(Reuters) - Two studies published on Tuesday in PLoS Biology suggest that papers posted on preprint servers before undergoing formal peer review do not change significantly before publication in peer-reviewed journals.
One study compared more than 180 reports posted during the first four months of the pandemic on the preprint servers medRxiv and bioRxiv to the versions eventually published in peer-reviewed journals. Roughly 83% of COVID-related papers and 93% of non-COVID-related papers did not change from their preprint to final published versions, they found. When the researchers did identify changes, in the majority of cases those changes did not qualitatively change the conclusions of the paper, they said.
The other study used machine learning to analyze the relationships between nearly 18,000 preprints on the bioRxiv server and their published versions. Most manuscripts had only modest changes in wording during the peer-review and publication process, the researchers found.
SOURCES: https://bit.ly/3rnNz5d and https://bit.ly/32Wjnoi PLOS Biology, online February 1, 2022.
Reuters Health Information © 2022