The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
How Many People Did NIH Director Francis Collins Kill?
He ordered Moderna to slow down vaccine trials to recruit more "people of color" for no valid scientific reason
ARONCZYK: People of color were disproportionately getting sick and dying from COVID. And at this point in time, Dr. Collins is overseeing the pharma companies that are in phase 3, that are recruiting tens of thousands of people to try out the shots. And he looks at who Moderna has been signing up.
COLLINS: They felt this enormous pressure to recruit quickly 'cause it's a public health crisis, and people are dying. And if you're trying to recruit quickly, you recruit the people who are most likely to say yes. And that tends to be white people, especially young, healthy white people.
ARONCZYK: Dr. Collins goes to Moderna's executive team, and he's like, what are you going to do about this? And he says that their response was less than satisfying.
COLLINS: I mean, it was hand-waving.
ARONCZYK: Right.
COLLINS: And this is where I got fairly directive.
(LAUGHTER)
COLLINS: And I made a little speech about, OK, if that's the strategy you're going to pursue, you may have a vaccine that turns out to be safe and effective for white people, but you will have failed, and we will not defend you.
HOROWITZ-GHAZI: So at this moment in the pandemic where it feels like every second matters, they pump the brakes. They don't want to go forward with the trials because, Dr. Collins says, if the people testing the vaccine don't represent the American public, the public won't trust the vaccine. Moderna then recruits more people of color.
ARONCZYK: So at this moment, you make this request to diversify the trials. What happens? Does it slow things down a little?
COLLINS: (Laughter) It, in fact, did have a modest effect of that sort.
ARONCZYK: But, Dr. Collins says, just by a week or two.
How many people died because Collins delayed Moderna's vaccine "just by a week or two?" I don't know, but how ever many it is, Dr. Collins bears responsibility for their deaths--a "request" from the director of NIH in this context is really a command. And this "modest effect" is certainly nothing to laugh about.
Let's be clear on several things: (1) There was (and is) no scientific reason to think that the Modern vaccine would act differently on people of different genetic backgrounds; (2) Even if there was reason to think it would, the categories the NIH requires researchers to use--African American, Asian American, Hispanic, Native American, and White--are extremely internally genetically diverse.* Asian Americans, for example, can be Austronesians, Caucasians, or East Asians, and there is much internal diversity within those subcategories. Hispanics can be any mixture of European, Indigenous, African, and Asian. And so on. There is no *scientific* reason to use these categories as proxies for genetic diversity; and (3) If Americans wouldn't "trust" a vaccine that didn't have "enough diversity," that's largely because government authorities like the NIH insist that vaccines aren't trustworthy unless they have been tested on a "diverse" population. If the NIH and other authorities consistently said that socially and legally constructed racial and ethnic categories are not scientific in nature and have no bearing on vaccine efficacy, then the public would be much more likely to believe it.
*When the federal government promulgated these categories, it warned that these "classifications should not be interpreted as being scientific or anthropological in nature." Yet NIH and FDA have adopted them anyway, for no good or even stated scientific reason. Directive No. 15, Race and Ethnic Standards for Federal Statistics and Administrative Reporting, 43 Fed. Reg. 19,260 (1978).
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
With Biden running on "there won't be a Trump vaccine, and if there is i wouldn't take it", there's no way that either Pfizer or Moderna were going to give Trump a pre-election bump by publicly announcing their vaccines "worked".
So Collins didn't kill anyone.
The heads of Pfizer and Moderna, who both announced within a week of the election, did kill people, by their decision to delay all announcements about results to when they could no longer help Trump.
I don't think a word of that is true.
Except maybe the part that "Collins didn't kill anyone." It's really odd to pin deaths on a single agency official.
The last six words of your comment are entirely extraneous.
You can live in whatever fantasy world you want. But very word of that is true
Trump did a good job dealing with Covid…then Republican governors went nuts without Trump around in 2021 to spank them into line and got a lot of Republicans killed. Deep State Republicans are back in charge and next chance Deep State Republicans get they will ship your job to China and send your sons to slaughter Muslims.
Wait, Greg. I thought the vaccine was useless and you could just take ivermectin or something.
Funny.
You can take HCQ, Zinc, and Z-pac, you can take Ivermectin, you can get monoclonal antibody treatment.
All of them appear to be far better than getting a "vaccine" that allows you to catch the disease, carry it asymptomatically while giving it to others, be sick with it, be hospitalized with it, and be killed with it.
But acknowledging treatments that work takes away fear, and then the Democrats wouldn't be able to abuse the "emergency". So everythign they push fails to make things better.
because makign things better would end their fun.
Note: If the vaccine wasn't crap, there would be no mask mandates for the "vaccinated"
I thought race was a social construct as was gender...
Race and gender are also social constructs.
Why are you so certain there's no scientific reason to think a vaccine might have any detectable efficacy or safety variation among people of different ethnicities? That might be true, but it seems odd for a law professor to voice such certainty when trying to pin deaths on someone.
Is this just accepted fact?
Guess what cancer has been cured?? The cancer that only kills whites…skin cancer!
This is very much debated in the pharma community, but the conclusion I have drawn is that it doesn't matter.
It is undeniable there are some differences. For example, Jews have higher rates of certain genetic diseases. But these differences almost never actually manifest themselves in clinical trials.
China has an interesting policy where you are only allowed to use clinical trials performed in China for releasing a medicine there ... the US sporadically flirts with doing the same, its really fucking stupid and kills people but whatever.
The point that the FDA doesn't understand here is that they don't just control medicine for the US. Other countries trust them. So if a medicine is approved in the US, most countries around the world thinks that, ok it works in Americans, therefore it works in Japanese, Koreans, Nigerians, Israelis, whatever.
It defies logic to think that medicine somehow works in white Americans, doesn't work in black Americans, but works in Africans. It's just idiotic. So if an African country thinks a trial of mostly white people is good enough for its citizens, if genetic differences do not matter in that case, it shouldn't matter here.
"It defies logic to think that medicine somehow works in white Americans, doesn't work in black Americans, but works in Africans."
Or rather...Africans dose it at higher levels, or don't dose it.
The best story here is that of Codeine and CYP2D6. Codeine is required to be metabolized by the human enzyme CYP2D6 in order to form morphine. But...people don't all have the same activity level of the enzyme, due to different genetics. Roughly speaking, you can be an "ultra-metabolizer," "normal," "intermediate metabolizer" or "poor metabolizer". Someone who is an ultra-metabolizer will need a lower dose. Someone who is an intermediate metabolizer will need a higher dose. And someone who is a poor metabolizer just shouldn't be given Codeine, it won't work.
Here's the key part. The distribution of these genetics isn't even. Asian Americans and African Americans tend more towards intermediate metabolizers, while Caucasian Americans tend towards normal metabolizers. And Middle Easterners have the highest proportion of being super-metabolizers.
https://www.ncbi.nlm.nih.gov/books/NBK100662/
What does that have to do with an mRNA vaccine? Has some study shown that a supposedly social construct ("race") determines how much mRNA is transcribed into protein?
The problem is you don't "know"-- since there are instances of variation in response it is necessary to determine whether that is the case here. You do that by making sure your sample represents the target population adequately.
The problem is holding it up when it could begin saving lives.
As this is an emergency, you could provisionally inject other races amd monitor closely.
But holding it up for all because a politician wants virtue signalling points, is murder.
This guy probably murdered several hundred people.
This is part of the larger mass deaths caused by FDA delays of years of lives that would otherwise have been saved, vs. a few extra saved because a dangerous drug got to market too fast (which would be realized as they track that, too. It just amounts to a large trial.)
Point is lives lost due to delays >> lives saved due to delays. But that doesn't work for politicians. A few sob stories in front of cameras vastly outweighs 100,000 dead in this country alone due to a heart treatment delayed 5 years.
"The problem is holding it up when it could begin saving lives."
Or it could cost lives. You don't really know.
Let's pretend, for example, there was a critical interaction between GPD6 deficiency and the mRNA vaccine that caused a dramatic immune overresponse in 1% of patients with the GDP6 deficiency.
12.2% of African American males have GDP6 deficiency. 0.3% of Caucasian males. Let's take your 30,000 patient trial. Assume it's all Caucasians. 15,000 males, 45 have the GDP6 deficiency. If it's only 1% of patients, it could be missed.
Instead, do a representative trial that includes 1500 African American males. Now you have 183 who have the GDP6 deficiency, and statistically you should see at least one or two patients with the dramatic immune overresponse.
Let's pretend 12.3% of asteroids in the asteroid belt are made of green cheese, because there's exactly as much reason to believe that as to believe that this vaccine has anything to do with G6PD (note spelling) deficiency.
How many space stations full of people would that feed, and for how long?
This is scientifically incoherent. "Asian Americans" includes South Asians. South Asians are Caucasians. "Asian Americans and African Americans tend more towards intermediate metabolizers, while Caucasian Americans tend towards normal metabolizers." So the government is requiring South Asians to be measured as Asians, even though ethnographically they are Caucasians.
"Middle Easterners have the highest proportion of being super-metabolizers." But Middle Easterners aren't identified as anything in these studies other than as whites.
So if these differences exist, they still don't track the categories researchers are required to use. And none of what you wrote has anything to do with MRNA vaccines.
OK...
Point 1: There are genetic variations among the human population. This is well understood
Point 2: These genetic variations can affect the safety and efficacy of various drugs. (For example CYP2D6 activity or G6PD deficiency.)
Point 3: These genetic variations are not evenly distributed among the population. They are more overrepresented in some racial populations, and underrepresented in other racial populations. (G6PD is a good example, being vastly more present in the African American male population, than among the Caucasian male population)
Point 4: For a safety and efficacy trial, you want your trial population to match your planned target population as closely as possible. If it doesn't match, it potentially risks missing a critical drug - enzyme interaction that is primarily present in a certain subpopulation.
Point 5: For a novel therapeutic...ie mRNA vaccines...you don't know a priori what the potential genetic risk factors are. There may be some. There may not be. But you don't know until you test. Which is why you run the clinical trial.
Point 6: If your priority is saving lives, you'll release it for the subgroup that it's been adequately tested on, while expanding the trial group so that it can be more widely tested.
Say you do testing on a vaccine, everything looks good, and you realize that, oops, you didn't test it on pregnant women. Do you normally say, "Men can't have it until it's been tested on pregnant women!"? No. Men can't get pregnant, (Barring extreme measures that no ethical doctor would take.) so you release it to men, and maybe to women not at risk of pregnancy.
Why should this be different for race? Because not discriminating on sex is less important than saving lives, but not discriminating on race is more important than saving lives?
"This is scientifically incoherent."
Professor, look in a damn mirror.
Because when you look a the the effects of COVID-19 on people of 100 different nationalities around the world, you see very little differences. And because, what is race anyhow?
We know that now. We didn't know that then.
I published that in the medical literature last September when the vaccines were still in trials.
In Trials. Not before the trials started. Which is when recruitment was.
That's not what he said.
While it is theoretically possible that a vaccine could have differential effects among people of different ethnicities, there was (and still is) no evidence of a detectable difference for this vaccine. He further says that even if there were a detectable difference, the definition of "ethnicities" mandated by the government would be inadequate to detect or confirm it.
DeathSantis killed more Americans than Osama Bin Laden by attacking public health officials and encouraging anti-vaxxers to move to Florida.
You know, if you're going to go after DeSantis for Florida's DeSantis numbers, at least do a little check on the current status.
Right now, Florida has the lowest Covid numbers of any state (6.6 per 100,000), well below the rates in deep-blue states such as California (15.3), Massachusetts (23.5), and Vermont (51.9).
I don't agree with how he handled everything, and the overall numbers compared to other states is still pretty high (though falling). But you may want to consider whey the are actually at before spouting off.
