This is a great illustration of how dangerous our new #doyourownresearch population of Facebook doctors and Google phds truly is.
Even some of the smart folk will latch onto “the median person taking horse paste got better two days faster” and just sort of trail off during the part where it wasn’t significantly significant and was highly probably just happenstance.
And now we got millions of people deworming, so I think it’s ok to be pretty darn precise vis a vis evidence based medicine.
I guess the idea of getting plague rats back on their feet fast enough to go around spreading COVID sounds good to the Emily Oster slay qween (literally!) crowd.
i’ve skipped 95% of this dumb derail but it seemed to me that the point chrisv was trying to make was “if somehow it turns out that horse paste actually does have some benefit as a covid therapy we’re never going to hear the end of what a fucking genius trump is and how he personally discovered this in the lab” etc. and this will lead to even more outcomes that are extremely bad as a bunch of shitbags get credibility and their new shitbag theories are given some sort of credence which will lead to additional bad outcomes down the road
this is a semantic game indeed. since the observation and the real effects are different things. they could have been observing a random positive outcome, rather than observing the real effect. since the size of the study turned out to be too small for such a small statistical improvement, it just gives no meaningful additional confidence about treating covid.
we can likely rule out that ivermectin is a miracle cure, or even a solid treatment, although it does appear to be safe enough to take even without a positive covid effect. but we could guess that about any drug before any studies with 99% success rate.
chris originally mentioned that ivermectin is a legit parasite treatment, and we shouldn’t badmouth it, so there’s an argument that because of news coverage fewer people will take it for parasites thinking it’s not for humans. but the act of continuing to push ivermectin for covid is still like selling snake oil which probably won’t kill. so much better than bleach for covid obviously, but not any better than hcq, which was used very widely (possibly too widely) before concluding it didn’t work. it was desperate times and no other options were available, i would probably try anything at that point as well.
I don’t know if “semantic” is correct or “game” is correct, but indeed he was trying to make a point that wasn’t anything about whether or not ivermectin is an effective treatment for covid. Are all indirect points in a discussion games?
eta: Actually I think “yes” is a fair answer there because foruming is basically a game, but it doesn’t mean bob’s point was wrong/bad/out of place.
This is not semantics. This is me repeatedly trying to explain the basics of evidence based medicine after doing things like developing and implementing a longitudinal curriculum for residency. While I’m admittedly not an expert, you don’t need to be one to correct these basic misunderstandings.
Sigh. Terms have specific meanings, particularly in technical fields. When lay people try to assign their own common language definitions it almost always goes astray. We went down this road with herd immunity.
Apologies but in a semantics debate on technical terms, the specific scientific definition wins.
I run afoul with legal terms for this reason all the time.
It’s more than that though. ITT you have people taking that misunderstanding of terms to criticize professionals using the terms properly. (I don’t me by that, I’m talking about the criticism of the general medical community for saying that ivermectin doesn’t work. Which I’m still kind of shocked was seriously done here)
Interesting that you bring up the law; I was just thinking that the presumption of innocence for a criminal suspect is a lot like the presumption of failure for a prospective covid treatment.
The difference is that overcoming reasonable doubt can be subjective, but overcoming statistical doubt should be transparent and purely objective.