Read on Twitter
My deep dive into “The Only #Hydroxychloroquine Study You Need to Read” by @AmeshAA
You can also consider this a deep dive into HCQ RCTs
Long thread coming - I’m not numbering so keep reading https://leapsmag.com/the-only-hydroxychloroquine-story-you-need-to-read/">https://leapsmag.com/the-only-...
You can also consider this a deep dive into HCQ RCTs
Long thread coming - I’m not numbering so keep reading https://leapsmag.com/the-only-hydroxychloroquine-story-you-need-to-read/">https://leapsmag.com/the-only-...
First I would like to establish something important
Most accounts of HCQ working are with early treatment
CDC says that the virus is dead a week after symptoms arise - dead viral shedding
Hence, if HCQ works as an anti-viral then it needs to be given during the viral phase https://twitter.com/gummibear737/status/1287094571445493760">https://twitter.com/gummibear...
Most accounts of HCQ working are with early treatment
CDC says that the virus is dead a week after symptoms arise - dead viral shedding
Hence, if HCQ works as an anti-viral then it needs to be given during the viral phase https://twitter.com/gummibear737/status/1287094571445493760">https://twitter.com/gummibear...
Now to the article
First he establishes the net superiority of RCTs over Observational Studies
Fair point - I agree
First he establishes the net superiority of RCTs over Observational Studies
Fair point - I agree
Moving on, he tries to show the problems with Observational studies by pointing out the Henry Ford Study - 50% reduction in death
Note:
Epidemiologist Samia Arshad: “We attribute our findings that differ from other studies to EARLY treatment”
But what about the steroids?
Note:
Epidemiologist Samia Arshad: “We attribute our findings that differ from other studies to EARLY treatment”
But what about the steroids?
Well it turns out that the study included a propensity matching analysis
Meaning that they selected 190 patients from each group to eliminate the discrepancy in corticosteroid treatment - same positive result
I go into detail about it in this thread https://twitter.com/gummibear737/status/1290685648266027010">https://twitter.com/gummibear...
Meaning that they selected 190 patients from each group to eliminate the discrepancy in corticosteroid treatment - same positive result
I go into detail about it in this thread https://twitter.com/gummibear737/status/1290685648266027010">https://twitter.com/gummibear...
Next he cites two observational studies from New York
I covered the NEJM study (which was problematic) in the linked tweet - also late treatment
And the second study was again for late treatment
But...
https://twitter.com/gummibear737/status/1283840201241047044?s=21">https://twitter.com/gummibear...
I covered the NEJM study (which was problematic) in the linked tweet - also late treatment
And the second study was again for late treatment
But...
https://twitter.com/gummibear737/status/1283840201241047044?s=21">https://twitter.com/gummibear...
If we’re looking at observational studies in NY, why not also include this one?
Study of 6500 patients - more than double the size of of the others 2 studies combined!
Result? Same as Henry Ford Study
Decrease in mortality by 50% with HCQ https://link.springer.com/article/10.1007/s11606-020-05983-z">https://link.springer.com/article/1...
Study of 6500 patients - more than double the size of of the others 2 studies combined!
Result? Same as Henry Ford Study
Decrease in mortality by 50% with HCQ https://link.springer.com/article/10.1007/s11606-020-05983-z">https://link.springer.com/article/1...
Then he goes with “the most important of these studies”
1) the subjects were late stage very sick patients and there was very high mortality in both groups
2) they gave 2400mg in first 24 hrs and 800mg every day after - this is a dangerously high dose - 4000mg is lethal dose
1) the subjects were late stage very sick patients and there was very high mortality in both groups
2) they gave 2400mg in first 24 hrs and 800mg every day after - this is a dangerously high dose - 4000mg is lethal dose
Next up two Chinese RCTs
First study: very late treatment (16.6 days from onset) and they used viral shedding as a measure. It has been established that after 7-10 day u are shedding dead virus
Second study had only 30 mild patients - none died
But...
First study: very late treatment (16.6 days from onset) and they used viral shedding as a measure. It has been established that after 7-10 day u are shedding dead virus
Second study had only 30 mild patients - none died
But...
Since we’re looking at Chinese RCTs, why not mention the positive Chinese RCT?
62 patients
13% progressed to severe disease in control
0% progressed w/HCQ
16% improvement in pneumonia in control
61% improvement in pneumonia with HCQ
A positive RCT! https://www.medrxiv.org/content/10.1101/2020.03.22.20040758v3">https://www.medrxiv.org/content/1...
62 patients
13% progressed to severe disease in control
0% progressed w/HCQ
16% improvement in pneumonia in control
61% improvement in pneumonia with HCQ
A positive RCT! https://www.medrxiv.org/content/10.1101/2020.03.22.20040758v3">https://www.medrxiv.org/content/1...
Next up is Dr Boulware’s first Minnesota RCT
Methodology very messy
But the kicker is that despite 49%, 29%, 16% improvement if given on day 2, 3, 4 respectively - they concluded that there was no positive effect
I cover it in this thread - worth a look - much more info https://twitter.com/gummibear737/status/1285285305730437126">https://twitter.com/gummibear...
Methodology very messy
But the kicker is that despite 49%, 29%, 16% improvement if given on day 2, 3, 4 respectively - they concluded that there was no positive effect
I cover it in this thread - worth a look - much more info https://twitter.com/gummibear737/status/1285285305730437126">https://twitter.com/gummibear...
