My RedState colleague Betsy Vaughn wrote a detailed piece Thursday that talked about how French Dr. Didier Raoult, who had previously conducted a study that showed a 91% effectiveness rate in Wuhan coronavirus patients treated with hydroxychloroquine and azithromycin, alleged that “scientific misconduct” had taken place in the VA hydroxychloroquine study that came out late last week.
Dr. Raoult cited a number of issues with the VA study, including the fact that the hydroxychloroquine and hydroxychloroquine-ZPAK combo were given to patients who had already been intubated. That’s a pretty significant point against drawing definitive conclusions about hydroxychloroquine from the VA’s study.
The French doctor is not the only person who found serious issues with the media-touted study after doing a careful analysis. In fact, it doesn’t even take a doctor to read the fine print on the data to conclude that not only was the study flawed, but that the mainstream media – in their eager rush to slap down Trump’s promotion of the drug as a treatment for the Wuhan coronavirus – completely botched their reporting on it.
Former CNN producer Steve Krakauer reviewed the details of the study and posted this Twitter thread noting how badly the media got it wrong:
Since we’ve gone 24 hours and I haven’t seen an accurate report on the hydroxychloroquine VA study, guess I’ll do it.
27 patients out of 98 died with HC, 25 patients out of 113 with HC and AZ, 18 out of 158 without any. Feel comfortable drawing massive conclusions from this? 1/
— Steve Krakauer (@SteveKrak) April 22, 2020
Also most reports say there was no difference between groups in ventilator need.
However it found 7% who had both hydroxycholoroquine and azithromycin needed ventilation, while more than 14% who had none needed it – double the amount. 2/
— Steve Krakauer (@SteveKrak) April 22, 2020
Most important though in the “VA hydroxychloroquine study” – it was not peer reviewed, not randomized, and… not even a study! It was a "retrospective analysis of data.”
This is a data point – nothing to draw conclusions from, either way. Full report: https://t.co/TPZ0z74PVz
— Steve Krakauer (@SteveKrak) April 22, 2020
“It could actually be harming patients,” said @JakeTapper about hydroxychloroquine and the VA results. This is ridiculous. The patients in this “study” were not actually in a “study.” They were simply patients – at who knows what degree of their recovery.https://t.co/feSpqWkokm
— Steve Krakauer (@SteveKrak) April 22, 2020
Forget “randomization” – these patients could have been different ages, different health risks, at different points in COVID-19 recovery. This was NOT a clinical trial. Just a study of "retrospective” data. @drsanjaygupta told viewers none of this. https://t.co/TPZ0z74PVz
— Steve Krakauer (@SteveKrak) April 22, 2020
I read Krakauer’s tweets and then the newsletter he sent Thursday night going into further detail on it, and my mouth dropped open. So I went and looked at the study myself, and found that his claims were correct.
I found this statement from the study (starting on the bottom of page 13) to be especially important to note (bolded emphasis added):
Despite propensity score adjustment for a large number of relevant confounders, we cannot rule out the possibility of selection bias or residual confounding. Our study cohort comprised only men whose median age was over 65 years. Therefore, the results may not necessarily reflect outcomes in women or in younger hospitalized populations, nor can they be extrapolated to pediatric patients. Our findings may also be influenced by the demographic composition of patients in our cohort, the majority of whom were black. Disproportionately higher rates of Covid-19-related hospitalization among the black population have also been reported in the United States as a whole.
Wow! So in other words, this “retroactive” study doesn’t really tell us anything at all about how the drug would work on people from other demographics who are under 65 and may not be as sick as some of the patients whose cases were reviewed for this study.
The VA even admitted as much when asked about it:
Veterans Affairs’ Robert Wilkie on yesterday’s hydroxychloroquine study: "That’s an observational study. It’s not a clinical study … We know the drug has been working on middle-age and younger veterans. And the gov of NY was just in the Oval Office yesterday asking for more" pic.twitter.com/uMRx6Z22Yh
— Tom Elliott (@tomselliott) April 22, 2020
Other doctors, too, have weighed in and found the study lacking, drawing similar conclusions to Dr. Raoult and Krakauer:
Here’s what an infectious disease doctor tied to a major nyc hospital told me about this study:
“This is probably the worst of study designs: retrospective analysis, with no stated methods as to how patients were assigned one of three treatment arms.”
https://t.co/cEnnPlS5dq— Buck Sexton (@BuckSexton) April 22, 2020
Sanjay Gupta reviewing VA "study" on #hydroxychloroquine
Says no benefit of HCQ & 2X as many people died.
Fails to mention that it is not a randomized study & that the differences could be due to confounding (sicker people got HCQ).
We must do better in communicating science
— C. Michael Gibson MD (@CMichaelGibson) April 21, 2020
So with all that in mind, let’s go a prime example of a typical media hot take on the study:
Hydroxychloroquine, long touted by President Trump, is of no use treating Covid-19 and actually causes more deaths than standard care, per a large new study. https://t.co/eCpEH9DGS0
— Ben White (@morningmoneyben) April 21, 2020
Note the garbage headline from the Politico link he cited, which was courtesy of a very flawed AP report on the study.
Unbelievable. And they wonder why we don’t trust them?
