This story is kinda complicated. It’s simple, but it’s complicated.
The simple part is the basic story, which goes something like this:
– In 2020, a study was done at Stanford–a survey of covid exposure–that I publicly criticized: I wrote that this study had statistical problems, that its analysis did not adjust for uncertainty in the false positive rate of the test they were using, and that they did something wrong by not making use of the statistical expertise that is readily available at Stanford.
– In 2023, one of the people involved in that study wrote a long post about that study and its aftermath. In that post, my comments from 2020 on that study were misrepresented–hence the title of the present post.
– Just last week someone pointed me to the 2023 post. I was unhappy to have been misrepresented so I emailed the author of the post. He didn’t respond–I don’t take that personally: he’s very busy, the holiday season is coming, the post came out over a year ago (I just only happened to hear about it the other day), and my complaint concerns only one small paragraph in a long post–so, just to correct the record, I’m posting this here.
The more complicated, and interesting, part, involves the distinction between evidence and truth. It’s something we’ve talked about before–indeed, I published a short article on the topic back in 2020 using this very example!–and here it has come again, so here goes:
And now, the details:
Stanford professor Jay Bhattacharya wrote about a covid study from 2020 that he was involved in, which attracted some skepticism at the time:
Some serious statisticians also weighed in with negative reviews. Columbia University’s Andrew Gelman posted a hyperbolic blog that we should apologize for releasing the study. He incorrectly thought we had not accounted for the possibility of false positives. He later recanted that harsh criticism but wanted us to use an alternative method of characterizing the uncertainty around our estimates.
On the plus side, I appreciate that he characterizes me as a serious statistician. He also called my post “hyperbolic.” He doesn’t actually link to it, so I’ll give the link here so you can make your own judgment. The title of that post is, “Concerns with that Stanford study of coronavirus prevalence,” and I don’t think it’s hyperbolic at all! But that’s just a matter of opinion on Bhattacharya’s part so I can’t say it’s wrong.
He does have two specific statements there that are wrong, however:
1. It’s not true that I “incorrectly thought their study had not accounted for the possibility of false positives.” In my post, I explicitly recognized that their analysis accounted for the possibility of false positives. What I wrote is that they were “focusing on the point estimates of specificity.” Specificity = 1 – false positive rate. What I wrote is that they didn’t properly account for uncertainty in the false positive rate. I did not say they had not accounted for the possibility of false positives.
2. I never “recanted that harsh criticism.” What I wrote in my post is that their article “does not provide strong evidence that the rate of people in Santa Clara county exposed by that date was as high as claimed.” But I also wrote, “I’m not saying that the claims in the above-linked paper are wrong . . . The Bendavid et al. study is problematic if it is taken as strong evidence for those particular estimates, but it’s valuable if it’s considered as one piece of information that’s part of a big picture that remains uncertain.” And I clarified, “When I wrote that the authors of the article owe us all an apology, I didn’t mean they owed us an apology for doing the study, I meant they owed us an apology for avoidable errors in the statistical analysis that led to overconfident claims. But, again, let’s not make the opposite mistake of using uncertainty as a way to affirm a null hypothesis.”
I do not see this as a “recanting,” nor did I recant at any later time, but I’m fine if Bhattacharya or anyone else wants to quote me directly to make clear that at no time did I ever say that their substantive claims were false; I only said that the data offered in that study did not supply strong evidence.
This bothers me. I’d hate for people to think that I’d incorrectly thought they had not accounted for the possibility of false positives, or that I’d recanted my criticism. Again I emphasize that my criticism was statistical and involved specification of uncertainty; it was not a claim on my part that the percentage of people who’d been exposed to Covid was X, Y, or Z.
The good news is that this post from Bhattacharya appeared in 2023 and I only heard about it just the other day, so I guess it did not get wide circulation. Maybe more people will see this correction than the original post! In any case I’m glad to have the opportunity to correct the record.
Stanford contagion
I have no problem with Bhattacharya making arguments about covid epidemiology and policy–two topics he’s thought a lot about. It’s completely reasonable for him and his colleagues to say that their original study was inconclusive but that it was consistent with their larger message.
Bhattacharya also writes:
In the end, Stanford’s leadership undermined public and scientific confidence in the results of the Santa Clara study. Given this history, members of the public could be forgiven if they wonder whether any Stanford research can be trusted.
He doesn’t fully follow up on this point, but I think he’s right.
A few weeks before the above-discussed covid study came out, Stanford got some press when law professor Richard Epstein published something through Stanford’s Hoover Institution predicting that U.S. covid deaths would max out at 500, a prediction he later updated to 5000 (see here for details). I’ve never met Epstein or corresponded with him, but he comes off as quite the asshole, having said this to a magazine interviewer: “But, you want to come at me hard, I am going to come back harder at you. And then if I can’t jam my fingers down your throat, then I am not worth it. . . . But a little bit of respect.” A couple years later, he followed up with some idiotic statements about the covid vaccine. Fine–the guy’s just a law professor, not a health economist or anyone else with relevant expertise here–the point is that Stanford appears to be stuck with him. In Bhattacharya’s words, “Given this history, members of the public could be forgiven if they wonder whether any Stanford research can be trusted.”
This sort of guilt-by-Stanford-association would represent poor reasoning. Just cos Stanford platforms idiots like Richard Epstein, it doesn’t mean we shouldn’t trust the research of serious scholars such as Rob Tibshirani and Jay Bhattacharya. But I guess that members of the public could be forgiven if they show less trust in the Stanford brand. Just as my Columbia affiliation is tarnished by my employer’s association with Mehmet Oz, Robert Hadden, and its willingness to fake its U.S. News numbers. And Harvard is tarnished by its endorsement of various well-publicized bits of pseudoscience.
Reputation goes both ways. By publishing Epstein’s uniformed commentary, Stanford’s leadership undermined public and scientific confidence, and then Bhattacharya had to pay some of the price for this undermined confidence. That’s too bad, also too bad that he ended up on the advisory board of an organization that gave the following advice: “Currently, there is no one for whom the benefit would outweigh the risk of these [covid] vaccines–even the most vulnerable, elderly nursing home patients.”
The challenge is that legitimate arguments about policy responses under uncertainty get tangled with ridiculous claims such as that covid would only kill 500 Americans, or ridiculous policies such as removing the basketball hoops in the local park. On one side, you had people spreading what can only be called denial (for example, the claim that the pandemic was over in the summer of 2020); on the other side were public health authorities playing the fear card and keeping everyone inside.
So I can see how Bhattacharya was frustrated by the response to his study. But he’s missing the mark when he misrepresents what I wrote, and when, elsewhere in his post, he disparages Stephanie Lee for doing reporting on his study. We’re doing our jobs–I’m assessing the strength of statistical evidence, Lee is tracking down leads in the story–just like Bhattacharya is doing his job by making policy recommendations.
> he’s very busy
Yeah, he’s very busy being appointed to lead the NIH, so I think it’s important to set the record straight because, at least if memory serves, even in the revision, they didn’t seem to understand or deal with this issue.
Yeah, the interesting thing about this post is that it doesn’t mention Bhattacharya’s selection for NIH. I suppose it was written before that. What makes him scary in this position is not his particular views on Covid or even his apparently limited statistical understanding, but what might be called his Lysenko tendencies. What made Lysenko a disaster was not his mistaken understanding of evolution (esp. since our understanding has continued to evolve!), but his identification of particular research methods and outcomes with political commitments. That’s how research gets suborned. It appears that Bhattacharya and his colleagues are libertarians and therefore offended by public health requirements. OK. But that bias should be kept separate from research, and if someone can’t do that they shouldn’t be in a leadership position over more conscientious researchers.
Peter:
I don’t mind if Bhattacharya’s libertarian political views (if he indeed has such views; I haven’t looked into this at all) affect his take on public health policy. I’m sure there’ve been many public health policymakers with strong social-democratic political views, and that will affect their take too.
My view of the episode described in the above post is that, back in 2020, Bhattacharya and his colleagues had strong views about covid policy, they did this study which they analyzed quickly and without the help of statistical experts, giving them a sense of overconfidence which motivated them to promote their ideas heavily, and then they were surprised by the backlash. I don’t think the authors of that 2020 paper were trying to understate the uncertainty in their results; statistics is hard, and I’m guessing they were doing their best in a compressed time frame–remember that, back then, everything felt very urgent–and they messed up. In any case, the negative reaction came as a surprise, and I can see how it would be natural for Bhattacharya and his colleagues to not take the critical statistical analysis and reporting at face value but instead to think of it as being politically motivated. Then when he wrote that thing in 2023, I’m guessing he hadn’t thought about the episode for awhile and he was just recollecting a false memory that I’d incorrectly thought their study had not accounted for the possibility of false positives and that I’d recanted my criticism. Had I heard about that post when it came out, I would’ve notified Bhattacharya right then; as it is, I only heard about it last week so that’s when I informed him.
You’re very charitable, Andrew! Maybe I’m jumping to conclusions, but it looks to me like the initial biases in B’s research and then his stubborn attachment to it in the face of reasonable criticism reflects an instrumental attitude toward the politics of research. After all, if people just can’t acknowledge and reflect on the criticism that comes their way, either they have ego issues or are in the grip of some ulterior motive. Like you say, that can be found all around the political spectrum. And — I think this is also your position — a certain amount of ulterior motivation is probably a good thing, since it gets people through the grind part of the process. A theme of this blog has been that over-motivation, so to speak, becomes destructive however when it prevents people from listening to criticism and admitting error. In this case, we also have someone who is being elevated to a position from which he will be able to reward or withhold resources from others, which makes it that much worse.
