Monday, August 9, 2021

Covid incompetence

 WWII started badly for the United States. Our tanks blew up. Our torpedoes were duds. Our airplanes were outclassed. Many commanders were incompetent, soldiers green, supplies chaotic. We lost a lot of battles.  But we learned. The lessons of each mistake were incorporated, incompetent commanders sacked, soldiers learned their terrible craft. 

Delta is the fourth wave of covid, and amazingly the US policy response is even more irresolute than the first time around. Our government is like a child, sent next door to get a cup of sugar, who gets as far as the front stoop and then wanders off following a puppy. 

The policy response is now focused on the most medically ineffective but most politically symbolic step, mask mandates. All all-night disco in Provincetown turns in to a superspreader event so... we make school kids wear masks in outdoor summer camps? Masks are several decimal places less effective than vaccines, and less effective than "social distance" in the first place.* Go to that all night disco, unvaccinated, but wear a mask? Please.   

If we're going to do NPI (non pharmaceutical interventions), policy other than vaccines, the level of policy and public discussion has tragically regressed since last summer. Last summer, remember, we were all talking about testing. Alex Tabarrok and Paul Romer were superb on how fast tests can reduce the reproduction rate, even with just voluntary isolation following tests. Other countries had competent test and tracing regimes. Have we built that in a year? No. (Are we ready to test and trace the next bug? Double no.) 

What happened to the paper-strip tests you could buy for $2.00 at Walgreen's, get instant results, and maybe decide it's a bad idea to go to the all night dance party? Interest faded in November. (Last I looked, the sellers and FDA were still insisting on prescriptions and an app sign up, so it cost $50 and insurance "paid for" it.) What happened to detailed local data? Did anyone ever get it through the FDA's and CDCs thick skulls that even imperfect but cheap and fast tests can be used to slow spread of disease? 

Last summer, we were talking about super-spreader events, and the idea that you don't have to have disastrous lockdowns of everything but maybe packed all-night disco parties are a bad idea? (Reopen smart, I wrote at the time, for example here) Today, silence. Masks. Nice big symbolic masks.  Period.

And then we indulge another round of America's favorite pastime, answers in search of a question. Delta is spreading, so... extend the renter eviction moratorium. People who haven't paid rent in a year can stay, landlords be damned. Usually our government mandates A shall give to B because it isn't willing to spend the money directly. Hilariously in this case it's because the government is simply unable to shovel rent money out the door, even to the scammers who have gotten so much unemployment money.  Somehow throwing people who don't pay rent into "the street" is our major source of covid spread, but releasing a wave of illegal undocumented migrants into "the community" in the next story does not, and the same people free to go to all night disco parties is not a danger. (Disclaimer: I'm all for immigration, but not for hypocrisy.) By and large the Administration and commenters can't even bother to pretend that stopping evictions has anything to do with stopping the spread of covid, the only source of government's authority. (One example I just happened to hear as I was writing this, PBS' interview of Secretary Fudge.) The rent is too damn high is all you need to know.

In the talk "pandemic of the unvaccinated," I hear, basically, resignation. We offered you vaccines. If you won't take them, fine, we're done. We're back really to what quite a few people argued for and were pilloried for in March 2020. Let it sweep through, get to herd immunity, it peters out, bury the dead and go on with life. 

The good news. A reproduction rate of 6 means Delta will spread really fast, peak really fast, and decline fast. The bad news: a lot of people will needlessly get sick and quite a few will die. The economy will slow down as people voluntarily pull back. Evolution got one more step ahead of bureaucratic bungling. A variant that transmits even more easily through vaccinated people can't be far behind.  

It did not have to happen.  The vaccine was in hand, the lines were done, anyone could walk in and get the vaccine. All we had left to do was get pretty much everyone vaccinated before the new variant hit, and it would have been pretty much over. Something like $5 trillion dollars of extra debt, the economy closed for a year, thousands dead, thousands unemployed, huge pain and now we lose sight of the ball when all we had to do was pay a little attention, get some small incentives going for everyone to get vaccinated. (Aaron Stupple suggested, just pay people $1,000 to get vaccinated. 100 million people times $1,000 is $100 billion. Couch change in what the pandemic has cost us, or the upcoming $3.5 trillion "infrastructure" bill. That's not much more than rail and transit subsidies alone. ) 

Just walk over the finish line and don't get distracted by that puppy -- or, in the case of our politics, get distracted by whether extending the wonderful job the public schools are doing to free pre-k for every toddler in the country is a good idea and counts as "infrastructure,"  whether we should refer to global warming as a climate "emergency" or "catastrophe,"  just how many trillions of dollars the government should print and spread around, and all the other brouhahas that keep Washington busy and distracted from basic governance, like stamping out the embers of a pandemic.

