It's a little far out to be speculating about this but I wonder what the unintended consequences of fast-tracking regulatory approval for more modular vaccine platforms will be. It's something we obviously ought to be doing, but I wonder what the world of vaccine, or even drug development, looks like when modular platforms are orders of magnitude faster to get to market.
I've heard a lot about Vitamin D's efficacy (or lack thereof) in treating COVID. Right now, my overall takeaway is something like "Being deficient has very bad correlations with outcomes, mediocre studies tentatively suggest that supplementing vitamin D could have very good causal effects on outcomes, for some reason we have yet to do non-mediocre studies on this yet." I've been taking 6000 IU per day, since I'm not exactly basking in sunlight these days so it'll improve my bone health if nothing else and even a 10% chance of it being protective against COVID has massive ROI. What evidence have other folks seen? Opinions by knowledgeable medical professionals / replication-crisis-y folks especially welcome.
> Prediction: 50% chance that sometime in 2021, the FDA grants a pharmaceutical company general approval for coronavirus vaccines which can adapt to changing virus strains without going through the entire FDA approval process again, and that whatever fast-track lane they get takes less than three months between creating the vaccine and it being approved for general use.
I think this prediction is far too low: the initial emergency use authorization for the Moderna vaccine took ~5 months from completion of the first trials, and that was for a new technique. Now that it's a proven technique, I'd expect that it'd take far less time the second time around.
I want to give feedback that the big grey boxes with blocky text look fairly ugly to me. At first I thought they were straight-up bugs, or inserted screenshots from elsewhere. I think I’d strongly prefer the reading experience of blocks-of-italics. (But not sure!)
I've seen a lot of reports of brutal side effects, particularly from shot #2, of the mrna vaccines. It's possible, I suppose, that this would go down with more careful dosing...but as is, I think this alone nukes any possibility that everyone takes a Moderna/Pzifer shot yearly. (And once everyone isn't going to be doing that, game theory says you shouldn't.) Not least because the accumulated total risk to serious reactions goes up, as far as I can tell, and both autoimmune/allergic things seem (to my non-expert view) to only get worse for most people. I think it's highly unlikely we all get our yearly covid shots.
That is to say: either we crush it right now with the first round of vaccines or *the game is over* and we should give up. And we've missed the chance.
It seems to me there are two overwhelmingly important variables for determining how "normal" things get in 2021. The first, which you mention but I think still underplay, is the extent to which vaccinated people are protected against severe disease and death even if/when the virus mutates to possibly be more (re)infectious.
The second, which it doesn't look like you mention, is the extent to which treatment protocols continue to improve, particularly treatments which can keep people out of hospitals rather than just saving their lives once they get sick enough to be hospitalized. The Treat Early folks are quite bullish on fluvoxamine and ivermectin, for example, and have a longer (and regularly updated) discussion here:
"There are whole countries whose cuisines are still built around weird decisions they made as part of World War II rationing."
You can't just throw this out there without a link! Total nerdbait. Simple searches just tell me how tough it was in a 1930s car with less gas available - I have to know!
my main question is how long until I'm allowed to make new friends again? right now some people are doing online dating, but seemingly nobody (in Blue Tribe circles anyway) is making new non-romantic friends. And due to moving right before COVID started, I'm no longer geographically near all my old groups of friends.
It's painful to imagine that even if restrictions on public gatherings eventually ease, I might still have to basically live a totally solitary life for many more years, as nobody who would be in my social circles will be willing to interact with a new person. I really hope you are overestimating the effect the new strains of the virus are going to have.
If this is true I might have to consider either trying to move to a country like Taiwan or New Zealand where they can manage it more successfully, or else moving to a red state where people just don't care.
How much were the three waves in the US over the last 11 months separate waves hitting the same people three times versus how much were they waves hitting separate people once? I really don't know the answer to that question.
The first wave hit some big cities, especially New York, and was very bad for blacks and Hispanics (especially, I'm guessing, Caribbean Latinos).
The second wave was worst in the Sunbelt air-conditioning belt of Arizona, Texas, and Florida, and hit Hispanics (especially Mexican Latinos) hardest.
The big third wave started in the Upper Great Plains and moved south with the cold weather. It hit non-urban whites hard for the first time, along with Hispanics for the third time. Blacks were not hit as hard in the third wave. For awhile, whites had a higher rate of excess deaths during the third wave than blacks did, although last I looked blacks had pulled even.
My guess is that a combination of being closer to herd immunity and "once-bitten, twice-shy" learning from experience makes places and groups that were previously highly vulnerable more hardened targets for the next wave.
By now, covid has has hit almost all of the country hard except, perhaps the northwest coast from the Bay Area to Seattle and upper New England.
Hey Scott, Sorry if someone has already suggested it, but have you considered putting (/ having an intern put) the old layout back on the blog archive? It's a bit easier to read, and also has some nostalgia value.
Hi - new member here, still learning the ropes. 1) the efficacy rate for flu vaccine is between 30% and 50%, which doesn't matter a great deal as the disease for many sufferers is not severe. It would suck if the rate for Covid vaccine stabilises around this number. 2) There was something last week (sorry, I wasn't tracking sources last week) about the vector used to propagate the Astra Zeneca vaccine. It is a chimpanzee vector. The story was, the human body when vaccinated learns how to prompt Covid19 anti-bodies the "chimp way" and that learning sticks - and when a new human-derived mutation arrives, the human body doesn't respond effectively.
Perhaps a prepper-like splinter culture of "CoViD recluses" will develop: people who either move to remote rural locations, or to "filtered communities" with elaborate protection including domes, UV lamps, air filtration, and so on.
Ethnic minorities will be stereotyped as infectious ... wait, that happened before.
It looks like we will have between 125 million and 150 million vaccinated after Biden's first 100 days. Beyond that somewhere between 60 and 100 million already have the antibodies. At the low end 185 million are resistant. At the high end, 250 million. So, there are fewer and fewer people who can spread a March bump.
Does anyone around here (or maybe on Metaculus) have predictions for whether or not there will ever be challenge trials (in the US or elsewhere)? Besides the obvious effect challenge trials should have on all other related predictions, this also potentially impacts my personal decision of whether or not to get a vaccine when it becomes available to me (as I would like to volunteer for a challenge trial).
i'm very curious what people here think about repurposing of already-approved pharmaceuticals for early treatment and pre-/post-exposure prophylaxis. To start the discussion off, i know there's some data for TMPRSS inhibitors: ambroxol/bromhexine and also pretty much like, every antiandrogen since AR affects TMPRSS2 expression. TMPRSS inhibition has direct antiviral activity. Also people have been looking at drugs with immunomodulatory effects such as fluvoxamine and ivermectin. The immunomodulatory part is important because AFAICT the actual damage caused by the disease seems mostly to come from severe immune system dysregulation and its consequences (immunothrombosis, hyperinflammation) so tweaking the immune system's operation so it doesn't fall down that path is very much desired. Doubly so if the immunomodulatory drug in question doesn't affect viral clearance. There's some trials for ivermectin showing faster viral clearance and quicker cessation of anosmia and also an RCT for fluvoxamine with really quite good results. There's also a certain endosomal entry inhibitor that became a culture-war subject but honestly my understanding is that without blocking the TMPRSSes or at least TMPRSS2, blocking endosomal entry is pretty much worthless.
It seems to me that fear is causing us to massively over correct to try to prevent unlikely and unrealistic failure cases. For example:
People are afraid that unsafe vaccines will get approved because the procedure is getting rushed. Instead of trying to maintain the most critical parts of the testing process, and skip the rest, we are scheduling one month of thumb twiddling between when we get the data and when we meet to discuss it.
People are afraid of line cutting so we are prosecuting doctors who distribute vaccine doses out of order instead of letting them get thrown out.
People are afraid of price gouging, so we all decided not to do it, which removed the incentives to scale out production as quickly as possible.
Pretty surprised at the confidence of B117 resulting in a 4th wave so soon. That seems to under emphasize seasonality/immunity and over emphasize behavior to my mind. >10% of the us is currently had at least one shot and a partially overlapping ~10-20% (depending on how much under testing you think there has been) have been infected and gained some level of immunity. If you ballpark guess that in march something like 20% have some degree of protection does that not compensate for the variant being more infectious, or are my estimates way out of line?
The second-to-last prediction would be more interesting to me if it gave a probability for people wearing masks indoors (say, a grocery store). Outdoors, people are much more cavalier about masks already, and are already drifting down towards mostly-not-wearing (plus generous distancing) in my Silicon Valley neighborhood. Indoors, they're still at nearly 100%. And this fits with what I've heard about transmission-- that indoors is by far the greater risk.
