Preprints: An underutilized mechanism to accelerate outbreak science

Johansson, Michael A.; Reich, Nicholas G.; Meyers, Lauren Ancel; Lipsitch, Marc

The statement coincided with a proliferation of Zika research in response to the epidemic in the Americas (Fig 1A) . Between November 2015 and August 2017, we identified a total of 174 preprint manuscripts with Zika in the title or abstract in 5 recognized public preprint repositories, including 4 general repositories: bioRxiv (124, http://www.biorxiv.org/), arXiv (31, arxiv.org), F1000Research (12, with 1 other that was originally posted in bioRxiv, f1000research.com), PeerJ Preprints (5, peerj.com/preprints/), and the World Health Organization Zika Open repository (2, www.who.int/bulletin/online_first/zika_open/), which was established explicitly for the Zika epidemic (S1 Dataset). These likely represent the majority of Zika preprints, though others may have been posted in ad hoc, lesser-known, or laboratory-or university-specific webpages or repositories. Over a similar time period in the Ebola outbreak (May 2014 to January 2016), there were 75 Ebola preprints ( Fig 1B) . There were many more publications in peer-reviewed journals over these time periods: 1,641 and 2,187 publications indexed in PubMed (http://www.ncbi.nlm.nih.gov/pubmed, S2 Dataset) with Ebola or Zika, respectively, in the title or abstract and of type "Journal Article" (257 and 417 additional publications were classified as letters, reviews, etc., respectively). This increase in publications (33%) cannot explain the increase in preprints alone (132%). Over the same time period, there was a general increase in preprints for health sciences [3] ; for example, bioRxiv submissions with the word "outbreak" rose approximately 10-fold between 2014 (24) and 2016 (232). Thus, the increase in preprints related to the Zika epidemic likely reflects this general trend, some increase in research, and possibly the influence of the statement on data sharing.