Yes, we are now, in the middle of it, slowly getting serious about vaccination. But, as my best post from last year pointed out, if you react to the level of disease, it's like turning the shower from hot to cold when it gets cold, and vice versa. You end up with waves. We ended up with waves. The fight against a virus has to anticipate the next wave, stamp out the embers during the lull. This is not new knowledge. You don't stop taking antibiotics when you feel better. Serious efforts to stamp out infectious diseases -- small pox, polio -- put huge effort in to the dwindling cases. 

Were anyone watching, it was easy to see that getting full vaccination was not going to be a cake-walk. Delta was known by spring, and its high reproduction rate was known. That pushes up the limit for herd immunity. And it was known that after around 60 or 70% of the population vaccinated, it gets really hard. As Stanford's Robert Kaplan reports in WSJ

A poll completed in August 2020 showed that about 20% of the population reported they were very unlikely to take a vaccine even if the evidence suggested it was safe and effective. Another 15% said they were unlikely to take it.  

(Update: I deleted a misleading graph from the CDC and surrounding discussion.)   

The internet is awash with misinformation about the vaccines. But data distortion seems to have little effect on vaccine uptake. Well over a year ago, 35% told us they wouldn’t take the vaccine, and they have kept their word. The recent increase in vaccine uptake appears to be among people who initially said they would “wait and see.”

This isn't "misinformation" cured by even more internet censorship. Indeed, it is the opposite. People know that the internet is censored, that public authorities have been lying to them all along.  After mask/no mask, Wuhan lab/no Wuhan lab, FDA explicitly trying to manipulate public psychology in delaying approval and the J&J pause, and so on, people have lost trust. CDC flip flopping on masks clearly under political pressure did not help. And none of them present facts.  The cry of wolf falls on deaf ears.  

Even the vaccine discussion now hews to the joke that everything must be either mandatory or banned. We jump to vaccine mandates, cops coming to your door to hold you down and jab you. That was not necessary and still is not. But not until this week or so is anyone talking about allowing even minor incentives to get vaccinated. Allow cruise ships to say, you're free not to get vaccinated, but you're not free to step on this private property without a vaccine! That's slowly happening, but way way too late. This needed to happen before the wave. 

Of course, only a few weeks ago did the FDA even think that maybe it's time to speed up full approval of the vaccine. The anti-vaxxers have a point. It's mighty hard to talk about vaccine mandates when it only has an emergency use authorization. 

What's going on with the Great Forgetting? Well, as with so many other things, it is to nobody's institutional interest to remember just how many mistakes were made the first time, how much better things could have been. To learn from the mistakes, and institutionalize better responses would mean to admit there were mistakes. One would think the grand blame-Trump-for-everything narrative would allow us to do that, but the mistakes are deeply embedded in the bureacracies of the administrative state. Unlike bad admirals in WWII, nobody less than Trump himself has lost their job over incompetent covid response. The institutions have an enormous investment in ratifying that they did the best possible job last time. So, as in so many things (financial bailouts!) we institutionalize last time's mistakes to keep those who made them in power in power -- which means we do not learn from mistakes. 

Delta will blow through in a couple of months. It looks like it takes about 6 months for immunity to die out, and the same for a new variant to adapt. See you in February or so, and let's see just how many more basics of public health our government can forget in the meantime. Yeah, Grumpy is grumpy today. 

More: A spectacular piece on the FDA by Scott Alexander, which I plan to blog shortly. (HT Russ Roberts) But worth it now. The piece is particularly good for diagnosing the FDAs awful incentives, which is how good people get wrapped up in a dysfunctional institution.  Zeynep Tufekci  and separately Jeneen Interlandi on the CDC, in the NYT no less (HT marginal revolution). MR on the NIH in the pandemic:

● Of the $42 Billion 2020 NIH annual budget, 5.7% was spent on COVID-19 research

* On masks. Of the many things we don't know, just how much masks help or don't help is one of them. You think with $42 billion dollars one could find out. Of the studies I have read and seen cited, I see a guesstimate of 20% reduction in reproduction rate. So if Delta has R0=6, masks might reduce that to 4.8. Even if a vaccine is only 50% effective in stopping transmission, then R0 among the vaccinated is 0.25. 