90% confidence in "not much outdoor masking" seems to imply a 10% probability that our mid-2022 masking norms aren't just paranoid and security-theater-y, but actively irrational. One would hope there's a somewhat higher risk that we'd still be wearing masks in grocery stores.
I think this is too pessimistic. The math on vaccination and herd immunity suggests that once you get to over 50% immune... The r value drops and keeps dropping in an additive manner.
I think this is fairly plausible, and here’s a question not discussed there: Where in this multi-year to permanent cycle do schools again educate all of the kids they were, in the buildings they were, where having 25-35 in a classroom, very often closer than 6 feet apart, was the norm?
One big question is when high risk activities, such as major concerts, sporting events, and conventions happen at 100% capacity. Whole industries, businesses, and urban developments are built around these kind of events happening on a regular basis, and they can't be put on pause forever. At some point things need to resume or people need to figure out what to do with all those useless arenas, stadiums, and hotels.
I've read that they're saying the vaccine might not stop the spread: vaccinated individuals could become asymptomatic carriers. To what extent is this a real possibility, versus an abundance of caution?
Like, I understand they haven't run specific trials on that for these specific vaccines, but for all the vaccines we've had in the past, how effective have they been in preventing transmission?
How much worse is it for places to want everyone to wear masks than for places to want men to wear underwear and pants nowadays, or for places to want everyone to wear hats or ties at some other points in time? It'll at least be a weird fashion restriction that serves some function in lowering the overall burden of respiratory diseases.
"Sorry, things may get worse before they get better."
Depends on how you define "things". I believe the only metrics we should really be looking at are hospitalizations and deaths (yes, Long COVID is a problem, but a lesser one).
Cases will quickly start to decouple from deaths and hospitalizations. So a spike in cases is not as big of a deal, and really shouldn't result in many additional restrictions, at least not ones similar to the winter spike.
We are soon entering a period where not getting infected will become a matter of personal preference, rather than an altruistic endeavor to avoid killing someone or causing them serious illness.
I'd predict that the cultural effects will be influenced by economics, culture-war factors, or a desire for signaling. Profitable gatherings will continue to be encouraged by advertising; more spontaneous or altruistic ones will stay online. Public group activities will be tilted towards the political factions that are now more likely to engage in them. I'd give a 70% chance that at least one NFL stadium reaches 90% capacity during the 2021 season, a 90% chance that food service workers in Democrat-governed cities will remain obligated to wear masks through the end of 2021 (and beyond), and a 75% percent chance that mass transit ridership stays below 2019 levels for a decade (unfortunately for the climate).
What I have wondered about for a while, is whether a system would arise (or be proposed) for allowing people entry to mass gatherings *only if they provide proof of vaccination*. The Super Bowl this year was eager to announce that it had allowed a few thousand vaccinated health care workers to attend in person. It hasn't come up yet because vaccination rates are still low, but once they are generally available and rates get around 50%, there will be eagerness to push those rates higher, impatience to drop restrictions, and concern about the portion of the population that refuses and the resulting persistence of the virus. Someone will get the bright idea to offer incentives, and someone else will get the bright idea to claim it's the Mark of the Beast; it would be challenging to avoid economic or racial disparities in such a system as well.
Back in March or April, I counted the "it's an annual disease like the flu" as the awful-tier outcome. I stand by that, but it's looking even more likely now than it was then, and I have less hope that I was wrong in my estimates.
New reader here. Excellent article, thanks Scott. Very curious about your predictions on the long term impacts on young children. Like your WW2 analogy, will there will be a whole generation of kids who are fearful of any human contact even after things return to normal(ish)? Or will the extreme social distancing learned behavior fade as we become more integrated into the world again?
This reads like a very Bay Area view of Covid and response - or a very myopic view of a particular part of the culture.
I've flown at least 12 times since the pandemic started. We've hosted many visitors from out of state, we go to restaurants multiple times a week - they are fairly full - and most of our friends do the same. We gather indoors and out for various occasions. Family is the same.
We're fragmenting into two covid subcultures, as others have noted. But the simple fact is that the data does not justify anyone who is under 70 and healthy changing the way they live now vs. how they lived in 2019 and prior.
I would really like to know about the likelihood of long-term effects from mild covid cases in people who have been vaccinated. I am okay with taking the reduced risk of catching mild covid post-vaccination if we're just talking about the ~two weeks of acute symptoms, but if "long covid" is still occurring frequently (I recall seeing estimates of 10-50% in current inoculated cases) then I'm back to dreading the virus, even post-vaccine.
In the SUPER long-term, does COVID seem like something we could completely eradicate, like we did with smallpox? Or is it more similar to the seasonal flu?
And why? Why is that smallpox never mutated beyond control? Why is COVID more or less prone to mutation?
Note that the annual flu vaccine update does not go through the entire FDA approval process that a brand-new vaccine does. I'm not sure what happens, I'd think there is some FDA involvement, but we manage to get out an updated vaccine for new flu strains every year without the FDA getting in the way of that. So I'm relatively confident we can do the same for coronavirus. (Also, my understanding is that coronaviruses don't mutate as fast as influenza.)
A very California-centric post, short on data and attitudes from other states much less other countries. There are already indications that severe lockdowns cause more harm than good. The fact that vaccinations are touted as no guarantee of anything or a return to maskless lifestyle, the 99.5+ survival rate even without the vaccines, the politicalization and incompetence of the vaccine rollout schemes mean even the most brain-dead CNN watcher will stop listening sooner rather than later. It may be worse than the Flu, but that's not saying much.
"Public health experts say there are key lessons for FDA in the global, long-held approach to influenza. Shots made yearly to battle the flu rely on a foundational vaccine that is then altered to fight the particular variants that pop up each season. Drawing from flu vaccine practices would mean that manufacturers could skip monthslong trials that enroll thousands of people and instead prove safety and effectiveness in smaller studies that track a few hundred volunteers for weeks.
"Peter Marks, FDA’s top vaccine official, signaled Friday that the agency is leaning toward this approach. [...]"
I just don't understand how, even after a year-long major life disruption keeping this issue constantly on everyone's mind, the number of people willing to accept a free vaccine against a deadly global plague is STILL only two-thirds.
I like the idea of having the FDA give categorical approvals for modular vaccines. What can we do as citizens to make that happen: write to Joe Biden? call our members of Congress? something else?
I feel like people aren't talking enough about rapid testing. It seems like some California schools have managed to open using testing, which seems promising:
Michael Mina (https://twitter.com/michaelmina_lab) has been flogging rapid tests (https://www.rapidtests.org/) for a year and I am among those baffled by the lack of take-up. I assume it should work against any variant, and by driving the incidence of cases to near zero, vastly reduce the opportunities for variants to emerge in the first place. And then, why not use it for the flu, cold, norovirus (cruise ships ahoy). Use it on pigs, use it on cows, use it on chickens, use it on anything that moves. A virtual population immune system.
This is a really helpful piece, but are we *sure* that this approach to vaccines is a good idea? Simply "changing the password" as often as possible and speeding it past normal processes? We don't know how reinfection affects people yet, and immune systems are complicated. It seems *possible* this could increase vaccine-resistant mutations or just somehow become a problem. We're moving really fast without a lot of information, and there isn't precedent for this. There's also the issue that, if we're going to make it like the flu shot, why don't we just do that now, and let those willing to take the risk interact normally, while giving vaccines to those who want them as quickly as possible, and helping them isolate? As you say, only a minority is likely to keep up with yearly injections. It scares me that our public decision-making basically operates, on the surface, according to a zero-risk logic. This is why I'm paranoid as you describe--if we continue validating this logic, it's not that hard to imagine things not going back to normal for decades, especially given the damage to the economy and social trust. I agree that, if I was thinking of it from a purely rational perspective, this would be like the flu in a few years. That's what I thought we'd decide back in March. But we didn't. And maybe that was reasonable given the unknowns early on, but it doesn't seem reasonable to me now. We're not operating like treating it as we do the flu is a comprehensible option, so my purely rational assessment is obviously false. This could change, but it worries me. From my perspective, the practical limitations on our ability to control this long-term, and the extraordinary accumulating costs that would come with trying, were very obvious from the beginning, and definitely not worth it. (Other than sensible, targeted mitigation measures, of course--I'm not saying do nothing, but anything close to eradication is not going to happen.) In the short-term, they seem worth it, but when you think about what would have to happen to keep it going long-term, the game is just unwinnable, and trying to win it just eats up the resiliency and ability to plan we could have used to cope with it and make necessary transitions. Society is a complicated system to be messing with---we don't know what we're disrupting in the long run. That's when you end up with no concerts forever---people may adjust to the risk and treat it like a flu, but there's no money left to sustain the concert industry, and a lot of people have grown to see it as an unnecessarily risk indulgence. If you don't come to a relatively clear-eyed acceptance of the situation, you can just get stuck drifting along as though it is a permanent emergency, even if most people want to live normally. They won't speak up against the others if we're operating under a logic that ignores the question of what the endgame is, and whether any of this is worth it long-term. I'm very uncomfortable with how this has proceeded and is proceeding. I appreciate that covid poses a real danger to some people, but I can't imagine any situation in which I would advise the response we've chosen. If I was high risk, I would want to shelter and be given the resources to do so, but I'd be terrified to have all of society "hibernate" like this. I would not see that as likely to protect my future, but likely to jeopardize it, by disrupting the available resources, conveniences, and stable society that make sheltering possible.