“I don’t mind if Bhattacharya’s libertarian political views (if he indeed has such views; I haven’t looked into this at all) affect his take on public health policy.”
First, you SHOULD mind – a libertarian approach to health policy is *deadly.*
Second, you should read more about this guy before assuming any *hint* of good faith in his arguments/pronouncements. Are you really serious that you haven’t heard the BS he’s been spewing in courtrooms and the media for the past 5 years straight (??) Judges have thrown out his testimony *repeatedly* on issues related to COVID.
The narcissistic injuries he sustained when the medical community reacted in horror to the Great Barrington Declaration (of which he was an architect) and the Santa Clara study have clearly turbocharged his psychopathy. He’s proven himself, repeatedly, to be a complete monster- slipperier than an eel, unfathomably egotistical, and utterly unqualified to be anywhere near the levers of power. An absolute *disgrace* to the medical profession. Many people will die if he’s given any sort of health policy-related leadership role. You don’t owe him any apologies.
I expect there was a lot of bad science in several directions. Look at the wiping down of shopping carts for instance. Or the “social signal” masking. (The very best study, with large N and strong controls showed no benefit. But somehow meta-analysis of multiple small crappy studies showed masks were great.) I remember telling my sister who is a physician that it was interesting how public opinion seemed to be able to put a toe on the scale.
Confirmation bias is powerful. And the sense of emergency tends to lead to emotions trumping thinking. Look at post 9-11 with the Iraq. That intel wasn’t just wrong, like picking a bad horse. It wasn’t adequately “murder boarded”.
(I was actually heartened to read that Rumsfeld had listed the possibility of WMD being wrong, before we went to war. He was at least trying to overcome Dunning Krueger, not just be a midwit trend follower like Shrubya. He didn’t stop the rush to a mistake. But he at least had some intellectual humility.)
“Or the “social signal” masking. (The very best study, with large N and strong controls showed no benefit. But somehow meta-analysis of multiple small crappy studies showed masks were great.) I remember telling my sister who is a physician that it was interesting how public opinion seemed to be able to put a toe on the scale.”
What you didn’t do, of course, was try to convince your physician sister that masks don’t work. Because for her, actually paying attention to how the world works is much more important than posturing as a brilliant iconoclast.
Next time you are scheduled for surgery, tell the surgical crew that you don’t want them to wear masks. I’m sure they won’t mind. After all, those masks are not intended to send a social signal that they are protecting each other, the masks are intended to send a social signal that they are protecting YOU!
“Or the “social signal” masking. (The very best study, with large N and strong controls showed no benefit.”
Get a grip. How do you think so many physicians, treating a tsunami of patients with COVID managed to avoid infection? By holding their breath the whole time? Of *course* the right masks work, and *very well.* What *doesn’t* work is telling a selfish population, for whom being *told* to mask is a fate *worse* than death, that they *must* mask. Fix the culture, save lives.
Regarding “social signal” masking, this may not be a good blog/thread to get into this, but at the start of Covid, I got deeply interested in masking and all the different types of masks there were (remember, this was before N95! N95! N95!). Notably, all the places then talking about masks tended to be, hard to describe, maybe culturally right-leaning. It was DIY home renovation guys, woodworkers, men who worked in construction, etc. It would probably be wrong to assume they were all Trump voters, but it sure wasn’t a hipster vibe. It was really weird as masking got politicized, that all this practical knowledge just seemed to be stuck at this grassroots level, versus culture-war sloganeering.
匿名:
Reading through Bhattacharya’s post, I think this is an evidence-vs.-truth situation. Lots more information has come in since that early Stanford study, and Bhattacharya is convinced that the population inferences from his early study are correct, so it makes sense that he would feel that, retroactively, the criticisms of that study are invalid. I have not followed all the epidemiology here, so I offer no comment on the correctness of the claims. I think their paper overstated the evidence from their study, that’s all.
I interacted with Jay online – maybe around six months ago – and he informed me that after the initial criticism of the Santa Clara study, they subsequently published material proving that the criticisms of the study were wrong, and there weren’t any statistical errors in the initial publication.
I will note that based on the initial publication, he and Ioannidis went on a mostly rightwing media tour to argue the COVID IFR was on the order of the seasonal flu. That they drew that conclusion from the Santa Clara study suggests to me that there were, indeed, scientific problems with the study even if they weren’t strictly statistical in nature. Personally, I think the errors were largely the result of their fundamentally flawed sampling methodology, as discussed here by biostatisticians:
https://youtu.be/NTXgbN6uB1I?si=b5G5V7-S9JupboGJ
Their sampling errors seemed so basic, I was astounded that scientists of their stature would make them.
I’ll also note that they later published a similar study that sampled from MLB employees and where the results weren’t fully consistent with the Santa Clara study (the results of that latter study suggested a higher IFR). It was amazing that they, themselves, downplayed the results of the MLB study based on issues with the representativeness of the sampling. Amazing because they basically ignored the (non) representativeness of the Santa Clara study as they used the results of that study to go on their national tour downplaying the virulence of COVID.
It should also be noted that there was another big (non-statistics related) problem with their treatment of false positives. In their recruitment they offered participants immunity passports, meaning that if someone tested positive they could go to visit grandma without concern of transmitting COVID. If I recall correctly because of that basic ethics problem, some of the researchers on the team quit the research because they felt strongly that anyone testing positive should be given repeat tests as confirmation, and there was no provision in the study for doing so. I have no idea how that ever passed an IRB. It’s mind-boggling to me that it did.
Vaccinations and vaccines are in the news, but we should not lose sight that there are distinctions to be made. Covid is sort of an oddball disease because it is new and there seem to be legitimate, thoughtful disagreements as to how it spreads and how to treat it. By this last sentence I do not wish to include the conspiracy meanderings of the lunatic right regarding the Chinese-intentional origin and the desire of the you-know-who conspiracy to enslave the nation.
The push to use Covid as an all-out weapon to undermine vaccines in general is puzzling/terrifying. My knees buckle when public figures seriously put forward policies to outlaw all those childhood vaccines that have been absolutely proven to be safe and effective. What planet do these people live on? Unfortunately, the answer is, mine. It is also the planet of my parents who often told me about their siblings who died in childhood because the vaccines did not exist back when they were young.
Paul, you might not be aware, but general anti-vaccine lunacy has been bubbling in the right-wing fever swamp for a long time before Covid. There was (and still is) the whole vaccines-cause-autism nonsense, for example. What Covid did for this movement was sort of a dark version of the saying “never let a good crisis go to waste”. They made a mutual alliance with the Republican “libertarian” fanatics who want to destroy public health because it’s “the government”, and, well, here we are in a campaign to Make Polio Great (in numbers) Again.
Note that while in theory someone could have a principled Libertarian belief that vaccines are one of the greatest health benefits in human history, but the government should not be involved with them at all because that’s simply not the job of the government, in practice the politics does not work out that way. Because the ideology holds that the government can never do good, hence if public health even just provides vaccinations, then by this mindset, vaccinations must be bad. Of course the result seems absurd, but the conclusion does follow from the premises.
This is all a point in favor of the people who say that facts are don’t really matter much (maybe not absolutely zero, but pretty minor in the overall scheme of things), and it’s all about blasting out your side’s “message”. I don’t like it, but the results here seem extremely troubling.
I find it odd, to say the least, that you are categorizing vaccine nonsense like vaccines-cause-autism as a right-wing failure. Maybe that isn’t what you meant, but it’s what you wrote.
Will:
This has come up on the blog from time to time. Historically there’s been no correlation between anti-vax attitudes and political partisanship. The vaccines-cause-autism thing has been a failure of fringe anti-science/anti-business/anti-government groups, not associated with the left or the right. In recent years, though, this fringe has been embraced by the political right. Indeed, some on the center-left have argued that the Democrats made a political error by pushing vaccine deniers out of the party rather than welcoming them as allies in the fight against big business.
As of 2024, you could say it’s been a right-wing failure to embrace the anti-vax movement (thus achieving political gain, but at the cost of weakening public health and trust in valuable institutions) or that it’s been a left-wing failure to reject the anti-vax movement (thus avoiding endorsing dangerous conspiracy theories, but at the cost of giving up voters and losing political power).
I don’t see the poor phrasing in what I wrote. I don’t deny I might have phrased things in a way which could be misread, but I can’t grasp your objections. The idea is that the right-wing anti-vaccine campaign is not new and from Covid. Rather, it goes back a long time, and has been anti-vaccine in general from the start, e.g. the “autism” nuttery.
Or you could say it’s been a right wing “success” enabling Trump to get elected, and appoint a swath of fascists including an anti-vaccine guy to the position of health secretary (Robert F Kennedy Jr.).
Calling someone a fascist who isn’t a fascists is a blood libel indented to justify violence against them.
Not coincidentally, there has widespread support from the left for the deconstruction, violence, and murders during the antifa riots, widespread support for trumps assassination attempts, and now widespread support of the use of murder against a insurance CEO.
Its more that some people have higher standards for medical care than others. And those “higher” standards were normal ~1950.
People have been adopting ever lower standards for both the evidence and even what treatments (eg, vaccines) are supposed to accomplish for decades now. It is partly due to propaganda/marketing from government/corporations but also a kind of learned helplessness (Stockholm syndrome).
If they just collectively stopped giving money for dubious medical interventions, then they would get treatments that worked 10x better.
Eg:
https://www.bmj.com/content/373/bmj.n1503.full
That said, I have no problem with people trying whatever crazy moonshot medical intervention they want. But forcing others to do so (or more often, to pay for your gambling) is a problem.