That’s because DeathSantis allowed Delta to run rampant through Florida and many died but those that survived now have natural immunity from Delta. So Louisiana and Florida show that herd immunity is very close and we just need to get through all the states having Delta. But if you compare North Carolina with a Democrat governor you can see had Florida just masked these last several months thousands of lives would have been saved. You can’t mask for 2 more months to save over 10,000 lives??
We are nowhere near herd immunity (around 80% HIT with Delta now), and vaccination doesn’t help get us there.
What is going on all across the south, from FL and GA through TX, is that they are now over their seasonal peak for respiratory viruses. That is because they are no longer cooped up inside air conditioned buildings all the time, but can now spend time outdoors, where the virus cannot congregate, but is, instead, killed by the UV. The north, as well as the Rocky Mountain west, on the other hand, are entering their peak season for respiratory viruses for the opposite reason - it is now too cold to spend much time outdoors, so everyone is spending their time indoors where it is warm.
I don't know precisely what you meant to convey with your statistics, probably that the current cases per 100,000, I guess?
But the more relevant number seems to be that Florida is #8 in the US in terms of deaths per 100,000 (283). Moreover, 8 of the top 10 states (in terms of deaths per 100,000) have Republican governors. This doesn't necessarily mean a whole lot without more rigorous analysis such as population density (though the red states with high numbers tend to be less dense than the blue states (NY and NJ), age (here the red states probably tend to have the older more vulnerable population), climate (presumably warmer climates tend to be less conducive to transmission given winter spikes, so red states have a natural advantage), etc. So the top number statistics are probably a pretty weak way to argue about who has been most effective.
But, surely, the overall death rate is a more significant indicator of the effectiveness of a particular state's Covid policies (and/or messaging) than what the cases are "right now". On that score, Florida and other red states have done quite poorly, compared to blue states. (Which could, in the end, be due almost entirely to higher vaccination rates, I don't know. But that itself would be an indictment of red politician's messaging.)
"surely, the overall death rate is a more significant indicator of the effectiveness of a particular state's Covid policies"
Maybe but then you have to ask why. What is the number of ICU beds in a state, what are the percentage of elderly in a state, what is the number of cancer patients and other immuno-compromised people? one can go on.
A better measure might be the numbered of confirmed cases in a state.
It is very easy to make simplistic statements and hard to prove most politically motivated assertions, including DB's implied accusation
I was right about Vermont’s low level of natural immunity being a problem—medical community is now admitting that low natural immunity is a problem during Delta. Still, Vermont shouldn’t have nearly the Delta surge of the southeast.
Oh, and I got banned from Berenson’s substack even though I predicted a population with low natural natural immunity but high vaxxed immunity would see a Delta surge.
That factor was likely the case in Israel. I have written from the authors of the oft-quoted study for addition information but have not heard back yet
Yes, the key is having an open mind and just not getting too invested in the “science” of today when the “science” is not settled due to an ongoing pandemic with variants.
You are, of course right. I would just point out that I explicitly noted that the differences in that top-line number could be due to many of the same factors you identified.
"So the top number statistics are probably a pretty weak way to argue about who has been most effective."
Looks like Florida is doing pretty good versus New York...now if we can just stop New Yorkers (especially those from NYC) moving to Florida (they all seem to go to Miami area don't they)...keep the little commies in NYC and Long Island where they can live in wokedom and tell each other their farts smell great while they drive their electric cars...fuck them
I don't know that Florida is doing "pretty good" compared to New York, even by raw numbers. They're actually pretty close. But, as I already pointed out, "the top number statistics are probably a pretty weak way to argue about who has been most effective" given confounding factors like population density, population demographics and age, climate, etc.
But you just wanted to rant about how you hate New Yorkers. Carry on.
How many? The world may never know;
https://www.youtube.com/watch?v=5ZtbCOpx8Sk
But probably way less than others who've made (and are still making far worse decisions.
Awaiting the decision of the Cali licensing board about a doctor who delayed the introduction of the vaccines by a couple of months, to prevent the election of Trump. He killed tens of thousands of people.
https://www.technologyreview.com/2020/10/19/1010646/campaign-stop-covid-19-vaccine-trump-election-day/
"If the NIH and other authorities consistently said that socially and legally constructed racial and ethnic categories are not scientific in nature and have no bearing on vaccine efficacy, then the public would be much more likely to believe it."
Just because you assert it doesn't make it true. If Republicans didn't believe "trust us, we're from the government," why would POC?
Trust doesn't matter. Parties don't matter. Delays matter.
Because success convinces people much faster than blabbering does.
This is fascinating.
I missed the Bernstein post questioning how many people Trump, wannbe-Trump (Fl), She-Trump (SD), or all the my-body-my-choice MAGAs have killed.
Perhaps he was too busy with his alt-CRT research.
All you have to do is compare North Carolina with a Democrat governor with the rest of the SE. The Republican governors in the SE killed tens of thousands of Americans by not allowing public health officials to order mask mandates. NC doesn’t even have a good vax rate and still has significantly fewer Covid deaths than other SE states.
"Dr. Collins says, if the people testing the vaccine don't represent the American public, the public won't trust the vaccine."
Sigh. This is precisely the issue with public health officials. They try to anticipate the behavior of the American public without actually understanding how people think.
Precisely zero people are gonna care about this when it comes time to take the vaccine. Ordinary people aren't public health officials! No one is seriously examining the results of the clinical trial, and, more to the point, if people have the knowledge to seriously examine the data in this much detail ... they know better! People rely on what they see in the real world.
If I get black community leaders to take the vaccine, maybe even pay them to, and have people watch, that is far more effective and cheaper and faster than this delay.
People want to know how their friends and family members reacted to the vaccine, and if people are fine, they themselves will get it.
Like, for all of this diversity nonsense in clinical trials, black people were skeptical of the vaccine anyway! It didn't matter! It was never going to! Once it was actually out there then people responded by seeing other people get it.
The same exact shit happened when Britian approved the vaccine. Dr. Fauci went on TV and started casting doubt about the vaccine because he felt people couldn't know if the vaccine worked unless the FDA specifically said it worked ... its so stupid. The vaccine doesn't suddenly start working when the FDA says so.
The same thing happened again when out of nowhere the government paused the J&J vaccine over myocardities, not because the vaccine was unsafe, but because the paused was supposed to restore confidence. It didn't! It was one of the stupidest decisions that killed people and forced the entire medical community to waste a great deal of time explaining to people why mycardities isn't an actual issue, which could have been resolved so much easier if the FDA didn't give anti vaxxers both gasoline and a lighted match.
And the response people were given was to simply "trust" the FDA. If your only response to criticism is that you need to trust harder, that is not a response! Explain to people why the vaccine works. If you can't do that, hire someone who can. People in the actual industry don't trust public health officials! Most people, outside cameras, regard them as idiots! Why should the American people trust them?
The job of the FDA is to figure out if the vaccine works and is safe. Not to do all this other nonsense to try to manipulate people. As much as I get annoyed at anti-vaxxers, I kinda understand with all the outright lying and manipulation done by public health officials how we got to this point.
There are people who understand marketing. Those people actually do understand how people think. If the government hired an ad agency instead of having the FDA do a job its not qualified for, then a lot if issues could have been avoided. Pharma companies pay for marketing for a reason. There are legit criticisms of that ... but this exact issue is a good reason for it.
And no one will learn anything from this mess. I used to work on the regulation side of pharma ... I mean, even as an intern / associate, I have so many stories of boneheaded decisions like this that are much less
high profile. This isn't new. I had a clinical trial rejected because the FDA didn't understand how baysian analysis works. Its extremely frustrating and has to change.
That will never change, it's been that way since 1950somethin', but if they got another bucket of money they could subcontract it.
David, you're way...way...way off base here. For multiple reasons.
1. You need to appreciate how unbelievably fast this vaccine development process was. Initiation to vaccine in less than a year. Typically this development is a multiyear process.
2. Keep in mind that while the trials were going on, the manufacturers were also upscaling their vaccine production. A temporary delay in the trial would not affect the vaccine production upscaling. And the limiting factor really was vaccine production ultimately.
3. There can be differences in race for pharmaceutical treatments. Most commonly in protein plasma binding rates to alpha1-acid glycoprotein (AGP). To not have a representative sample in a mass phase 3 clinical trial, and potentially miss something critical would be folly.
https://pubmed.ncbi.nlm.nih.gov/10706191/
1. It is true that the development was extremely fast. But it is totally fair to criticize it for not bring nearly fast enough.
The sample was identified 2 days after the DNA strand for covid was released. 2 days. The reason vaccine development takes so long is because of research ... all of the research was done up front.
2. Limited supply was a problem until the beginning of April. Then it was a demand issue. The problem with the delay was that it increased uncertainty was to what was going on, introduced totally legitimate accusations of impropreitary among public health officials, and delayed the first shipments.
3000 people were dying a day at that point. We wouldn't have saved 27000 with a week sooner shipments, again that's the supply issue. But we would have saved a significant fraction of that 27000, simply by reducing spread earlier.
And an earlier approval would have allowed Pfizer to allocate more doses to the US. What increased manufacturing was the government guaranteeing it will buy shots ... a better guarantee can be made with faster approval.
Also, once we actually started shipping this, we discovered that a lot of the supposed problems weren't actually problems. The while cold storage thing turned out not to be an actual big issue. Half doses are equally effective. The second dose can be delayed. We would have discovered this earlier with earlier approval and more flexibility.
Also we had shots from other people. Astrazeneca has still not been approved ... if supply is what matters, why the delay there? Millions of doses had to be thrown out because they expired.
3. Thats is very true, and Bernsteins dismissal of them entirely is very wrong. But in vaccines specifically there is very little scientific evidence it matter. And, nonetheless, the official didn't offer any! He have a reason that is not motivated in science is any way at all, pretending to be a marketer. judging public confidence, when he is not.
No one proposed a legitimate scientific medical reason why the vaccine would work in white people but not black. No one.
Granted, I do have a different view, I used to work more on pharmacokinetics than clinical trials so my focus is the mechanism, but ... like its no different from the flu vaccine. The mRna is the only difference ... which only lasts in the body for a couple days. There is no legitimate mechanism here why race matters.
"No one proposed a legitimate scientific medical reason why the vaccine would work in white people but not black. No one."
Do you know how the vaccine is metabolized? Or potentially metabolized? You're aware there can be significantly different genetics for many of the critical CYP enzymes in the population? And they aren't distributed equally by "race"?
I have no idea what you are talking about.
How what is metabolized? mRNA? That differs on race? Or anything, really? We all produce it ... did the NHS director mention this?
The spike protein? Its the same spike protein in covid, and dosage doesn't really matter because the body will be producing it ... so what difference does that make?
"I have no idea what you are talking about."
Neither does the Armchair. He is just babbling on
An Ad Hominem argument....typical and non-informative.
Why shouldn’t the spike proteins be slightly different by race? There are sharply different incidences by race, according to CDC figures.
The spike protein comes from the virus, or from the vaccine, not the infected person. The virus and vaccines don't have a race.
No, it's a statement of plain fact. You cited a survey that found ethnically correlated differences in drug responses, but missed the context that it was talking about different delivery and effect mechanisms than this vaccine. You speculated about "how the vaccine is metabolized", but didn't provide any specificity about what you meant. When people pointed those things out, you changed the topic rather than answer the challenges.
"How what is metabolized? mRNA? That differs on race?"
Critically here, the vaccines aren't really naturally occurring mRNA. They have synthetic nucleotides in them that do not naturally occur. Not to mention the adjuvants.
Do you know how they are metabolized? Do you know if there are any genetic differences that would affect how they are metabolized? Or alter the immune response? If there is POTENTIALLY a genetic alteration overrepresented in a sub-population that would result in lack of efficacy? Or would result in an immune response that is too dramatic?
