Scientific consensus has been wrong in the past, so it's worth exploring
There’s an old Carl Sagan line: “The fact that some geniuses were laughed at does not imply that all who are laughed at are geniuses. They laughed at Columbus, they laughed at Fulton, they laughed at the Wright brothers. But they also laughed at Bozo the Clown.”
The Establishment gets a lot of things wrong. In the pandemic, for instance, institutions were wrong about airborne transmission, and about masks. They were wrong about herd immunity. Lots of nerdy outsiders got things right when the Establishment got them wrong.
But, to ape Sagan, the fact that the Establishment believed some things that were wrong does not imply that all things that the Establishment believes are wrong. The world’s stupidest man may say the sun is shining, but that doesn’t mean it’s dark outside. The Establishment also thinks vaccines work and hydroxychloroquine doesn’t, and it turns out the Establishment is right about that.
The latest thing-the-Establishment-might-be-wrong-about is ivermectin, an antiparasitic drug used to kill various worms and flukes and such. Various prominent people, including Bret Weinstein, the evolutionary biologist, say that it has been shown to improve Covid outcomes and should be used; the US FDA, on the other hand, says it should not.
Now, the PRINCIPLE trial, a large-scale clinical study at Oxford University, is investigating it. If it turns out to be effective, it wouldn’t be the first time an off-label drug showed good results against Covid: remdesivir, dexamethasone and tocilizumab have all been shown by the RECOVERY trial to reduce patient mortality.
So is ivermectin another hydroxychloroquine, or another remdesivir? Will it save lives or waste time? My unconfident suspicion is that it won’t be effective — a meta-analysis regularly quoted in support of it that finds positive results seem to be affected by publication bias (here’s what I mean by that), while this one finds no effect. A scientist friend says this seems to be the best RCT so far carried out and it finds no result either. I suppose I ought to make an explicit forecast and say I think it’s 60% likely that PRINCIPLE will find ivermectin does not have a statistically significant improvement over placebo.
That said: I’m wary of the FDA’s stance as well, and here’s why. Back in early 2020 I might have said I was 60% confident that masks didn’t work. If I had said that, I hope I would have also said “that means there’s a 40% chance that they do work, and a 40% chance of a good outcome at low cost is probably worth taking”.
Ivermectin is, apparently, cheap, and very safe for everyone other than those on blood thinners. The FDA reports that some people have been self-prescribing massive doses and making themselves ill, but if doctors were allowed to prescribe it for hospitalised Covid patients, they wouldn’t do that. (Huge doses of a drug and small doses of the same drug have very different effects.)
If Ivermectin really is very safe, and if there’s a reasonable chance that it would have some non-trivial impact on survival, then doctors should be able to make the decision to use it at their discretion. Currently it can be prescribed off-label, in both the US and UK, but some sort of temporary authorisation for use in Covid seems reasonable.
The risk is that ivermectin becomes politicised: believing that it works marks you out as a particular kind of right-wing Covid sceptic or anti-vaxxer, and the “correct” view is that it doesn’t work. But as this Italian doctor writing to the BMJ says, medical practice, especially in a fast-moving situation like a pandemic, has to work on best guesses and clinical judgment as well as rigorous RCTs. When you have a drug that might work but might not, and you know that it’s safe and cheap, waiting for perfect data might cost lives.