This paper reanalyzes the Boulware data:
“We conclude their RCT presents statistical evidence, at 99% confidence level, that the treatment of Covid-19 patients with HCQ is effective in reducing the appearance of symptoms”
Wow, a second positive HCQ RCT!
https://arxiv.org/abs/2007.09477 ">https://arxiv.org/abs/2007....
“We conclude their RCT presents statistical evidence, at 99% confidence level, that the treatment of Covid-19 patients with HCQ is effective in reducing the appearance of symptoms”
Wow, a second positive HCQ RCT!
https://arxiv.org/abs/2007.09477 ">https://arxiv.org/abs/2007....
Next up is the Cavalcanti RCT
Its a later stage RCT - median time since symptoms was 7 days
“it is conceivable that interventions that may limit viral replication (e.g. HCQ) may be more effective earlier in the course of the disease.”
Not blinded
Authors admit limitations
Its a later stage RCT - median time since symptoms was 7 days
“it is conceivable that interventions that may limit viral replication (e.g. HCQ) may be more effective earlier in the course of the disease.”
Not blinded
Authors admit limitations
But then the author inexplicably links to the retracted Lancet - Surgisphere article
Why on earth he would touch on the work of Sapan S Desai is beyond my comprehension
NYT Article on Sapan https://www.nytimes.com/2020/07/27/science/coronavirus-retracted-studies-data.html
https://www.nytimes.com/2020/07/2... href=" https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)31180-6/fulltext">https://www.thelancet.com/journals/...
Why on earth he would touch on the work of Sapan S Desai is beyond my comprehension
NYT Article on Sapan https://www.nytimes.com/2020/07/27/science/coronavirus-retracted-studies-data.html
Then, links to Gilead press release - HCQ makes remdesivir less effective
Given how Rmdvr’s clinical trial evolved (retracted WHO report saying it didn’t work) and their mediocre results (less recovery time, no effect on mortality), consider me skeptical
https://www.gilead.com/news-and-press/press-room/press-releases/2020/7/gilead-presents-additional-data-on-investigational-antiviral-remdesivir-for-the-treatment-of-covid-19">https://www.gilead.com/news-and-...
Given how Rmdvr’s clinical trial evolved (retracted WHO report saying it didn’t work) and their mediocre results (less recovery time, no effect on mortality), consider me skeptical
https://www.gilead.com/news-and-press/press-room/press-releases/2020/7/gilead-presents-additional-data-on-investigational-antiviral-remdesivir-for-the-treatment-of-covid-19">https://www.gilead.com/news-and-...
Another Boulware study
I covered this in my thread before
But interesting to note that hospitalization was half for HCQ arm but not statistically significant because cohort too small
In his infographic he rounded 1.886% up to 2 and 3.791% was rounded down to three
I covered this in my thread before
But interesting to note that hospitalization was half for HCQ arm but not statistically significant because cohort too small
In his infographic he rounded 1.886% up to 2 and 3.791% was rounded down to three
That’s it for evidence
He argues that HCQ proponents are moving the goal posts by requiring early treatment
But Dr Raoult, Zelenko are on record since March saying that early treatment is essential
And the best study to test this was cancelled by NIH
https://twitter.com/gummibear737/status/1289263452574355459?s=21">https://twitter.com/gummibear...
He argues that HCQ proponents are moving the goal posts by requiring early treatment
But Dr Raoult, Zelenko are on record since March saying that early treatment is essential
And the best study to test this was cancelled by NIH
https://twitter.com/gummibear737/status/1289263452574355459?s=21">https://twitter.com/gummibear...
What is missing from this article are the 45+ studies which show a benefit to HCQ
The author argues that you need RCTs to prove and he’s right
But you can’t parade a few flawed RCTs as proof that HCQ doesn’t work
And I’ve shown that 2 RCTs are positive https://c19study.com"> https://c19study.com
The author argues that you need RCTs to prove and he’s right
But you can’t parade a few flawed RCTs as proof that HCQ doesn’t work
And I’ve shown that 2 RCTs are positive https://c19study.com"> https://c19study.com
Finally he argues risk/benefit
Problem is that if you treat with HCQ and it turns out not to work then you have done a little harm
But if you don’t treat and it turns out to work then you have let alot of people die
Problem is that if you treat with HCQ and it turns out not to work then you have done a little harm
But if you don’t treat and it turns out to work then you have let alot of people die
I am not going to try and convince anybody that HCQ works
What I will do is continue to show that there is evidence that it may work
The idea that the science is settled is very strange because it obviously isn’t
What I will do is continue to show that there is evidence that it may work
The idea that the science is settled is very strange because it obviously isn’t
In this context, people should have the right to use HCQ
Inform them of the risks but leave the decision to the patient/doctor
Can you imagine what will happen if HCQ is proven to work?
What if tens of thousands could have lived but didn’t because of this partisan nonsense?
Inform them of the risks but leave the decision to the patient/doctor
Can you imagine what will happen if HCQ is proven to work?
What if tens of thousands could have lived but didn’t because of this partisan nonsense?
Regarding the article, I have no animus toward @AmeshAA and I’m not going to question his motives
The conclusions of his article are the result of a very superficial analysis of cherry-picked data
Its easy for somebody that has actually looked at all the data to prove him wrong
The conclusions of his article are the result of a very superficial analysis of cherry-picked data
Its easy for somebody that has actually looked at all the data to prove him wrong
Again, I don’t think it is bad faith but rather the result of living in an echo chamber where outside ideas/opinions are rejected outright
The left has already left the HCQ conversation and they consider it a waste of their time to actually study the data objectively
The left has already left the HCQ conversation and they consider it a waste of their time to actually study the data objectively
