Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

nature.com

Post reply on HN
Print view View original

131–140 of 493 posts

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#131

Post by h2odragon »

shadowgovt wrote: post #123
h2odragon wrote: post #108

Earlier quoted context omitted.

> (I would never consume a medicine made for animals) Why not? I either pay $150+ all up for a doctor visit, prescription, etc for antibiotics, or $20 for a bottle of "fish antibiotics" at the feed store. Either way I get amoxicillin; but from the feed store I get 3x the amount and in smaller capsules which makes adjusting dose easier. We stock the "horse paste" ivermectin for our half dozen hounds, I've been known t…

Fish antibiotics might be fine, but since they aren't regulated for human consumption it's more of a gamble. Plenty of chemicals that don't hurt fish that the manufacturer therefore won't be testing for that can be harmful to a human. But more importantly in the case of antibiotics: how do you know amoxicillin is the right one for what ails you? There are multiple antibiotics for multiple bacteria, and applying the w…

> how do you know amoxicillin is the right one for what ails you?

In our case, that's what the doctor prescribes for my wife's chronic sinus infections whenever she asks one about them, and I've been stocking it as a goto first antibiotic for veterinary use since i started caring for my animals 30yr ago.

In anyone else's case, there's no great barrier to educating yourself about what you put in your body and why, and taking responsibility for your own health. Doctors are consultants, not priests.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#132

Post by brandonmenc »

cma wrote: post #106
brandonmenc wrote: post #54

Earlier quoted context omitted.

> People are literally taking horse formulations of ivermectin People are taking the horse formulation because most pharmacists won't even fill an off-label prescription for the drug, which - even if it has no effect on COVID - is safe and taken by a quarter billion people every year. Let people take it. It's not harmful. Or, make a big huge deal about people taking it and, well, then it becomes a big huge deal.

> is safe and taken by a quarter billion people every year. Usually yearly or bi-yearly. Not bi-weekly. There is much less safety data on doses that frequent.

> There is much less safety data on doses that frequent.

You can say the same thing about COVID boosters.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#133

Post by acdha »

gonational wrote: post #85
acdha wrote: post #53

Earlier quoted context omitted.

> I hope the more science-literate proponents of IVM If they valued and were capable of reasoning about the science, they would not be “proponents” — they'd recognize the current scientific understanding (no known mechanism or benefits, significant side-effects) and wait for a solid study before recommending anything. The people you do see advocating it are almost always doing so for political or financial purposes a…

Yeah, because nothing says "recognize the current scientific understanding" like dismissing hundreds of studies because of a couple bunk meta-analysis. Also, nothing says "recognize the current scientific understanding" like injecting an untested medicine into your veins to avoid what amounts to the common cold for younger people; while ignoring the fact that it does not help prevent the spread, and according to the…

You're packing enough untrue things into that comment that it's hard to believe you're acting in good faith but:

* It's dismissing the studies which show effects due to errors, which leaves the studies not showing beneficial effects. Science is built on evidence and thus far it hasn't been shown that Ivermectin has a benefit for COVID.

* Similarly, it's untrue to the point of being a flagrant lie to claim that the vaccines are untested when they went through full clinical trials and the high positive effects and safety rates shown in those trials have been backed up by data from many millions of people

* Similarly, it's flat out wrong to claim that COVID is similar to the common cold — even if you compare it to influenza, which is far more serious, that's not true — we've seen more children die in the last couple months than we typically see in an entire flu season.

* Vaccination does help reduce the spread, and that's even shown in the Israeli data — it's not 100% but if you look at the data it's very clear that vaccinated people at a significantly lower rate. Israeli comparisons are tricky because they have substantially different vaccination rates across age categories and a vaccinated 80 year old is still more at risk than someone younger, especially when those younger people are not taking precautions, which brings me to:

* Israel is nowhere near having the highest vaccine rate — that's a tie between Portugal, Malta, and the U.A.E. currently — and at a paltry 62% they're not even in the top 25. That's only slightly better than the United States at 55%.

* The Iraeli data does not show that vaccination increases spread: this is a lie spread by people who are comparing unlike things in the knowledge that many people will reshare those claims without checking them.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#134

Post by avhon1 »

scoopertrooper wrote: post #41
ashtonkem wrote: post #26

The other takeaway is that preprint publication is a pathway ripe for abuse. The major paper that drove most of this situation, and the meta analysis, appears to have been a straight up fraud. But it got circulated a lot as a preprint to people who didn’t understand the difference between a preprint and a peer reviewed paper. If the preprint had been more private, this might have been averted.

What's the motivation for this sort of thing? It seems the very thing they crave (attention) will invariably bring them undone as attention invites scrutiny.

For one, money. There is an organized group that charges $90 online for hocus-pocus covid medication prescriptions [1]. The more people that can be convinced that this is a worthwhile purchase, the more money they get.

https://americasfrontlinedoctors.org/treatments/how-do-i-get...

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#135

Post by sgt101 »

brandonmenc wrote: post #119
sgt101 wrote: post #104

Earlier quoted context omitted.