Re: "yearly COVID shot". I just got my second shot out of the two-component vaccine. That means me potentially being immune to two different vaccine vectors, hopefully for less than how long the COVID immunity lasts, but possibly longer. Is there a technology that would help us avoid running out of vectors for the foreseeable future?
We need to consider just how different humanity is now compared to when Spanish flu arrived. It was allowed to take its natural course because we didn’t know any different. Today if we were told that allowing a virus to run its natural course would likely result in less deaths than fighting it (allowing for more mutations to proliferate), we couldn’t make the call. International travel, coupled with the fact that we are so easily triggered in today’s online world, the ‘panic and fear’ factor could make things worse.
Based on tele-work, tele-medicine and tele-everything else: the current nerdopolis that is the Silicon Valley will start losing its dominance, steadily and surely. Not sure how to quantify it, though.
You've described what will happen in the US which seems plausible. Here's what would happen in developing countries: people collectively decide to ignore COVID as-is, it kills off 0.3-1% of the population (depending on the median age) allowing the rest to reach herd immunity. New mutations kick in but they're not perfect at beating natural immunity and in any case the most vulnerable people have already been killed at that point, so the IFR consistently hovers around 0.1% which is low enough to let people ignore COVID. So no vaccinations, no "adjusting the vaccines each year", no masks, no social distancing, nothing. Society moves on as it did with every pathogen in human history after suffering some casualties. India is already there: their case counts have been falling for 6 months now despite having very few restrictions in place, all thanks to herd immunity.
I am affraid you are far too optimistic about going roughly back to normal. Especially in Europe where lockdowns have been stronger and more general than in the state, with much less political opposition. With the economic crisis looming, governments will not revert back to full democracies naturally, there will need to be popular pressure. At this point, there is a risk to fall to real dictatures, as the power in place feels the risk about having to render counts for last year decision threaten so many political, police and media heads.
The counter power at this time is mostly judiciary, parliament is absent / aligned with executive, and only having the judiciary guard against dictature seems small...
This worry me much more than the covid itself, delation, police violence, scapegoating of tourists / lockdown parties, arbitrary fines linked to ever changing illogical regulations... All of this is up regardless of the epidemy evolution (it keep increasing even when vaccination started and the epidemy stagnate).
Prediction about the future are notoriously hard, but there are more signs about the west transitioning from democracies to fear - powered autocracies than signs that the plagues are back.
Humans don't dominate the planet. Plant life does, and it's not even close in terms of cell count. We should expect more viruses to be selected for infecting the social plants, assuming the virus's "goal" is only to increase its copy count.
Early victims included a couple of Broadway figures in their middle-aged primes, gifted songwriter Adam Schlesinger of Fountains of Wayne and "Crazy Ex-Girlfriend" (who had just signed a big contract to compose a Broadway show with his creative partner Rachel Bloom) and Tony-nominated actor Nick Cordero. So, eleven months ago I figured covid was going to be pretty bad for our human capital.
But, since then, most of the fatalities among semi-famous people have been concentrated in either the elderly or the long-term sickly. I periodically look at the Americans who have died in the Wikipedia list and I don't see much evidence that our culture will deprived of too many great works going forward.
That sounds brusque, but when we are making trillion dollar decisions, we really ought to look at all the evidence, even impolite evidence.
Your impressions may differ, so I encourage you to look at the Wikipedia list or the New York Times' list of people who have died.
Is it safe for people to get additional vaccinations in rapid succession to beat virus variants? I realize that there isn't enough vaccine for that to happen much now, but I wouldn't be surprised if the vaccine process gets sped up.
One more area to hope for progress is for more genome-checking to keep track of mutations.
I am skeptical about the fast track. Any time you change the antigen in the vaccine, it could, by coincidence, resemble some normal body protein and trigger autoimmunity. Or do something else unintended. Not that it is super probable.
I will be glad to be corrected, I am not super sure about this.
As an Israeli who has been vaccinated, I can confirm that they plan to vaccinate every 6 months. When you get vaccinated, they make it clear that you will have to do it again soon.
Another point: Many many health giants around the world have spent billions on new Coronavirus vaccines. They will want to put them to use. I expect that by 2022 there will be an oversupply of such vaccines covering every extant strain.
What about Covid treatments? So far we had a few proposals but the results were a bit underwhelming (chloroquine, remesdivir, favipinavir, etc), but dexomethason seems to work at least for some severe cases. Now we have the monoclonal antibodies and nano antibodies - so far the results are are still a bit disappointing but in theory antibodies should work and at some point we should be able to produce them in mass quantities cheaply.
And by the way - what do you think about amantadine? I have already asked that question here: https://astralcodexten.substack.com/p/hidden-open-thread-1595#comment-1236759 - but with no replies. I think it is an interesting case of auto-immunology disease in science - that is I think the allergic reaction of the scientific establishment to dr Bodnar is overblown.
A minor terminological point on "If vaccines made R0 go to 0.5 or whatever, we would loosen some restrictions until it was back at 1 again." R0 is defined as the reproduction number in a completely susceptible population; the reproduction number when some people are immune (through having had the infection or vaccination) is called R or Rt (see e.g. https://www.nature.com/articles/d41586-020-02009-w).
All I have to say is I have zero regrets for every midnight movie I've ever seen. You don't know how much I miss them, and I treasure every hour I've spent in a theater after midnight. I'm devastated that they may never come back.
YOU'RE TEARING ME APART 'RONA!!!!1111
I also miss my family too. But can quarantine for a few days and see them.
Maybe I really am that paranoid type. I promise I don't have a political axe to grind. But I am just really curious why you would put the chance of rock concerts not happening again or mask wearing staying majority social pressure at "less than 1%" chance? I think it's literally a 50/50 chance.
You even compared it to child kidnapping panics which I agree with. I personally like to compare mask wearing to no smoking in airplanes. No one can smoke in airplanes anymore. It's never coming back. Same thing with going to a rock concert without a face mask. It's never coming back (except in very exceptional one-off cases). I could be wrong obviously, but I would certainly bet money against anyone putting it at "less than 1%" odds.
I have the feeling that a lot of peaople assume that new virus-strains generally should be less lethal than old strains, not due to previous infection but because viruses will tend to mutate towards an equilibrium were it does not kill the host (because there will be a selection-pressure to let the host live and further spread that virus-strain). Is this a valid assumtion? Couldn't it be that "equilibrium-viruses" is selected for over (long) evolutionary time, and that they also depend on the virus in question acting as a selection-pressure on the host? So that, any given mutation is random and might as well be much more (or less, or equally) lethal compared to Corona Classic. Only in the long run will this pan out to be a more flu-like situation. (Confidence in this speculatory line of thought is low.)
I find it strange that people have stopped questioning our response to COVID19 and whether or not it's been a) justified, b) effective and c) cost effective. Now that the vaccines are here, those questions have been sidelined and stigmatized. I think we should continue to interrogate every aspect of our response to the pandemic.
How should we view clearly costly policies that are not supported by the overwhelming body of empirical evidence? I can see a case being made for lockdowns being so intuitively effective and studying them so fraught that we should proceed regardless of the evidence. But I think this case needs to be made explicit so it can be interrogated. Currently that's not what's happening.
Presumably the virus is going to become and remain endemic, and we'll see perpetual annual or semi-annual vaccination for the most likely variants, just like (and along with) flu vaccines.
Question about different vaccines: To what extent do they "stack" and to what extent do they make each other less effective?
E.g., if I get two shots of AZ but J&J turns out to be more effective, is it possible the antibodies in my body will render an additional J&J shot ineffective?
If so, is it clear whether this cannot happen with mRNA vaccines?
I think the first prediction is not well defined. What is a "wave", i.e. what is the y-axis? Case numbers (as everyone knows well) are dependent on testing which changes over time. A more reliable metric (although it changes over time as well) is hospitalisation rates. I would predict that we will not see another wave of hospitalisation in the US similar to the previous 3: the majority of older people will have been vaccinated and the protection against severe illness is very high
For now I'm not sure what to think about all those "mutants". Because it seems researchers "threw" something in petri dishes and made some conclusions with that. But I don't really see any "real world" implications with that... Great-Britain - you know, this country, that "named" one of those highly-infections mutants - got their daily infections down from 60k/day to now around 10/k day in just a month! You could theorize if it was the "old virus" that their lockdown did that even faster. But that is basically impossible to prove in either way! So ... *meh* you mutants, I don't care!