And are you prepared to simply inject a large subpopulation, in the millions of people, who haven't been tested for any of this potential?
Do you know if there were any South Asians in the studies? People of 100 pc North American indigenous origin? South American? Ashkenazi Jews? Arabs? We certainly did inject millions of members of subgroups without any data about them.
"Critically here, the vaccines aren't really naturally occurring mRNA. They have synthetic nucleotides in them that do not naturally occur."
That led to some fascinating reading, I have to say.
But this technique has been around long enough you'd think they'd already know if it caused different reactions in specific groups.
A couple of points:
(1) Bernstein has always been a second-tier contributor here, but a least he used to offer a little diversity in his posts. There was some on Israel, some on constitutional law, some on the right-wing fad of the moment. Now? It's race, race, race, race. The professor has sadly become a Johnny-One-Note. It's kinda funny how often right-wingers seem more obsessed about race that the Lefties they criticize (for being obsessed).
(2) There's a real question to be asked, but DB probably isn't honest enough to do it: How many people did Trump kill? You'll respond he presided over an exceptionally efficient vaccine effort, and - yes - I gladly grant him that.
But take a look around and it's clear today's Right is the Pro-Covid party. Seeking a bit of political gain, its leaders consistently sold birtherism-style disinformation about the pandemic. They programed their followers to think the disease isn't serious, its effects are exaggerated, its spread overstated, its recorded deaths conspiratorial lies, the measures against it irrelevant & driven by secret agendas, the public health officials who fight it evil villains, and the "real" treatments against it secrets hidden from the public.
And almost all of that started with Trump. He presided over both the creation of Covid vaccines, and the creation of a whole political party of anti-vaxx loons. A few months back there was a guy leading one of those groups that has fought every measure against the disease. When he got sick, he refused to go to the doctor because he didn't want to add to the Covid statistics (per his widow). Instead he dosed himself with horse dewormer & zinc until it was too late.
Did Trump's political bullshit kill him? That's difficult to say, but it's far more likely true than Bernstein's nonsense above.
You're butthurt because Bernstein didn't write a different post?
Not only different, but more honest ..... less trolling.
How is this dishonest? How would the critique of Trump that you want him to write be more honest?
People have explained above how the change in trials didn't delay delivery of the vaccine, which was a production issue. Among other reasons. Trump and other's messaging on "it's just like the flu", masks are a secret government program to control you, and similar nonsense did, actually, affect people's decisions (including whether to get the vaccine). Consequently, it is much more honest to claim Trump, DeSantis, etc. killed thousands of people, than this one guy whose efforts, apparently, didn't delay delivery of the vaccine (but may actually have helped messaging).
That's why.
Were you living in some alternate universe where Trump wasn't bragging about accelerating the vaccine development, and promoting its use? Maybe one where Democrats weren't saying the vaccine was suspicious because Trump had promoted it?
It's sad that such a tiny pianist must have such a tiny heart that it is incapable of pumping blood to what must be your tiny little brain
Y'all know I have a book coming out about racial classification, which includes a chapter the mandated use of race in medical research, right? And I am therefore something of an expert on this issue, but not on whatever other vaccine related issues you would prefer I blog about?
Well, funny if you are an expert but, as some commenters noted and a quick google search seems to confirm, any delay in clinical trials was inconsequential due to the bottleneck being production.
If you wanted to write an informative post about "the mandated use of race in medical research", then maybe leave out the hackery of "How many people did NIH Director....kill?" In addition to the above, you also had to dishonestly waive away the fact that he was facing a public who, rightly or wrongly but not due to Collins' actions, Collins believed needed to be reassured about whether the clinical trials were valid for all "races."
And hackery like this, David, is why you aren't taken more seriously by more people. Write a legit post on your subject of self-proclaimed expertise rather than click bait.
NOVA lawyer, I shouldn't really respond, but I'm so amused when people are so busy blustering and insulting that they miss the obvious. Vaccines were ready to be put in arms as soon as they were approved. Moderna was approved maybe two weeks late. That means that the first people who got the vaccine, medical staff and the like, didn't get the vaccine until up to two weeks later than they would have. Some small fraction of these people got covid in the meantime, and passed it on to others. And given that one constraint was simply having the personnel to administer the vaccines in hospitals et al, this continued for a while. Both these factor are completely aside from whether the constraint was eventually supply. We can agree, however, that in the scheme of 700K Americans dying, however many extra deaths this led to was "insignificant." But whether "insignificant" in this context means dozens, or, hundred, or low thousands, they were all unnecessary and died because of this diversity idiocy by Collins. But thanks for playing.
David,
I shouldn't respond, because, although you seem confident that the approval did delay some people from getting the vaccine until maybe two weeks later (which presumably you have researched, I have not), you don't address the entirely faulty logic you use to accuse Collins, specifically, of killing people. It's just poor form, in addition to bad reasoning. You concede that Collins' justification for ensuring there was diversity (so the public, particularly including people of color, would trust the trials) may have been valid, but, again, hand wave away that for the reason that, essentially, it was people like Collins who created that environment. But it wasn't Collins. Which means, making it specifically about Collins "killing" people is click bait hackery.
For example, actual scientists were telling the public that racial diversity in the trials was important: Angela Rasmussen, a virologist at Columbia’s Mailman School of Public Health. “It is particularly important to make sure that Black and Latinx participants are represented equitably in the trials and the vaccine shows protective efficacy in different populations of people.”
Whether this is due to messaging or valid scientific concerns, doesn't really matter so much at the time Collins made his decision. If either is legit, then ensuring diversity in the trials ensured more people (particularly including people in communities that, for whatever reason, are harder hit) would choose to take the vaccine. Thus, without acknowledging that Collins therefore likely saved some number of lives by ensuring the clinical trials were diverse, your math doesn't work because you are only subtracting lives for an up to two week delay in completion of the trials but not adding lives saved due to more effective messaging (at least). That you did it and still won't acknowledge it says what needs to be said about whether you are a source to be trusted or just a guy with an agenda.
Basically, it's a subject that interests me (racial classifications in medical research and care; e.g., AHA guidelines that added 3 points for nonblack cardiac patients, basically making black patients less likely to get treatment than nonblack patients with similar objective test results). An informative post previewing the chapter in your book likely would have gotten me to read it. But you've revealed you are an ideological hack with political axes to grind, so no thanks. I'll get my information from someone who plays it straight.
Great comment NOVA, agree fully.
I haven't read many Bernstein posts lately, but if this Enquirer-worthy headline (with only slightly-above-Enquirer-level supporting argument) typifies his recent blog output, I am disappointed. I can imagine several possible drivers of this post's tone and quality, but disinterested scholarship (or just plain truth-seeking) aren't among them.
Bernstein being bright and Collins being bright and superlatively knowledgeable in his field, I imagine that if the two ran into each other just after Collins read this post, it is Bernstein who would stand to learn something, *if* he kept an open mind. Collins would listen charitably and respond constructively, calmly, and instructively, as he is a thoroughly decent man. Bernstein might even update his post. The world would be a slightly better place.
Oops, I should have said "extremely knowledgeable in his field," not "superlatively" (which arguably applies to his direct report Fauci).
Thanks!
The world would be a slightly better place if David updated his post.
"NOVA lawyer, I shouldn't really respond, but I'm so amused when people are so busy blustering and insulting that they miss the obvious."
Holy shit, I thought your articles were obliviously stupid, but you are really doing the hard work in the comments. To think that you are a professor and students are supposed to learn from you. Even more sad to think that they pay for it.
I think you've been unfair to Collins. You repeatedly posit that race is mandated in medical research. If that is true, why would you accuse Collins of responsibility for killing them. Are you advocating for the right of public servants to nullify the law if they think they know better? Is that the concept of rule of law you believe in? That should prove interesting!
(1) This was an emergency authorization, you may recall.
(2) The government doesn't define ex ante how many of each group constitutes "enough" in any individual study. You don't know until you submit the data. The safest thing is to simply try to match USA demographics, even though that's obviously not a scientific standard, but a political/social one. But Collins could have simply determined that in this case, whatever level of "diversity" Moderna had was "enough," certainly for an emergency authorization.
Authoring a book doesn't make one an expert.
To be fair, it was a chapter in a book. But why burst his self-inflated bubble?
"Did Trump's political bullshit kill him? "
No, his own damned stupidity did and that did his wife a favor. Imagine being married to such a damned fool.
A tendentious counterfactual in service of an accusation of killing?
Is this World Net Daily? Come on, this is crap.
What's tendentious? Are you suggesting that the more diverse sample was worth a two-week delay in the vaccines? What are you, some kind of anti-vaxxer?
Hindsight about risk mitigation is bullshit.
Sturgeon's law: 90% of everything is crap. Except CDC and NIH risk mitigation, which is 99% crap because they care very strongly about not approving a dangerous drug, and almost nothing about rejecting effective ones.
Unless their bosses want the vaccines badly enough the safety wasn’t an issue.
You want more dangerous drugs approved so that effective ones are not delayed? Take that up with the public.
[Effective drugs are rarely rejected under the current scheme - you can always go back for more efficacy trials]
Hah, right. My father, a doctor, recently (as such things go) advised a pharma company on a Phase 2 or 3 trial. He advised them that they should have X participants because the rate of spontaneous remission for the target condition was fairly high. They got a second opinion advising them that about 60% of X would be enough, and they did that because getting and using more participants is expensive.
The study came up one case -- in either the positive or negative group -- from showing efficacy with statistical significance. The drug company decided not to pursue another trial because of cost.
The drug company decided not to pursue another trial because of cost.
So, not exactly rejected at all.
In fact, the cause for the rejection you laid out is that the company didn't follow the protocol your father suggested, and thus did a useless study, and then made a business decision.
And what I posted about how you can always add in another trial is correct.
And you also have wandered away from the original thesis that dinging someone in retrospect for their risk mitigation does not understand how time works.
To be clear, this comment is about the philosophy of decision making in general, not anything to do with this particular decision.
What makes a good decision? I'd define it something like 'making an accurate assessment of the odds (and consequences) and choosing the best course of action'. In poker terms, drawing to an inside straight is usually a bad decision, while drawing to an outside straight might be a good decision, depending on what other information is available.
But one can absolutely criticize a decision, in hindsight, for being too timid or reckless, and that criticism is valid as long as you only consider the information the decision maker had at the time.
A couple of classic examples are the amphibious invasions at Gallipoli and Anzio. An amphibious attacker has the advantage of concentrating a large force in a small area; the defender, not knowing where the blow will fall can't be strong everywhere. Once the landing happens, it's a race to see who can reinforce quicker - the attacker over the beach or the defender overland. At both Gallipoli and Anzio the landings encountered negligible opposition, but in both cases overly timid generals halted their forces just inland, from a fear of spreading too thin, before the Turks or Germans arrived in force. Those were very costly mistakes, and it is absolutely valid to criticize the generals involved for making bad decisions.
For a more recent example, the Bush II decision to invade Iraq was a very poor reading of the odds. It's not that it was inconceivable that deposing Saddam would result in the flowering of a vibrant democracy, it's that the country devolving into viscous sectarian strife, egged on by Iran, was a lot more likely. In poker terms, they drew to an inside straight, and it's fair to criticize that decision even in hindsight.
Do you dispute what others above are citing that the delay in the trial didn't actually delay production of the vaccine (which continued apace during the trial) or delivery of the vaccine (which was dependent upon production, not approval)? If that's true, then even the factual underpinnings of Bernstein's chain of logic (and yours) is utter bullshit.