Seth wrote: “Paul, you might not be aware, but general anti-vaccine lunacy has been bubbling in the right-wing fever swamp for a long time before Covid.”
My reply: It would be impossible to NOT be aware of general anti-vaccine lunacy that has been going on for decades. On the other hand, H.L. Mencken may have been right.
I’m going by memory, which at my age is dangerous, but I was following Covid fairly carefully when the 2020 study came out, and my recollection is that they estimated a considerably lower death rate than others had at the time. I recall that my guess for people like me was about 10%, so I thought they were low-balling it because they did not like the implications of a higher rate for restrictions on peoples’ activities. A recent paper (https://wwwnc.cdc.gov/eid/article/30/6/23-1285_article) estimated case fatality rates by age and sex for the period. For guys my age, the rate was a bit under 15%. Here is their abstract:
Few precise estimates of hospitalization and fatality rates from COVID-19 exist for naive populations, especially within demographic subgroups. We estimated rates among persons with SARS-CoV-2 infection in the United States during May 1–December 1, 2020, before vaccines became available. Both rates generally increased with age; fatality rates were highest for persons >85 years of age (24%) and lowest for children 1–14 years of age (0.01%). Age-adjusted case hospitalization rates were highest for African American or Black, not Hispanic persons (14%), and case-fatality rates were highest for Asian or Pacific Islander, not Hispanic persons (4.4%). Eighteen percent of hospitalized patients and 44.2% of those admitted to an intensive care unit died. Male patients had higher hospitalization (6.2% vs. 5.2%) and fatality rates (1.9% vs. 1.5%) than female patients. These findings highlight the importance of collecting surveillance data to devise appropriate control measures for persons in underserved racial/ethnic groups and older adults.
Jay Bhattacharya was one og the three co-signers of the Great Barrington Declaration. This seems to have said that there is no need for any specific health measures to prevent Covid-19 infection other than hand washing and staying home when sick. There was a need for (mainly unspecified) precautions tho protect the elderly but herd immunity would soon happen. IIRC, herd immunity really worked well for smallpox. /sarc
Great Barrington Declaration
Jkrideau:
From that link: “The authors say that, instead of protecting everyone, the focus should instead be on ‘shielding’ those most at risk, with few mandatory restrictions placed on the remainder of the population.”
My recollection is that in early/mid 2000, people were not so hopeful that an effective vaccine would be coming, and if there was not going to be a vaccine, it could’ve made sense to just give up regarding disease spread, in which case it also would’ve made sense to focus on protecting the most vulnerable people (which was mostly the elderly).
In retrospect, this seems like a reasonable idea, and I can see how Bhattacharya et al. felt blindsided by the attacks on it.
In practice, though, I don’t see how its proposals could’ve been implemented, at least in the U.S. Back in 2020, many people were scared and didn’t want to send their kids to school. It’s my impression that some of the lockdown rules were implemented as a solution to a coordination problem: if a large percentage of kids aren’t going to school, and a large percentage of adults aren’t going to work, then it can make sense to have a policy on this to coordinate people’s behavior.
The other problem, as discussed in the above post, was that Stanford and the Great Barrington team had connections to complete wackos such as Richard “500 deaths” Epstein, the people who said that “even the most vulnerable, elderly nursing home patients” should not take the vaccine and who said that the pandemic was already over in summer of 2000. Bhattacharya et al. were pushing against crazy maximalists who were trying to keep everyone indoors, while at the same time allying themselves with crazy deniers.
It’s really easy to forget that one of the main reasons for the so-called lockdowns was to ensure that the medical infrastructure didn’t collapse. As is the pandemic years weren’t all that good for health care workers: (https://bhw.hrsa.gov/sites/default/files/bureau-health-workforce/state-of-the-health-workforce-report-2024.pdf
Andrew –
In retrospect, this seems like a reasonable idea, and I can see how Bhattacharya et al. felt blindsided by the attacks on it.
The problem was that the GBD had logistical issues from the very start. There was no substantive policies on offer for HOW the highly vulnerable, and there was certainly no advocacy for such policies to be implemented. It seemed to me that glaring problem with the “declaration” was that the people advocating for it were often the ideological kin (if not the very seem people) who would fight tooth and nail against the kinds of top-down, government implemented interventions that would allow the vulnerable to stay at home while the virus was spread among the less vulnerable to reach “herd immunity.”
I don’t why anyone reasonable would feel blindsided by criticism of the obviously impractical and poorly thought out aspects of the policies they proposed.
That’s leaving out the questionable aspects of the “let it rip” advocacy that undergirded the GBD. They weren’t the only people who wrongly thought that the way out of the pandemic was through reaching a “herd immunity threshold,” so they don’t deserve a special criticism there. But it is disturbing that they don’t seem willing, in retrospect, to own up to how flawed that advocacy was. They also do deserve special criticism for pronouncing that a herd immunity threshold had been reached early in the pandemic, hen it was clear that no such threshold had been crossed.
I’ll also note that this:
My recollection is that in early/mid 2000, people were not so hopeful that an effective vaccine would be coming, and if there was not going to be a vaccine, it could’ve made sense to just give up regarding disease spread…
Isn’t consistent with what Jay said about the vaccine, and presumably doesn’t capture the beliefs underlying the GBD. In fact, in Nov 2020 (a few weeks after the GBD was signed) Jay said he was extremely confident that the vaccine was coming fairly soon – and the GBD was the recommendation for what to do prior to the vaccine.
Andrew –
Just ran across this quote from exactly 4 years ago – which illustrates the thinking that the GBD was based on.
Martin Kulldorff – 12/24/20
If the young live normal lives, some will be infected, but their risk is less than lockdown collateral damage. Pandemic will then be naturally over in 3-6 months. Lockdowns just prolonged the pandemic, while completely failing to protect the old.
Sorry for belaboring the point, but I think anyone promoting that viewpoint from an elevated scientific platform during an ongoing pandemic shouldn’t feel blindsided by pushback – no matter whether it’s inappropriately vitriolic. And that’s before you even begin to address how impractical their approach was with regard to “protecting the old.”
Certainly, this is at least comparable to publishing questionable scientific research based on flawed statistical methodology – particularly when you consider that they have shown zero willingness to address the analytical approach that undergirded their clearly overconfident analysis.
Given that Jay will soon hold such a powerful position in our society, I find his lack of accountability appalling.
Speaking of the “Great Barrington Declaration”, I wonder why it is that the libertarian right has a tendency to come up with such ridiculously pompous names for things? (Perhaps I should ask someone from the Mont Pelerin Society…)
When I first saw the name it made me think of climate change and environmental damage: “Great Barrington” -> “Great Barrier reef”. Kind of a bad choice for a name when it distracts attention from its topic.
This is one part of the curious puzzle that I believe we can solve right here about those uppity libertarians. The town where the three men met and wrote down their thoughts on what the public conversation should look like was named (wait for it) Great Barrington, Massachusetts.
Based on looking at your article, they were estimating that in early April 2-4% of santa clara county had been infected already. This was, more or less, 2-3 weeks after the widespread stay at home orders in mid March.
Let’s take a look at the *country wide* excess mortality in the US during that time period.
https://ourworldindata.org/grapher/excess-mortality-p-scores-projected-baseline?time=earliest..2020-05-03&country=~USA
at its peak in that time period, April 12 2020, country wide we had 1.39 times the mortality rate expected based on averages over previous years. However, as everyone probably remembers, much of the mortality in that time period was in NYC where they were burying people in mass graves on some island because of the enormous numbers.
Deaths in this period by month and county are available if someone wants to take a look.
https://data.ca.gov/dataset/death-profiles-by-county/resource/89d787cf-64c5-4850-8d65-3d2242fe68e4
The death rate curve in the county should tell us about the disease spread dynamics. Was it reasonable to think that in April 2-4% of the population had been infected already? Unlikely, but I’m not willing to take the time to do that analysis on my christmas break. Maybe someone with more motivation could do some estimates of infection rates based on now known rates of fatality among age groups, the age demographics of the county, and the known death curve. I think we’ll find that they were off by an order of magnitude or so.
I’m not seeing any my comments in that April 2020 thread that I disagree with.
Eg, that the “gold-standard” PCR was unreliable, and the response (lockdowns/ventilators) was killing people.
Was this ever done, besides just putting covid as the proximal cause?
https://statmodeling.stat.columbia.edu/2020/04/19/fatal-flaws-in-stanford-study-of-coronavirus-prevalence/
Anon,
I think you answered this a couple of years ago but I’ve forgotten your answer: is it your claim that COVID did not kill very many people, and that almost all excess deaths that are attributed to COVID are actually due to other factors, especially suicide and drug overdoses brought on by lockdowns and other socially isolating actions?
I would appreciate a simple declarative sentence. I often have trouble interpreting your comments.
Me too. And I was thinking about this again when I looked at some mortality data. Overall mortality rates jumped sharply during Covid. I didn’t delve too deeply into the cause of death data, but It was pretty much respiratory diseases that jumped. Interestingly, that same jump does not appear in the data for Austria (which had stricter lockdowns, among other differences). I’ve often been confused whether Anon was just pointing out difficulties with the data or whether it was a belief that Covid didn’t kill many at all. So, I’d also like a clear declarative statement.
As I have said here multiple times, elder abuse has now been normalized and institutionalized. It is even considered a valid medical intervention.
https://www.apa.org/topics/aging-older-adults/elder-abuse
There are many more aspects (stress -> heart attacks, ventilators with ~1% survival rate which also filled up the hospitals, overdosing on HCQ), but abusing the most vulnerable members of society must be the most egregious idea to ever come out of “evidence-based medicine”.
