Dr. Elisabeth Bik is a microbiologist and world-renowned scientific integrity expert. After receiving her PhD at Utrecht University in The Netherlands, where she worked on epidemic Vibrio cholerae strains, Dr Bik conducted research on the microbiomes of humans and marine mammals for many years. Writing on her blog ScienceIntegrityDigest, Twitter, and PubPeer, she is an expert in detecting inappropriately duplicated or manipulated photographic images and plagiarised text in biomedical literature.
Her investigative work has been featured in many international portals, including Nature News, the New York Times, and the Washington Post.
Kaustubh: The emergency with which the COVID-19 crisis has hit us has also led to an explosion in research about it, and in many cases, such work is hastily published without thorough peer-review or large sample sizes. Do you think this has been particularly problematic in terms of scientific integrity or research quality?
Elisabeth: Yes, I do. In light of the pandemic, many scientific journals have shortened the peer-review process, to make new studies more rapidly available. Normally, it might take weeks or months before a paper passes peer-review, but many recently published papers on COVID-19 and/or SARS-Cov-2 have been published within days. It is great that journals want to quickly share results, but the short or seemingly absent peer-review process has resulted in several published studies with flawed research designs and some suspected cases of fabricated data.
In addition, many papers being published now in scientific journals are opinion or regurgitation pieces that bring little insight or have a low novelty content. It seems that a lot of authors are trying to get a COVID-19 paper out, even if they have not worked on the topic themselves.
Kaustubh: You have been a vocal critic about the French study on chloroquine, both about its research and publication ethics. Can you please tell our readers about your main concerns with this study?
Elisabeth: I have many concerns with this study, which I have shared on my blog. Here are my main concerns. First, there were important differences between the treatment and control groups in the study, such as recruitment in different hospitals, underlying medical conditions, and outcome format of the viral detection test. A good clinical trial will assign patients randomly to the treatment and control groups to limit such differences, but unfortunately, Professor Didier Raoult, the senior author of this study, does not believe in randomisation.
Secondly, six patients in the treatment group were not included in the final analysis. Most of these did not fare well during treatment - three were admitted to the ICU and one died - but that does not mean a researcher should just remove them from the study.
There were also issues with ethical approval, viral detection, and clinical outcomes. All these problems should have been caught and discussed during peer review. But here is my biggest concern with this paper: it was published within a day with very little changes, and in a journal where one of the authors is the Editor in Chief. That is perceived as a huge conflict of interest.
Kaustubh: Is our traditional approach to scientific research too slow when facing such crises? Both in terms of addressing the problem, and on being exploited by dishonest scientists.
Elisabeth: For published, peer-reviewed papers, yes, scientific research is too slow. Good research just takes time, and in these COVID-19 times, there is a great and immediate need for new knowledge and treatment options.
But there is another way to rapidly share biomedical findings: by preprint on e.g. bioRxiv or medRxiv. Those preprint servers allow researchers to quickly upload and share data, typically within 24 hours. Preprints are not yet peer-reviewed, and not all research posted on preprint servers might be of high quality, but it is a fast way to share exciting new research with other researchers. These sites allow others to comment on the studies and are quick to take down papers that appear to contain flawed or falsified data.
Kaustubh: There has also been an explosion on fake news circulating on social media, from diverting blame to touting unverified cures. What are your thoughts about it?
Elisabeth: I am sad to see how gullible and not-critical some people are. It seems that suddenly everyone is an expert in viral epidemiology and anyone who says something online about a new way to solve the COVID-19 crisis will quickly gain a lot of followers.
Unfortunately, scientists who have years of experience in this field are not always heard or believed, and some of us have been attacked on social media for criticising poorly executed studies, fake news, and conspiracy theories. I am especially sad that some political leaders are not critical thinkers or are ignoring scientists, and how little social media platforms do to fight fake news. There have been some big announcements, for example, that Twitter would fight medical misinformation, but there is no way yet to report bad or even dangerous medical advice on the platform.
Kaustubh: How can the general public spot such fake news? Is there any way to improve awareness?
Elisabeth: It is extremely rare in medicine that a drug has a 100% preventive or healing effect. So the general public should not believe such stories, especially if they were done on a small study group. Anyone online who makes you first afraid of something and then wants to sell you their supplements or books should be distrusted. If you ask them to show a scientific paper that supports their claims, and they don’t answer, that is a huge red flag. And any conspiracy theory that has not been covered in trusted media sources should not be believed. In short, any story that makes you think ‘wait - what?’ should not be immediately believed but first be fact-checked.
Kaustubh: Do you think such fake news can lead to fewer people following the government guidelines, thereby causing an adverse effect on the public health situation? How do you think this can be controlled?
Elisabeth: Yes, there are some people who claim that the COVID-19 pandemic is not too bad. Some are ignoring shelter-in-place orders, demanding that sports events, cinemas, and other large gatherings will reopen. They do not realise that the situation might have been far worse without these social distancing rules, or that it could become worse if we ‘open’ the country too soon. I am not sure how this can be controlled. It seems that ignorance is everywhere, and it is especially hard to fight ignorance if political leaders agree with it.
Kaustubh: There has also been substantial criticism of the official reporting of numbers in various countries. For example, in the UK the official figures only report hospital deaths due to COVID-19, not deaths in other places such as care homes, which are suspected to be substantial. Does it erode public trust in official data and integrity, thereby giving an opportunity to false reporting and fake news?
Elisabeth: These are unprecedented and chaotic times, so it is not surprising that the numbers of death have been absolutely overwhelming in some places and have not always been recorded correctly. Nursing homes and deaths in private homes have not consistently been included in mortality rates across areas, so we will likely see increases in death rates as other numbers are starting to be included. I am not sure if that will erode public trust; it seems expected that death tolls are hard to know in a particular moment.
• Wondering how experts conduct research and modelling of COVID-19? We have another interview with statistician Dr Sayantan Banerjee, who explains in more mathematical terms, how such modelling is being done, and its application to the situation in India. Click here.
• To watch a thorough discussion and Q&A session on COVID-19 with experts from the OU STEM faculty, click here.
• If you are interested in the numbers around COVID-19, such as the meaning for the various terms, the modelling that's being used to obtain estimates, or the reliability of the statistics, then head over to ‘A statistician’s guide to coronavirus numbers’ by the Royal Statistical Society.