And what is certainly tendentious is Bernstein waiving away the objection that having a diverse sample would help ensure more people chose to take the vaccine by saying "If Americans wouldn't 'trust' a vaccine that didn't have 'enough diversity,' that's largely because government authorities like the NIH insist that vaccines aren't trustworthy unless they have been tested on a "diverse" population." Which is an admission that Collins' reasoning might have been valid and Collins didn't create that situation, so it is odd (tendentious) to blame Collins for deaths that are due, allegedly, to bad messaging by other people regarding the need for diversity in clinical trials.
I have no dog in the fight as to whether Collins' decision was a good one or not (whether from a scientific viewpoint or from a messaging standpoint), but for Bernstein to post this poorly reasoned accusation when he hasn't said peep about Trump, DeSantis, etc., etc., well, what little credibility he had is evaporating. He's apparently trying to compete for stupidest nonsense hackery with Josh.
I think Trump did a wonderful job here in getting the vaccine production done in as fast a manner as possible.
Given what the "experts" were saying, I think if Biden was in charge, it would've taken far longer.
Of course you do.
What precisely do you think Trump should've done differently with Operation Warp Speed? Again, it was perhaps the best single run government operation in the last 20 years. An amazing success.
Meanwhile, Biden's current administration is having execution failure after execution failure on multiple programs. Why OWS would somehow be different if Biden was in charge? I don't know.
The Covid vaccine development was a success.
The first case of H1N1 (swine flu) was detected in the US on April 15, 2009, the US declared the H1N1 a public health emergency eight (8) days later, a vaccine first became available five (5) months later, and was available to the general public in December 2009, just over (7) months after the first US case.
The first case of Covid-19 was detected in the US on January 21, 2020, Jan. 22 Trump says "We have it totally under control. It’s one person coming in from China, and we have it under control. It’s going to be just fine."; Feb 2 Trumps says "We pretty much shut it down coming in from China.”"; the US declared Covid a public health emergency on February 3, 2020; Feb. 10, Trump says, "“Looks like by April, you know, in theory, when it gets a little warmer, it miraculously goes away."; February 12, Trumps says, "“In our country, we only have, basically, 12 cases and most of those people are recovering and some cases fully recovered. So it’s actually less.”; Feb. 26, Trump says, "“When you have 15 people, and the 15 within a couple of days is going to be down to close to zero, that’s a pretty good job we’ve done.”; Mar. 5, Trump says, "The April 2009-10 Swine Flu, where nearly 13,000 people died in the U.S., was poorly handled.”; Mar. 23, Trump says, “America will again, and soon, be open for business — very soon — a lot sooner than three or four months that somebody was suggesting. A lot sooner. We cannot let the cure be worse than the problem itself. We’re not going to let the cure be worse than the problem.”; Apr. 27, "“LIBERATE MICHIGAN!”; Dec. 11, first shipments of the vaccine, eleven (11) months after the first US case was detected; rollout to the general public took another few months.
So, yeah, the Covid-19 vaccine development was rapid and impressive. But it was slower than the H1N1 vaccine development and rollout. And Trump criticized the H1N1 response with about 13k dead as opposed to well over 500k dead during his "management" of the covid pandemic.
Trump downplayed the virus, discouraged commonsense safety measures pending release of the vaccine ("LIBERATE MICHIGAN!", and otherwise indicated little urgency from his office (as opposed to the people more directly involved in the development and logistics of the vaccine).
Trump is no hero on this, or anything else. As in everything from spending to foreign relations to ethics, he suffers terribly in comparison to Obama.
"So, yeah, the Covid-19 vaccine development was rapid and impressive. But it was slower than the H1N1 vaccine development and rollout."
Do you think the difference is because management of the covid development program was inept, or because we develop new flu vaccines every year, and we hadn't developed any coronavirus vaccines before?
"Trump is no hero on this, or anything else"
No disagreement there.
I'm just pointing out, the nearest analogy that we have is the H1N1 situation which Trump criticized during the Covid-19 pandemic when he thought 13k deaths was horrible....and here we are, over 700k deaths later.
Yes, the vaccine development issues were different. You will search in vain for where I claimed they weren't. But my point is, if someone like Armchair is going to sing the praises of Trump for personally managing the vaccine development because he happened to be President at the time (rather than focusing on what actions (including pronouncements) he actually took and how those actions actually affected the pandemic, then the least he can do would be to praise Obama for his more consistent and effective response to H1N1.
I think it's fair to say that the Trump administration didn't grossly eff up the development of a vaccine. I'm open to correction from the better informed but my sense is that they threw enough money at it, and didn't throw any sabots into the gears. That's not always true, see e.g. LBJ and Vietnam or Bush II and Iraq.
Similarly, I don't see that the Obama admin made any great mistakes with H1N1, other than not restocking the mask stockpile afterwards.
But fundamentally the 13K vs 700K numbers, IMHO, have a lot to do with biology and not much to do with politics. We tend to blame/credit presidents for the economy and now pandemics. In general, neither makes a lot of sense to me.
"We tend to blame/credit presidents for the economy and now pandemics. In general, neither makes a lot of sense to me."
I couldn't agree more on that point. That's essentially why I pointed out that, if Armchair is going to blame Obama for the state of the Strategic National Stockpile in 2020, then surely he should give Obama credit for the 2019 economy.
Of course the difference in numbers is due to biology. But relevant to that discussion is (1) Trump first raised the 13k issue as a disaster (if you're going to set up the petard, prepare to be hoisted), (2) Trump explicitly downplayed the particularly dangerous biology of covid. That's something he, specifically, did in terms of the response to the pandemic. The vaccine development, as the Obama example illustrates, was nothing special on Trump's part. It's easy enough in the midst of a pandemic to throw all necessary resources at quickly developing a vaccine. I think anyone claiming any President wouldn't have has the burden of proof there. But he does get credit for not messing that up and for touting it as a priority (again, I think anyone would have done the same and, of course, the UK is the first to give a non-clinical trial vaccine...well, discounting Russia's very questionable Sputnik vaccine)
Of course, you buy into the shortage of N95 masks as "Obama's" fault. Why? Did he not ask for more funding for the CDC than he got? Did he personally choose to use limited resources to purchase limited supply, life-saving medicines instead of replenishing the supply of N95 masks? Was that, ex ante, a bad choice? And, again, how is Obama to blame for the failure to request money to replenish the supply in the FY 2018, 2019, 2020 budgets? Seems to me, someone else had three shots at the ball, if we're saying this was a President's decision. At best, replenishing the SNS is something both Obama and Trump failed to do (if we're blaming Presidents instead of Congress), so kinda weird to blame just Obama.
You're comparing H1N1 to COVID. In terms of vaccine development, the two aren't comparable.
H1N1 is a strain of influenza, also known as the flu. Flu vaccine production is quite well precedented. We do it every year, with new strains. The infrastructure is in place. We don't do full clinical trials for the flu vaccine every year for the new strain.
COVID is notably NOT a strain of influenza, but from a completely different family of viruses. It's not the same "grow it in chicken eggs" type of deal used for influenza vaccines. It required the full range of clinical trials.
If we're really comparing the two though, one element is important. And that's the national strategic stockpiles of protective equipment. Obama decided to use the stockpile that Bush built up during the Swine Flu...then ALSO decided he couldn't be bothered to replace it. So when the materials were needed for the next epidemic, the proverbial cupboard was bare. That is on Obama.
"So when the materials were needed for the next epidemic, the proverbial cupboard was bare. That is on Obama."
You can lie, but that doesn't really win except in your own head.
As you know, when a Democrat is in the White House, Republicans become spending hawks. As a result, when Obama sought money for CDC generally and the stockpile specifically, he received less money than he asked for. Also, they implemented discretionary spending limits which impacted later budget requests. And, in his last year (2016), the Republican controlled Congress approved less than half what he requested. He got what he could and the CDC had to prioritize.
Further, his last full year in office was 2016. Notoriously, they had a pandemic plan that was ignored by the Trump team.
But, most importantly, the pandemic hit in 2020. Trump had three full years of "the best economy ever" and a ridiculously spending-compliant Congress. If ensuring that the strategic stockpile had enough N95 masks was the responsibility of the President, why aren't you excoriating President Trump for sitting on his hands for 3 full years without replenishing it? Unlike Obama, he didn't have pushback from a spending averse Congress.
It's just bizarre that three years into his Presidency, Trump is blaming the state of the SNS on Obama, while, in contrast, he has been taking credit for the economy since before he even took office. Weird.
And you fell for it. Both times.
On the N95 stockpile:
1)FWIW, USA Today says:
"We rate this claim TRUE because it is supported by our research. There is no indication that the Obama administration took significant steps to replenish the supply of N95 masks in the Strategic National Stockpile after it was depleted from repeated crises. Calls for action came from experts at the time concerned for the country’s ability to respond to future serious pandemics. Such recommendations were, for whatever reason, not heeded."
2)I buy N95 masks, retail, for about a buck each. I've used up my allotted link, but search for "The H1N1 influenza pandemic of 2009 triggered the largest deployment in U.S. history of the Strategic National Stockpile, the federal government's last-resort cache of drugs and medical supplies. The stockpile distributed 85 million N95 respirators". I dunno how much cheaper they are wholesale, but given the scale of federal spending, pleading poverty for not spending something under $85M ... sounds like someone had other priorities.
To be clear, it's not like I'm saying Obama is an arch-villain. I really doubt there was a meeting where someone said 'Mr. President, we used up most of the mask stockpile and we need to restock' and Obama said 'Eff that, I don't care about pandemics, spend the money on my inaugural ball instead'. I expect it just dropped through the cracks.
Preparing for Bad Things is not something everyone values. I lived for years in Seattle, which one of these days will have the Mother Of All Earthquakes. And I had neighbors who wouldn't even bother keeping 3 days of water around. I don't get that, but it's common enough.
"But, most importantly, the pandemic hit in 2020. Trump had three full years of "the best economy ever" and a ridiculously spending-compliant Congress. If ensuring that the strategic stockpile had enough N95 masks was the responsibility of the President, why aren't you excoriating President Trump for sitting on his hands for 3 full years without replenishing it?"
And to address that, when you just distributed the stockpile, making a note to self 'replenish stockpile' seems a little more obvious than taking office years later and waking up one night and wondering 'hey, do we have a mask stockpile in case of a pandemic'.
To be clear, if someone went to Trump and said 'Mr. President, we depleted the stockpile years ago, and we should restock it' and he replied 'Eff that, I'd rather spend the money on whatever', then he was wrong. But my guess was that it just never came up.
1. The USA Today article differs from several other accounts, including those that provide specific numbers for funding requests for the SNS that were denied by Congress in 2010. Thereafter, because of budget controls, Obama did seek lower amounts for the CDC given availability of resources which it said at the time. But, again in 2016, they asked for more money for CDC and the SNS, but were denied. And a Republican Congressmen explicitly acknowledged Congress's role in the lack of replenishment of the SNS (in terms of N95 masks, because, again, it wasn't a bare cupboard but they prioritized hard to acquire life-saving pharmaceuticals rather than what they thought would be relatively (compared to limited availability drugs) easy to acquire masks).
2. The fact that you buy them retail for a buck each, relatively easily, is probably one reason the people who actually made the prioritization decisions chose hard to manufacture and/or acquire pharmaceuticals over masks.
3. "when you just distributed the stockpile, making a note to self 'replenish stockpile' seems a little more obvious than taking office years later and waking up one night and wondering 'hey, do we have a mask stockpile in case of a pandemic'."
Again, the Obama administration was warned and the Trump administration was warned about the need for more money for the SNS stockpile. I doubt either Obama or Trump personally made the call not to buy more masks and instead use the limited funds to buy pharmaceuticals. Which you acknowledge, but then somehow weirdly buy into the notion that Obama was personally responsible. And, to repeat, the Obama administration tried to get more money for the SNS and was denied by Congress.