Can industrial-scale elder abuse explain ~15% increase in mortality all on its own? Seems plausible to me, but we will never know because no one is going to volunteer for that RCT (because it is so obviously bad, which is also why it is not a prohibited “treatment”).
Just blaming all these deaths on a virus doesn’t even rise to the level of pseudoscience (even people using a ouija board try to come up with other explanations). The testing and intervention are just as, and possibly much more, important.
In every example I’ve seen, assertions that NPIs caused”l morbidity and mortality during the pandemic are based on empty counterfactual assumptions about what would have happened absent the NPIs.
As just one example, the real world alternative to less family visiting visiting elderly people would have been more elderly people getting infected sooner, and before there were vaccines. For some reason, some folks are locked into a binary mindset: NPIs were associated with negative outcomes, therefore we wouldn’t have experienced negative outcomes had NPIs not been implemented.
Anoneuoid
I guess you are incapable of making any simple declaratory statements. From your comment, I understand that you think COVID policies were responsible for many deaths (especially among the elderly). I don’t disagree, nor do I disagree that these policies caused deaths among the non-elderly. But stating “blaming all these deaths on a virus” doesn’t help me understand your position. I never blamed “all” these deaths on a virus, nor would that be a credible view. However, I do blame “many” deaths on the virus. I think the sharp increase in respiratory deaths in 2020, but not in other cause of death categories points towards COVID as a major factor. Also, the death rate sharply increased in all age groups except for children and teenagers in 2020 – pointing towards COVID as a major factor (especially when it is confined to respiratory illnesses).
I’m sure a case can be made that these observations are not conclusive – and I don’t claim that they are. But they make it hard for me to believe that COVID caused “few” deaths or that the strict policies caused more deaths than COVID. Are those your claims? Why can’t you answer that question without espousing your many beliefs about poor policy and poor medical studies?
Phil wanted a simple declarative sentence. That doesn’t really count. So here’s my interpretation of your claim:
“The excess deaths attributed to COVID in Anoneuoids opinion appears to be that it was largely elderly people dying of sudden massive increase in elder abuse through isolation and institutional neglect”
I’d just like to say that’s the most ridiculous thing I’ve heard in a long time. And I am one of your bigger supporters because you have lots of valid insights into biomedical issues.
But seriously, deaths of elderly who had respiratory symptoms were widespread *worldwide* including people who weren’t institutionalized, and people who were living with their families (for example in Italy early on). Deaths of middle aged people were widespread. The leading cause of death of police officers (who were not isolated, they were on duty) for several years in a row was COVID. Your explanation is completely ridiculous.
The simple declarative sentence is:
To model covid deaths you also need to model testing/diagnosis and anything that ensues due to a positive test.
This is bare minimum stuff to be considered science.
If you believe elder abuse has little/no effect on mortality rate, then state that explicitly. Right now the default is assuming zero.
I think Daniel’s summary is good. As for Anon’s response, I don’t know whose default for “elder abuse” is zero, but not mine – although I also find it non-credible that it is a major component of the total. While anything is possible with less than perfect measurement, I don’t think many of the “elder abuse” deaths would be labeled as “respiratory” illnesses. Also, I’m not sure how necessary it is to model “testing” to say something about death rates from COVID. “Diagnosis” is more relevant, but I do find it hard to believe that misdiagnosing COVID will change the overall picture (it will certainly have “some” impact, but I don’t believe enough to render COVID deaths minimal or less than deaths caused by COVID policies).
It would help if Anon could refrain from taking his somewhat valid concerns about RCTs, medical “science,” and measurement issues, and using them as evidence that COVID has killed fewer people than the COVID policy response – the latter view which I find “completely ridiculous” as Daniel has said.
In general I agree with that. But we have situations like large numbers of elderly people coming down with a disease with respiratory symptoms and then dying at home in Italy with their family in early 2020. It’s impossible to explain that by some kind of elder abuse or ventilators, or HCQ overdoses or anything.
Yes, in the first say 3 months deaths were likely higher than they should have been because good treatment protocols weren’t in place and that guy with this HCQ claims distracting people, but it’s pretty striking that both excess deaths ramped up RAPIDLY after people started getting sick, and that they peaked and plummeted after people started staying home and using non-pharmaceutical intervention to mitigate spread, and that they then subsequently increased during periods when people stopped doing nonpharm interventions (like the Sturgis motorcycle rally)
https://ourworldindata.org/grapher/excess-mortality-p-scores-projected-baseline?time=earliest..2021-01-10&country=USA~GBR~CAN~ESP~ITA
It takes enormous motivated reasoning to ignore the fact that a simple model of “people get sick with COVID and then some significant fraction of them die” works really well.
Dangit that link didn’t work. So here’s An HTML link to the our world in data excess death graphs
Heres one guy who bothered to give an estimate:
https://apnews.com/article/pandemics-us-news-coronavirus-pandemic-daac7f011bcf08747184bd851a1e1b8e
So 5% increase from elder abuse alone (15% for that plus viral disease per se). I note that is assuming the covid patients did not also get this treatment.
Whats the plausible range? Up to 1% increase in mortality? 20%?
Here’s some suggestive data (source: https://www.mortality.org/ – they also have detailed cause of death data which I haven’t looked at in detail other than to see similar patterns for respiratory deaths compared with overall death rates):
Age 2019 death rate (per 100K) 2021 death rate (per 100K)
40 213 293
50 398 541
60 910 1144
70 1828 2243
80 4673 5630
Presumably not many of the 40 or 50 year olds were in nursing homes. The surge in death rates from 2019 to 2021 appears in every age except ages below 15. Of course, COVID policies may have increased non-COVID death rates for a variety of reasons, with depression/loneliness, lack of treatment for non-COVID medical problems, and poor nursing home care accounting for many deaths with different specific causes for different age groups. But it’s much harder to paint a high level picture that makes more sense than the simpler story that the disease was more deadly than the policy response prior to vaccination.
I do wish there was a way to format tables in the comments!
Where has that come from? Has to be said that deaths associated with mechanical ventilation (MV) is a conspiracy theorist’s dream since it is the most severely compromised patients in ICU’s that require(d) ventilation and these are most likely to die irrespective of MV or non-MV treatments – a “correlation” isn’t unexpected. Even so, survival rates with Covid-related MV are way higher than ~1%. Analysis of deaths in a large US health care network in the first part of the epidemic in 2020 found a mortality rate of 37% in the MV cohort (73% survival) [1]. In fact, the cohort that was initially treated with noninvasive respiratory support (NIRS) had a higher mortality rate (39%). In a study of 58 ICU’s in Spain from the start of the epidemic through end-August 2021 that was designed to assess the effect of initial (within 24 hours) and delayed intubation (after 24 hrs), the within-hospital mortality rates were 27.3% for MV within 24 hrs and 37.1% for delayed MV, and for 90 day mortality: 30.9% for initial MV and 40.2% for delayed MV [2]. A meta analysis of a large number of studies from around the world indicates an average Case Fatality Rate (CFR) of around 45% (with quite a bit of understandable variation) for Covid patients receiving MV [3].
There are lots of studies on outcomes following MV, but I’m not seeing any data that supports your assertion of a ~ 1% survival rate. The evidence as far as I can see indicates that MV in the early days of the epidemic was an unfortunate necessity in many cases and in fact early MV seems to have had better outcomes compared with non-invasive ventilation treatments at least in some studies.
BTW it wasn’t “ventilators with ~1% survival rate which also filled up hospitals” but very sick and dying individuals that filled up hospitals: the aim of lockdowns and other NPI’s was to limit the overloading of health resources until the disease could be better understood and improved treatments and vaccines became available. Some of the improved interventions reduced the use of ventilation in later periods of the epidemic.
[1] Acute Respiratory Failure From Early Pandemic COVID-19: Noninvasive Respiratory Support vs Mechanical Ventilation. CHEST Crit Care. 2024:100030. doi: 10.1016/j.chstcc.2023.100030
[2] Effects of intubation timing in patients with COVID-19 throughout the four waves of the pandemic: a matched analysis. Eur Respir J. 2023 61(3):2201426. doi: 10.1183/13993003.01426-2022
[3] Case Fatality Rates for Patients with COVID-19 Requiring Invasive Mechanical Ventilation. A Meta-analysis. Am J Respir Crit Care Med. 2021 203:54-66. doi: 10.1164/rccm.202006-2405OC
Dale –
You say:
, I understand that you think COVID policies were responsible for many deaths (especially among the elderly). I don’t disagree, nor do I disagree that these policies caused deaths among the non-elderly.
For the sake of argument let’s say there was a congregate housing facility for 1,000 people where during a pandemic, they strictly isolated the residents. None were infected and 100 died with a failure to thrive diagnosis.
But say isolation hadn’t been enforced and all the residents got infected and 103 died “with covid.”
Obviously, that’s a very simplistic, contrived rhetorical scenario (for example, some would likely have died from “failure to thrive” without being isolated, and some % could have died “with covid” but in actuality not from covid).
But in such a situation, I don’t get why people would say that the intervention was responsible for the deaths.
People were going to die either way during the pandemic. Of course it’s important to try to evaluate whether more would have died with or without the interventions, but I just don’t get why people say the interventions were responsible for deaths if we don’t know whether actually, fewer people died, differentially, because the interventions were implemented.
What if a seatbelt worn in a high speed crash resulted in fatal injuries, but surely the person would have died from other injuries had the seatbelt not been worn. Does it make sense to say that it was a “seatbelt death” as people use the term “lockdown deaths?” I think most everyone would just say that the person died because of a high speed crash. The parallel would be to say that a person died because of the pandemic, but people seem to have a different way at looking at pandemic deaths.