Masks do much more to stop you from giving it to someone else than to protect you. Cloth masks are close to useless. Well-fitting N95 masks work much better in both directions. Neither comes close to vaccination. Wearing cloth masks outdoors, far from other people, in the wind, as is the fashion in Palo Alto, is just part and parcel of the pointless virtue-signaling so prominent here. If you do go to that crowded all-night disco, wearing an N95 mask might be a good idea. Of course if you’re even thinking about wearing a mask, you’re not going in the first place, which is why the whole mask-mandate business is a bit silly.

I welcome comments pointing to good numbers on this question. 

Update: Tweeters seem to think I argue here that masks do no good. No. For the record: Masks work a bit. Not getting in contact with other people works better. And vaccines work a lot better. 

Update 2: On the effectiveness of masks, a good summary of evidence. Note mask mandates may "work" where masks themselves don't, because forced to wear masks many people choose not to go out in the first place. But mandatory dunce caps would work as well for that. 

The real news of the link is this: we don't have any reliable controlled-trial evidence on the effectiveness of masks against covid-19. Stop and process that for a second. Maybe they work. Maybe they don't. But our policy wonks are imposing mandates, and focusing on masks as the central tool of pandemic policy, based on essentially no evidence whatsoever. 



48 comments:

  1. Early on (quoting myself) it was predicted COVID-19 would be the public health sector's Vietnam. Always more money needs to be spent,to get around the next corner towards the light in the end of the tunnel. Endless threats and darkening outlooks are constantly cited.

    But what is winnable in theory cannot be won by large-scale government on the ground, in a real world of human beings, with human strengths and weaknesses.

    As in Vietnam, declare victory and quit.



    ReplyDelete
  2. I die a little inside every time someone brings up Tuskegee. First from https://www.jhsph.edu/news/news-releases/2005/brandon-tuskegee.html,

    "The Hopkins researchers completed a telephone survey in the summer of 2003 of 277 African-American and 101 white adults, aged 18-93 years, living in Baltimore, Md. Only two-fifths of those interviewed had heard of the Tuskegee Study. Of those who had heard of it, most knew few details of the study."


    Second from https://www.buffalo.edu/ubreporter/archive/2010_04_14/tuskegee.html,

    "Results of this random survey of 356 African Americans, 493 whites and 313 Puerto Ricans show that 89 percent of respondents across the ethnicities could not name or identify the Tuskegee study."

    ...

    "There was no difference between ethnicities in 'likelihood of participation' in medical research, the study found, despite the fact that blacks reported being 1.8 times as likely as whites to fear being used as 'guinea pigs.'"


    Third from https://www.kqed.org/news/11861810/no-the-tuskegee-study-is-not-the-top-reason-some-black-americans-question-the-covid-19-vaccine,

    "The conclusions were definitive: While Black people were twice as 'wary' of participating in research, as compared to white people, they were equally willing to actually participate. And, there was no association between knowledge of Tuskegee and willingness to participate."

    ...

    "Tuskegee was not the deal breaker everyone thought it was."

    "These results did not go over well within academic and government research circles, Warren says, as they 'indicted and contradicted' the common belief that low minority enrollment in research was the result of Tuskegee."


    It's one of those things people bring up because it's a readily-available "educated person" narrative that can be autopiloted into discourse without having to do any of the hard work to get real understanding.

    ReplyDelete
    Replies
    1. The conclusions of these studies actually frighten me for different reasons. The fact that so few people remember this absolutely horrendous study suggests either people forget quickly or education is still terrible for most people.

      I shudder to think of the day the Nazis and the Soviets become an antiquated part of our history and largely forgotten by the public

      Delete
    2. Agreed. Apparently we are close already: https://www.pewforum.org/2020/01/22/what-americans-know-about-the-holocaust/

      Delete
    3. The Tuscagee study wasn't horrendous. The blacks at Tuscagee were the control group, they received no treatment for syphilis. The other group received the standard of care at the time, topical treatment of mercury, and that failed to cure syphilis, it made them sicker. So, the blacks, the control group, got a better deal because doctors at the time were morons. After the study had been going 20 years, a worthwhile treatment was invented and was not given to the control group. Obviously, this was a horrendous mistake but it didn't happen until the study was 20 years old.