You seem quite reasonable. It just seems odd to blame Obama for not buying N95 masks for the SNS in a situation where his administration asked for more money but didn't get it, and he personally probably had no input whatsoever in whether to buy surgical masks, N95 masks, pharmaceuticals, or other equipment/supplies. Ditto for Trump. I am not blaming him for not replenishing the stockpile, but you can't really blame one President and not the other, especially when there is publicly available evidence that both were told more attention needed to be paid to the SNS. And Trump's administration was specifically warned on masks.
To be clear, I am not blaming Trump for the lack of N95 masks, but he had ultimate control of the CDC and the SNS for three full years. If there were problems, he had as much responsibility as Obama for them by 2020.
But race doesn't matter....
Oh wait, it DOES matter but only for certain things....
If you point out the OTHER things though, that is "racist".....
And, then race doesn't matter once again....
I'm so confused....
Skin cancer—race matters with respect to skin cancer.
So races are different?!?!?
Or is it only for purposes of cancer that they are different, but for everything else they are the same????
I'm still very confused.....?????
"I'm still very confused.....?????"
That's because you aren't very bright. Skin tone makes a difference in susceptibility to sunburn and skin cancer. Shocker. Things not affected by skin color, pretty much everything with which you seem to secretly want skin color to correlate.
OK so race matters for cancer then. Got it.
When else should it matter and not matter?
I'm not sure, but are you trying to get at the argument that if there are racial disparities in, like, skin cancer, then there must be racial disparities in intelligence?
Actually what matters is the melanin content in the skin.
But the sickle cell trait is a predominately in those of African lineage.
If I were a dermatologist I would have my office in a country club or a yacht club…that’s where the skin cancer is at!
A business savvy choice, I'd say
Race matters for dying from COVID-19. There are significant differences between the races in their death rates.
There's also a political question. Suppose, after wide vaccination, one particular ethnic minority group had a higher "breakthrough" infection rate than others. Perhaps for a cultural reason - members of the group living in closer quarters than others, or whatever. The companies, and the regulators, would have to be able to protect themselves from suspicions of having failed to test the vaccine adequately in this group. Saying "the immune response does not depend on race!" wouldn't be adequate, even assuming it's true.
Anyone who knows the history of vaccines knows that producers and suppliers are exceptionally vulnerable to junk-science class-action lawsuits decided by juries who don't know the first thing about the science. (That is, in fact, why vaccine manufacturers have a special court-system-within-the-court-system: back in the day, they were getting hit so hard by scientifically invalid lawsuits being decided by juries who didn't know the science and therefore decided according to their feelings for the plaintiffs, that they (the suppliers went to Congress and said: either you protect us from this bull-govno, or we're gonna stop supplying USA with vaccines altogether, and you will go back to watching your kids get crippled by polio. The government, very wisely, capitulated.)
Or it could be that the virus was designed at the Wuhan Institute of Virology to minimize the damage to Han Chinese. And, so far, it does seem to hit Blacks worse than some other ethnicities.
Okay, Bruce.
Bye.
Yes, Bruce, it's a Chinese bio-weapon, and the reason why the Chinese unleashed it when they did is very obvious to anyone who was up-to-date on the trends in classical music.
There's an opera-- Puccini's last opera-- which is very challenging and not performed all that often, except by the really top-level companies, because it's so difficult for the singers. Puccini never finished it, but other lesser composers have composed endings. It's called TURANDOT and it is set in ancient Peking, or rather, in Puccini's fantasy of what ancient Peking was like.
Now, TURANDOT was experiencing renewed popularity early in 2020; it was a big fad in the opera world. The Met was scheduled to do it in April, and several other opera companies like San Francisco Opera and Chicago Lyric Opera were gonna do it, and even the "little-grand-opera" companies like Regina Opera in Brooklyn and West Bay Opera in Palo Alto -- companies which cast young, still-unknown singers to give them a start on their careers, and which perform in small venues for audiences of fewer than 150 people, were getting ready to do it.
It was also trendy in Europe. (These trends come and go.)
This opera TURANDOT is very offensive to Chinese nationalists, because it depicts the Chinese people as superstitious, bloodthirsty barbarians ruled by a sadistic tyrant. But because of COVID-19, the companies had to cancel their performances. This was obviously what the Chinese government was hoping to accomplish by unleashing the virus at that particular time-- to prevent TURANDOT from being performed in Europe and USA.
That is cui bono in this case.
Nevertheless, Turandot is an incredible piece of operatic art.
I was fortunate enough to see it performed with Luciano Pavarotti in his prime
" Suppose, after wide vaccination, one particular ethnic minority group had a higher "breakthrough" infection rate than others. Perhaps for a cultural reason - members of the group living in closer quarters than others, or whatever. "
You might suppose that, but you'd be supposing wrong looking over many dozens of nations across the globe
Bureaucrats care about numbers on spreadsheets. The lives of people don’t matter unless they appear as numbers on spreadsheets.
And even when they do appear, bureaucrats will ignore any other harm to any other people who aren’t analyzed and don't appear on spreadsheets.
That’s how you get long Covid lockdowns that keep people from going to the hospital for treatment. Only Covid showed up on the spreadsheets. If people died from lack of care for something else, their numbers weren't counted.
"If people died from lack of care for something else, their numbers weren't counted."
This doesn't even make sense. There are a lot of dead people whose deaths weren't counted? No one filled out death certificates? Does SSA still count them as alive?
But, it's nonsense. It is easy to google that from March 2020 to August 2021, Covid deaths were counted at about 605k while excess deaths were roughly 808k. Which might mean some people died because of lockdowns, or that some people died because hospitals were full of Covid patients which rationed the supply of medical care. You're gonna have to do more work to support your hypothesis, but saying the deaths "weren't counted" is bizarre.
Indeed, NOVA.
In some countries the "excess deaths" were even negative.
Ok, I thought is was very clear, but I will spell it out completely, because anything implied will apparently be (intentionally?) misunderstood.
---
If people died from lack of care for something else, their numbers weren't counted in the particular spreadsheets used to decide to continue the lockdowns..
Bureaucrats who enact policies do not care about unintended consequences. They're not represented in the spreadsheet model so they don't determine the policy in any way.
Which leads to the answer to Bernstein's question: How many people died for racial diversity in the Moderna trials?
The answer: so what? The numbers of people of each race were written down and tabulated so those numbers mattered. The deaths caused by the delay were not, so they didn't.
---
And the death toll from victims of the Covid policy will be felt for many years to come. Someone who dies of cancer in 2023 because the lockdowns delayed the diagnosis isn't counted as having died in 2020.
But of course I don't expect you to care any more than bureaucrats do.
You still haven't done the work to show that the lockdowns caused more deaths than they prevented.
Scenario 1 (your implied hypo): There was a set number of Covid-19 infections and associated deaths, the lockdowns did not prevent any new Covid-19 infections but did prevent people from going to the hospital (not because the hospitals were overwhelmed and certainly not because people with life threatening illnesses were forbidden from going to the hospital) because.....unclear....and, therefore, people died unnecessarily.
Scenario 2 (much more likely): The more people social distanced (i.e., followed the lockdowns), the less Covid-19 was transmitted and, so, the fewer people died of Covid-19 AND hospitals were less overwhelmed than they otherwise would have been (see Italy for an example of what happens with truly overwhelmed hospitals), therefore health care had to be rationed less than it would have been under Scenario 1 and, so fewer people died due to Covid-19 and fewer people died due to lack of care for other conditions than would have been the case in Scenario 1.
Your analysis is kind of an effort at pure logic: Bureaucrats have incentives, here their incentives (at least some of them) were primarily to manage covid numbers, so they made suboptimal decisions compared to considering all relevant factors.
The irony, you have definitely failed to consider all relevant factors and show an unwarranted degree of certainty that Scenario 1 (or some variation thereof) is "how it is". And that's entirely due to your own incentive structure. You want to prove the lockdowns were counterproductive and "bureaucrats" messed everything up.
While I think Scenario 2 accounts for far more variables and, therefore, is much more plausible than Scenario 1, I can't say for certain under which Scenario there were fewer non-covid deaths. I can say with near certainty, that there were fewer Covid deaths in Scenario 2. Therefore, even if you prove that there were more non-covid deaths in Scenario 1, you still have to prove that the decrease in non-covid deaths if we had followed Scenario 1 is greater than the increase in covid deaths if we had followed Scenario 2.
My post wasn’t about one number being larger or smaller than another. At no point did I make any claim about any number being more or less than any other.
The correct policy is for government to be truthful and to only act when necessary, and only as long as necessary. Not to have one set of spreadsheet numbers or another different set and to bully people (often to death) to try to optimize based on whichever spreadsheet is deemed politically or emotionally or otherwise satisfactory.
More or fewer than the virus-flouting, lethally ignorant, antisocial Republicans and conservatives — your disgusting political allies and friends — have killed, Prof. Bernstein?
Right-wing ankle-nippers are among my favorite culture war casualties. Pathetic.
The Covid vaccine has done more for white supremacy then actual Jim Crow ever did. Now that is something to think about...
Huh?
I have no expertise in the subject, but it's my understanding that it's not uncommon for disease prevalence and vaccine response to vary based on "race." See, e.g. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5325335/ While there's no such thing as race, we all have different genetic makeups, and if most of one's ancestors are from sub-Saharan Africa, one's genetic makeup will be more similar to other persons whose ancestors are from that part of the world, and a little less similar to a person whose ancestors are mostly from Europe or China. The differences are subtle, and absolutely do not justify discrimination based on those differences in everyday life, but if certain diseases were more prevalent in China than they were in Europe, for example, people in China may have evolved slightly differently as a result. For example, persons from sub-Saharan Africa are more likely to get sickle cell disease, and Ashkenazi Jews are prone to get Gaucher disease. Thus, it seems to me that it may be important, when conducting vaccine trials, to try to have persons whose ancestors are predominantly from a variety of places, because the responses to the vaccine may vary. Skin color is not a perfect way of achieving that variety, but I suspect it's a reasonably accurate way of doing it. Thus, it seems to me that Dr. Collins was right and Bernstein was wrong.
(1) Ashkenazic Jews are a great example. They had a genetic bottleneck that makes them genetically distinct in various ways that affect certain diseases. Like other distinct, genetically isolated populations, the government doesn't require or even encourage researchers to look at them separately in things like vaccine trials, but it does make researchers look at "Hispanics" and the like, who are genetically very, very diverse.
(2) mRNA is very different from other vaccines.
(3) Collins, you will note, did not justify his decision on a scientific basis about differences in vaccine response, but in how the public would or would not accept the vaccine.
(4) When the government determines whether "enough" minorities are subjects of studies, they don't look at whether there is a sample size that allows for statistical significance, but whether the sample size more or less matches the demographics of the USA. I.e., the subjects should, crudely speaking, "look like America." This is obviously not a scientifically based methodology for determining who should be subjects.
1) The different racial/ethnic groups are all diverse, with different relative distributions several critical genes. We should endeavor to have the best distribution compared to our target population as possible.
2) It is. It's in fact, novel. We just don't know any potential interactions a priori. Which is why it's important to test as widely as possible, and not leave our major racial groups.
3) Collins, bless his heart, was likely worried about a differential response in a major untested ethnic group. He was well aware of the fact the different ethnic groups have different relative distributions of certain genes. If one of those genes had a critical interaction with the COVID vaccine...and it was missed because the trial just didn't bother to test that group. It would be an ethical disaster. As well as potentially costing thousands of lives....
4. In a clinical trial, your cohort should match your target population. That's just good science. To leave out large chunks of a grouping from your trial, but then include them in the target population is bad science, and leads to poor results.
I want to address point #4 here.
Bernstein's point about statistical significance vs population is what is important. If the subgroup sample size is sufficient, then it does not matter what that subgroup's size is in relation to other groups.
