What is a biomedical preprint?

A preprint is a complete scientific manuscript that researchers post to a public server before submitting it to a peer-reviewed journal. The manuscript is not peer-reviewed, edited, or typeset before going online. It undergoes only a basic screening process — checking for offensive content, non-scientific material, and potential biosecurity risks — and then becomes immediately and freely readable by anyone.

The concept is not new. Physicists and mathematicians have used arXiv (hosted by Cornell University) since 1991. What is new is the adoption of preprints in biology and medicine. bioRxiv launched in 2013; medRxiv followed in 2019. Together they transformed how biomedical findings reach the research community.

Authors retain full ownership. They can submit revised versions at any time, but every version is permanently archived — nothing disappears. And because preprints carry a DOI, they are citable from the moment they go live.

bioRxiv and medRxiv — how they work

Cold Spring Harbor Laboratory operates both servers. bioRxiv (pronounced "bio-archive") focuses on basic research in the life sciences — molecular biology, genetics, neuroscience, ecology, bioinformatics, and related fields. medRxiv (pronounced "med-archive") focuses on clinical and health sciences research and is a partnership between Cold Spring Harbor Laboratory, BMJ, and Yale University.

Both were inspired by arXiv, the physics/mathematics preprint server that demonstrated decades ago that rapid, open manuscript sharing accelerates the scientific enterprise.

The submission process is straightforward: upload a manuscript, provide metadata, and the server's screening team reviews it — typically within 48 hours. Once cleared, the preprint goes live with a permanent DOI. Authors can transfer the manuscript directly to a partner journal for formal peer review, and many journals now accept submissions via bioRxiv's direct-transfer pipeline.

Do preprint conclusions survive peer review?

This is the question that matters most. If preprint findings routinely collapse under peer review, their utility drops. The data says otherwise.

Abdill and Blekhman analysed all 37,648 preprints uploaded to bioRxiv in its first five years and found that approximately two-thirds of those posted before 2017 were subsequently published in peer-reviewed journals. Downloads correlated with the impact factor of the eventual publishing journal — preprints that attracted more attention tended to land in higher-profile outlets (PMID 31017570).

A separate study in PLoS Biology tracked what actually changed between preprint and published versions. Brierley et al. found that only 7.2% of non-COVID-19 preprints had a discrete change to their abstract conclusions after peer review. For COVID-19 preprints that figure rose to 17.2%, but even among those, the majority of changes were not qualitative reversals (PMID 35104285).

The COVID-19 preprint explosion

Nothing tested the preprint ecosystem like the SARS-CoV-2 pandemic. Researchers flooded bioRxiv and medRxiv with manuscripts at a rate the servers had never seen. In the early months of 2020, unreviewed preprints initially outnumbered peer-reviewed COVID-19 articles — and they became the primary information channel for clinicians, public health officials, and journalists (PMID 33094906).

This brought both the value and the risk of preprints into sharp public focus. On one hand, preprints enabled immediate sharing of viral sequences, epidemiological models, and treatment data — weeks or months before journals could process them. On the other hand, some poorly designed studies reached policy-makers before anyone had checked the methods. Bagdasarian et al. documented specific cases where unvetted results influenced public health decisions prematurely (PMID 32586327).

Añazco et al. tracked over 5,000 COVID-19 preprints posted in the first five months of 2020. Only 5.7% had been published in journals within the study window — an unusually low rate likely driven by the sheer volume of submissions overwhelming editorial pipelines. Published preprints received significantly more citations than their unpublished counterparts (PMID 33717688).

Benefits and risks of biomedical preprints

The literature on preprint benefits and risks is substantial. Vlasschaert et al. reviewed 53 papers discussing the topic from the researcher, publisher, and end-user perspectives (PMID 32680914). The consistent themes:

The critical point: preprints shift the burden of quality assessment from reviewers to readers. For trained scientists with domain expertise, this is manageable. For journalists, policy-makers, and the general public, it is not — and that gap is where misinformation enters.

Sever, co-founder of bioRxiv and medRxiv, has argued that the future of biomedical publishing lies in a more decoupled system: manuscripts disseminated first via preprint servers, then evaluated by journals and independent content-vetting initiatives. The goal is to separate the act of making results public from the act of certifying them (PMID 37788235).

Preprints, open access, and the future of biomedical publishing

Preprints are open access by definition — free to read, free to post, permanently available. This makes them a structural part of the open science movement, not a marketing exercise.

The numbers matter. Approximately half of all biomedical literature is now open access, and the share grows every year as funder mandates tighten. The NIH, Wellcome Trust, Gates Foundation, and EU Horizon all require open access to funded research. The US White House mandated immediate public access to all federally funded research results. Plan S and cOAlition S continue to expand across Europe.

Preprint servers accelerate this shift. Every manuscript posted to bioRxiv or medRxiv is immediately open — no embargo, no APC, no institutional subscription required. For the most current research, which is what most scientists need, preprints offer the fastest route to full-text access.

The traditional publishing model — where scientists produce knowledge, corporations own it, and institutions pay to access it — is under pressure from every direction. Preprints are not a replacement for peer review; they are a parallel channel that ensures results reach the community regardless of how slowly journals move.

How BioSkepsis handles biomedical preprints

BioSkepsis treats preprints from bioRxiv and medRxiv as first-class content. They are fully indexed, searchable, and available for AI-powered synthesis alongside peer-reviewed articles from PubMed Central and OpenAlex.

This matters because ignoring preprints means ignoring some of the most current research in any field. A tool that only searches peer-reviewed publications will always be months — sometimes years — behind the frontier.

The goal is not to blur the line between preprints and peer-reviewed articles. It is the opposite: give researchers access to everything, label it honestly, and let them make informed decisions. That is what evidence-based research practice looks like.

Frequently asked questions