      Delete
  3. If it is any consolation, the US does not seem to have botched the situation as badly as Australia has:

    https://quillette.com/2021/08/09/covid-zero-was-it-worth-it/

    ReplyDelete
    Replies
    1. As an Aussie I feel compelled to offer a defence, but really, only a bit. A gilded cage is still a cage. A few months back we congratulated ourselves on the lack of the widespread loss of life, that's true. Today, it's a different kind of bureaucratic crapheap to that in the US, but it's no less infuriating. On balance I'd probably prefer the Australian COVID experience to the US one though.

      But it's also a function of the casual attitude of many voters to COVID as well. We had a little outbreak at Christmas last year which cancelled many holiday plans at the time, but little outbreaks don't concentrate the mind to urgent and meaningful action the way that a raging pandemic does. Australia today is like Europe in March 2020: fearful, angry, and short of vaccines, and that fear is driving rapid vaccination figures. And just like in the US, some parts of the country seem to be very worried (Melbourne), while others seem to be pretty relaxed (New South Wales), and others are so complacent they won't realise that it's a catastrophe until they're personally infected (parts of Queensland and South Western Sydney).

      I expect that this will turn around fast though. Bear in mind that Europe went from locked down in April to partying in July, only a few months apart. But if you thought we had the world prize for isolation and complacency, have a look at New Zealand.

      Delete
  4. In addition to Scott Alexander's original post on the FDA, he has published two more posts amplifying one part of his original narrative:

    Details Of The Infant Fish Oil Story
    https://astralcodexten.substack.com/p/details-of-the-infant-fish-oil-story

    Contra Drum On The Fish Oil Story
    https://astralcodexten.substack.com/p/contra-drum-on-the-fish-oil-story

    Original Story:

    Adumbrations Of Aducanumab: Is the FDA too lax? Do bears go to the bathroom in spotless marble-floored lavatories? Is the Pope an Anabaptist?
    https://astralcodexten.substack.com/p/adumbrations-of-aducanumab

    ReplyDelete
  5. minor side note: Those Abbot BinaxNOW antigen tests have been available OTC for months now at pharmacies like walmart walgreens, CVS, amazon etc... roughly $10/per test however

    ReplyDelete
  6. Most restaurants don't demand vaccine proof for entry. Whether it's because they fear loss of revenue or can't afford the fixed costs for checking, it's not like the private sector is being cajoles by the government.

    And unless the government literally makes it illegal with punishment being jail time, there will always be anti vaxxers who will keep the current and next virus hanging around.

    Masks is the most painless, most convenient policy in terms of optics. I think if they could get away with lockdowns again, they would have passed that too.

    ReplyDelete
  7. "It did not have to happen. The vaccine was in hand, the lines were done, anyone could walk in and get the vaccine. All we had left to do was get pretty much everyone vaccinated before the new variant hit, and it would have been pretty much over. "

    I agree - it was clear that if everyone went and got the vaccination we'd be in great shape.

    What stopped that from happening? Your article doesn't seem to explain how things went wrong. Did the government put obstacles in peoples path? Or did people choose to not get vaccinated?

    ReplyDelete
  8. > Our government is like a child, sent next door to get a cup of sugar, who gets as far as the front stoop and then wanders off following a puppy.

    But a human being cannot live his life moment by moment; a human consciousness preserves a certain continuity and demands a certain degree of integration, whether a man seeks it or not. A human being needs a frame of reference, a comprehensive view of existence, no matter how rudimentary, and, since his consciousness is volitional, a sense of being right, a moral justification of his actions, which means: a philosophical code of values.
    -Ayn Rand

    ReplyDelete
  9. > Our government is like a child, sent next door to get a cup of sugar, who gets as far as the front stoop and then wanders off following a puppy.

    But a human being cannot live his life moment by moment; a human consciousness preserves a certain continuity and demands a certain degree of integration, whether a man seeks it or not. A human being needs a frame of reference, a comprehensive view of existence, no matter how rudimentary, and, since his consciousness is volitional, a sense of being right, a moral justification of his actions, which means: a philosophical code of values.
    -Ayn Rand

    ReplyDelete
  10. Yes, government has done a poor job of managing the pandemic while sowing distrust among the public with inconsistent messaging. But at this point first order externalities are solved – the vaccines are highly effective in preventing serious illness or death. Those who don’t get vaccinated are highly limited in their ability to harm the vaccinated. First order, they choose to risk only harming themselves. I agree, it’s needless suffering and people will die, and the government should allow more private incentives, but these are second order effects now.