For example, suppose the vaccine was tested on 100,000 whites and 100,000 blacks. Results come back, and look perfect! Would you object, and demand that another 700,000 whites be tested?
Note: Leaving a subgroup out entirely would be bad, which is why I mentioned that the subgroup sample size needs to be sufficient above. This also leads to Bernstein's other point - that the subgroups do not actually measure all the known different genetically differing ethnic groups.
Additionally, you seem to be assuming that your "target population" has the US demographics. But biology has no borders, and for a vaccine that will be sold worldwide, why are you looking on at one nation? Under your suggestion, wouldn't you want to represent the world's demographics, rather than just an outlier nation like the US?
Now, there are places where your sample subgroups needs to match the population you draw from. However, that's the case where your sample is used as a whole to represent the entire population - opinion polling is an example.
Most experimental design does not do this, in fact due to differing treatments, different subgroupings cannot be combined in many experiments.
Re: Subgroup size and target population.
The issue here is, the groups are not heterogeneous. They include various subgroups...White Americans include Jewish Americans, Arab Americans, Italian Americans, and Irish Americans (among others). That doesn't even count the relative distribution of rare mutations in enzymes. It's impossible to accurately assess "all" subgroups. You do the best you can, with the resources you can. The best way is to use a representative sample.
Re: Target population.
Keep in mind, this clinical trial is for U.S. approval by the U.S. FDA. Because of that, the target population for approval, for this clinical trial...is yes, the U.S. population. If other countries want to jump on it and not bother with their own clinical trials, then that's their choice. But the US FDA has a responsibility not to the world population, but to the US population.
Here I have to agree with Toranth:
" If the subgroup sample size is sufficient, then it does not matter what that subgroup's size is in relation to other groups."
The appropriate judgement is whether the sample size is sufficient in size and whether the overall trial is appropriate designed to assure statistical significance and absence of significant systematic bias.
... are you supporting Bernstein's argument, now?
As he has pointed out, and you have now repeated, "whites" and "blacks" do not actually correspond to distinct homogeneous genetic groups. If Collins's desire was for additional data on underrepresented genetic groups, demanding more "blacks" would not solve the problem.
Remember, the entire point of this article is to point out that Collins delayed approval of the vaccine, because he wanted a more "diverse" set of subjects, even though the government (which would have included him if he was basing his decision on science) knew that the categories he was using had no biological basis behind them. From the article:
And remember, the FDA uses data from foreign medical trials - 10 years ago, more than one-third of all trials used by the FDA were conducted outside the US, and the percentage had been increasing every year. To suggest the FDA should be deliberately requiring testing to model the US demographics as of this year (and not, say, 1962) while rejecting outside trials seems contrary to both existing policy and the trends over the last several decades.
What utter nonsense.
"1) There was (and is) no scientific reason to think that the Modern vaccine would act differently on people of different genetic backgrounds; (2) Even if there was reason to think it would, the categories the NIH requires researchers to use--African American, Asian American, Hispanic, Native American, and White--are extremely internally genetically diverse.*"
Firstly, of course there is; such genetic differences are common, and are why we balance samples. Secondly, the 'required categories' are guidance to ensure a representative sample is obtained.
It is absolutely unquestionable that it was essential, as a matter of basic science, that the group chosen as a sample was representative.
Yes, lives were lost while we were proving vaccines work. But that doesn't mean we just skip the process.
This is a truly execrable post from Prof Bernstein, and he should retract with an apology. A classic example of an expert in one field being a layman in another.
"Firstly, of course there is; such genetic differences are common, and are why we balance samples."
LOL! Really? Tell me why we would expect a multiracial group like "Hispanics," with origins in Africa, the Americas, Asia, and Europe, to be a group that will have genetic differences from, say, Americans--who have origins in Africa, the Americas, Asia, and Europe.
That's an argument for finer grained diversity policies in drug trials, not against having them at all.
One of the fun facts we learned when my wife got breast cancer was that you don't just have 'breast cancer'. They take the tumor and sequence its genes, and you find out whether your particular flavor is 'estrogen positive', 'triple negative', or what. There is a whole taxonomy, and the prognosis and treatment are very different for the various types.
So medicine absolutely needs to realize that people come in a rainbow - or maybe Jackson Polllock painting - of genotypes, and that that variation can make a huge difference in appropriate treatment. I don't think this is controversial at all.
However, to follow the breast cancer example, if you are 'triple negative', you need a certain kind of treatment. It doesn't matter whether you are Italian, Irish, or Inuit, Muslim or Methodist. And I think Mr. Bernstein's point is that the categories people are usually talking about when they say 'diversity' aren't very good proxies for the genotypical differences that matter.
(as an aside, also, the 'finer grained' can be pretty statistically challenging, alas. We don't really have a big enough sample size to know whether the vaccines carry a small risk of myocarditis in general, much less whether they have that risk for some rare genotype. It's a noble objective, but one you may not generally be able to accomplish while trying to roll out a vaccine in a timely manner during a pandemic)
The chapter in one book on the general subject argues that the government's (NIH's and FDA's) focus on "diversity" in terms of crude Directive 15 categories has slowed progress toward research and medical practice focusing on relevant genetic differences. So, yes, genetic diversity should be studied. But genetic diversity isn't "diversity" in current FDA and NIH regs.
David,
You seem driven by political considerations to overlook the nuances of experimental design to minimize systematic error in the final determination.
The FDA and NIH diversity rules had and have no underlying scientific rationale, so there is no theory of why they should be the way they are to overlook. At most, one can defend them as necessary to increase public confidence that scientists are not neglecting the health concerns of the official minority groups, bc people incorrectly believe that they need to be “represented” as subjects of studies.
The whole thing is stupid. It starts with the presumption that the vaccines are good for people. They are dangerous, and the government has been trying to downplay the danger. It appears to be cumulative. It looked suspicious when immune system component assays would crash after the second jab (but not significantly after the first). But now, people are starting to drop like flies after their third jab. It appears to be hitting athletes maybe the worst. It appears to be an accumulation problem. One of the inventors of the mRNA vaccines theorizes that the vaccines cause gross overreaction of the immune system when getting the vaccine for those with natural immunities, which seems to make sense. For those already with full immunity to the virus, getting hit by a massive number of spike proteins from the virus (created by the vaccines) would logically trigger a massive response.
The other insanity here is that there is a very significant racial component as to the fatality rate, according to the CDC figures. If I remember correctly, the rank order is (worst to best results): Pacific Islanders: Native Americans (and their Alaskan brethren); Hispanic; Black; White; and Asian (including the Chinese who almost assuredly created the virus). What I haven’t seen is the racial breakdown on fatality rates, factoring in known comorbidities, and in particular obesity and diabetes. That may explain much of the racial difference in death rates. On the flip side, tens of millennia of separate environmental based evolution has caused significant differences in the psychologies of the different races. For example, NE Asians (Orientals) developed compact frames to protect against prolonged cold. Caucasian, and esp N Europeans, were not far behind. Blacks, on the other hand, evolved in a more resource rich environment, where size could be advantageous. Etc. Moreover, there do appear to be slight racial differences in the critical ACE2 receptors. The reality is that race exists, we all instinctively know it, and it is silly to pretend like it doesn’t. We can hope that these racial differences disappear over time, but we aren’t there yet.
I maybe should have been more clear. There appear to be two issues with the vaccines, and they both appear to involve immune system responses to the spike proteins generated by the vaccines. The purpose of the vaccines is to teach our immune systems to recognize the virus through those spike proteins. Each vaccination operates effectively as a large dump of these spike proteins. In the case of 2nd and 3rd jabs, our immune systems are already sensitized to these spike proteins, and react appropriately. And since the number of spike proteins indicates a massive viral attack, our immune system acts, or, rather, overreacts, appropriately. And, it very much appears that something similar occurs when those with natural immunity are vaccinated - the massive spike protein dump results in an overzealous immune response.
One of the compounding factors appears to be that the spike proteins are not staying in the muscles where they were injected, but appear to also be migrating to other inconvenient locations in our bodies, where they accumulate (to attract an immune response, in immune systems that already recognize them). Some of these inconvenient locations appear to be the heart muscles and germ cells. This isn’t, of course, a problem with the first jab to a COVID-19 innocent immune system. But appears to be, when it isn’t.
Much of this is beyond me. Here is a link to a recent CME conference. There are a number of interesting lectures included. In particular, I recommend the one by Dr Richard Malone, one of the invents of the mRNA technology utilized in the vaccines.
https://globalcovidsummit.org/news/watch-talks-from-leading-physicians-at-the-florida-covid-summit
My point there is not that the sky is falling, but rather that there is no real consensus in the experts, and that the vaccines may well be more dangerous than we are being told.
Among the links from your 'source' is a youtube video talking about Ivermectin as an effective and cheap COVID treatment.
Maybe you should choose your sources more carefully lol.
Or maybe you should. Ivermectin does appear to work (look at India, and several other countries around rather world) - as a prophylactic, and not as a therapeutic, which is how it was tested. The problem is that it apparently requires a week or two of taking the drug to build up the levels required to fight viruses, which means that it does not work when you start administering it after someone is admitted to the hospital, already testing positive for COVID-19.
Keep trying.
"It works, but not how the studies have tried to use it."
LOL. Nope.
https://www.bbc.com/news/health-58170809
Richard W. Malone? The guy who wants to sue Pfizer, CNN and Sesame Street under RICO for promoting the vaccine (that he also claims isn't FDA approved)?
Professor Bernstein, the title of your blog, How Many People Did NIH Director Francis Collins Kill? is harsh, and there is an implied accusation that Dr. Collins has 'blood on his hands'. In our tradition, the murder of a man's reputation is nearly as bad as murdering his body. Is it possible that Dr. Collins simply made a mistake, absent any intentionality?
I would ask that you consider changing your blog title to something less harsh, if the possibility of Dr. Collins having made an honest mistake is a plausible one.
I see your point, but the fact that he *laughed* about this in his NPR interview, ho ho ho, it was only a week or two delay, and we got diversity!, leaves me lacking empathy, just as he apparently lacks empathy for the victims of his decision.
"just as he apparently lacks empathy for the victims of his decision."
That is a crass character assassination attempt, unworthy of a serious scholar, which in this area I am sure that you are not.
Exactly this, Don Nico. At least any ambiguity about who David was has been removed. He's this guy.
He didn't used to be.
So much more the pity. He should never have tried to out Josh, Josh.
No, race could not matter as a matter of biology.
With conventional medications there might be differences in uptake, distribution and clearance (pharmacokinetics) or in biologic response to a given effect site concentration (pharmacodynamics). But even with aspirin, antibiotics, and beta blockers, the mean difference between any two randomly selected individuals is likely to be greater than the mean difference between any two randomly selected race populations. Inter-individual response differences can be important, but race is a poor way to predict such important differences.
With an mRNA vaccine, the effect is to cause host cells to manufacture proteins according to the mRNA coded sequence.
This action is conserved across pretty much any organism, and efficacy of this vaccine is clearly demonstrated in any number of mammalian species(just google veterinary and zoological use of covid vaccines). If a house can can develop robust neutralizing antibodies as readily as a human, then no reason to spend great resources looking for politically motivated differences between racial categories of humans.
Second big problem of including race sub-groups within a study is that it will virtually always result in insufficiently small numbers of sub-study participants to make meaningful statistical comparisons.
A study has a pre-designated number of participants in order to measure an effect size (e.g. get vs not-get symptomatic covid) based on how big of an effect is deemed important (80% target was the goal), and how often the event of interest (getting covid) occurred. Covid was an infrequent event, in the order of a few percent of the study population during the study period of a few months.
So the problem with sub-groups is that with AA at 12% of the US population, one would have 1/4th the likelihood to discern an effect due simply to being a smaller total number than the total study population (statistical effect moves with the square root of the number studied, hence 1/4 and not 1/8). This would lead one more often to Type II error, incorrectly stating that an intervention had no effect, more often in small populations.