No one else seems to be bothered by this much, so maybe it’s just my issue.
I still have no idea what Anon thinks about how dangerous covid infections are to old people. I’m pretty sure he is unconvinced that they carry a substantial risk of death. If true, I’m with Daniel: that’s nuts.
I also thought some of the more extreme ‘lockdown’ policies at nursing homes were nuts. But that doesn’t mean I deny that covid was very dangerous to old people. As far as I can tell, in the whole world only Anon thinks that maybe it wasn’t.
The proposition that elder care failure accounts for a substantial part of the excess deaths fails as there was no increase in excess deaths among the very young who are just, if not more reliant on others.
Perhaps a T-cell immunologist could help you out with this? I am not one, but my paltry understanding is that “The Young” have fully functional naive T-cells that can be safely recruited to fight infections in whatever capacity is called for without overreacting and destroying the host’s organs. The elderly have T-cells that are correspondingly elderly and frequently no longer can rally to mount a proper immune response which is balanced.
If the elder individual has metabolic disorders, in particular, history of poorly controlled blood sugars, they are not going to fare well with this Corona virus, are they? If the elder already has vascular insults and are high risk when they contract a virus which targets the endothelial lining of the vascular system, then, they are not going to have a very good outcome. If the elder is overweight, clots easily, is confused, is partially immobile, or any of the states of being which make us high risk.. then careful oversight with early intervention is critical yet probably not going to happen. Isolation with repeated cycles of re-upping isolation with each new diagnosis in a facility may lead to dehydration and delirium and ultimately a DNR and death.
I don’t think small children and infants who are reliant on a different type of constant monitoring and care by families or professionals fall into the same life experience category as the elderly. They are not on the brink of deterioration every single day upon arising. They are not kept in isolation for their own good during a pandemic.
If you’ve ever listened in on conversations happening in Canada you would be quite surprised at what is thought about how they interpreted Covid policy. I’m not saying that the concept of ‘targeted protections’ was weaponized … I’m not saying that.. but what happened around the world with ”Covid policy” was not wholly for the benefit of mankind.
Moving on..
If it matters, I am vaccinated.. but never endorsed mandates. From Day One.
I am not a wild-eyed libertarian but am a registered Democrat in the U.S.
I am adjacent to the non-profit BiosafetyNow! dot org (as was Dr. Jay Bhattacharya until very recently) and Dr. Jay knows exactly what went down with ”biodefense” and decades of funding from our NIH/NIAID.
*** In fact, Dr. Jay was on the board of BiosafetyNow. ***
THIS DOES NOT MAKE HIM POPULAR
* follow the university and lab funding *
None of this makes him dangerous or anti-science or anti-vaccine or anti-research. It simply makes him want to see regulations enforced by a third party for Risky Research related to our civilian biodefense program.
You’re welcome to ask Dick Cheney why he started this endeavor with the NIAID back in the 2,000’s and why Dr. Fauci agreed to go along with it.
We did not appreciate the last/current pandemic and are not looking forward to worse ones.
Dr. Jay will be wonderful as head of our amazing NIH.
*************************************************************************
To Jay Bhattacharya M.D.
Dr. Jay, if you’re reading this, please be strong .. but.. humble and from now on, ask for help with statistics in case you actually were too overconfident. It takes a wise man to admit this might be true. I know you possess the grace to be this man.
Everybody made bad calls in the beginning. Even the W.H.O. insisted that SARS2 was “not airborne”. What the heck, right? Talk about killer proclamations in Public Health. Significant.
And Citizens were given the impression that we didn’t need to wear masks. Except, later, if we did wear masks, as in, any mask? .. just stay 6 feet behind others in the store,, toes behind the masking tape on the floor. Between the W.H.O and THAT? That’s how you science in making Public Health Policy?
Stay indoors ..unless you were protesting. Wear a mask at the beach, alone, or the Police would chase you – (yes, this happened). Send elderly Covid+ pts. back to their care homes if you were in certain states that will remain nameless. (yes, unbelievably, this happened).
Oh, right, that’s not elderly abuse at all.
And be censored (per government request) if you didn’t follow the narrative.. if you simply wanted open discussion. How fringe was this? The expectation of free speech in America. Ha ha.
So, if you were overconfident – learn from it, Dr. Jay. You know I want you for Head of NIH.
God speed good Sir.
Andrew:
” or that it’s been a left-wing failure to reject the anti-vax movement (thus avoiding endorsing dangerous conspiracy theories, but at the cost of giving up voters and losing political power).”
This is one of those cases where we need to be careful about using “left,” “liberal,” and “Democrats” interchangeably. Democrats have been much more likely to marginalize the paranoid fringe. I’m not sure we say that about the left wing.
Andrew, let’s not forget that another prominent co-author was publicizing their study interpretations to social media as early as April 17th 2020, a fact I remember commenting during our discussions of the Santa Clara study back then. You write that they did not have statistical experts on the team, but they did have the most cited methodologist of all time, and it seems that none of them have ever shown signs of acknowledging that the uncertainty in their variables, if properly propagated, would have blown up their IFR range considerably.
https://m.youtube.com/watch?v=jGUgrEfSgaU&pp=ygUTSlAgSW9hbm5pZGlzIGNvdmlkIA%3D%3D
“By publishing Epstein’s uniformed commentary,…”
I’ve heard that “a language is a dialect with an army” but never that “a travesty is a commentary in a uniform”.
You may be on to something, though…
Guys lets hit the high points- Jay Bhattacharya is about to lead the NIH short of something tremendous happening. He has shown via the book We Want them Infected, that he is an absolute monster. He will continue the denial, virus minimizer, and hard libertarian tactics as NIH chair and punish those who disagreed- with actual power this time in withdrawing lab funding as political punishment to literally shut labs down. This is why his career needs to be ended yesterday.
Andrew,
The GBD was in no way “reasonable” because it had no mechanism for “focused protection” and was also premised on authors clinging to the false (and very unlikely even at the time) IFR. Also, given the true IFR is a gradient from most to least susceptible, without any plausible mechanism by which to focus protection, and with the founders being against any type of mandate be it vaccine or mask, the GBD was essentially a question begging PR exercise. Even before their study came out, here were the authors prominently (and with essentially no data) making case in a WSJ opinion piece that COVID not really that bad. I believe this was the piece making a back of the envelope argument from positive NBA players. https://www.wsj.com/articles/is-the-coronavirus-as-deadly-as-they-say-11585088464
What was the pipeline to get them this real estate? The scientific arguments from hindsight are incredibly powerful that the Santa Clara study was wrong. The evidence since the study is also quite extensive that political philosophy was motivating this non-epidemiologist to dip his toe in the water and that the result was based on heavy thumb on scale methodology.
Just read the blog post. In some ways it makes me somewhat more sympathetic to Jay’s case. On the other hand, the lack of serious treatment of legitimate criticisms is troubling when I anticipate the results of his high position in the coming administration.
In her email, my wife wrote that a positive antibody result implied that a person was immune to reinfection and could probably resume work and normal life. My wife is not a clinical researcher and did not know that these prophetic thoughts were forbidden at the time. Nor did she understand that alleging infection provided immunity was deemed misinformation by the press.
Prophetic? It was “prophetic” to say that a positive test result would mean someone is immune to reinfection? And apparently he thinks there’s no valid reasons to question whether stating that infection provides immunity is misinformation? Has he never run across anyone who’s been infected more than once?
It’s exactly that kind of clearly wrong, and hyperbolic rheotic that justifies skepticism about Jay’s science.
I’ll also note that he says they fully expected the MLB results, which wasn’t at all evident in their statements when the MLB study came out.
I read the blog post as well. I was struck by 2 things. First, the language sounds a lot like Trump’s constant whining of being singled out for investigation. I noticed at least 3 uses of the word “hyperbolic” and the characterization of the “inquisition” which all make it sound like critics are unfairly targeting him. Also, lots of appeal to his credentials, the ultimate publication of the study (his gold standard), and suspicions of ill-will of his critics. That is the second thing I noticed: there appears to be lots of Stanford politics involved. Ill-will, accusations of bad incentives (on all sides), and ego battles between famous researchers. That all amounts to airing a lot of dirty linen in public – since I have no personal knowledge of the people involved, I find that unseemly and boring.
So, I don’t find myself “more sympathetic to Jay’s case” although I will admit that he has perhaps been treated unfairly. As with so many of these debates (and most political battles these days), I only find that almost everybody looks bad – in this case, Jay, his collaborators, his critics, and the Stanford administrators.
One minor nit: while the Hoover Institution is located at Stanford, it has an independent Board of Directors and it’s own funding stream. Epstein was on the Law School faculty at University of Chicago when he first became affiliated with Hoover. He retired but came back to teach at NYU (right down the road from Andrew). There are lots of Hoover affiliated people from all over that do not have an affiliation with Stanford University.
I was doing a daily Covid newsletter, mainly for friends who I had worked with in the biopharma industry, and remember the Santa Clara study well. I was skeptical of the results and remember the discussion on this blog which I did reference at the time. Bhattacharya is in for an adventure at NIH, assuming he is confirmed. He will quickly find out that the director does not have all that much power as the major institutes have their own directors and specific appropriations from Congress. He can tinker around the edges a bit and we’ll see what that brings. Of greater concern is Kennedy who wants to put a pause on all infectious disease research for the next 10 years. Fauci’s former institute, NIAID, is the second highest funded on behind the National Cancer Institute and has a satellite campus in Montana that studies tick-born diseases among others. Of course, they will have to convince Congress to stop funding this line of research. There is an incredibly strong patient’s lobby that will push back on all this nonsense.
