Every once in a while, we are reminded of the wisdom in ancient tales. During the current pandemic, we are specifically reminded of one of the most famous tales of Aesop’s fables, the “tortoise and the hare”, which gave rise to the famous idiom “slow and steady wins the race”.
The metaphorical race, in this case, is in the area of scientific publications. By becoming the most important area of research in 2020, Severe Acute Respiratory Syndrome-Related Coronavirus-2 (SARS-CoV-2) has prompted an unprecedented number of publications on a single topic within such a short time-span.
While the explosion in scientific publications pertaining to the virus was seen as a necessary measure to spread any information that could save lives, it is now clear that the accelerated submission and approval of such data was vulnerable to seriously flawed results.
This can be seen in the growing number of retracted SARS-CoV-2 papers, with retraction alerting the readers of an unreliable paper with erroneous content.
COVID and 2020 were fertile grounds for an “extraordinary year for science”, as dubbed by Nature. One of the publishers, Elsevier, saw 270,000 submissions between February and May of 2020 (a 58 per cent increase compared to the same time-point in 2019).
Health and medicine titles, in particular, saw a massive 98 per cent increase in submissions, with estimates claiming that papers published about the coronavirus pandemic, ranging from the disease itself to mental health effects, were between 100,000 and 200,000.
Approximately 30,000 of those articles were originally published as preprints (articles being uploaded online before being peer-reviewed), which put considerable pressure on journals to accelerate the process of reviewing and publication.
In other words, it wasn’t just the volume of articles that saw a huge jump but also the speed at which those articles are processed and published. It’s therefore not surprising to learn that such a publishing frenzy could result in many papers being later retracted.
‘Retracted watch’ estimates that 73 COVID-related papers have been retracted, an additional ten were retracted due to an error from the journal’s part, and five more papers are currently under concern. Looking at pure numbers, the proportion of published papers to those that have been retracted is about the same as the proportions seen in research in general.
What is quite unusual, is the speed upon which those retractions happened, ordinarily it would take around three years for a paper to be retracted after its publication. The important question to ask then is why were papers retracted so quickly?
The answer to that could vary widely. In the case of high profile studies such as the ones that relied on electronic health records from Surgisphere (a US medical analytics company), doubt was cast onto them after the company refused to allow auditing of their health data. Other cases of retraction came as a result of astounding technical failures, such as Elsevier publishing several papers twice. That being said, it would be interesting to consider whether the chase after a higher number of citations could have led some researchers to attempt to publish SARS-CoV-2-related papers in a short time-frame, contributing to the publishing frenzy.
The necessity for more data was critically high during the height of the pandemic, but should that mean that we must take the risk of higher data contamination in dire circumstances?
Regardless, it seems that it is yet too soon for any deep analyses to be done on whether SARS-CoV-2 papers (published in 2020) are more likely to be retracted.