Covid-19 Research Scandals Illustrate What’s Wrong With Science
A new book takes science to task, and it couldn’t come at a better time
As a science journalist and former researcher, I was terrified by British psychologist Stuart Ritchie’s new book, Science Fictions. Ritchie lays out all of the ways in which modern science has failed, with a plethora of shocking and embarrassing examples, many involving famous studies. He lists negligence in scientific methods, bias in the search for answers, hyping up of a study’s results, and flat-out fraud as being science’s four horsemen of the apocalypse, and he cites scandal after scandal in the fields of medicine, biology, and especially his own discipline of psychology as evidence. These error-filled and misleading publications confuse and delay genuine scientific progress, waste vast amounts of human and monetary resources, and, in some cases, even cost people their lives.
Ritchie, who holds a doctorate and is a lecturer at the Institute of Psychiatry, Psychology and Neuroscience at King’s College London, argues that the academic science industry, with its infamous “publish or perish” mantra, has pitted scientists against each other for dwindling resources in the form of grants and tenure-track jobs. As a result, corners are cut during the research process — either deliberately or accidentally — in order to produce ever more positive and impressive results, regardless of whether they’re real or not.
The book, which was published in July, could not have come at a more apt time. Every problem Ritchie raises in it has played out in the research on the novel coronavirus. Elemental spoke with Ritchie about how the pressure cooker of a pandemic has made the situation worse and what it means for the public’s trust in science at this critical time.
Elemental: Taking a Covid-19 lens to the book, it seems like a lot of what you describe is accelerated with the novel coronavirus. The pressure to publish is even more heightened right now, and we’re already seeing papers being retracted. What’s your overarching view of the scientific process during the Covid-19 pandemic? Are you concerned about the results that have already come out, or could the pandemic be beneficial to the scientific process in some way?
Stuart Ritchie: You can basically find a Covid story that sits nicely alongside any of the stories I told in the book, whether it’s negligence or hyping up stuff or bias. Fraud I actually haven’t seen a clear case of, I guess because usually fraud investigations take a longer time, but I suspect there are a couple of fraud cases as well. I think, overall, it fits very well with the thesis, but that may just be confirmation bias on my part.
Is there any silver lining to all of this? As you’ve just said, everything that you’ve written about has been replicated with Covid-19, but is there heightened scrutiny so that the problems are caught faster too?
The reason that a lot of stuff doesn’t get caught is because no one really cares about the science of it. Covid is a double-edged sword — we really care about this science, but the reason that we care about it is because we have to live through a global pandemic, and I’d rather we didn’t have that. But I think it tells us that a lot of the other stuff [that’s not related to Covid-19] that’s being put out into the scientific literature, whether it’s biased or fraudulent or whatever, doesn’t garner enough attention, and no one really cares enough about it for it to get retracted quickly. Whereas with Covid, everyone’s reading every single study, it pops up on Twitter, and there’s a whole bunch of discussion around it.
The situation you paint is so dire, and one of the depressing recommendations you make in the book is that scientists should be more distrustful of their peers. Do you think that’s really necessary? Could that harm scientific collaboration, or do we have such a big problem right now that we have to take that drastic measure?
I think it’s like banks asking for a security check with a personal question when you call them up. They know that the vast majority of calls are going to be the actual person who’s asking about their bank account, but they still have to check because in some cases, someone’s going to try and commit fraud and steal the money. I think we have to have systems in place in science that check this stuff. And everyone has to agree that “Yes, I’m going to give you my data. I’ve got nothing to hide, so you can have it,” and that’s just the norm.
The Harvard researchers who published the Surgisphere data, they clearly didn’t even look at the data once before they published those papers [on hydroxychloroquine and blood pressure drugs]. They completely trusted this dodgy guy at Surgisphere, who, by the way, has got loads of image duplication in his PhD thesis, which implies that maybe he’s not the most reliable researcher. They just totally trusted him, and he gave them all this data.
My own personal theory is that this happened around the time that Donald Trump was coming out and saying hydroxychloroquine is really good, and I feel like they just wanted to get one over on Donald Trump and publish a paper that showed that he was wrong. I suspect that’s the reason — the results were too good to check. “Hydroxychloroquine, recommended by Donald Trump, is actually killing people! You’re more likely to die if you take this!” Any little bias like that could push you towards coming up with incorrect results. That’s a case where they should not have trusted their co-author as much as they did.
