Since the start of the COVID-19 pandemic, there’s been an explosion in the production of scientific articles and data. Publishers have geared up to publish content at an unprecedented rate, and to provide open access to certain scientific publications. Researchers, meanwhile, are increasingly opting for online prepublication platforms. Will the pandemic change the research culture or how results are made available in the future?
“Open access to scientific articles is not a new thing,” says Vincent Larivière, who holds the Canada Research Chair in the Transformations of Scholarly Communication at Université de Montréal. But he believes that the current crisis has demonstrated the effectiveness of open access to scientific publications.
By making their articles freely accessible, publishers are setting a precedent. According to Dr. Larivière, they are effectively acknowledging that “barriers to access are an obstacle to the advancement of knowledge.” The scale of the current health crisis and its impact on global health make it essential to share research results quickly with the largest possible audience. “But at that point, how do you justify not making research on cancer or cardiovascular diseases freely accessible as well?” asks Dr. Larivière.
The growing power of prepublication
Since the start of the pandemic, Dr. Larivière has observed “almost exponential growth” in the use of prepublication servers. The lack of peer review for the results posted to these platforms is frequently mentioned as a critical weakness of prepublication servers, but he says he doesn’t think this is really a problem.
Leslie Chan, an associate professor, teaching stream, at University of Toronto Scarborough, shares this opinion. “There are plenty of examples showing that peer review doesn’t prevent unreliable publications or even outright fraud,” notes Dr. Chan. Rather, he believes that prepublications accelerate the spread of knowledge, and that releasing the results to the largest possible audience makes it possible to detect errors earlier in the process. “This is a test period,” he argues. “After the pandemic, we’ll need to compare prepublications with the traditional publishing system. We’ll see which approach is more effective.”
For example, the prepublication that suggested similarities between SARS-CoV-2 (the virus that causes COVID‑19) and HIV was withdrawn 48 hours after it was published on the BioRxiv server and immediately faced criticism from researchers from around the world. By contrast, it took fully two weeks for The Lancet to withdraw a controversial study on the use of hydroxychloroquine to treat the symptoms of COVID-19.
Access to data
The controversy over the Lancet article also highlights the importance of sharing raw data. According to Dr. Larivière, the discussion also “reopened the debate on privatization of medical data.” Jean-Baptiste Poline, an associate professor in the department of neurology and neurosurgery at McGill University’s faculty of medicine, notes that “collecting and sharing data takes time.” He says he regrets that this process is not more enthusiastically supported by large funding agencies and universities.
The health emergency has also led to increased collaboration and data sharing among researchers. “The scientific community doesn’t just need articles, it also needs the computer codes, protocols and raw data that are used for research,” Dr. Poline says. He believes that researchers should share their data, but also make it accessible and reusable by other researchers – and by machines as well. “That’s what’s going to make research truly effective,” he argues.
China as a driver of change
China, where the coronavirus first emerged, could be a source of major changes in the system of scientific publications. Dr. Larivière explains that Chinese authorities, who had previously been very attached to publications in high-impact journals, are now realizing “the adverse effects of such an evaluation mechanism.”
The first studies on SARS-CoV-2 were published in English-language Western scientific journals by researchers from the Chinese Center for Disease Control and Prevention in Beijing. As a result, news of the discovery of an emerging virus travelled around the world before reaching across China.
In response, the Chinese ministry of science and technology requested in February 2020 that research institutions stop paying publication bonuses to researchers. In the context of the pandemic, Beijing also imposed a requirement that a portion of research on the novel coronavirus be published in Chinese, to ensure that it could be understood by the Chinese medical community.
Dr. Larivière believes that in light of China’s large population and its political importance, these decisions could tip the scales and hasten the end of the English language’s hegemony and impact in scientific publications.
Your example of the quick condemnation of a preprint publication suggesting similarities between SARS-CoV-2 and HIV, is spot on! Post-publication review, be it quick (a preprint publication) or slow (a formally peer-reviewed publication), has numerous advantages. But please do not step into the publishing industry trap of referring to preprint publications as “prepublications.” Publication means make public, or, in your words, “to share research results quickly with the largest possible audience.”
“Will the pandemic change the research culture or how results are made available in the future?” That very much depends on whether the publishers will be able to appropriately transform their business model. The sad bereavement of PubMed Commons (National Center for Biological Information) after a 5-year run of illuminating post-publication, non-anonymous, review, hints negatively.
Regardless of how dominant China becomes in science (and that can easily be and frequently is overstated) there is little to no chance that the hegemony of English in scientific publications will be broken. First, English has a huge scientific vocabulary based on Greek and Latin roots; Chinese has neologisms that result in very complex technical and scientific terms. Chinese is neither as precise or descriptive as English as a technical language, or German was before it with its compound nouns. Second, Chinese is one of the most difficult languages in the world to learn, and 75% of the world’s researchers would have to learn Chinese as adults, a nearly impossible task. It doesn’t even have an alphabet, but requires the memorization of thousands of characters for fluency. Third, while China is now certainly a scientific power of the first rank, Europe and North America, as well as other major countries in Asia from Japan to ANZ to India, remain critical loci for research advancement. That’s not going to change anytime soon.