My thinking is more along the lines of this diagram in the German wikipedia about the spanish-flu:
We are currently in this second, really big and deadly "wave". We will have a third one, that is smaller, but I don't think it will be the "same" months as in this diagram. Because at least here in Germany nothing happened last years summer and we basically opened up "everything": bars, restaurants, cinemas, even concerts happened (but with fixed seating). And "our" second wave started in october/november; so it seems very seasonal! And sorry, for now I don't think these mutations change anything about that! Maybe time will prove me wrong.
But important about this diagram: it ends after the third wave! And I think that is something we do too with Covid19!
I was hoping to see a discussion of the so-called "Long Covid", since as far as I can tell from picking through various studies over the past 6 months, still has limited evidence that it actually exists as a unique effect of the coronavirus.
Studies all seem to fall into two camps: either a very small study of like 50 patients who were hospitalized, are elderly & obese, and demonstrate lung damage in scans taken 1 or 2months after being discharged. Or a survey of people from the "Long-Covid Support Group" on Facebook, who self-report suffering varied symptoms several months on, such as Fatigue, Tiredness, and feeling worn out.
Neither of these are remotely what we are interested in though. There have been enough studies, even without large samples or controls, that suggest lung damage in survivors that I expect there is a real effect, but if it mostly appears in hospitalized patients, who have pre-existing conditions, are elderly, etc. then chances are it's very rare overall. And that's assuming it's actually long term and doesn't go away after 6months.
The surveys get towards what we're interested in: what are the chances that an average person who gets infected with coronavirus has to suffer long term issues, even if the disease itself isn't a danger. But self reports of very vague terms like fatigue are just way too unreliable, and again we want to know if these are long term, not just for one or two months. One particular survey did ask for effects at 12 weeks and found sufferers fell away to almost nothing. At that point you have to ask whether this isn't just standard post-viral syndrome?
Ultimately I can't spend much time tracking down medical papers so perhaps there has been more definite evidence emerging recently as to the prevalence of long term sequelae and the possible effects.
A thing I recently wondered about: Why are there no illnesses that do GOOD things to you?
Like a flu that makes you feel awesome and full of energy.
There are literally hundreds of disease strains that kill you, which is, we should not forget, a bad ending for the disease, in the coffin, it will not spread. If a disease made you feel great, you'd go out and meet people, which is great for as disease, and we would not spent billions to develop vaccinations and social distancing and such against it. From an evolutionary perspective, there should be massive advantages for a disease that makes you feel good vs. ones that make you feel bad. Yet I know of not a single one.
Now, there are a few obvious counterpoints.
1) Most of the symptoms we have when we are sick are not the disease, but the body fighting the disease. A body would fight a foreign substance no matter whether we like ot or not. But there are a lot of diseases that the body cannot successfully fight, from HIV to Herpes, that if you have them you are stuck with them. Why only those that will eventually kill you? And not all illnesses are so noticable as the flu, the body fights other sicknesses with us barely noticing. So a sickness where the signs of the body fighting it are low and the effect of the sickness making you feel great is at least thinkable.
2) If it was good, it would be a symbiote, and the body would just incorporate it, like the colon bacteria. True, but only over millions of years. Bacteria and viruses evolve much faster, so one evolving into a feelgood plague could happen spontaneously without cumbersome human evolution having had the time to give it an estabished niche.
Also, circumstances change. An illness that makes you feel awesome would have led our anchestors to be eaten by tigers or starving in winter, because they didn't build up stores in time. In our modern time, the same illness would just improve how you feel living your life.
Personally, I feel the reason for this is to show us that there are no loving gods, that, if something can be created to be good or bad, it most often is created to be as bad as possible (I kid, partially). But I really wonder. There are diseases that can manipulate an ants' brain to make the ant walk up a leaf of grasss and bite down at the end and wait for a cow to eat it. Some are that sophisticated. But none has the effect of releasing whatever hormone or neurotoxin would make you feel better, despite the huge evolutionary advantages that would have for it? It is easier for a disease to jump from bat to human than to be nice for humans?
Teeny tiny nitpick from someone involved in SARS2-sequencing: The British lineage is actually B.1.1.7 -- not B117 -- since there is also a B.1.117 which co-circulates (at least here in Germany).
That already led to some confusion in meetings and emails, and an often repeated clarification that one-one-seven and one hundred seventeen are actually two very different things.
Perhaps I have a bit of a different perspective on the political aspect as I'm from elsewhere (UK), but my biggest concern is the impact on Governance. My concern is that the total suspension of many civil liberties has now been established as acceptable - and what in the future will also be seen as an acceptable reason to implement these.
Also specifically here in the UK, its changed the dynamic of our parliamentary system to something a lot more command and control by the prime minister and his cabinet, with a reduction in public scrutiny by parliament - this concerns me, and I'm one of those actually same side of the political spectrum then them.
And for those who say these are temporary emergency measures, call me tin-foil-hat-wearer but these things have a horrible habit of becoming permanent. I'm still waiting for the 'temporary' income tax to be removed that was put in place in 1799 to cover expenditure during the Napoleonic wars. It would be interesting to predict which government tools become part of daily life, and which will go back on the shelf at least until the next pandemic.
As an expat currently living in Taiwan who was really hoping to return to the US this year: This feels seriously depressing. It sounds like there is a very good chance I won't be able return without getting some variant of the coronavirus at some point. I'm not in any super high risk group, but I'm really not looking for long-term brain or lung damage. And I don't have quite as much faith in American institutions as does Scott.
Still, it's good to get a sobering dose of reality. Now I can consider what I would need to do in order to stay here longer.
Does the FDA approval process matter all that much (or is its slowness typical for other vaccine-producing countries)? The rest of the world also creates vaccines. If the FDA is so slow that the rest of the world has vaccinated everyone already whilst they are still playing golf, then traveling US citizens would simply get vaccinated on arrival or be sent back at the border. Assuming they wouldn't have to be quarantined and tested for new strains first, if the US remains a permanent refuge for ever-changing mutations.
So new reader here. Excellent post with which I mainly agree although not sure if we get new vaccines down to 2 months - if we bet early enough based on Southern Hemisphere spread may be okay if it’s a bit longer. But a flu vaccine like model could work. Three other points 1. Therapeutics- if we get some highly effective measures against capacities that are required for the virus that changes the equation. Some interesting studies including one reverse engineering existing drugs at UCSF but not proven yet. Second excellent review recently in Science (sorry don’t know how to do the link thing yet) on how coronaviruses become endemic - they continue to hit young children who have mild disease and are not immune while the rest of the population is as you describe. There were some reports of a tilt to this pattern in the UK and right now in Boston persons between 0-19 are the highest group with almost no hospitalizations thankfully. Is this the beginning of wave 4 or is the UK variant already more widespread than we know ?
Right now Moderna and A-Z have x capacity to make the mRNA vaccines. Thanks to the crazed DIY vaccine article from last week, it appears that making and getting the peptides for the "classic" version of the spike is not hard.
So how do we create 5x capacity? Is it the lipid sheath that holds us back? Lack of trained personnel to run the factories?
We know this works. How do we scale it?
We also need a way to push this out to the 3d world or they will always be a disease reservoir.
I like your approach of making testable predictions. Much better than just stating opinions. I am still proud that I landed a good prediction of the number of deaths at the end of March in Germany. I am a bit less proud that in summer, I considered the whole second wave of Covid infections "rather unlikely".
In the interest of keeping myself honest, here are three things that I've gotten definitely wrong in, like, April, about covid:
1) I expected that countries which did well in a first wave will also be relatively fine in second and subsequent waves. Ups. Some of them were, but many weren't.
2) I expected substantially more deaths per capita in Europe than in the US. There are huge reporting issues in parts of postcommunist Europe but nevertheless this was also clearly wrong.
3) I definitely didn't expect that countries in historically Chinese cultural orbit (to which I include also Japan and Singapore) would do as a bloc so much better than the rest of the world, despite big differences between their political systems.
I haven't heard about convalescent plasma therapy in a while. Shouldn't this be easy to scale up now, considering we have so many recovered?
Also can you transfer vaccine protection via blood plasma? Would be lovely, if I could simply let my grandma donate me some plasma, instead of waiting till September when it's my turn.