I get where you are coming from, but I think that there is a nuance that you have missed. Like many, many drugs - and vaccines - ivemectin has some unpleasant and dangerous side effects in some people rarely. You can see these if you look the drug up : https://www.medicalnewstoday.com/articles/ivermectin-oral-ta... for example. The point is that if this drug did help people make it through covid (like the vaccines do…

All of this is for the patient and their doctor to decide. Not the pharmacist. Not the FDA.

I agree about the pharmacist - but the drug regulator really does have to have a role. There is a serious information imbalance between patients and doctors and both big pharma and lunatic talkshow host millionaries. Regulation is important to help prevent junk getting pushed into kids arms to make bastards money. It's not perfect and they get a lot wrong, but without it you and I are at the mercy of some very mercy short people.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#136

Post by sgt101 »

rubatuga wrote: post #127
sgt101 wrote: post #109

Earlier quoted context omitted.

I would like to meet you in a casino. Type "Invermectin side effects" into any search engine, read the links. These are "bad effects". The fraudulent studies (y'know, including dead patients, patients that never existed, drugs that weren't administered and so on) have shown a good effect. The real studies unfortunately showed no effect. My advice to you; don't gamble. Ever.

You need a better argument than that to convince somebody to never gamble lol. Remember, it was also a gamble to trust the WHO not to use masks early in the pandemic. We are making gambles everyday.

>trust the WHO not to use masks early in the pandemic

That one was a very interesting story. The best informed people made rational decisions to mask up, semi informed believed the bad advice to do without, the least informed acted randomly.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#137

Post by tptacek »

rory wrote: post #57
tptacek wrote: post #28

Earlier quoted context omitted.

It matters a lot if ivermectin doesn't do anything to stop C19, even if it is otherwise safe (it surely is, dosed professionally). People take ivermectin instead of vaccinating or complying with NPIs. If ivermectin doesn't work, that is a very, very big deal.

> People take ivermectin instead of vaccinating or complying with NPIs Doing one thing but not another doesn't imply you're doing one thing instead of another. Do you truly believe the people ODing on ivermectin would be lining up for vaccinations if it didn't exist?

I don't know what this is supposed to mean, but a close family member took ivermectin instead of being vaccinated, and caught a serious case.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#138

Post by tzs »

hackingforfun wrote: post #99

With all the criticism of the mRNA and other COVID vaccines, which actually do have science backing them up, what is the science that would make ivermectin actually work? It's an antiparasitic drug. I see a lot of criticism of the COVID vaccines, especially from the right, but then they are fine taking hydroxychloroquine or ivermectin if there is any glimmer of hope that they will work. I just don't get it. Maybe, ju…

Here is an article from May 2020 about 15 drugs [1], including hydroxychloroquine and ivermectin, that were being looked into to see if they might help COVID. It goes into the mechanisms that led researchers to believe they were worth looking at.

[1] https://www.nature.com/articles/d41591-020-00019-9

Edit: Oops...forgot to link the article. Added.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#139

Post by rubatuga »

pjc50 wrote: post #126
mechE321 wrote: post #71

The Bayesian gambler in me wants to think this: The studies on ivermectin seem to be split between "good effect" and "no effect," and there don't seem to be any (by my extremely informal review! going off of memory here) in the camp of "bad effect." Seems reasonable to take ivermectin as a decent gamble to me while we wait on the dang science to get its head out of its butt.

Be very careful how much you take: https://www.npr.org/sections/coronavirus-live-updates/2021/0... Remember that the in vitro study that gets cited a lot used a concentration that would be lethal in humans.

Poison control center calls are not poisonings. There were no significant increases in hospitalizations due to ivermectin. It's a fear-mongering article.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#140

Post by eynsham »

sjwalter wrote: post #45
ashtonkem wrote: post #26

The other takeaway is that preprint publication is a pathway ripe for abuse. The major paper that drove most of this situation, and the meta analysis, appears to have been a straight up fraud. But it got circulated a lot as a preprint to people who didn’t understand the difference between a preprint and a peer reviewed paper. If the preprint had been more private, this might have been averted.

> The major paper that drove most of this situation, and the meta analysis, appears to have been a straight up fraud. Umm, wat?! The vast majority of the studies about Ivermectin have been observational, run by front-line clinicians. Prior to 2020, basically everyone on earth agreed (including the WHO, who STILL agrees) that front-line clinicians and observational studies are excellent signals that can lead to scient…

Consider dexamethasone.

> There is, no, no large-scale RCT for IVM. However, it is inherently obvious that none will happen, none that give it a fair shake.

Well, the gold standard RCT Recovery (https://www.recoverytrial.net/) had a look.

> The entire public health apparatus in the West has a huge desire to treat all illness with only on-patent, new medications.

Dexamethasone is a cheap steroid. (For that matter, vaccines are incredibly cheap and yet nobody seems to have stopped them in favour of, say, monoclonal antibodies.)

> Do you not thin that this incentive influences what gets into the news?

Evidently not terribly much, given that dexamethasone was at least in all the British newspapers.

> Do you think the public health authorities in Uttar Pradesh are lying?

Quite plausibly. UP is perhaps the worst governed state in India and has been under all political parties. The case of Kafeel Khan is rather illustrative.

More to the point, I don’t care whether they’re lying. The whole article is full of vague statements that are hardly a good basis to believe anything about ivermectin.

Post reply on HN