When analyzing data from all the significant studies (that weren't retracted)
However, if you remove the Elgazzar paper from their model, and rerun it, the benefit goes from 62% to 52%, and largely loses its statistical significance. There’s no benefit seen whatsoever for people who have severe COVID-19, and the confidence intervals for people with mild and moderate disease become extremely wide.
Where does that leave us on the question of whether ivermectin works for COVID-19? Well, firstly, once you exclude Elgazzar from the research pool the current best evidence shows a fairly consistent lack of benefit. There are still one or two small, very positive trials, but in general ivermectin does not appear to reduce your risk of death from COVID-19.
"I've been talking about ivermectin a bit recently, and every time I mention it someone will link me to this odd website - ivmmeta dot com
So, a bit of a review. I think this falls pretty solidly into the category of pseudoscience"
---
"The semi-anonymous site claims to be a "real-time meta analysis" of all published studies on ivermectin, collating an impressive 60 pieces of research
It's flashy, well-designed, and at face value appears very legitimate"
---
"The benefits that this website show for ivermectin are pretty amazing - 96%(!) lower mortality based on 10,797 patients worth of data is quite astonishing. Sounds like we should all be using ivermectin!
Except, well, these numbers are totally meaningless"
---
"Digging into the site, you're immediately hit with this error. That's not how p-values work at all, any stats textbook will show you why this statement is entirely untrue"
---
"Most of these dotpoints are wrong in some way (heterogeneity causing an underestimate is particularly hilarious) but this statement about CoIs is wild considering that there are several potentially fraudulent studies in the IVM literature"
---
"Going back to the heterogeneity point, this is the explanation from the authors about why heterogeneity is not a problem in their analysis. They appear to have entirely misunderstood what heterogeneity is (hint: this is more about BIAS than heterogeneity)"
---
"Also worth noting, I've previously shown the heterogeneity is high in meta-analysis of IVM for COVID-19 mortality, and that's almost entirely because there are 2 studies that show a massive benefit and a bunch of studies that show no benefit at all"
---
"Anyway, back to the website - the authors then present this forest plot of effect estimates
Each dot is a point estimate, and the lines around the dots represent confidence intervals"
---
"Now, any data thug will immediately notice something wildly improbable about this forest plot (H/T @jamesheathers
)
Can you see the issue? "
---
"While you have a think, here's a graph I made replicating these results. Not very pretty, but the final result is the same (with some minor rounding differences)"
---
"Ok, so back to the question - why does this look problematic?
It comes down to confidence intervals. When you've got a bunch of very wide confidence intervals from different studies, you expect the point estimates to move around inside them quite a bit"
If these are legit, I mean, if there are so many blatant errors in Ivermectin research, for example, things like
>"Digging into the site, you're immediately hit with this error. That's not how p-values work at all, any stats textbook will show you why this statement is entirely untrue"
>This interview left me with a variety of thoughts which I haven’t managed to collate – but here are some impressions. Dr. Garegnani didn’t seem anywhere near as well informed on the subject as Kory – and although we shouldn’t judge a scientific discussion as a boxing match, in my opinion Garegnani seemed to be completely outclassed and out of his depth. More significant though was a deeper difference between the two which I find difficult to describe – I’m tempted to call it a generation gap, or a mismatch of values and priorities. Garegnani seemed to view the topic purely as a question of theory and methodology (although to be fair that may be his speciality), and appeared unmoved even after numerous compelling points made by Kory. In contrast, Kory appears much more pragmatic, much more experienced and well-informed, and more passionate about saving lives. Of course these differences do not necessarily prove one position to be correct, but I’ll let you judge this conversation for yourself.
So the question is, if there are so many errors as purported by these cherry-picked twitter allegations, what are people not putting them forward in an actual debate..
Anything that falls in line with mainstream narrative gets boosted.
If you follow the links of doctors who're seeing first hand difference between Ivermectin use and not, including to meta-analysis and more recent A/B testing of Ivermectin, the conclusion is opposite to these articles cherrypicking which studies to include.
This article isn't saying it doesn't work. It is saying we do not yet have enough evidence to use it outside of a trial setting.
The medical field requires a high bar of evidence before allowing a treatment for general use. This is what they mean by "first, do no harm".
From your linked article:
> Several other repurposed medications have shown promise in early smaller trials for example sofosbuvir/daclatasvir, colchicine and remdesivir but the benefit was not seen later in larger trials. This meta-analysis of 24 RCTs in 3328 patients showed a 56% improvement in survival, faster time to clinical recovery and signs of a dose-dependent effect of viral clearance for patients given ivermectin versus control treatment. This benefit needs to be validated in larger confirmatory trials.
The thing I don't understand is some people's implied or sometimes explicitly expressed belief that medical researchers are suppressing promising COVID treatments for some reason.
Clinical trial research is perhaps the most complex and challenging type of research to get right. It's necessary to be critical of clinical research and meta-analyses, so pointing out serious research flaws doesn't mean someone is rooting against a safe, effective, cheap COVID treatment. The VAST majority of candidate drugs don't work well even for the intended indication, let alone have serendipitous effect in a different disease.
>Clinical trial research is perhaps the most complex and challenging type of research to get right
Cool, now compare the length, breadth, and historical use of ivermectin, hcq, etc, and experimental never-before-used-on-humans-en-masse mRNA injections.