I think you are interpreting the "neverending lockdown" concerns too narrowly. Mask mandates on subways are one of the least bad scenarios that could happen. "Long Lockdown" in a wider sense can include things like:
- Independent and diverse restaurants going bankrupt and ceding space to chains. This won't be easily reverted, as most customers will be eager to eat out *somewhere* and thus end up supporting the chains.
- Public schools never going back to their pre-pandemic functionality, as parents who care about face-to-face instruction have left, and those who remain are either those without the means to find an alternative or those less concerned with the quality of instruction. Evaporative cooling, to use the language of this blog.
- Women leaving the workforce for good, partly because of the previous point. It remains to be seen whether this will be permanent or will last until the next economic upswing. I don't think men's wages will rise anywhere near enough to make up for this; I expect to see a lot more broken families and delayed pregnancies than successful returns to trad.
- The panda in the room: A new-found trigger-happiness of Western governments to pre-emptively suspend basic civil liberties (freedom of movement, assembly and commerce) based on opinion from medical experts or -- even worse -- epidemiologists. After all, we can't wait for the virus to come to us, can we? This sort of thinking has become popular after COVID-19, to the point that experts are now berating Western politics for not taking previous epidemics (SARS-1, swine flu, avian flu, ebola) seriously enough. But what would it mean to take them seriously? Border closures during travel season? Mandatory quarantines? Countries like Italy and France have taken the axe to civil liberties in ways that no one had seriously entertained since the 1960s; inter-European borders were closed for months. Demonstrations were outlawed in several countries, and at least one (Poland) has used this occasion to ram an unpopular law through. While the most stringent measures that have been actually taken in the US were restrictions on commerce in blue states, the European model was being seriously discussed in media and often positively contraposed against the American laissez-faire one. Unsurprisingly, a general feeling of "it could happen here" has emerged. Meanwhile, no one has bothered to delineate the level of danger that warrants extreme measures like stay-at-home orders and border closures; by the standards that are emerging, we could have a pandemic scare every 3 years, not to mention that a less scrupolous government could easily pull one out of its hat whenever convenient.
All of these concerns would have looked paranoid some 10 years ago, where the typical Westerner believed liberal democracy was self-sustaining at least in the ever-growing Dar al-Democracy, and whatever mistakes politicians would make the market would fix with its gentle invisible hand. Suffice it to say, one thinks of solid ground differently after an earthquake.
I wonder what you predict/think about Covid "long-haulers" and whether that will distinguish Covid precaution from flu precaution. The anecdotal evidence about the persistence of symptoms and the unknown long term damage to lungs, heart, etc. significantly increases my concern about getting it. If vaccination prevented long-haul Covid, that would be great news.
1) Genomic surveillance is so spotty in the US. What are the chances that other variants of concern with similar properties to B.1.1.7 or B.1.531 are currently circulating in the US? I read a paper (can't find it now) suggesting that B.1.1.7 wasn't increasing in relative proportion at quite the same rate in California as it was in the UK. One suggestion I think was that there are other variants that are similarly more infectious out in California. (On the whole, if this is more likely then that should decrease our expectation of a large fourth wave?)
2) To what extent is it possible to form an expectation of a fourth wave without mathematical modeling? I haven't seen a tone of modeling that attempts to take vaccination, seasonality and emerging variants into consideration. I think Youyang Gu does this, and his fourth wave is very moderate: https://covid19-projections.com/path-to-herd-immunity/
So, question: are there other attempts to model this mathematically rather than heuristically?
3) In the endgame where SARS-CoV2 becomes endemic, does it have persistently low IFR because of immunity from prior immunity? And more broadly is this what happened with other currently circulating CoVs? Suppose that you took one of the other "common cold" CoVs and removed it from human circulation for 150 years, and then reintroduced it. Would that cause a pandemic, because of weak prior immunity? And would it be similarly lethal?
It turns out antidepressants greatly mitigate Covid symptoms. Nature is an impeccable source. The magnitude of the effect reported is great. I don’t know why this is it getting more publicity
Those on reddit.com/r/skincareaddiction say you shouldn't buy skincare products from them because their problem with fake products. People have said that they've bought from trusted sellers, received brand name packages, and only discovered the product was a knock off because the consistency/color was different from their last bottle (or something worse happened like they got a chemical burn).
My prediction: 100% chance that Joe Biden is going to OWN a coronavirus vaccine failure if his administration does not get its act together and create a national corona vaccine appointment scheduling system. The current every-grocery-store-for-itself mess of vaccine appointment methods is far far worse than the Obamacare rollout site ever was, and is far more infuriating than not having the vaccine at all (ala #trumpolini times)
Bioprocess engineer here, I wanted to comment on your proposal here:
"They allow anybody to manufacture vaccines and charge market price for them, subject only to usual safety restrictions. If Amazon wants to get into the vaccine distribution business, for God's sake, let them."
This seems like a really unlikely outcome to me, for a few reasons. For a bit of context on drug manufacturing, a drug approval also requires approval of the manufacturing process and the analytical methods testing the quality of the drug. If a drug manufacturer wants to make major changes to their manufacturing process, then they may need to perform new clinical trials to justify the new manufacturing process. This is based on the concept of "quality by design," usually abbreviated as QbD, which is often compared to the older, crappier paradigm of "quality by testing." Quality by testing just means that you ensure product quality by running a bunch of analytical tests on important properties. For the mRNA vaccines, that's probably going to be stuffy like number of copies per dose of the mRNA transcript by qPCR, nanoparticle size by DLS or some other sizing technique, etc. The problem with this approach is that there is always the potential that there is some super important quality attribute that you aren't aware of and don't test for. If you have some critical quality attribute you aren't testing for, then any changes in the manufacturing process that also impact our hidden quality attribute could negatively impact drug safety or drug quality. Therefore, the QbD paradigm was invented, which involves control of drug quality through control of the manufacturing process, so that even if you aren't testing every critical quality attribute, you know you are making a consistent product because your manufacturing process is consistent. All modern drug processes rely to an extent on both testing and manufacturing design for quality control.
Because of the expectations around QbD, it's going to be really hard to open up vaccine manufacturing to all potential manufacturers. You will either need to 1) force the original vaccine maker to transfer all information on their manufacturing process and analytical methods to all prospective manufacturers, 2) make new manufacturers do clinical trials on their drugs or 3) abandon the QbD paradigm and allow manufacturers to develop their own manufacturing processes to make existing vaccines, without a requirement for new clinical trials.
#1 is never going to happen, because the innovators are going to scream bloody murder about having their patents violated. Even potential manufacturers would likely avoid this, because they don't want to set a precedent that the FDA can just seize your manufacturing IP.
#2 is essentially the existing biosimilar pathway, just starting before the existing vaccine patents expire. This is probably feasible in theory if the FDA and Congress nullified all patents on the vaccine drug substance, or they bullied the manufacturers into allowing competition. Even then I can say from experience that making a biosimilar still takes a ton of time and resources. The requirement for new clinical trials will delay the release of the biosimilar vaccines significantly, probably to the point that the biosimilar vaccine is no longer effective against newly circulating strains. I don't really see anybody taking this route.
#3 is basically a supercharged biosimilar pathway where you abandon QbD principles to avoid doing new clinical trials. Without QbD control, you would almost certainly need extensive testing of the biosimilars using the analytical methods of the innovators to ensure comparability. However, you would still need to force vaccine innovators to share their analytical methods, which seems unlikely. And you are still voiding all of the initial vaccine patents to allow early biosimilar competition. And the FDA is super risk averse and is not likely to ever approve of an approach like this.
The more likely scenario for expanded vaccine manufacturing is for the government to subsidize voluntary contract manufacturing relationships. This is already quite common; Lonza for instance is acting as a contract manufacturer for Moderna, for instance. In these relationships, the innovator maintains full control over their IP, and so they can share their full manufacturing process without worrying that the contract manufacturer is going to steal their IP. It's important though that this be voluntary and the innovators get to choose who to work with. While you can sign agreements saying "IP is protected," a contract manufacturer still learns a ton by making a drug. For instance, let's pretend Merck has a fledgling mRNA development program. Moderna is most definitely not going to want to use Merck as a CMO, since that would be giving Merck a huge leg up in terms of manufacturing experience for their potential new mRNA vaccines.
At the moment, with our government continuing to impose lockdown while promising that it'll all be relaxed down the line (there will be no St Patrick's Day parades this year, which doesn't have me weeping bitter tears to be honest), I have no idea. Every time something new comes along, e.g. they relaxed restrictions for Christmas, people travelled home from the UK, and we got the new mutant variant and up went the infection rates once again and we're back at Level 5 restrictions https://www.gov.ie/en/publication/2dc71-level-5/
So my natural tendency is to go "things will ALWAYS get worse" but will they get better? No idea.