As a Stanford alum, I say Stanford can either host the Hoover Institute or not house it. Not both. The decision has consequences. Dogs, fleas, etc.
Reading through Bhattacharya’s article, I am actually quite impressed with the further work he and his group did to validate their results after the initial criticism.
A lot of researchers who face criticism have a tendency to bunker down, but you have to give Bhattacharya props for instead choosing to further verify their work. Replicating their results in Los Angeles (with better sampling) and doing further tests of their test’s false positive rate was a lot of effort and really helped resolve a lot of concerns about the initial study.
I still hold that the statistics in the original study were bad (https://statmodeling.stat.columbia.edu/2020/04/19/fatal-flaws-in-stanford-study-of-coronavirus-prevalence/#comment-1305950 is my favorite way to demonstrate that), but I have a much better opinion of Bhattacharya now as a scientist and I hope he does a decent job at the NIH.
Ethan –
I’m trying to reconcile your reaction and Dale’s above.
I was also impressed with the extent and seriousness of their efforts to confirm their results, which I wasn’t aware of. So I’m with you there. And I think the post makes a reasonable case that some of the criticism was unwarranted. On the other hand, I’m with Dale on the self-victimizing attitude that runs throughout the post. To the extent that some of the criticism might have been unwarranted, the defensiveness and lack of humility or accountability from Jay was pretty striking.
That kind of polarized engagement characterizes so much of where science surfaces at the public level these days. Jay is just one of the many who engaged hyperbolically regarding COVID themselves, and then played the victim card when there was hyperbole engagement from their critics. Perhaps the best example of that pattern was when public health policy critics like Jay complained that they were being silenced, even as they met with Trump, appeared all over conservative TV media like Ionnides, were well represented in powerful positions by people like Scott Atlas, or published oo-eds in the Wall Street Journal like Marty Makary.
It’s just all so juvenile and banal. And depressing.
The ventilator issue was largely resolved by May 2020, so your papers are not really relevant. More precisely, they dilute the data showing the problem until it is no longer visible.
It was those days when covid had the weird symptom of low SpO2 (low blood oxygen, at least as measured through the skin) but little/no dypsnea (shortness of breath). For whatever reason this bizarre and novel aspect of the illness later disappeared from discussion. I have never seen an explanation for why.
At the time, “early intubation” referred to using the ventilator *at this (low SpO2) stage of the disease*. It had little to do with with time since admitted to the hospital (eg, 24 hrs like used in your refs). But it was those patients who had their lungs damaged and filled up the hospitals for weeks, causing knock-on problems throughout the rest of the system.
What happened is critical care doctors in NYC/Italy raised the alarm in late March, then the word spread and the covid “guidelines” based off anonymous rumors from china were disregarded:
https://vimeo.com/402537849
https://pubmed.ncbi.nlm.nih.gov/32228035/
You can find discussion of the state-of-affairs in Spring 2020 here:
https://www.thelancet.com/journals/lanres/article/PIIS2213-2600(20)30079-5/fulltext
https://www.sciencemediacentre.org/expert-reaction-to-icnarc-report-on-the-first-reported-775-patients-critically-ill-with-covid-19/
https://jamanetwork.com/journals/jama/fullarticle/2765184
Mortality like in NYC was never seen anywhere in again:
https://www.cdc.gov/mmwr/volumes/69/wr/mm6946a2.htm
Hmmm… I’m not seeing any evidence to support your assertion about other “aspects” including “…ventilators with ~1% survival rate which also filled up the hospitals” in the articles you linked to.
You linked to five articles and a video (which I didn’t watch because I don’t wish to sign up to your linked site). Of the articles none are anywhere close to supporting your assertion about “ventilators with 1% survival rate”. The papers are in order:
Am J Resp Crit Care Med – zero data on ventilator-related mortality. The letter does give some potentially useful advice on ventilator pressures (PEEP) as experienced at that moment in time early in the epidemic in Italy.
Lancet: 52 extremely critically ill patients in Wuhan China very early in the epidemic. 37 required mechanical ventilation – 18% of these had survived at end of study period.
ICNARC report: UK Covid patients up to 26 March 2020. 775 patients – overall mortality 10% (79 out of 775). Of those in ICU’s (165 patients requiring critical care) 52% survived 48% died. 35% of those receiving advanced respiratory support survived.
JAMA: 5700 patients hospitalized in NY March 1 – April 4, 2020. 373 required ICU care. For patients receiving mechanical ventilation 24.5% died (76.5% survived) by end of study period.
MMWR: nothing on relationship between mechanical ventilation and survival/mortality.
So I’m seeing nothing to support your assertion about “..ventilators with ~1% survival rate…”. I’m surprised you consider my papers “not really relevant” because “The ventilator issue was largely resolved by May 2020”. Perhaps if one is pursuing non-scientific agendas, it does make sense to focus on periods of greatest uncertainty and confusion which are the fertile arenas for agenda-led misrepresentation! However if one wishes to understand important issues (like effects of ICU mechanical ventilation), a clearer perspective guided by good evidence is appropriate. This short article, for example, is useful for helping to understand issues around assessing “ventilation-associated” mortality at very short periods early in the epidemic when consequences of infection and treatments haven’t fully played out:
Mechanical Ventilation in COVID-19: Interpreting the Current Epidemiology. Am J Respir Crit Care Med. 2020 202:1-4. doi: 10.1164/rccm.202004-1385ED
BTW if “The ventilator issue was largely resolved by May 2020”, how does one account, for example, for the very large double spike in US Covid deaths in Autumn 2021-Winter 2022 (see WorldinData site Daniel Lakeland linked to above)? Obviously not due to use of ventilators. More likely due to low vaccine uptake in the US (compared e.g. to the UK where these post-vaccination-period mortality spikes didn’t occur).
I thought this article was surprisingly good for illustrating the scenario in April of 2020. A point made that I hadn’t considered before was the psychological effect of a context where a “normally” high mortality rate of very few patients on ventilators would seem very different than the same mortality rate with ICUs filled with people on ventilators.
https://www.reuters.com/article/us-health-coronavirus-ventilators-specia-idUSKCN2251PE/
I emailed Andrew, maybe he will do a post.
But in general, we keep disagreeing due to the level of data aggregation. There is a tendency to aggregate over larger and larger populations (symptoms, time, geography) that obscures the details. This is standard behavior in the medical literature.
Averaging is a form of lossy compression. It works great if the variation is “random”, but not when it is due to dynamics and individual differences. The latter isn’t “noise”, it contains the most interesting details.
Covid warriors: Is Trump’s hair color golden yellow or yellowish gold?
@Phil
Here is a plot I made before “covid” was even a term. From Mar 11, 2020:
https://i.ibb.co/983SC13/covplot.png
It compares US life tables (https://www.ssa.gov/oact/STATS/table4c6.html) to data from this paper:
https://pmc.ncbi.nlm.nih.gov/articles/PMC8392929/
We also had data from the Diamond Princess cruise ship at the time (https://en.wikipedia.org/wiki/COVID-19_pandemic_on_Diamond_Princess).
Afaict that has stood the test of time as (at least) an upper bound on mortality due to the virus.
Clear as mud. Are you being purposefully obtuse or do you really not know how to clearly respond? From what you have linked, it appears that you know COVID was dangerous for old people and that it was very contagious with a fairly low mortality rate. I believe these are accepted facts. But it is quite a leap to go from that to claims that the excess mortality rate is low, perhaps even zero (or negative) due to large numbers of deaths from COVID policies relative to deaths from COVID. You also have a habit of cherry-picking references that you discover – analysis of Chinese data depends on the accuracy of those reports and the Diamond Princess data, while informative, is a single case that may or may not be representative of anything. For that matter, you can probably dig up some reference to what happened at a particular nursing home in NY or WA and cite that as evidence that the mortality rate for the elderly is extremely high.
Can you tell that your style exasperates me? And, you never ever back down, show any hesitation or reflection, or willingness to stay on point. There should be a term for these traits – which are becoming increasingly common in political circles as well as academic ones.
Reality is “clear as mud”, not me. I have been consistently saying the same type of thing since before anyone heard of covid (eg, regarding cancer + caloric restriction due to chemo).
Here is the simple declarative sentence from above:
Also, re:
Don’t know how you got that from my post, which gives an upper bound. Those results are obviously covid + response (strict lockdowns + whatever else a positive covid diagnosis entailed in China at the time).
The problem is you think someone knows how deadly covid is in the elderly. We don’t, we only know covid + response. Then it is assumed (but not by me) that the effect of the response was minimal. This I do very much doubt due to prior knowledge and various aspects of the data.
Eg, here is part of what I emailed Andrew*:
https://i.ibb.co/k1CS0zf/NYCmort.png
Can you come up with a model of viral dynamics alone (ie, assume negligible effect of the response) that explains that singular spike in mortality? Obvious variables are population density, R0, then the dynamics of mixing rate, and percent immune. It would also be nice to find all-cause mortality for other cities, rather than state-level (which mixes urban and rural pops).
* One thing I forgot about in the email is Cuomo mandating that covid patients be sent to nursing homes. There were so many hysterical aspects of the response it is hard to keep track of them all. Andrew was in NYC at the time so can probably point out more.
I looked at some county data from New York. The huge spike from your graph certainly points to the issues you raise about nursing home policies. I would note that the other COVID waves (after that ill-fated response) are still evident. I also can see those other COVID waves in rural counties in NY. I didn’t do an exhaustive analysis by any means, but I’d suggest that the NY data shows many deaths due to the initial policy response (combined with COVID-caused deaths) and also shows that each wave of COVID (pre-vaccine) had similar peaks in deaths across all geographic areas. So, my belief is that COVID was in fact deadly for the elderly, while some policy responses also caused deaths (whether these were unnecessary or not is harder for me to say, given the knowledge at that time). The extreme peak in your graph also seems like a singular event, not found in many places.
