by sufficient documentation to enable reuse and, wherever possible, deposited in repositories that provide persistent identifiers (see Box 1). This expectation mirrors our long-standing commitment to open data and open materials, and reflects the reality that code is as essential a research artifact as the data it analyzes.

We recognize that legitimate ethical or legal considerations, for example, patient privacy limitations or contractual restrictions, may sometimes constrain what can be shared. In such cases, authors should be transparent about these constraints and describe how code can be accessed or evaluated under appropriate controls. Editorial and peer review will continue to evaluate compliance thoughtfully and fairly, with the shared goal of maximizing openness without compromising ethical obligations.

A cultural shift, not just a policy

Implementing this policy is not merely a procedural change—it signals a transformation in how medical research communicates and collaborates. Medicine tackles complex human challenges where every decision can have profound consequences. The computational tools and analyses that underpin these decisions should be visible, interpretable, and reusable, not confined to private scripts or inaccessible silos. By embracing open code, PLOS Medicine is committing to a future where clinical and translational research is not only open to readers but open to re-use, scrutiny, and community innovation.

We are grateful to the many authors, reviewers, and editorial board members who have championed open methods and supported us with developing this policy. Many in our community already share code generously, with 53% of articles published in 2025 sharing code [5]; this policy formalizes that culture and encourages broader adoption of best practices. We will continue to provide guidance and support to authors as they navigate these changes, and we look forward to learning from our community’s experience as this policy takes effect.

Our commitment to code sharing is a milestone in a larger journey toward fully open and accountable science. Over the course of 2026, we will run a soft rollout, offering support and learning opportunities for authors, before making code sharing mandatory in 2027. We will continue to listen, learn, and refine our approach and support authors, ensuring that the policy strengthens reproducibility, fosters innovation, and accelerates scientific discovery.