I find it very hard to care whether covid is eradicated or not, compared to how much I care that the lockdowns end. I feel like people forgot that these are extreme measures that we shouldn't use unless absolutely necessary.
I remember in January when I heard about the lockdowns in China and thought "that's absolutely barbaric, I'm so glad that can't happen in America." Then in March there was talk about lockdowns, and everyone was duly skeptical, but after looking at graphs and statistics and checking the numbers and careful argument we were convinced that it was necessary. Now when I point out that the specific conditions which made the lockdowns rational (despite the incredibly high cost) might stop being true well before the virus is eradicated, I feel like I'm taking crazy-pills. "You can't stop sheltering in place, you might get covid." I'd much rather catch covid and take my chances than spend another year inside, it's only the structural issues around R0 and hospital capacity that I'm worried about.
The only question I want anyone to ask is "when will the risk of overwhelming hospitals go away, when will the death rate be less than 100k/year, even without restrictions?" If you're wondering how long lockdowns would be needed to eradicate the virus, I assume you've lost track of your priorities.
Nicely put, Scott, as usual. What's missing is the public health impact, not of the Rona, but of the response. The illness and death caused by closures and lockdowns. First order effects like suicide, alcoholism, drug abuse. Second order effects like domestic violence, child abuse (includes school closures), depression. Third order effects like unemployment, poverty, and stress. I posit that the public health impact of our response far outweighs that of the virus itself. On the whole, Rona is an opportunistic disease that attacks mostly those about to die soon of something else. All the rest of what I mention impacts healthy, middle age people and, especially, children.
Well done Scott - I like the way you share your thoughts. I would love to hear your thoughts on "What is the Goal?" or "What exactly is the problem we are trying to solve?" I think all would agree COVID has created many problems, so all the more important that we define the biggest problem clearly. I head a wise saying once that "a problem well defined is more than half solved."
I agree with Scott's sense that COVID will always be with us -- a sort of lingering flu like illness that burns hot in certain regions of the country during coronavirus season in Nov/Dec/Jan -- but I think his introductory predictions that we will have further "waves" in March and next Fall extremely pessimistic and unlikely.
First, what do we mean by wave? At this point, everyone knows that case numbers do not quite equal infections and that the scarcity of testing early on in the pandemic makes our first wave look artificially small. The best estimate for true infections that have occurred in the US come from Youyang Gu, who has run the remarkably excellent website COVID-19-Projections and essentially outperformed anyone else in the COVID modeling game. (Link: https://covid19-projections.com/).
Gu's Super Fancy Model estimates that our peak number of daily infections during wave 1 was ~300k; during wave 2 we hit ~250k; during wave 3 we hit ~660k. I will somewhat arbitrarily define a "wave" as a period in which the US exceeds 200k true infections per day. My prediction (70% confidence) is that this will never happen again.
The news cycle is currently saying California = Heavy Lock Down but Florida = No Lockdown and Both Have Same Outcomes. What are the best arguments for and against this kind of thinking? Preferably not mechanistic argument (masks and lockdowns HAVE TO do something so this CAN'T BE RIGHT).
What I do not get in this whole Covid mess is why no country so far pushes well engineered comfortable masks with a good seal and long lasting efficient filtes with low breathing resistance in a much more aggressive way, espcially since the general effectiveness of masks is no longer really in doubt. Instead we are stuck with cheap one way products that most people either wear wrong or not at all and that get quite uncomfortable after an hour or so and essentially cannot be worn during any kind of serious physical exertion.
It becomes increasingly clear that between mutations that evade the antibody response and the fact that only 70% of people in most countries are willing to be vaccinated vaccination will not be the silver bullet we hoped for. Thus I think implementing such a strategy, at least as backup for the next winter is a very sensible and comparatively cheap strategy that would also protect us from potential future pandemics after Covid.
Does anyoe have a good idea how to make this idea more public?
"Instead we have a patchwork of approvals and I have 70m doses that I can’t ship because they have been purchased but not approved. They have a shelf life of six months; these expire in April."
1. Daily deaths is a far better visualization of the waves in the US. Confirmed cases make the first wave appear small, but when you look at deaths you can see the first wave was nearly double the size of the second wave. When you do check your first prediction, accuracy should be determined by checking daily deaths.
2. I highly doubt that once we reach some sort of equilibrium with the coronavirus that we'll need yearly updates to the vaccine. Right now it is "mutating quickly" because there is a huge number of replication events combined with the fact that the virus is still getting to know us. Once infections are down and the virus has discovered most of the "low-hanging fruit" mutations that provide large jumps in fitness, the rate of mutations will drop and the availability of mutations that provide large jumps in fitness will also drop. In fact, my prediction is that the only thing needed to drive the risks of covid down to the levels of the flu is a single exposure to the original strain of the virus, whether that exposure comes from the virus itself or the vaccine. After that, exposure to the newest variants might cause symptomatic infection, but the "old password" will still give previously exposed immune systems a major head start.
2b. My point above isn't a counterargument to your prediction that public health officials will recommend a yearly vaccine for the newest covid strain. Just saying that I think a yearly update will be major overkill after ~2022.
3. Many of your predictions have some vagueness that will make them difficult to check. For example, the prediction about mask usage on an average street in SF Bay Area seems particularly difficult/expensive to check. Maybe change it to focus on something you can easily check, like the proportion of businesses that still require masks (easily checked by viewing signs on doors). Also, this prediction seems to be a prediction about wearing masks OUTSIDE, which could change if people finally get the message that covid doesn't spread well outside. So, its entirely possible that your prediction is "correct" even though we're still living with ridiculous mask requirements.
Nice info as usual! Does anyone have a prediction about when schools get back to normal or what they'll look like in the fall? My 7yo is going to in-person school with restrictions and here's what those restrictions look like - Everyone is masked, which still makes me so sad to see. He gets 15 minutes to eat lunch because that's how long they're allowed to remove the mask at all. Many of the afterschool clubs and extracurriculars have been cut. A lot of typical activities are proscribed at recess and there aren't so many fun or "special" activities to spice things up. Obviously there are no field trips They're still doing a couple of special in-class events but parents are not welcome. Logistical arrangements were required to e.g. hold more class time outdoors, which obviously isn't feasible in lousy weather. Otherwise they stay in the same classroom all day rather than traveling to different rooms for specials like music or art. It sounds like such a grind to me, for both the students and teachers. It all can't possibly be sustainable long-term, and to my perception there's a lot less joy and anticipation at present.
Frankly, a lot is obscured in looking at “the peaks” across the whole us. At geographically finer resolution, nowhere that had a bad first peak (use deaths not cases) had a bad second peak, let alone 3 peaks. What looks like new peaks is simply the virus reaching new populations. the speculation about strains causing big new peaks are a bit dumb, seeing as the UK and SA covid cases have both fallen 75% over the past month.
One plausible trajectory is that of the Russian flu of 1889-90. Not only is it now suspected to have been a coronavirus, but the patterns of infection and disease were very similar - severe/deadly in older people, almost no effect on children. Speculation is that children built up immunity early on, so as they grew older they had only mild symptoms; these symptoms now present as a 'common cold'. Given that covid-19 also has very few symptoms in children, it is entirely plausible that they will develop a partial immunity that protects them from all but mild symptoms as they grow older. No vaccine necessary.
I noticed this curious pattern emerging from last week's NERVTAG updates.
1. Using the SGTF bio-marker as an identifier for the B.1.1.7 variant, the B.1.1.7 variant is now at 90% of the strains detected across the UK. Yet UK case rate continues to fall, and the hospitalizations rate continues to fall.
2. By case *percentages* B.1.1.7 seems to have a significantly higher R0 value than the "classic" D614G strain. However, some epidemiologists predicted that there would be a new case surge when B.1.1.7 reached 70 percent penetration in UK's virus pool. B.1.1.7 reached ~70 percent penetration the final week of December in most areas of the UK. By those predictions, the UK is past due for another surge.
I've been discussing this with an immunologist friend, and his only explanation is that people are masking and social distancing better now in the UK. I don't know of any way to accurately measure this variable. And I'm far from convinced.
Has there been any writing/speculation that the mutations we're seeing now represent "low hanging fruit" in the evolutionary space of the COVID spike protein? A few of the point mutations in the different new sub-strains are the same, which may lend credence to this idea but I'm not sure.
I read someone say that humans triumphed over the animal kingdom because while they adapt over evolutionary time, human ingenuity adapted at an exponentially growing rate. But it may be that the tables have turned briefly - the virus evolving new strains faster than we can develop, approve and manufacture vaccines for it.
I've been saying all along that COVID will be with us essentially forever (for the next century at least), so I feel a bit vindicated now.
