This is a long and very thorough review of the studies listed on ivmmeta.com.
https://astralcodexten.substack.com/p/ivermectin-much-more-than-you-wanted
It's very good, and very fair -- much more fair than I have been, which is a good critique of me.
It also has a surprising potential confounder explanation for Ivermectin success disparity, as well as some very insightful commentary on science, sociology, and politics - including speculating why the consternation exists over it, in particular the correlation between anti-vaccine beliefs and pro-ivermectin beliefs.
You really should read the whole thing - it's worth it - but if you're going to skip, when he says scroll down to the section called "The Analysis" do it, but read the rest.
Here's the final four bullet points along with the author's assigned confidence (this is a rationalist thing - you attempt to gauge your personal confidence in a belief, rather than live in a world of binaries - not a statistical number)
https://astralcodexten.substack.com/p/ivermectin-much-more-than-you-wanted
It's very good, and very fair -- much more fair than I have been, which is a good critique of me.
It also has a surprising potential confounder explanation for Ivermectin success disparity, as well as some very insightful commentary on science, sociology, and politics - including speculating why the consternation exists over it, in particular the correlation between anti-vaccine beliefs and pro-ivermectin beliefs.
You really should read the whole thing - it's worth it - but if you're going to skip, when he says scroll down to the section called "The Analysis" do it, but read the rest.
Here's the final four bullet points along with the author's assigned confidence (this is a rationalist thing - you attempt to gauge your personal confidence in a belief, rather than live in a world of binaries - not a statistical number)
- Ivermectin doesn't reduce mortality in COVID a significant amount (let's say d > 0.3) in the absence of comorbid parasites: 85-90% confidence
- Parasitic worms are a significant confounder in some ivermectin studies, such that they made them get a positive result even when honest and methodologically sound: 50% confidence
- Fraud and data processing errors are of similar magnitude to p-hacking and methodological problems in explaining bad studies (95% confidence interval for fraud: between >1% and 5% as important as methodological problems; 95% confidence interval for data processing errors: between 5% and 100% as important)
- Probably "Trust Science" is not the right way to reach proponents of pseudoscientific medicine: ???% confidence