Supporting Post-publication Peer Review: Journals with New Peer Review Models

F1000 Research

F1000 Research is the flagship journal of Faculty of 1000, a company that provides publishing and research services primarily for life scientists and clinical researchers. It was created to combat key weaknesses in the publication system:

• Publication Timing Is Slow: Whether author or fellow researcher, you want to communicate and learn about the newest findings in your field as soon as possible. Conventional publication, even open-access, puts a long waiting period on publication due to peer review.
• Reviewers are Anonymous: Reviewers of a paper are anonymous. This secrecy invites conflicts of interest, as researchers in the same field could have an advantage by slowing down publication of competing research labs’ work.
• Gatekeepers are Not Experts: Conventionally, the final publication decision rests on journal editors. It’s highly unlikely that a journal editor will have comparable expertise to the author in the specific sub-field, especially for journals with wider scopes.
• Bias in Article Selection: Journals are notorious biased towards publishing “significant” and positive research findings: for example, findings that X is an effective treatment for a disease is prioritized over findings that X isn’t an effective treatment for it. This skews our understanding of science to include only positive findings, even though negative findings are sometimes just as important.
• Journals are Selfish: Journals are rated by impact factor, just like papers and researchers: this creates a conflict of interest between publishing research in the public interest and publishing research that they believe will be cited more.

F1000 began as an organization that invited scientific researchers who are in their respective fields to serve as a panel of experts. Starting with 1000 members, each invitee was asked to name the 10 most important research topics in their specific subject, and then the 10 best experts in those topics. Though its name has changed, F1000 now boasts membership of over 10,000 leading experts, both seasoned Nobel Prize winners and newly minted PhDs, who recommend scientific papers (the F1000Prime service) and serve as peer reviewers (F1000Research).

F1000Research addresses the problem above as follows:

• Publication is Immediate: Submissions are published immediately, following a cursory check for scientific relevance and policy adherence.
• Reviewers are Transparent: Another revolutionary step is that authors choose their own reviewers from a list. While conflicts of interest are forbidden (for example, a co-author cannot be chosen), the author can be assured that because reviewers are members of F1000, they have been vetted as experts in their stipulated fields. If there are not enough suitable reviewers, the author can make recommendations.
• There are No Gatekeepers: Each published paper is given an approval status: Approved, Approved with Reservations, and Not Approved. This approval status is visible for all to see, as is a detailed report written by the reviewers outlining their logic and findings.
• No Selection Per Se: All papers are published, irrespective of whether their findings are positive or negative. The ‘significance’ of each paper is established through the review process and commentary.
• The Journal Has No Say: All papers are published, regardless of whether they’ll be cited highly or less highly.

PubMed Commons:

If you work in medicine or the life sciences, you’ve no doubt accessed PubMed. Maintained by the United States National Library of Medicine (NLM) at the National Institutes of Health, PubMed is a search engine for the MEDLINE database, which contains over 26 million records ranging from the 1950s to the present. After an article is published in any one of 5,600+ publications covered, MEDLINE indexes its abstract, citation information, and (for open-access articles) links to its full text.

Enter PubMed Commons. Also maintained by the NLM, PubMed Commons enables authors to share opinions and information about scientific publications in PubMed, providing a forum for scientific discourse that is integrated with PubMed. Any author of a publication in PubMed is eligible to join and post comments to any citation. Successful applications of this forum to date include:

• Reporting corrections and retractions to papers
• Discussing alternative and conflicting interpretations of data
• Enhancing reproducibility by providing links to datasets, code and software, or publicly accessible texts

We checked the website on a random day, and found the following new comments featured on the homepage. Wouldn’t you agree that this information is important? Yet, you wouldn’t not find it in a traditional journal.

• An Australian journal club’s praise of a well-conducted randomized clinical trial comparing hypertension interventions, with accompanying criticism of its external validity.
• A British physician’s rebuke of suspicions of a conflict of interest with the tobacco industry, in the context of tobacco harm reduction.
• An American scientist’s report linking readers to follow-up reports on a European genomics study of ancestry.
• An American scientist’s announcement that she was refused access to data from a French environmental contamination study whose findings were cited in the text of government regulations, in violation of open-access rights for its journal of publication (PLOS ONE).

Public Service Announcement: PubMed Commons is great for journal clubs

PubPeer

PubPeer is not a journal per se, but basically a repository of comments on published journal papers, both peer reviewed and not. Comments can be accessed via the PubPeer website, or through a browser extension (available for Chrome, Firefox, and Safari). When you view a site with the browser extension enabled, you’re provided with another layer of commentary, not only the paper itself.

As a result, while most papers are still going through the traditional peer review system, PubPeer gives the paper a wider context. Criticisms, praise, and comments are all intermingled together to provide new insights, collaboration, and research directions.