And thus it is, if we were to take the original data, and run the numbers on any small sub-population, that results could go from clearly robustly effective overall to mathematically non-significant as a result of too small a study population. So just like AA might be mistakenly labeled as ineffective simply due to being a small population, so too would any other biologically implausible sub-group. Left handed people(10%), vegans(5%), and methodists (3%), would all likely show the same Type II error simply because their numbers studied would be insufficient to show differences that rose above the background noise that the analysis is designed to differentiate.
Wait a minute.
IIRC, and I do, the accuracy of results from a sample depends on absolute sample size, not on what percentage of the population is sampled. IOW, a sample of say, 1000 voters, is just as good or bad in a national election as in a statewide one, despite the fact that there are many fewer voters in the latter.
So f you want to be sure you've measured the effect on some group you just want to be sure you have "enough" - in absolute numbers - members of that group in the trial. I don't know what "enough" is in this context, but I bet there are those who do.
So let's say 5000 is enough. Then if you have 5000 Blacks, or whoever, you have enough, whether you have 5000 whites or 50,000. If you do have 50,000 whites your results for whites will be more reliable - smaller error - but the only way to equalize that is to have 50,000 Blacks as well. This gets impractical in a hurry.
Further, all the talk about whether genetic diversity is worth worrying about misses a point. Different races) may vary in ways other than genetic variation - cultural practices, socioeconomic condition, types of jobs, etc. It's worth checking that, because it may be a clue to something other than race that affects outcomes.
You know this, but the "accuracy" depends on both. If your absolute sample size is only 1, but your population is 1, then your accuracy is 100%.
And for this case, it may depend most of all on the quality of your assumptions around homogeneity of your population-- which after all is the point being argued: could blacks be different from whites for the purpose of testing a new vaccine?
This article was a hit piece. You can't dish up statistically problematic problems as red meat. I don't like government hacks, and Collins probably is one, but the actions Bernstein decries here as virtually indefensible are not. They have some arguable merit, for all the reasons others have posted.
" You can't dish up statistically problematic problems as red meat. "
Welcome to the Volokh Conspiracy, first-time visitor!
(The Five Satins supplement Southside there)
the "accuracy" depends on both. If your absolute sample size is only 1,
Yes. Of course I know that. I was talking about the situation where the population is large relative to any practical sample size, which is usually the case when we rely on sampling.
And for this case, it may depend most of all on the quality of your assumptions around homogeneity of your population-- which after all is the point being argued: could blacks be different from whites for the purpose of testing a new vaccine?
I don't think I ignored that. Note that I said you needed enough members of any group you are interested in looking at.
"So f you want to be sure you've measured the effect on some group you just want to be sure you have "enough" - in absolute numbers - members of that group in the trial. I don't know what "enough" is in this context, but I bet there are those who do."
The easiest, though not always the only, way to satisfy FDA and NIH bureaucrats, is to closely match your study numbers to the US population. So "enough" Hispanics is 18% (with no differentiation between, e.g., Hispanics who are 100% European and those who are 100% Indigenous), African Americans 14% (no differentiation between, e.g., Ethiopians, sub-Saharans, people with mostly Caucasian ancestry but who are socially considered African American), 7% Asian (no differentiation etc) and Native American 1% (you get the drift on differentiation). This obviously has nothing to do with science.
David,
Every time you respond, you reveal an deep ignorance of experimental design and your passion for focusing only on US political considerations. That combination certainly detracts from any credibility you might have on the topic.
But to restore our confidence, please reveal the degree to which you have studied epidemiology and biostatistics. Then we'll judge for ourselves.
Yup.
"But to restore our confidence, please reveal the degree to which you have studied epidemiology and biostatistics. Then we'll judge for ourselves."
Says the anonymous commentator...
This isn't about epidemiology or biostatistics. Around 20 years ago, Congress, responding to political pressures, ordered the FDA and NIH to require researchers to collect data on research subjects by race and ethnicity, which meant they also had to recruit people on that basis. FDA and NIH *could have* proceeded, in a scientific manner, to figure out how to comply with this directive in a way that would be consistent, or as consistent as possible with good science. They did not. They held no hearings, did not research, consulted on experts in genetics. Instead, they took the politically easy way out of just adopting the standard Directive 15 categories, which were never meant to have scientific purchase, and do not, and imposed them on researchers. They have never sought to defend this decision scientifically. The best you will get if you ask is some mumbling about how this is necessary to ensure faith in the system by minority groups. And yet, what these agencies are actually doing is reinforcing the notion that these scientifically arbitrary categories in fact reflect biological reality--why else would they need to be used in scientific studies? So it's a disaster scientifically, and ideologically (if you are against racism).
In short, if you can find me any reputable scientist who actually thinks that the Moderna mRNA vaccine could plausibly be thought to have distinct effects on the categories of people we call "Asian," "Hispanic," "whites," etc., then we can debate their scientific presumptions. But since no one, including Collins, ever believed this, we are not debating science, but politics and ideology masquerading as science.
David,
First, you're the one the raised the issue, so the burden is on you.
Second, how did you go from "I wrote a chapter in a book, I'm an expert!" to "well, I'm not aware of any reputable person that thinks X"? Are you an expert or not? Of course, not, but it'd be nice to hear you acknowledge it.
Third, even accepting that you happen to be right about the science, you continue to assiduously avoid the fact that Collins faced not only a scientific issue but also a messaging issue and so he had to make the decision based on the environment as it was, not as it ought to be or as you think it ought to be. Hence, ensuring the clinical trial included a diverse population likely increased vaccine use by people of color which saved lives. You utterly fail to account for this (other than saying Collins is somehow guilty of creating that environment despite multiple reasons black people are wary of medical establishment, e.g., Tuskegee).
Fourth, actual scientists do disagree regarding whether it is scientifically useful to have diversity in clinical trials, including for reasons other commenters have pointed out (self-identified race may correlate with genetics, geography, culture, environmental exposures, etc.). See Oh SS, Galanter J, Thakur N, Pino-Yanes M, Barcelo NE, White MJ, et al. (2015) Diversity in Clinical and Biomedical Research: A Promise Yet to Be Fulfilled. PLoS Med 12(12). Table 1 provides numerous examples of how research on diverse populations revealed both genetic and non-genetic differences in disease development/response/treatment between self-identified racial groups.
Fifth, it's clear you knew you couldn't throw down a broader gauntlet, so you asked for evidence that there was a known difference in response to mRNA vaccines for Covid-19 vis a vis broad racial categorizations. But that's the point of clinical studies, you don't necessarily know until you do the study. And there are examples where odd, "racially" specific response were found precisely because diverse populations were studied.
Stop embarrassing yourself defending this stupid post.
If you actually knew anything about how and why racial classifications are used in medical research via government dictate, you would realize it's you who is making embarrasingly ignorant comments.
David's standard when caught out: Deflect and insult. Avoid engaging on substance.
My comments had nothing to do with how and why racial classifications are used in medical research, though, for what it's worth, the origins far predate the Congressional response in the early 2000s which you raised and do, in fact, include concerns about scientifically valid, evidence-based medicine particularly including whether the homogeneity model resulted in disservice to various populations. And it did.
Notwithstanding there were scientifically legitimate reasons to revise the white-male homogeneity model of research, there are valid points to be made about the right way to ensure clinical trials reflect the diversity of the country and ensure the medical issues of all subpopulations are adequately considered and served, but you are doing a really, really bad job of it. Mostly because you can only see it through your own ideological lens and feel the need to do it in the most insulting way possible.
edit: "My comments had nothing to do with how and why racial classifications are used in medical research via government dictate..."
The only relevant how and why in this particular case is whether there was a scientific basis to believe different races/ethnicities may be affected by the covid vaccines and, crucially, the fact that the world as it stood when Collins made his decision was one in which it would be more difficult to convince certain populations that the vaccine was safe for them if people like them were not included in the study. You can bellyache about how its terrible that the general public is scientifically illiterate, but the fact remains. And Collins didn't cause it.
Including a diverse population almost certainly increased confidence in the vaccine which almost certainly saved lives. Now, whether it saved more lives than the two weeks during which some people would have been vaccinated (allegedly) but for that particular change is open for debate, I guess.
But you're the type of person who accuses someone of killing people when they made a practical decision (which very well may have, on net, saved lives) based on the world we actually live in and, when called on it elsewhere in this thread, said you don't mind so much that it was terribly unfair because you, David Bernstein, had determined that he was a bad person because he laughed in an interview once. Gross.
David,
Duck, divert, and duck again.
You are the one who proclaimed yourself an expert.
You are the guy who obviously have no expertise in experimental design, statistics, or epidemiology.
You are the one who criticized Collins take on the statistical makeup of the trials.
You are the guy who ignores the fact that Collins is a distinguished scientist with true expertise in the matter
You are the guy who launched character aspersions
So what that you don't know my name. I am on a faculty of a better university than you-- not that that matters.
So get off your high horse and show a tiny bit of humility
Just to make one point:
"You are the guy who obviously have no expertise in experimental design, statistics, or epidemiology."
Francis Collins, former director of NIH, has a degree in molecular genetics from 1974, and has been a bureaucrat for at least 40 years. Assuming he has, or ever had, any expertise in "experimental design, statistics, or epidemiology" is quite generous.
It is very plausible that Bernstein could be self-taught and still have more expertise than Collins.
I'm an expert on how and why racial classification is used by the government, including in medical research. In the latter, the way it's used has no basis in science, and no one in authority has ever claimed that it has such a basis, and no one would, unless they would be willing to defend the proposition that genetics maps the arbitrary categories used. So you are wildly missing the point.
"I'm an expert on how and why racial classification is used by the government"
Not true.
"the way it's used has no basis in science, and no one in authority has ever claimed that it has such a basis"
Not true. (See my citation elsewhere and it is very easy to find multiple quotes from actual scientists who have claimed it has a scientific basis.)
"no one would, unless they would be willing to defend the proposition that genetics maps the arbitrary categories used."
This is too stupid to even bother saying it isn't true. Elsewhere in this thread it has been pointed out multiple times by more knowledgeable people than me and research relating diverse populations in scientific studies has shown, for at least some types of research, it is quite important to have diverse samples for reasons having absolutely nothing to do with genetics. The fact that you are unaware of this easily googleable fact demonstrates, beyond doubt, that despite your impressive credential of writing a chapter in your own book, you are not an expert in this subject.
You are confusing the issue of the desirability of having a diverse pool of research subjects with the desirability of using the percentage of Americans who belong to arbitrarily defined official “race” categories as a proxy for diversity. Just for example, one satisfies the government if one uses only basques from northern Spain for the Hispanic category, only Somalis for the black category, only Armenians for the whie category and only Koreans for the Asian category. Not only are the categories themselves arbitrary on terms of diversity considerations, but there is no requirement or encouragement of seeking diversity within the categories. You can easily wind up with a less genetically diverse pool this way than if you just took people who signed up at random. (Indeed, I recently asked the author of a study on “Asians” and Covid whether he knew the ethnic subgroup of the Asians. They were 100 pc Chinese. Zero reason to think the results would apply to Pakistanis or ethnic Malays.
David,
I didn't confuse anything. I never said it was "desirable" to use "race" as a proxy for scientifically relevant diversity.
My main focus has always been that Collins' decision, as he says, was primarily motivated by the fact that certain population groups would be less likely to get vaccinated if the clinical trials did not include people like them. He was almost certainly right about that. Therefore, that aspect of the decision saved lives. Likely many thousands of lives. You give disingenuous reasons for not including that in your equation.