That said, COVID is more deadly than the flu, and it disproportionately affects old overweight diabetic cardiovascular-compromised people -- which is a sizeable chunk of the American population, which is getting larger every day. For this reason, I still expect mask requirements to become the norm in most public indoor spaces (e.g. grocery stores). I also expect public gatherings of any kind to become a lot less frequent, and of course the same goes for nationwide/international travel.
Alternatively (or perhaps in combination), society might slowly change, so that COVID-vulnerable people are expected to stay indoors all the time. If a person dies from COVID while being old/fat/etc., the public sentiment would be, "well, he knew the risks, so it's sad but it's kind of his own fault".
When I switch back to an ACX tab, my computer is busy for ~10 seconds re-rendering the page. I have a super-nice high-end computer so this must be horrible for people with normal or below-average hardware.
It seems the same whether or not I am using the "ACX Tweaks" extension.
I'm no expert at JavaScript debugging, not at all, but it looks like the "visibilityChange" event is triggering some extensive UI refresh. Can we ("we" meaning the community including me; SubStack Inc could do it too but they are busy and have a long QA process) write an extension that disables that functionality, and perhaps makes refreshing the page a manual event done by a button-press?
> "When the fifth wave strikes in late spring/early summer [...] another part of the population (~25%?) will have had the disease already"
Note that more than 25% of the US population has probably been infected already. My best point estimate through 2020 was 29%. See e.g. the CDC's estimate, COVID-19 Projection's estimate, etc.
(I only skimmed the post quickly, but this jumped out at me.)
This isn't working as an open thread. Everything or just about everything is about COVID. I suggest having open threads that don't lead off with highly engaging topics.
>The second wave was in July, when we got sloppy and lifted lockdowns too soon.
When I looked into this "second wave", back in the summer, I concluded that there was no second wave, it was the first wave, just in different places. New York and Massachusetts got hit hard early on, but then it calmed down there and, after a bit, picked up in other parts of the country. If you just looked at the stats for the US (which is a BIG place) as a whole, it looks like a second wave, but that's just an artifact of lumping the whole country together. If there was no second wave, that argues against the notion that "we got sloppy and lifted lockdowns too soon".
Prediction: lockdowns will continue to have no effect on the virus (compare Florida to California), nor will mask wearing nor social distancing. Also predicted: lockdowns will continue to harm the least-prepared, least-capable, and least-off people in our society. You know ... the people we don't listen to and don't care about.
Sorry if I'm cynical, but this ENTIRE last year has been filled with nonsense.
If you study the Covid Tracking Project's regional & state charts you will see that nearly everywhere in the country there have been two waves, not three. Now that the waves are in sync, it is unlikely there will be another in the next few months. Probably not another wave until next November.
Are we doing the world a disservice by labeling the variants by country of first expression? Is this just a virus with a fast rate of evolution and multiple possible vectors to do so? Would it be better to show how it may outpace a slow vaccination campaign?
> But even if they have it tomorrow, that's...what? Another four months for studies, one month before the FDA is able to meet to discuss an approval (you can't rush meetings!), two months to ramp up production
Shouldn't the approval process and ramping up production run in parallel? If a company is fairly confident that it's vaccine is likely to be approved, it should make sense to start manufacturing it before it's approved.
Indeed, why didn't this happen with the current vaccines against the original strain? At least I assume it didn't, or the initial rollout should have been much faster.
"I bet we'll pick up the Asian custom of wearing masks in crowded areas if we feel sick." Depends on who "we" are. Some people already skew masks. Also, couldn't wearing a mask after the pandemics ends make some people nervous, making them think the mask-wearer is very disease (maybe carrying a new COVID strain or something like that)? To be honest, I would like to see mask-wearing when one feels sick being mainstreaned, but I do not think it will happen.
I have seen polls that suggest nearly half the U.S. population may not opt to be vaccinated. Here Germany the government has tied re-opening to an r factor of 35 or less. The current German lock-down seems to be pushing the number in that direction but new variants have the government prepping the population for continued lockdown. I wonder how the vaccinated population percentage will affect government policy?
Kind of strange reading an article like this, praising lockdowns, not mentioning the data we have on their efficacy (very nearly none as far as the virus is concerned, but with staggering negative side effects). Why is that? Are you, Scott, ignoring the data?
Like most recent waves, this one began in an animal lab or a meat market, i.e. a site where we crowd together and brutally manipulate non-humans. From that filth & degradation, "zoonotic" viruses will continue to arise; and the new global system of control that takes advantage of that spread, like any good Disaster Capitalist, will continue to amplify fear (i e. the meta-virus) to maximize its own hold over us. Prediction: humans will continue to herd and slaughter, and when disease wafts off the kill pile, we'll cage and (medically) torture more animals to "solve" it.
"Over 900 UK health and care workers have died of COVID-19, most due to occupational exposure. They continue to die because infection control in health and care settings is built around an outdated 'droplet' model which ignores that #CovidIsAirborne. 1/"
I apologize if this is a dumb question. I am not a scientist but my sister is, which makes me think I have scientific genes ...
Is it possible that some significant portion of the population is simply not susceptible to COVID? And maybe the recent decrease reflects approaching herd immunity in the susceptible population?
How, at the human mechanical level, is it "more contagious"? What makes it spread more? Does it travel further in the air when sneezed? Does it make people more likely to cough? What's the mechanism?
Most promising predictions. But it depends, what happens in the future, of course. For example, there is no such certainty with vaccine and vaccination in general in Nordic after the situation which happened with 80 old people who vaccinated in Norway. Because of this, Sweden still disputing regarding vaccines. By the way, a strange trend happening because of this uncertainty. Swedish companies started renting additional warehouse space, such as https://www.matchoffice.se/hyra/industri-lager/malmo, in order to fill it up with products. Waiting for a more strict lockdown, I guess.
If it's a seasonal coronavirus like the flu or the common cold then you'll see another surge in late March / April because that's how seasonal coronaviruses behave. The waves we've seen so far also mimic the behavior of seasonal coronaviruses.
I thought y'all might be amused to know that Byrne Hobart (whose Substack is "The Diff") has written at least a short version of the article that Cade Metz originally claimed to be writing for the NYT. "Why Did One Internet Subculture Spot Covid-19 So Early?": https://diff.substack.com/p/why-did-one-internet-subculture-spot.
My favorite snippet: "[Rationalists are] a sort of distributed, mostly open-source monastic order, spending a lot of time contemplating the world and passing down important observations, but less time directly interacting with it."
My prediction: 80% chance that on an average day in mid-2022, inside of an average business (grocery store, hair salon, bookstore, etc.) in the SF Bay Area, more than 20% of people will be wearing face masks. (This of course assumes that the relevant business is open.)
This might or might not conflict with Scott's second-to-last prediction. I would be interested to hear what probability he would give to my statement.
The issue of people believing conspiracy theories is one thing, but the NYT's solution is 'don't try to think for yourself! just trust us! And Wikipedia!'
People forget that the “waves” get bigger because more TESTING became available. Not because there were actually more cases, and not all the tests gave accurate results. This is blown way too far out of proportion.
It's a little far out to be speculating about this but I wonder what the unintended consequences of fast-tracking regulatory approval for more modular vaccine platforms will be. It's something we obviously ought to be doing, but I wonder what the world of vaccine, or even drug development, looks like when modular platforms are orders of magnitude faster to get to market.
Great article! Regarding COVID becoming a new flu-like situation, I read this insightful take on it that bets that it won’t: https://twitter.com/tomaspueyo/status/1359170996008325123
I've heard a lot about Vitamin D's efficacy (or lack thereof) in treating COVID. Right now, my overall takeaway is something like "Being deficient has very bad correlations with outcomes, mediocre studies tentatively suggest that supplementing vitamin D could have very good causal effects on outcomes, for some reason we have yet to do non-mediocre studies on this yet." I've been taking 6000 IU per day, since I'm not exactly basking in sunlight these days so it'll improve my bone health if nothing else and even a 10% chance of it being protective against COVID has massive ROI. What evidence have other folks seen? Opinions by knowledgeable medical professionals / replication-crisis-y folks especially welcome.
> Prediction: 50% chance that sometime in 2021, the FDA grants a pharmaceutical company general approval for coronavirus vaccines which can adapt to changing virus strains without going through the entire FDA approval process again, and that whatever fast-track lane they get takes less than three months between creating the vaccine and it being approved for general use.
I think this prediction is far too low: the initial emergency use authorization for the Moderna vaccine took ~5 months from completion of the first trials, and that was for a new technique. Now that it's a proven technique, I'd expect that it'd take far less time the second time around.
Solid set of predictions and reasoning.