As for your “simple declarative sentence” is isn’t what I was hoping for. My understanding of your sentence is that without a complete model, you are not willing to believe anything about COVID mortality at all. If that is indeed your position, I think it is not widely shared nor does it make any sense. The requirement you suggest (the need to “model testing/diagnosis and anything that ensues due to a positive test”) suggests that there cannot be anything that you would find sufficient to claim that COVID (the disease, net of the policy responses) has led to excess mortality. Such a position is something that I think should be given a name. Rejecting all claims because there are no perfect studies is a tactic to support an extreme position without having to state that position (the statement of the position is what I was hoping for when you were asked to state simply what you believe).
Dale –
You say…
, while some policy responses also caused deaths (whether these were unnecessary or not is harder for me to say,
I’m curious which interventions you believe “caused” deaths and what evidence you use to reach those conclusions.
Let’s take the nursing home policy in NY off the table (even if there, we don’t know what would have happened had the infected older people been left in hospitals, even if deaths in congregate senior housing was high everywhere including on Sweden where infected older people weren’t even brought into hospitals for care, generally). Let’s say that policy “caused” deaths – which other policies fall into that category?
Joshua
I’m not advocating a strong opinion here, but I do believe the policy responses were excessive in some cases and did real damage. Perhaps the removal of basketball nets led to some people having depression and even committing suicide (sure I’m being absurd here, but somethings along those lines probably happened, people getting depressed by being stuck inside). More reasonable cases are people not getting diagnosed/treated due to COVID fears and lockdowns. I suspect that some other causes of death increased due to less treatment of non-COVID problems. I haven’t looked at the cause of death data carefully, but it is available should you want to examine it. So, the position that Anon seems to be articulating has some truth I believe. What I don’t believe is the extreme position they are seemingly not stating (how’s that for a double negative?). There are plenty of stories I can imagine about people that died due to policy choices (e.g. ventilators, isolation, lockdowns, etc. – none of which I claim any expertise in, or study of). I just don’t believe that these total more than the direct deaths due to COVID itself – a position that Anon either holds or isn’t willing to reject before seeing those perfect studies that “model testing/diagnosis and anything that ensues due to a positive test.”
Plenty of people died, in direct association with multiple waves of covid infections, in places where basically none of those interventions took place. And in some places where the interventions were significantly more strict (say New Zealand) there was no pattern of significantly more excess deaths.
The age-stratified gradient that remains consistent across interventions such as people being returned to nursing homes or not being returned to nursing homes, makes it highly dubious to conclude that the interventions themselves where significantly “causal,” let alone to a higher degree than COVID in itself. And of course I still bristle at the idea of causality from interventions if you’re concluding causality based on empty assumptions about the counterfactual of what would have happened without them.
And then there are other factors that complicate the idea of causality. There’s a clear signal of social determinants of health in COVID mortality and morbidity. So then how could you tease out the causality of impact from interventions themselves from the causal impact of poverty or
race/ethnicity or the health impacts of adverse childhood events? You need to do a sophisticated analysis, imo, where you try to adjust for the mediating or moderating or interaction effects of all those variables.
I don’t feel strongly that all the interventions were net positive, or wouldn’t have been more net positive had they been more restricted or ended earlier. But it really annoys me to see it become such “conventional wisdom” that the interventions “caused” deaths. As I’ve said above, all actions or lack thereof would have resulted in morbidity and mortality. I am concerned that a casual approach to causality, as if any action or lack thereof wouldn’t have been sub-optimal, has become so prevalent. I think it signals danger for us as a society going forward as it contributes to the broader trend of crippling institutions of public health in such a way that they will become even more ineffectual than they already are.
I also wanted to say that there’s plenty of evidence that the number of infections and resulting morbidity and mortality were negatively correlated with interventions.
I remain rather agnostic about that evidence, as I think that establishing causality (and not just correlation) requires crossing a very high bar that includes at less some “control” for important confounding variables that are very complicated to “control” for.
I watched this recent hearing/berating of Cuomo to see if there were any new details: https://www.youtube.com/watch?v=dZmpZK-UrxE
As expected, it was an ascientific circus. No data like provided above (don’t they have more resources than a random person downloading data from the CDC website?), but we did get giant posters of Trump and emails/tweets.
In particular, it seems Cuomo’s “help” was meant to get people out of the hospitals. But there was no discussion of why exactly these hospitals were so full to begin with. More specifically the ICUs were full while the other departments were far below capacity.
How many of those patients were misdiagnosed HCQ overdoses (methemoglobinemia), which would explain the “happy/silent hypoxemia” phenomenon (that seems to have eventually disappeared) and thus very poor response to the usual ARDs protocols?
Also, I would not bother with “covid deaths” and instead look at all-cause mortality. Eg, there is little/no “vaccine” signal in the all cause mortality data.
Apparently it is not just HCQ, but some antibiotics can also cause this:
https://journal.chestnet.org/article/S0012-3692(22)02256-5/fulltext
Anoneuoid –
Eg, there is little/no “vaccine” signal in the all cause mortality data.
You obviously haven’t done a thorough search for evidence of a vaccine signal in all cause mortality.
And BTW, love how you put vaccine in quotes.
Here are the proposed benefits of vaccines:
1) Eradicate disease to the benefit of current and future generations
2) Herd immunity to prevent infection of vulnerable populations
3) Expose body to minor case so severe disease is less likely upon later exposure
4) Allows people to have normal social interactions (eg, see family and friends)
The results:
1) Neither promised nor delivered
2) Promised but not delivered
3) Promised and maybe delivered
4) NEW! Unpromised but delivered
The “vaccine” signal here refers to a combination of points 3 and 4. Only number 3 is really due to the vaccine per se. Thus the combination is better denoted “vaccine” signal.
The three most salient aspects of the all-cause mortality data are:
1) Whatever happened in/around NYC in late winter and early spring 2020.
2) The seasonality
3) The clear onset in 2020 and end in 2022
https://i.ibb.co/k1CS0zf/NYCmort.png
There are various explanations for this, but the difficulty in seeing such a signal is surprising. It is at best some secondary factor. Maybe you can download the data from the CDC site yourself then plot it to highlight the clear vaccine benefit.
I also searched “all-cause mortality vaccination” on DDG, the first three papers all concluded “no increase in all-cause mortality” (ie, no benefit either). This is also surprising given the prevailing bias and narrative.
But if you would like to share the results of your thorough search (why does it require such a thorough search anyway?) I am interested.
Anoneuoid –
The three most salient aspects of the all-cause mortality data are:
1) Whatever happened in/around NYC in late winter and early spring 2020.
2) The seasonality
3) The clear onset in 2020 and end in 2022
That may well be the most salient aspects to you but you’re conflating salience with what you consider salient.
Meanwhile, there have been a slew of investigations looking for a “vaccine signal¥ in all cause mortality in across the globe over the course of the entire pandemic. Some people (of the sort who put vaccines in quotes) have decided that the vaccines killed tens of millions of people. Those who I find far more credible have found a signal associating the vaccines with saving millions of lives (and I’m sure on a association much morbidity).
There’s no way that you haven’t seen such evidence if you have done a thorough search.
Anoneuoid wrote:
” I would not bother with “covid deaths” and instead look at all-cause mortality. Eg, there is little/no “vaccine” signal in the all cause mortality data.”
Stripped of all the pretzel logic, the statement just comes across as dumb: “But look at this noisy number! EXPLAIN THAT!” But dumb is not the problem here.
The underlying rationale that leads to this kind of goofy stuff is that the covid vaxxed-vs-unvaxxed death count is the big number, and everyone looks at the big number. That means that the evil pharmaceutical companies will bribe doctors to fix that one number in a way that makes their placebo “vaccines” look good. All the doctors have to do to get the money is report any death by an unvaxxed person as covid, and any death by a vaxxed person as pneumonia. So you can never trust the big number because that is the one the bad guys always fix.
Only sheeple accept the big number.
[And BTW, all-cause mortality shows the effects of the covid pandemic and is not inconsistent with a significant vaccine effect. The fact that the curve is messy and buffeted by contingency should not be a surprise.]
1) It turns out very frail (eg terminally ill) people are very unlikely to get vaccinated. So the baseline mortality is much higher in that group. This is termed the “healthy vaccinee bias”. There are multiple papers that look at this and report the same thing (let me know if you need me to share them again). There are also multiple papers that ignore it (or put a sentence in the limitations about it as if it is some minor thing).
What you will not find is a paper that looks at the data, attempts to account for it, and does not find it to be a giant confound. It should be easy for you to find one if I am wrong.
2) Another aspect is that vaccination correlated with testing. At first the vaccinated were less likely to get tested, because unvaccinated still had mandatory testing. Then the mandatory testing ended, and the vaccinated were *more likely* to be tested due to greater healthcare seeking behavior. At this point data collection on covid cases/deaths largely stopped.
3) The same thing (no or even deleterious impact on all-cause mortality) was reported in the vaccine RCTs. However, those were done using especially healthy participants with below average mortality.
You will quickly notice these issues if you attempt to treat this problem scientifically, ie come up with quantitative models that account for known factors and explain the data.
Of course, applying actual science is now considered “extreme” or “conspiracy theory” and people prefer their healthcare to be based on questionable assumptions and speculations.