    There are 100,000 alcohol related deaths annually in the US every year. Going forward over the next year Covid might get into this ballpark for the unvaccinated – maybe, if the unvaccianted are completely intransigent. That’s the appropriate scale of the problem now. We don’t ban alcohol. Perhaps living with Covid and choosing to return to normal life is the best thing we can all do – let the rehumanization of society begin.

    ReplyDelete
  11. Small correction. I think the light blue bar is interpreted as: among those who completed all recommended doses in the last 14 days, what fraction are in each ethnic/race category. So comparing it to their fraction of the population, it does indicate that whites are taking the vaccine at a lower rate. This does suggest that Trumpers are still holding out.

    However, interpretation is subject to one-third of all not being classified.

    ReplyDelete
  12. John, could you please point to reasonably credible evidence (from your perspective/standards) consistent with the assertion that masks are "the most medically ineffective step"? Thanks

    ReplyDelete
    Replies
    1. Prof. Cochrane might have better data... Published literature on masks may be divided into three categories: 1: models computing that masks should work; 2: empirical studies and one RCT that find minimal effect; and 3: randomized controlled trials that show no protection from influenza with N95 masks in hospital settings.

      Category 1 isn't worth discussing, in my opinion.

      Category 2:
      https://docs.google.com/document/d/13Xt6pN_VASGOd3abMafH2Jj3Y2MMnj9NYF512KJLJ2M
      https://www.acpjournals.org/doi/10.7326/M20-6817
      https://wwwnc.cdc.gov/eid/article/26/5/19-0994_article
      https://www.aier.org/article/lockdowns-do-not-control-the-coronavirus-the-evidence/
      https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2020.00818 finds a 2% relative reduction
      etc

      Category 3:
      https://bmjopen.bmj.com/content/5/4/e006577.long
      https://wwwnc.cdc.gov/eid/article/26/5/19-0994_article
      (These are pre-Covid. Post-Covid reports require critical study.)
      etc.

      As Jay Bhattacharya points out in the interview referenced below, masks do have costs. They are terrible impediments to elementary and intermediate education. More important, they send a constant, false message that other human beings are bags of germs, mortal threats from which we must all guard ourselves - not people just like us, our friends and neighbors.

      Delete
  13. Please check out https://www.hoover.org/research/matters-policy-politics-delta-variant-dummies - Jay Bhattacharya discusses Covid-19 with your colleague Bill Whalen

    - The coronavirus vaccine - like influenza - is "protective," not "sterilizing." Vaccinated people can still catch and transmit the disease, but have morbidity and mortality 50-100X below unvaccinated people. Because the vaccine is not sterilizing (like vaccines for polio and smallpox) and because there are many animal hosts for Covid-19 (dogs, cats, deer, bats - possibly all mammals) (unlike polio and smallpox) this virus will never be "controlled," let alone eliminated.

    - Because vaccinated people tend to have asymptomatic infections, and because the vaccine offers amazingly good protection, it is hard to justify the social and economic cost of your continuous testing strategy. It's not clear it will offer any benefit at all. Similarly, because Covid-19 vaccination does not generate a positive externality (it doesn't protect other people) vaccine mandates are very hard to justify.

    The best answer is to get vaccinated and enjoy life.

    - No randomized controlled trial or any natural experiment shows any benefit from masks. Masking is justified by models based on untested assumptions. RCTs testing N95 masks in hospital settings found no effect on influenza transmission.

    -

    ReplyDelete
    Replies
    1. I believe vaccination does generate positive externalities. First, I believe recipients of the vaccine are less likely to get the virus if they’ve been vaccinated and second, they are likely to have much smaller viral loads, which should reduce transmission.

      Delete
    2. I think vaccination has significant positive externalities.
      1. If you are vaccinated, you are less likely to get the virus and therefore less likely to pass it on.
      2. If you are vaccinated and get the virus, your viral load will be smaller and you'll be less likely to pass it on.
      3. If you are vaccinated, you are less likely to get sick or die and therefore less likely to consume healthcare resources.
      4. If you are vaccinated, you are more likely not to get sick and therefore continue to be a productive member of society.
      Etc.

