«I Do Not Have Time»—Is This the End of Peer Review in Public Health Sciences?

Abstract

In a 2019 workshop of the Swiss School of Public Health (SSPH+) -the owner of the IJPH and PHR -all participants agreed that our journals should continue with thorough prepublication peer review, including revisions, point-by-point responses and final decisions made by science editors, to strengthen the quality of publications. But soon after, the explosion in Covid-19 research caused an unprecedented increase in submissions and thus demand for peer review in health science journals (1)(2)(3). During the first 6 months of the pandemic, total publications and Covid-19 related publications increased exponentially (2). Submissions to JAMA almost tripled (4) and to IJPH more than tripled. Elsevier's health and medical journals saw a 92% increase in submissions between February and May 2020 (3). The publication rates of all peer reviewed public-health related articles increased by 25% from 2019 to 2020, and by 21% from 2020 to 2021, exceeding the annual publication growth rates before the pandemic which were 4, 15, 9 and 19% from 2016 until 2019 (Web of Science, category Public, Environmental and Occupational Health).The growth in publications has brought the peer review system to the edge, as the commitment of scientists did not keep pace with the increase in manuscripts. For Elsevier's health and medical journals a 50% decreasing proportion of peer review invitation 37 acceptances have been disclosed for February to May 2020 compared to the same period in 38 2019 (3). The records of IJPH provide sobering facts, too. "I do not have time" has become a 39 leading response -and even worse: the majority does not respond at all. In 2021, 53% of 40 (repeated) invitations for peer review remained un-answered, whereas 38% actively 41 declined. Both, in 2021 and 2022, only 9% of all invited reviewers agreed to provide a 42 review. In contrast, during the years prior to the pandemic, we observed a rather stable rate 43 of 35-40% who would agree and deliver a review report.The consequences of this crisis are equally bad for our authors and editors. First, the 46 endless search for reviewers has slowed down the publishing process substantially. Second, 47 editors are forced to release automatic "mass-invitations" to 20+ potential reviewers at a time 48 and to remind those repeatedly, thus, flooding the stressed community with even more 49 invitation emails. Third, automated search strategies require artificial big-data search 50 engines. As those are of limited intelligence yet, a rather high rate of 25% -probably the tip 51 of an iceberg -of active decliners tell us "this is not my field". Fourth, we are forced to 52 discuss whether and under what conditions final decisions should be based on the 53 feedbacks of only one review. And last but not least, the workload for our handling editors 54 has also increased substantially, which in turn triggers their resistance to handle 55 manuscripts.The simple truth is: if we collectively "do not have time" to review manuscripts, there will be 58 no pre-publication peer review anymore. Thus, key questions emerge. Does public health 59 science serve authorities and the public sufficiently if research is published without formal 60 pre-publication peer review? Is post-publication peer review also appropriate to promote 61 good public health sciences? Will publishing on pre-print servers become the new standard 62given the abundance of innovation in this field (5). In the absence of clear evidence for the 63 opposite, we strongly believe that the multi-disciplinary public health sciences profit from pre-64 publication peer review. Thus, we need a solution to address the crisis.Given the ubiquity and magnitude of the problem, we call for concerted strategies also of 66 publishers to resolve it together with journal editors as they cannot do it alone. The literature 67 discusses various incentives for reviewers, among them discounts on publisher's products, 68 certificates, recognizing the best reviewers, increasing diversity in the peer-review process 69 and financial incentives (5-8). The development of reviewer recognition platforms (ORCID, 70Clarivate 's reviewer recognition platform) is a positive example.From a quantitative perspective, the solution looks rather straightforward in our typically 73 multi-authored science: if researchers agree to review as many papers per year as they write 74 as first or last author, the crisis would resolve. Indeed, the trends in scientific publishing 75where the number of publications double every ten years whilst the number of scientists 76 increases by only 21% ( 7) calls for such level of commitment. When reviewers are asked, a 77 vast majority (85%) find recognition and training will improve the efficacy of peer review and, that universities and employers should explicitly require and recognize the reviewing work 79 which should be career enhancing (9). 80 81Our two journals have adopted a feature to remove at least barriers of the peer review mode 82(blinded or open) on reviewers' willingness to review: while we run peer reviews double blind 83 to minimize biases, we leave it up to the reviewers to decide after peer review if they would 84 like to publish their name, the review report or both with the published article. We are aware 85 that reviews of rejected articles are not recognized this way and therefore, encourage 86 reviewers to register their reviews on reviewer recognition platforms. We invite excellent 87 reviewers who provided relevant information beyond the content of an article to comment in 88 an editorial. We publish reviewer acknowledgments. And -upon request -we provide 89 reviewer certificates. We offer an online course to train junior scientists in the editorial tasks 90 including peer reviewing. However, this is not enough.

Topics & Concepts