Just to inject some data into this discussion. Here is excess mortality in 4 developed countries, US, Spain, France, and Sweden. I just chose these because I started with the US and thought of 3 more developed countries likely to have decent data and comparable living standards, not by looking at the curves or anything. I call this “haphazard” sampling.
excess mortality
And Here’s percentage of “fully vaccinated” people in those countries:
“people fully vaccinated”
All of these countries had ~50% vaccine uptake by Sept 2021. Some of them went much higher than that, and of course there will be variation in booster uptake through time.
What we don’t know is the true number infected as a function of time and age. It does seem pretty clear though that the US had more cases and more deaths prior to Omicron in early 2023, and places like Germany and Sweden had a spike in mortality that US and Spain didn’t when Omicron showed up. Basically, either those most likely to die were already dead in US and Spain, or more people had the actual infection before then and so had a more robust immunity when Omicron came along.
One thing that’s clear is that the psuedo-periodicity is fairly consistent across all these regions… basically when the virus was spreading, people were dying. That’s why I think it’s insane to think that COVID wasn’t the cause of a lot of deaths.
I could easily imagine a comprehensive model that would probe how much we really know about vaccine effectiveness, case fatality rate across age, and infections across age and time… Unfortunately I don’t know of a good way to get funded to do that research, not being part of an organization that is eligible for NIH grants, nor am I convinced that something like a Kickstarter would get any real traction to fund a few people to work on this stuff for 6 months or so (200-300k$). That being said, I think the answers here are hidden in uncertainties enough that there isn’t a clear answer to questions like “overall, did vaccines substantially reduce mortality?” and/or chronic illness, etc.
My guess is vaccines work pretty well for 4-6 months. They keep people out of hospitals pretty well. But over the long run, without constant boosters they may not have protected against mortality as much as would be desirable.
Daniel Lakeland wrote:
“…there isn’t a clear answer to questions like “overall, did vaccines substantially reduce mortality?”
What’s wrong with the fact that unvaccinated people were nine times more likely to die of Covid than vaccinated people? The uncertainties are orders of magnitude lower.
It is hard to imagine a clearer signal in epidemiology.
Anoneuoid dodges the question because of the reasons I listed above, what is your story?
@Daniel
The goal of modelling it at this point would be more to find out which data needs to be collected *next time*, so something can be learned from it. There are too many free parameters to come up with anything really convincing for covid.
However, a few points stand out to me that could vastly improve the situation:
1) Cheaper/easier screening for methemoglobin (and anything else that may mimic severe respiratory problems)
2) Assessing frailty of vaccinees
3) The details of testing (eg, cycle thresholds and “the process” that actually determines the test result based on clinical suspicion in each lab/hospital)
@Joshua
Your tweet (why not post the paper next time?) is about a paper that concludes exactly what I am saying. Healthy vaccinee bias is multiple times larger (~3x here) than the putative vaccine benefit. Ie, the *main* thing being measured using this method is the baseline difference. Trying to measure vaccine effectiveness this way is like trying to use the error due to expansion/contraction of a metal ruler as a measure of temperature.
@Matt
The problem was already explained above. It is also described in Joshua’s link, or eg here:
https://www.nejm.org/doi/full/10.1056/NEJMc2306683
There are more papers describing this as well. In fact, it is 100% of papers that investigate the issue. Additionally, I suspect your 9x number is further confounded by differences in testing and date of diagnosis (eg, unvaccinated cases were more during peak of winter while vaccinated tended towards the spring).
@ Joshua and Matt
I am trying to figure out where the failure in communication lies. I would appreciate it if you answer the following question:
Anoneuoid wrote:
“I am trying to figure out where the failure in communication lies. I would appreciate it if you answer the following question:
A parking lot is full of cars […]”
Wut.
This has nothing to do with a failure of communication. If 60% of the population is vaxxed and 40% is not, and the vaccine has no effect on mortality, then we expect the hospital death rate to be of the same order of magnitude, 60-40 or 50-50 or 40-60 depending upon issues such as nonrandom sampling, misattribution of cause of death, whatever. But the actual death rate in hospitals was nothing like that. For every vaxxed person who died in a hospital with a concurrent diagnosis of covid, approximately 10 unvaxxed died with covid. The numbers from hospitals look like what you would expect if the vaccine was approximately as effective against mortality as reported from the clinical trials. In other words, a nice clean picture of vaccine efficacy against mortality.
Despite this, you want to engage me in quibbling about uncertainties like nonrandom sampling. Why would I want to do that when the vaccine mortality reduction effect size is an order of magnitude bigger than all of the uncertainties combined?
Despite filling post after post with drivel about other stats, you still have yet to even mention the hospital vaxxed vs. unvaxxed mortality, even though it is the only statistic where we expect to see a clear effect from the vaccine. That is because you have nothing that can even dent the numbers, all you can do when it is brought up is change the topic or hint at conspiracy theories that involve pretty much the entire medical profession lying and cheating about who had covid and who did not.
Thanks, very interesting.
Btw, regarding:
The all-cause mortality rate was actually ~10-15% higher in the *vaccinated* group during the RCTs. But it was 37 vs 33 deaths out of n ~ 30k per group with below average mortality for their age (because they excluded very unhealthy people).
This was pointed out at the time:
https://statmodeling.stat.columbia.edu/2022/08/23/i-agree-with-dean-eckles-that-this-study-effects-of-political-versus-expert-messaging-on-vaccination-intentions-of-trump-voters-is-too-noisy-to-be-useful-when-will-we-reach-the-point-where-res/#comment-2074130
What you will not find is a paper that looks at the data, attempts to account for it, and does not find it to be a giant confound. It should be easy for you to find one if I am wrong.
You are wrong. You clearly haven’t looked.
Here’s a more general discussion as well as a short discussion of the estimated HVE in one short paper:
https://x.com/jsm2334/status/1682452326576017411
Could anyone please publish the most important number, which is readily available?
What is the average age of persons dying of Covid as compared to average life expectancy?
If the average age of a Covid death is older than average life expectancy (which it is), it becomes absurd to believe we should all want to get infected with Covid so we can live longer.
The whole Covid thing is fubar.
Hmmm. That link was funky for some reason (went to the wrong tweet).
This isn’t quite the right one but it will do if you just look above where you will see the link to the paper.
https://x.com/jsm2334/status/1682452619900473363
For the latest on the Bhattacharya activities (as now Director of NIH), I found the budget hearing testimony quite interesting: https://www.c-span.org/program/senate-committee/national-institutes-of-health-director-testifies-on-2026-budget-request/660983. The questions from Congress were better than usual and far more bipartisan than usual (with the Exception of Senators Kennedy and Britt who exemplify what I don’t want elected representatives to be). Bhattacharya comes across (to me) as a capable researcher and well-intentioned. But he had no facts and repeatedly (almost every question) responded saying the budget request is a negotiation between Congress and the Administration – a far cry from a defense of the budget he is requesting or an explanation of how that budget would actually work. He also claimed his job will be to ensure that the money Congress ultimately appropriates (even if it exceeds what NIH is asking for) will be spent as intended, without any mention of why that should be believed given the Administration’s actions thus far.
My political views aside, if you listen to the testimony you will find a number of issues recently discussed on this blog: the “gold standard” of science, AI, DEI and medical research and keyword search, overhead rates on grants, and concentration of research in a few “elite” institutions. Regarding the last point, while I lament the disproportionate domination of Harvard (and a few other institutions) in medical research, I do wonder whether spreading research funds among a number of smaller institutions will constitute an improvement. I think too little is understood about the production function of medical knowledge to be confident about the relative effectiveness of where the money is spent or how much overhead is appropriate. I doubt that better understanding will come through the budgetary process despite the fact that the process will assume answers to such questions.
Dale:
From a political perspective, I think the goal of the administration is to have more control. Their problem with Harvard and other institutions is that they represent independent sources of power, and I get the sense that the current government would prefer to disburse funds on a patronage basis. The stuff about gold standards, overhead rates, etc., just seems like justification for the political goal of asserting more control.
Fascinating interview with Kevin Hall whose careful work on diet, the metabolic response, and differential equation models of energy balance has been posted here before.
https://www.youtube.com/watch?v=WBllzAb_vAk
At about the 2 hour mark, he gets into the process of being censored by RFK’s allies, quitting, and his subsequent interactions with Bhattacharya. Bhattacharya comes across as:
1. Hilariously credulous when it benefits him. He at first apparently did not believe what Hall was telling him about censorship under the Trump administration until presented with strong evidence.
2. Obsessed primarily with the media. Hall was a well respected researcher and pretty senior within the NIH, and attempted to contact Bhattacharya, his literal boss, before quitting. It was only after Hall’s departure was publicized that Bhattacharya even noticed and reached out on twitter.
Other than that, what Hall experienced in the first place is an interesting intersection of science and ideology. Hall seemingly should be ALIGNED with the general thrust of RFK’s MAHA movement. The advice you can take away from his research is on the margin pretty much the same as the MAHA USDA’s “real food” recommendations (which are also pretty much the same as the pre-MAHA recommendations, in substance if not in style). MAHA is all about “real food” and RFK is against ultraprocessed “fake foods”, and Hall produced some of the most convincing evidence that ultraprocessed foods are obesogenic.
However, even being broadly aligned was not enough. Hall’s research was not indicating that ultraprocessed foods are addictive in the same way or magnitude as psychoactive drugs like cocaine. It was also trying to probe at the specifics of “why do people consume more calories in ultraprocessed foods ad libitum” and “which kinds of ultraprocessed foods are the most obesogenic and which ones are not.” It seems like RFK’s stance is that industrial food manufacturing is an ontologically evil process, the problems with which cannot be broken down into constituent parts and the harms of which cannot be overstated, and no nuance can cross that party line even at poorly attended technical conferences for experts, let alone in public communications.