A large analysis of around 70,000 biomedical studies suggests that research papers posted as preprints are often more reliable than critics believe.

The study, shared on the preprint server bioRxiv, found that the main conclusions of most biomedical preprints usually remain unchanged after the studies go through peer review and are published in academic journals.

Researchers also discovered that studies first released as preprints are retracted at roughly half the rate of papers that were only published after traditional peer review.

The findings suggest that preprints can serve as a dependable source of scientific information, although some researchers caution that the results should be interpreted carefully.

Posting research as a preprint has become increasingly common in modern science, allowing scientists to share findings quickly before formal journal publication. However, concerns remain among some researchers who argue that preprints have not undergone the same level of scrutiny as peer-reviewed studies.

Ruslan Rust, a neuroscientist at the University of Southern California, said he frequently hears concerns about the reliability of preprints. Based on his own experience, he believed peer review often does not result in major changes to a study’s core findings.

To test whether this was true across biomedical research, Rust and his colleagues analysed thousands of studies from different fields.

The research adds to the ongoing discussion about the role of preprints in science, suggesting they may provide valuable early access to research while still benefiting from later peer review and validation.