Thread by @GidMK

Of all the terrible defenses of ivermectin fraud, I think the "but 20% of all medical research is said to be fraudulent!" angle is one of the weirdest

Firstly, it doesn't matter at all. If every house on your street is one fire, pointing at the other houses and yelling "they're on fire too!" doesn't extinguish the flames eating away at your wedding photos

Secondly, the 20% figure is a very extreme estimate. Now, I thoroughly respect Prof Ben Mol, who is the one who made that assessment, but I'm not sure I agree that the average rate is actually that high blogs.bmj.com/bmj/2021/07/05…

If we look at anonymous surveys of researchers, the rate of serious misconduct - plagiarism, outright fakery etc - is estimated to be somewhere between 3-14% journals.plos.org/plosone/articl…

It also varies widely by the area of research. Pharmaceutical companies, for all their other evils, likely almost never fake research because they are bound within a tight web of legislation

The cost to a pharma company that faked a single study could probably be measured in the billions, if not more. They do all sorts of dodgy shit, but outright fraud is, I think, extraordinarily rare

Conversely, there are areas of research where fraud is probably rampant, particularly where research is less well overseen by regulators

So while I agree that there is much more fraud than most people imagine, I don't know if I'd agree that 1 in 5 studies is fake. I'd guess it's more like 1 in 10 overall, depending on the area we're looking at

Anyway, the point is that if you care whether ivermectin works or not, it genuinely doesn't matter whether other people have committed fraud. The question is whether the fraudulent research in the ivermectin space has impacted our conclusions, and it clearly has 🤷‍♂️