      Delete
    3. That podcast was tremendous, thank you for bringing it to my attention.

      Delete
    4. How are there no positive externalities? Vaccination doesn't only turn symptomatic infections into asymptomatic ones: it all up *prevents* infection (up to its level of efficacy). This was shown fairly early on.

      Delete
    5. The Director of the NIH goes on Eunday talk shows to promote the use of masks. Is he a moron or is unaware of any credible RCT study on this? The study from Duke and the NAcOfSci are not credible enough?

      “focusing on masks as the central tool of pandemic policy” is not true. Administration has focused on vaccines, and masks given vaccines hesitancy. And I’m not dissecting the chart show above.

      Delete
  14. It would be useful to see the population of all unvaccinated divided into the categories you use (white non-Hispanic, etc.), both the absolute numbers and percentages.

    ReplyDelete
    Replies
    1. Good data at https://www.kff.org/coronavirus-covid-19/issue-brief/latest-data-on-covid-19-vaccinations-race-ethnicity/

      Delete
    2. The wording of that article bothers me. For example it makes statements about how many people have "received" vaccinations, as though people are passive and waiting for someone to hand them a vaccine. It should be making statements about who chose to go and get vaccinated.

      Delete
  15. Sorry for a late post on quantity, but my eyes glazed over, which spilled over into my brain.

    The CDC chart is presented in such a way as to measure the degree of equality of vaccination incidence, and even that is done opaquely.
    If I want to know what share of the Black population is vaccinated, I must multiply the Black share of all vaccinated by one over Black Population share in total population times share of vaccinated in total population. Or, 0.091 x (1/0.124) x 0.501 = 0.367.
    I won't calculate for the White population because my head hurts. 'Twill be higher.
    If I have made an error, please let me know.

    ReplyDelete
  16. Dr. Cochrane, deep down I think you're an idealist. Me too. It's easy to get grumpy when you see insanity unfold before your eyes. Human behavior triggers that easily.

    There's also the issue of COVID fatigue, too, I think. The vigilance we saw last year has largely died out. Though I still see people wearing masks, as I still wear one as well, overall it seems people have accepted this as a new normal but still hate it.

    What's tragic are the anti-vaccers now lamenting as they die that they wished they had gotten a vaccine. Hopefully few listened to their misguided advice.

    Silver linings? We got to see how messed up America is for many. It exposed a lot of structural problems in government and the economy. But will we learn is the ultimate question.

    Best,
    M

    ReplyDelete
  17. There is sufficient data to show that masks are effective in controlling the spread of Covid-19. I documented a number of studies last year when I was doing my daily newsletter of all the papers that were coming out. Most conclusive real life study was the Jena Germany study that was published earlier this year in the Proceedings of the National Academy of Sciences: https://www.pnas.org/content/117/51/32293 and just yesterday there was a NY Times story from Duke researchers showing the impact of masks in schools: https://www.nytimes.com/2021/08/10/opinion/covid-schools-masks.html there are other papers as well and one should not throw out comments when they can easily be disproven.

    Vaccines should have been mandated early on as anyone with a background in infectious disease realized that this was the one good way to stop the pandemic in its tracks. Things would be much different today and Arkansas would not be at a point where all ICU beds are occupied.

    The amount of confirmation bias among those with no expertise in infectious disease or epidemiology was astounding last year but I've grown accustomed to it now.

    ReplyDelete
    Replies
    1. Alan, I'm curious to get your take on https://twitter.com/davidzweig/status/1425131046782259204?s=21.

      I looked at the PNAS study you linked and I find it not very conclusive.

      I'm unclear what to think about masks. Intuitively, it seems like they *ought* to be effective --- the causal mechanism is pretty obvious. I certainly wear one frequently.

      Delete
    2. I do not know what to believe regarding masks. I followed the links by Skeptic above which suggest small or no effect. The PNAS study finds a 45% reduction in cases but uses synthetic control of aggregated regional differences, not an RCT. I suspect cherry-picking of citations.

      Is there a comprehensive lit. review?

      Delete
    3. ICU beds are occupied because the number is tightly limited. Emergency rooms must treat everyone who arrives as long as beds are available, they must even treat indigent people who can't pay. Hospitals respond to this by keeping a small number of ICU beds and occupying them as often as possible so that they can reject patients who can't pay.