16 comments
[ 546 ms ] story [ 81.8 ms ] threadMonash study, effect in vitro : https://doi.org/10.1016/j.antiviral.2020.104787
Pasteur Institute study, effect in guinea-pigs : https://doi.org/10.15252/emmm.202114122
https://www.nature.com/articles/d41586-021-02081-w
And on the subject of IVMMeta
https://twitter.com/GidMK/status/1422044335076306947
"I've been talking about ivermectin a bit recently, and every time I mention it someone will link me to this odd website - ivmmeta dot com
So, a bit of a review. I think this falls pretty solidly into the category of pseudoscience"
---
"The semi-anonymous site claims to be a "real-time meta analysis" of all published studies on ivermectin, collating an impressive 60 pieces of research
It's flashy, well-designed, and at face value appears very legitimate"
---
"The benefits that this website show for ivermectin are pretty amazing - 96%(!) lower mortality based on 10,797 patients worth of data is quite astonishing. Sounds like we should all be using ivermectin!
Except, well, these numbers are totally meaningless"
---
"Digging into the site, you're immediately hit with this error. That's not how p-values work at all, any stats textbook will show you why this statement is entirely untrue"
---
"Most of these dotpoints are wrong in some way (heterogeneity causing an underestimate is particularly hilarious) but this statement about CoIs is wild considering that there are several potentially fraudulent studies in the IVM literature"
---
"Going back to the heterogeneity point, this is the explanation from the authors about why heterogeneity is not a problem in their analysis. They appear to have entirely misunderstood what heterogeneity is (hint: this is more about BIAS than heterogeneity)"
---
"Also worth noting, I've previously shown the heterogeneity is high in meta-analysis of IVM for COVID-19 mortality, and that's almost entirely because there are 2 studies that show a massive benefit and a bunch of studies that show no benefit at all"
---
"Anyway, back to the website - the authors then present this forest plot of effect estimates
Each dot is a point estimate, and the lines around the dots represent confidence intervals"
---
"Now, any data thug will immediately notice something wildly improbable about this forest plot (H/T @jamesheathers )
Can you see the issue? "
---
"While you have a think, here's a graph I made replicating these results. Not very pretty, but the final result is the same (with some minor rounding differences)"
---
"Ok, so back to the question - why does this look problematic?
It comes down to confidence intervals. When you've got a bunch of very wide confidence intervals from different studies, you expect the point estimates to move around inside them quite a bit"
---
"Instead, look at thos...
>"Digging into the site, you're immediately hit with this error. That's not how p-values work at all, any stats textbook will show you why this statement is entirely untrue"
then I cannot explain this observation in this debate with the author of https://ebm.bmj.com/content/early/2021/05/26/bmjebm-2021-111...
>This interview left me with a variety of thoughts which I haven’t managed to collate – but here are some impressions. Dr. Garegnani didn’t seem anywhere near as well informed on the subject as Kory – and although we shouldn’t judge a scientific discussion as a boxing match, in my opinion Garegnani seemed to be completely outclassed and out of his depth. More significant though was a deeper difference between the two which I find difficult to describe – I’m tempted to call it a generation gap, or a mismatch of values and priorities. Garegnani seemed to view the topic purely as a question of theory and methodology (although to be fair that may be his speciality), and appeared unmoved even after numerous compelling points made by Kory. In contrast, Kory appears much more pragmatic, much more experienced and well-informed, and more passionate about saving lives. Of course these differences do not necessarily prove one position to be correct, but I’ll let you judge this conversation for yourself.
https://darryllrbetts.wordpress.com/2021/07/19/dr-pierre-kor...
The actual debate..
https://www.youtube.com/watch?v=DCpiHHG0R2k
So the question is, if there are so many errors as purported by these cherry-picked twitter allegations, what are people not putting them forward in an actual debate..
The same kind of thing that made some people realize that lying about masks early on in the pandemic was an irreversible mistake and breach of trust.
c19ivermectin.com
"Database of all ivermectin COVID-19 studies. 104 studies, 67 peer reviewed, 60 with results comparing treatment and control groups."
https://www.covid19treatmentguidelines.nih.gov/tables/table-...
If you follow the links of doctors who're seeing first hand difference between Ivermectin use and not, including to meta-analysis and more recent A/B testing of Ivermectin, the conclusion is opposite to these articles cherrypicking which studies to include.
E.g. https://covid19criticalcare.com/wp-content/uploads/2021/07/A...
The medical field requires a high bar of evidence before allowing a treatment for general use. This is what they mean by "first, do no harm".
From your linked article:
> Several other repurposed medications have shown promise in early smaller trials for example sofosbuvir/daclatasvir, colchicine and remdesivir but the benefit was not seen later in larger trials. This meta-analysis of 24 RCTs in 3328 patients showed a 56% improvement in survival, faster time to clinical recovery and signs of a dose-dependent effect of viral clearance for patients given ivermectin versus control treatment. This benefit needs to be validated in larger confirmatory trials.
Clinical trial research is perhaps the most complex and challenging type of research to get right. It's necessary to be critical of clinical research and meta-analyses, so pointing out serious research flaws doesn't mean someone is rooting against a safe, effective, cheap COVID treatment. The VAST majority of candidate drugs don't work well even for the intended indication, let alone have serendipitous effect in a different disease.
Cool, now compare the length, breadth, and historical use of ivermectin, hcq, etc, and experimental never-before-used-on-humans-en-masse mRNA injections.
Is the science so settled that we want to shut down any further discussion? I really don't understand