We agree that it is better to focus on actually relevant aspects of diversity (socioeconomic status, health conditions, age, geographic location, diet, and genetics, among others) than proxies for those factors. And it can also be desirable, in an appropriate case, to control as much as possible for certain types of diversity (because sometimes what you really need is for all people in the study to be as similar as possible). But it is a bridge too far to go from those propositions to the idea that the push for diversity in clinical trials had no scientific basis. It did. Has the push for social justice combined with scientific laziness overwhelmed the valid scientific justifications. Perhaps. It is also going too far that racial categorizations can never be useful proxies (given things are being recorded by race, again reiterating that race is a social construct that has no independent scientific meaning but may correlate with scientifically meaningful factors).
Your anecdote is great, but your focus on that to the exclusion of the faulty logic you used in smearing Collins' character (and stated indifference about your acknowledged unfairness) demonstrates that you have a similar blindspot to the people you are criticizing. The world isn't all hot or all cold. Racial classifications are generally bad and far more harmful than helpful. But they do exist in the real world and analyzing the data that is collected based on race can reveal actual problems or scientifically-valid concerns.
Be a better person and a more logical thinker.
"I'm an expert on how and why racial classification is used by the government"
As a self-anointed expert expert you will be pleased to tell us the citations for the peer-reviewed papers that you have published on this topic.
Toranth,
Collins was not made a member of the Academy of Medicine and the Academy of Sciences for being a bureaucrat. You may call leading leading the National Genome project being a bureaucrat, but you would be dead wrong. Collins has 780 publications with an h-factor of 180, and a total of 208,000 citations. !2,200 citations were in 2017 alone.
It is extremely implausible that Bernstein understands anything about experimental biology better than Collins.
Claiming that just lowers your own credibility.
"Different races) may vary in ways other than genetic variation - cultural practices, socioeconomic condition, types of jobs, etc. "
I don't get why race is the preferred proxy for cultural practices, socioeconomic condition, or types of jobs. If, say, vegans are more susceptible to covid-23 because their diets typically don't contain enough vitamin Z, shouldn't you be looking at vegans vs. non-vegans? If welders are protected from some disease because of the extra zinc they inhale, shouldn't you just analyze by occupation?
You are looking at sets (hispanic, vegan, welder), (hispanic, vegan, taxi driver), (hispanic, omnivore, welder), (hispanic, omnivore, taxi driver) and their not-hispanic counterparts and saying 'hey, let's look at the ethnic angle'. It's not clear at all that ethnicity matters more than diet or BMI or smoking history or amount of exercise or latitude or blood type or whatever.
Because it's available data, whereas say vegan largely isn't.
"What are you looking for?"
"I lost a quarter in the bedroom."
"Why are you looking in the kitchen?"
"The light's better in here."
"Because it's available data, whereas say vegan largely isn't"
So, when conducting a clinical trial, is it unusual to give patients a form asking about age, diet, exercise, smoking, medical history, etc? I would expect that to be the norm, but I have never run any clinical trials.
But if you do ask those questions, how is it easier to determine whether someone is Hispanic than whether they are obese?
What I think is going on is a confusion between public health and clinical medicine. Clinical medicine, sometimes to a fault, doesn't deal with demographics, it deals with biological systems. If culture or class doesn't change the uptake of Factor B, it's not going to be part of a clinical study. Race, as we're finding more and more, sometimes does have clinical differences.
Maybe this should change - I think we need more coordination between the public health and basic medical research communities - but that's what's going on.
"Race" does not have clinical significance. Genetic attributes that may be correlated with race occasionally do. Sociological factors that are correlated with race, but are also independent with race, also do. But race does not. There is in fact a movement among physicians to make race, as such, clinically salient, and it's extremely dangerous, for reasons I discuss in the book, some of which are obvious, some much less so, but you will have to read it to get the full story.
Genetic attributes that may be correlated with race occasionally do.
There's the rub - we don't know. That's why trials are needed!
I think part of this is an issue of scope. It has been generally established that medical trials that don't pay attention to race and sex have resulted in some deadly dosing issues at the very least. Maybe this is due to genetic correlation - we don't know. Parma studies are phenomenological, and don't seek causation.
So you're taking a broad clinical policy, created after some specific cases, and saying that because after the fact it didn't seem to matter for this specific case, that the requirement was incoherent. I don't think you can change the scope like that. Because again, we didn't know.
And then there is the fact that even if it is incoherent, it doesn't seem to have been related to any real world delays.
Here's the rub: the categories required by the government don't correlate to any plausible definition of "race". If you found out that "Hispanics" had a slightly worse reaction to the vaccine, what useful scientific information would that give you? Could be a Spanish thing. An Italian thing. A Jewish thing. A Miztec thing. A Kenyan thing. And so on. You don't know the internal genetic makeup of the Hispanic group, so you have no idea whether this was a very strong reaction from a small cohort within the group, or a mild reaction in a large cohort within the group, or a mild reaction among several subgroups... Or, just random chance. If you were really trying to determine whether different "genetic clusters" of people had different reactions, you would do that directly. This is just politics/pr.
But your OP doesn't say 'fix the demographics of NIH study design requirements.' I'm not sure you even verified what the NIH study design demographic requirements were.
Instead, you seem to take issue the very concept of racial demographics being a factor.
NIH and FDA never justified using race in this context on a scientific basis. They justified imposing race because HHS told them too, and HHS did so b/c Congress told them to, and if you trace it all back the primarily underlying issue is health outcome disparities among groups. So they wanted to make sure that when researchers study health issues, they also identify health disparities and hopefully study their causes and possible solutions. This is problematic for a variety of reasons, but is at least within the realm of reason, because it's using the racial categories as proxies for sociology. But the rules are the same for biomedical studies that in effect use the categories as proxies for genetics, which is not reasonable.
David,
This is the most cogent and reasonably point you have made in this entire thread. Well said.
Too bad you didn't lead with this instead of smearing Collins as a killer. With this statement, you demonstrate that you don't have to be a click-bait writing hack. Unfortunately, when you wrote this post, you consciously chose that route. A pity.
I remain unsure that racial classifications are what you think they are in these studies. NIH has a whole office dedicated to this issue of study design, so I'm not sure the classifications you're relying on apply.
What this looks like to me is HHS telling NIH which regulations it had to continue to abide by and which it could make exceptions for, not HHS creating whole new mandates.
It's available data because the government has required that it be collected. If the government required researchers to collect data on whether people are vegan, then that data would be available. And since occasionally by random chance studies would show that vegans are more susceptible to X, vegans would become keenly interested in ensuring that they were always represented in studies to ensure that things like "X" aren't happening. And then the government would justify its vegan rules on the grounds that if enough vegans weren't included, vegans wouldn't trust the studies.
Bernstein posts the most race-baiting moronic crap on this site and he had the nerve to pretend there aren't anti-semitic posts regularly in the comments here. What are you a professor of, Bernstein? Being a complete fucking moron?
Just for the hell of it, let me add another critique of the study methods. My rheumatologist tells me that clinical trial data are no help to me for predicting possible responses to Covid vaccines. Apparently, all people with auto-immune conditions such as mine were screened out from the trials.
I can see why you might want to do that. If you included me in the trials you would have a challenging task to sort from the results my likely idiosyncratic symptoms. Or, more pessimistically, you might encounter the challenge to explain why one identifiable class with adverse responses does not show a risk to others.
But it is frustrating to have to decide about taking a vaccine on the basis of a potentially dangerous guess, balanced against a likely valid and reassuring assessment of the balance of harms. Doing that does diminish trust, or at least it would if you didn't think (or hope) you understand why it was done.
I know someone with immune system issues in his immediate family tree who got the vaccine and soon thereafter acquired an immune system disease. His doctors think the vaccine is responsible, though of course you can never prove it in an specific instance.
It's been at least a couple of years since I asked, so here's my periodic check-in: David, do you still consider yourself non-partisan?
lol
At a right-wing publication whose flag includes "libertarian" but not "conservative," precisely how much do self-descriptions matter?
I have no idea what Collins' party or ideology is, and I have similarly no idea why you think criticizing him for insisting on the use unscientific "racial" categories in a way that delayed a life-saving drug is a "partisan" issue. Unless you are saying that I'm an anti-Republican partisan, because what Collins did occurred during the Trump administration, under the watchful eye of HHS secretary Alex Azar, who I believe is and has always been a Republican. Not to mention that I just covered the incident in a book on racial classification that I today sent to be copy-edited by the publisher. Your partisan thing has always been nutty (you should see what actual Republican partisans say about me over at Instapundit), but this one is especially dumb.
(Nutty bc you seem to insist that “nonpartisan” means “has no strong ideological priors on the issues of the day” which (a) is not what it means and (b) is not something that I have ever claimed)
A coward and a moron
If I told you I'm a blogger at Jacobin, and the readers there drag me for not being Marxist enough, would you assume I'm non-partisan? Or would you think that I'm probably quite left-wing, though maybe not enough to satisfy the kind of doctrinaire purists who consider everyone right of Bernie Sanders ideologically fungible?
Exactly. And that would be no less so if I'm a registered Independent and never express approval to any particular left wing party or movement; I just happen to send the overwhelming bulk of my condemnation and contempt in the other direction.
The fact that you think taking crap from the regulars at Instapundit, where you blog, is persuasive evidence of your non-partisanship is a bright flashing tell of the blind spot in your self-perception.
Almost forgot. I don't know Collins' party affiliation either, but why does it matter? What's Liz Cheney's party? Jim Comey's? The former VPOTUS who the 1/6 rioters wanted to hang for not joining their insurrection? Francis Collins is a charter member of the establishment elite which salt of the earth GOP populists like Cruz, Hawley and Cotton (which schools did they go to again?) smear as existential enemies of freedom.
Just out of curiosity, how many people do you think Tucker Carlson has killed by discouraging vaccination?
> There is no *scientific* reason to use these categories as proxies for genetic diversity
Even so, there is an *ethical* reason to consider diversity in clinical subject groups. It's the same reason why it would be unethical for Harvard to cross-reference a donor list when deciding who to admit to its clinical trial.
I should also point out that while categories such as "Black" are indeed highly genetically diverse, it does not follow that we should expect vaccines to be equally effective in Blacks as a whole.
Consider for example sickle-cell disease, which is far more common in Blacks than other categories. That's because the subgroups that happen to carry the sickle cell gene are all more likely to be Black. So if the vaccine reacted differently against that gene, then Blacks as a whole would also have a different response to the vaccine.
And in fact it is well known that Blacks tend to react differently to certain blood pressure medications. That doesn't mean that Blacks are all genetically similar, but it does suggest that there are Black subgroups with different responses that must be accounted for when developing new medications.
As an analogy, a drug trial that omits Blacks is like a presidential election poll that omits everyone on the East Coast. While it's true that East Coasters are highly diverse and any randomly chosen East Coasters might have nothing in common, you can't safely assume that East Coast opinions can be generalized from the rest of the US. East Coasters are diverse but nevertheless statistically different from other diverse groups, like Midwesterners. And the East Coast contains unique enclaves (Manhattanites, Miami Cubans) whose responses may not be correspond to any respondents outside the East Coast.
How would an extra two weeks of availability for the Moderna vaccine have had any effect? Pfizer's vaccine, also mRNA, was available in December, and the federal government distributed 12.4 million doses but only 2.8 million were administered. The bottleneck was in distribution, not authorization of another vaccine, and the blame is not where Bernstein would place it.
Points 1 and 3 both appear to be incorrect - several posters have suggested scientific reasons for a diversity requirement, and the obvious rebuttal of point 3 is from past experience like the Tuskegee syphilis study. Point 2 is questionable; barring an attempt to skew selection, racial diversity will probably give genetic diversity since it's unlikely that random members of a given race will all be within one of the subgroups genetically more similar to a different race.
Another dishonest partisan blast from David Bernstein, and as usual shredded in the comments.