      There are plenty of regular hospital beds available to treat everyone. In Colorado, 6% of the beds are currently occupied and other states are similar.

      Delete
    4. Have you read the studies you're citing? The Duke study that just came out had *no control group*. As such, it showed evidence of absolutely nothing.

      Delete
  18. Great article John.

    Pick your headstone: humankind, USA, Constitution, Economy, or pandemic under RIP they all say the same thing: We don't learn.

    Me grumpy too.

    ReplyDelete
  19. I recall Arnold Kling's admonition in the early days of the pandemic: Fire the peacetime bureaucrats!

    ReplyDelete
  20. Back in March of 2021 I read this AEIR paper. https://www.aier.org/article/masking-a-careful-review-of-the-evidence/
    I noticed the review of the Danish Study included enough data to estimate the efficacy of masks. Using the formula for efficacy of a treatment,https://en.wikipedia.org/wiki/Vaccine_efficacy, I came up with 0.15. This means that wearing a mask reduces your risk of getting covid to 85% of the risk from not wearing a mask. So, if the prevalence of covid in your environment is 5% wearing a mask reduces the chance of getting covid to 4.25%.

    Would the FDA have found any of the vaccines effective with this level of effeciveness? Has the FDA found any drug effective with this level of efficacy? If they haven't, then why does the CDC and others insist that people wear masks? Are those making the recommendation simply supporting quackery? Or, is the "sanity waterline" simply too low? https://www.lesswrong.com/posts/XqmjdBKa4ZaXJtNmf/raising-the-sanity-waterline

    ReplyDelete
  21. "But, as my best post from last year pointed out, if you react to the level of disease, it's like turning the shower from hot to cold when it gets cold, and vice versa. You end up with waves. We ended up with waves."

    Counterpoint:

    What we've generally had are 6-8 week waves, the timing of which line up very well with Hope-Simpson seasonality for respiratory viruses.

    The history of other respiratory viruses shows similar seasonal "waves". That's true for influenza pandemics, including ones where reactions were far less than what we've done in reaction to SARS-CoV-2. We also see seasonal waves with respiratory viruses such as seasonal influenza and common cold viruses, which result in little if any change in people's behavior.

    It's possible that some of the behavioral changes cited by John impact the ultimate magnitude of peaks and troughs. It is not compelling, however, as an explanation for why the waves occur nor for the timing of peaks and troughs. For one thing, I'd expect to see much more variability in the curves if John's explanation is the dominant driver of rate of spread. We'd see much more difference in timing - lots of increases (or declines) where a trend reverses after only a couple week move in a certain direction, places that bounce along at a peak level plus or minus 20% for a month or two (rather than sharper increases/peaks/declines), etc.

    In short, there's too much difference in *how* people in different areas react to think that human behavior is the dominant driver of such similar shape/timing for these curves.

    ReplyDelete
  22. I understand ethical issues, but surely we can strap masks on to ferrets, and give a couple of them covid.

    ReplyDelete
  23. Masking is liturgy. That's all it is.

    ReplyDelete
  24. Hi John, do you see any shift in public health policy towards the “nudge” type of incentives you talk about? Is there even room to do that - I recently read that while Walmart tried to mandate vaccinations for staff, it could not do so for the large mass of blue collar workers because of labour shortages (which I understand also has to do with misplaced incentives)

    ReplyDelete
  25. If 50% of the population is vaccinated and no longer susceptible, the basic reproduction number is half that when 100% of the population was susceptible. So, instead of R0 = 6 to 9, R0 will be 3 to 4.5 -- still higher than the original strain but less concerning than if none of population had been vaccinated. The Delta or (East) Indian strain showed up in India in October of last year. By the end of February of this year, the Delta strain was reported to be in the U.S.