I want to talk about that a little more because there’s a lot to unpack with the Surgisphere hydroxychloroquine debacle. You talk about the political bias, but obviously it plays both ways: Hydroxychloroquine was pushed because of a political agenda, but in a lot of ways it was also debunked so vociferously because of the opposite political bias.
It should go without saying that, obviously, the political bias was pushing it in the first place. There was this really weird thing in the right-wing media where they were all saying, “Hydroxychloroquine, let’s try it! It’s great, it’s the new thing!” They had all these doctors on TV who were saying, “My father got treated with hydroxychloroquine, and it saved his life” and all this really weird and irresponsible stuff. So, yeah, absolutely, the other side of the political equation was pushing it just as hard in the opposite direction.
What do you think the agenda was of those doctors who were pushing hydroxychloroquine? Was it money? Prestige? Politics?
The guy who originally published the hydroxychloroquine study, Didier Raoult in France, I suspect with him it’s a personal fame motive. He’s maybe interested in getting lots of attention, which he did, and now he has all these people online who are shilling for him. I think, as we find in the case of lots of fraudsters, there’s a personal self-aggrandizement thing going on.
“I think there’s an unhealthy reciprocal relationship there where scientists hype things up because they know that that’s what the press wants.”
An interesting and unexpected facet of Covid-19 that relates to the book is the downfall of Stanford researcher John Ioannidis. He was such a stickler for thorough research and being skeptical, and in the book, he’s cited as an example of what science should be. And yet with Covid-19, it’s the complete opposite — he’s publishing faulty datasets and has taken a pretty political stance on the pandemic. I’m very curious about what your thoughts are on his pivot and where it has come from.
I wrote a piece on why you shouldn’t have heroes in science that was about him. It was shocking to see in someone who we all really respected. He’s the guy who started off this whole discussion with his 2005 paper [“Why Most Published Research Findings Are False”] and has been contributing all the way through. I mention him four or five times in the book because there are just really good papers that have this great quality of nailing the problem.
I feel like it’s contrarianism taken too far. He’s clearly a contrarian in that he stood up and said, “Most published research is false.” And I think he saw that both the political and scientific consensus, to the extent there was one, was that lockdowns were a good idea and that Covid was really dangerous, and he went looking for evidence in the opposite direction, even if that meant accepting low-quality studies that had massive sampling biases and so on.
I’m curious about the role the media plays in all of this. There’s a lot of criticism of science media for contributing to research hype and oversimplifying things, but on the other side, investigative journalism can reveal some of these fraudulent cases or questionable funding sources. Especially now when there’s so much attention on scientific research into Covid-19, do you think the media has a net positive or a net negative on the process of science?
I think there’s an unhealthy reciprocal relationship there where scientists hype things up because they know that that’s what the press wants, and then the press hypes stuff up because they can say a scientist said it. So you have this back-and-forth relationship, which I think is not good for science because it feeds back into how scientists write their papers in the first place, and it’s certainly not good for the public understanding of science.
You talk about the benefit of preprint papers in the book as a way to hold scientists accountable and add a layer of scrutiny to a paper before it’s published. What is the role of preprints during Covid-19? Is this how science should be done, where everything is issued as a preprint first, and maybe the results are rushed out, but there’s a lot of discussion among scientists around it? Or has this gone too far, especially with the media covering so many preprints and the public potentially not understanding the difference between them and a peer-reviewed study?
I think science is making much faster progress [thanks to preprint papers]; it’s just that when it escapes out into the world, people need to be aware that this is a working paper, this is not peer-reviewed. That’s the fundamental problem. So I think science journalists need to do a better job of communicating that, and scientists need to do a better job communicating that. And you really shouldn’t be press-releasing your preprint, which has happened several times in this pandemic.
Is there a concern that airing the dirty laundry about all the problems that exist in academic research can contribute to the distrust in science right now among the public?
I think the fact that these concerns exist and that people are anti-science only makes it more important that we all raise our standards in science so that those people cannot criticize the science we’ve got. So I think let’s get everything out in the open, be completely transparent all the way. We’ve nothing to hide. We’re saying, “Look, these are problems, and we’re trying to deal with them, and here are the ways we’re dealing with them.” And we have to actually deal with them; we have to get on board and make these changes. I think that’s the way we gain trust in science.
The interview has been edited for length and clarity.