    Apart from masks, and vaccinations, 'practicing social distancing', and 'self-isolation', governments could have ring-fenced India and Pakistan by refusing to admit visitors and migrants from those two countries. This is the policy that Canada adopted, albeit somewhat too late--but better late than never. Another policy that could have been adopted in the U.S., if it had its act together, would have been to seize on any outbreak of the virus in a geographic area and saturate the area with vaccine stations and vaccination drives. This is what the Canadian province of British Columbia did early on when pockets of Delta-variant showed up in the Greater Vancouver metropolitan area and the Alpha-variant showed up at the Resort Municipality of Whistler, BC. At Whistler, the provincial medical authorities closed down all access into and out of Whistler--just shut it down, quarantined the entire population, and then airlifted vaccination teams into the town and began vaccinating the population. It worked. In the case of Surrey, BC, and Delta, BC, the provincial authorities worked with the leaders of the local immigrant population (South Asian) to overcome cultural barriers that were preventing the uptake of vaccinations in that population which had specific and large exposure to the Delta-variant due to visitors from South Asia mingling with that population. This approach was feasible, where geographic isolation and large-scale quarantining of entire towns and villages was thoroughly unworkable. In consequence, vaccination uptake soared, and the outbreak dwindled. That's not to say that it would have worked in the U.S. in the same way; but, without the political will and the medical capability to undertake 'drastic' measures, you get what you're currently experiencing in the U.S. southeast and south. Cultural norms matter--as was seen in the case of minority population enclaves in metro-Vancouver. Political leaders and public health officers have to be prepared to quickly adapt to changing circumstances and bring intense pressure against the virus wherever it breaks out. Cooperating with local community leaders is an important element in the strategic arsenal of government. 'Big data' is a tool that the provincial authorities in BC have been able to utilize advantageously. There are no impediments in the U.S., other than an excessive degree of political polarization at both federal and state levels, to adopting and adapting similar measures. As to the CDC, FDA, etc., in AD 2020 the government broke all taboos and brought vaccines to the people in record time. That took political will-power and vision. Something similar will be needed soon if the pandemic is ever to be reduced to the level of a mere endemic.

    ReplyDelete
  26. That's a strange figure.

    The gray bar gives the ethnic's group share of the total population. The red bar gives the ethnic's group share of the vaccinated population (>= 1 dose). It gives, P(race | vaccinated) rather than P(vaccinated | race).

    Fine, We can use Bayes rule and say P(Vaccinated | Race ) = P(Race | Vaccinated) / P(Race) * P(Vaccinated). A demographic group has above average vaccination if the red bar is longer than the gray bar. The only large group that's true for is multiple/other! In fact, the graph implies hispanic whites have a higher conditional probability of being vaccinated than non-hispanics whites.

    My only way to make sense of every group being at or below average (besides multiple races) is that we're seeing more people identifying as multiple races in the Covid survey or perhaps some kind of selection story. I think a lot of people may take a quick glance at this figure and not understand what it does or doesn't mean.

    ReplyDelete
    Replies
    1. The chart is useless. It does not represent what it purports to represent. First, it omits 36.9 pct. of the fully-vaccinated population. Second, it includes a category "multiple/other non-hispanic" that over-represents the demographic proportion of individuals in that category, i.e., the fraction of the population in that supposed category (2.3%) amounts to 7,662,968 individuals (census population of 2021) versus the purported number of fully-vaccinated persons self-identifying with that category 8,639,441 reported in the chart.

      Pretty pictures, pretty useless for any practicable purpose.

      Data:
      2021 Census population: ............................................. 333,172,543 (100% of POP)
      No. of persons represented on the chart: .............106,305,743 (63.1% of FVP)
      No. of persons not represented on the chart: ...... 62,056,315 (36.9% of FVP)
      Total number of persons fully vaccinated: ............168,362,058 (50.5% of POP)

      "FVP" = "fully vaccinated persons". "POP" = "population". 62,056,315 persons (18.6% of POP) did not disclose their racial category. The chart presents the distribution of a censored sample. No amount of manipulation will yield useful information as to the proportions of each census racial category fully vaccinated, or recently completing their full vaccination course.

      Delete
  27. Covid doesn't cause blindness, does it?

    ReplyDelete
  28. Somewhat agree with most these points, but with an order of magnitude less confidence. You are way, way over confident for someone with no expertise in medical science.

    ReplyDelete
  29. My doctor said the vaccine is good for Delta. Who should I believe?

    ReplyDelete
  30. I do not understand by what math you get to R0=0.25 among the vaccinated? 50% effective would divide 6.0 by 2, right? Are you first reducing infection of vaccinated by more than 10x and then cutting the infection window in half?

    You said: Even if a vaccine is only 50% effective in stopping transmission, then R0 among the vaccinated is 0.25.

    ReplyDelete

Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.