I want to give feedback that the big grey boxes with blocky text look fairly ugly to me. At first I thought they were straight-up bugs, or inserted screenshots from elsewhere. I think I’d strongly prefer the reading experience of blocks-of-italics. (But not sure!)
I've seen a lot of reports of brutal side effects, particularly from shot #2, of the mrna vaccines. It's possible, I suppose, that this would go down with more careful dosing...but as is, I think this alone nukes any possibility that everyone takes a Moderna/Pzifer shot yearly. (And once everyone isn't going to be doing that, game theory says you shouldn't.) Not least because the accumulated total risk to serious reactions goes up, as far as I can tell, and both autoimmune/allergic things seem (to my non-expert view) to only get worse for most people. I think it's highly unlikely we all get our yearly covid shots.
That is to say: either we crush it right now with the first round of vaccines or *the game is over* and we should give up. And we've missed the chance.
It seems to me there are two overwhelmingly important variables for determining how "normal" things get in 2021. The first, which you mention but I think still underplay, is the extent to which vaccinated people are protected against severe disease and death even if/when the virus mutates to possibly be more (re)infectious.
The second, which it doesn't look like you mention, is the extent to which treatment protocols continue to improve, particularly treatments which can keep people out of hospitals rather than just saving their lives once they get sick enough to be hospitalized. The Treat Early folks are quite bullish on fluvoxamine and ivermectin, for example, and have a longer (and regularly updated) discussion here:
https://www.quora.com/What-is-the-current-treatment-for-Covid-19/answer/Steve-Kirsch
I think you'd perform a public service by taking a look at their evidence and saying whether/how much you think that bullishness is justified.
"There are whole countries whose cuisines are still built around weird decisions they made as part of World War II rationing."
You can't just throw this out there without a link! Total nerdbait. Simple searches just tell me how tough it was in a 1930s car with less gas available - I have to know!
my main question is how long until I'm allowed to make new friends again? right now some people are doing online dating, but seemingly nobody (in Blue Tribe circles anyway) is making new non-romantic friends. And due to moving right before COVID started, I'm no longer geographically near all my old groups of friends.
It's painful to imagine that even if restrictions on public gatherings eventually ease, I might still have to basically live a totally solitary life for many more years, as nobody who would be in my social circles will be willing to interact with a new person. I really hope you are overestimating the effect the new strains of the virus are going to have.
If this is true I might have to consider either trying to move to a country like Taiwan or New Zealand where they can manage it more successfully, or else moving to a red state where people just don't care.
How much were the three waves in the US over the last 11 months separate waves hitting the same people three times versus how much were they waves hitting separate people once? I really don't know the answer to that question.
The first wave hit some big cities, especially New York, and was very bad for blacks and Hispanics (especially, I'm guessing, Caribbean Latinos).
The second wave was worst in the Sunbelt air-conditioning belt of Arizona, Texas, and Florida, and hit Hispanics (especially Mexican Latinos) hardest.
The big third wave started in the Upper Great Plains and moved south with the cold weather. It hit non-urban whites hard for the first time, along with Hispanics for the third time. Blacks were not hit as hard in the third wave. For awhile, whites had a higher rate of excess deaths during the third wave than blacks did, although last I looked blacks had pulled even.
My guess is that a combination of being closer to herd immunity and "once-bitten, twice-shy" learning from experience makes places and groups that were previously highly vulnerable more hardened targets for the next wave.
By now, covid has has hit almost all of the country hard except, perhaps the northwest coast from the Bay Area to Seattle and upper New England.
Hey Scott, Sorry if someone has already suggested it, but have you considered putting (/ having an intern put) the old layout back on the blog archive? It's a bit easier to read, and also has some nostalgia value.
Hi - new member here, still learning the ropes. 1) the efficacy rate for flu vaccine is between 30% and 50%, which doesn't matter a great deal as the disease for many sufferers is not severe. It would suck if the rate for Covid vaccine stabilises around this number. 2) There was something last week (sorry, I wasn't tracking sources last week) about the vector used to propagate the Astra Zeneca vaccine. It is a chimpanzee vector. The story was, the human body when vaccinated learns how to prompt Covid19 anti-bodies the "chimp way" and that learning sticks - and when a new human-derived mutation arrives, the human body doesn't respond effectively.
Perhaps a prepper-like splinter culture of "CoViD recluses" will develop: people who either move to remote rural locations, or to "filtered communities" with elaborate protection including domes, UV lamps, air filtration, and so on.
Ethnic minorities will be stereotyped as infectious ... wait, that happened before.
It looks like we will have between 125 million and 150 million vaccinated after Biden's first 100 days. Beyond that somewhere between 60 and 100 million already have the antibodies. At the low end 185 million are resistant. At the high end, 250 million. So, there are fewer and fewer people who can spread a March bump.
Does anyone around here (or maybe on Metaculus) have predictions for whether or not there will ever be challenge trials (in the US or elsewhere)? Besides the obvious effect challenge trials should have on all other related predictions, this also potentially impacts my personal decision of whether or not to get a vaccine when it becomes available to me (as I would like to volunteer for a challenge trial).
i'm very curious what people here think about repurposing of already-approved pharmaceuticals for early treatment and pre-/post-exposure prophylaxis. To start the discussion off, i know there's some data for TMPRSS inhibitors: ambroxol/bromhexine and also pretty much like, every antiandrogen since AR affects TMPRSS2 expression. TMPRSS inhibition has direct antiviral activity. Also people have been looking at drugs with immunomodulatory effects such as fluvoxamine and ivermectin. The immunomodulatory part is important because AFAICT the actual damage caused by the disease seems mostly to come from severe immune system dysregulation and its consequences (immunothrombosis, hyperinflammation) so tweaking the immune system's operation so it doesn't fall down that path is very much desired. Doubly so if the immunomodulatory drug in question doesn't affect viral clearance. There's some trials for ivermectin showing faster viral clearance and quicker cessation of anosmia and also an RCT for fluvoxamine with really quite good results. There's also a certain endosomal entry inhibitor that became a culture-war subject but honestly my understanding is that without blocking the TMPRSSes or at least TMPRSS2, blocking endosomal entry is pretty much worthless.
It seems to me that fear is causing us to massively over correct to try to prevent unlikely and unrealistic failure cases. For example:
People are afraid that unsafe vaccines will get approved because the procedure is getting rushed. Instead of trying to maintain the most critical parts of the testing process, and skip the rest, we are scheduling one month of thumb twiddling between when we get the data and when we meet to discuss it.
People are afraid of line cutting so we are prosecuting doctors who distribute vaccine doses out of order instead of letting them get thrown out.
People are afraid of price gouging, so we all decided not to do it, which removed the incentives to scale out production as quickly as possible.
Pretty surprised at the confidence of B117 resulting in a 4th wave so soon. That seems to under emphasize seasonality/immunity and over emphasize behavior to my mind. >10% of the us is currently had at least one shot and a partially overlapping ~10-20% (depending on how much under testing you think there has been) have been infected and gained some level of immunity. If you ballpark guess that in march something like 20% have some degree of protection does that not compensate for the variant being more infectious, or are my estimates way out of line?
The second-to-last prediction would be more interesting to me if it gave a probability for people wearing masks indoors (say, a grocery store). Outdoors, people are much more cavalier about masks already, and are already drifting down towards mostly-not-wearing (plus generous distancing) in my Silicon Valley neighborhood. Indoors, they're still at nearly 100%. And this fits with what I've heard about transmission-- that indoors is by far the greater risk.
90% confidence in "not much outdoor masking" seems to imply a 10% probability that our mid-2022 masking norms aren't just paranoid and security-theater-y, but actively irrational. One would hope there's a somewhat higher risk that we'd still be wearing masks in grocery stores.
I think this is too pessimistic. The math on vaccination and herd immunity suggests that once you get to over 50% immune... The r value drops and keeps dropping in an additive manner.
I think this is fairly plausible, and here’s a question not discussed there: Where in this multi-year to permanent cycle do schools again educate all of the kids they were, in the buildings they were, where having 25-35 in a classroom, very often closer than 6 feet apart, was the norm?
One big question is when high risk activities, such as major concerts, sporting events, and conventions happen at 100% capacity. Whole industries, businesses, and urban developments are built around these kind of events happening on a regular basis, and they can't be put on pause forever. At some point things need to resume or people need to figure out what to do with all those useless arenas, stadiums, and hotels.
I've read that they're saying the vaccine might not stop the spread: vaccinated individuals could become asymptomatic carriers. To what extent is this a real possibility, versus an abundance of caution?
Like, I understand they haven't run specific trials on that for these specific vaccines, but for all the vaccines we've had in the past, how effective have they been in preventing transmission?
Has anybody yet seen an authoritative estimate for Quality-Adjusted-Life-Years Lost due to covid? We have them for most other diseases.
How much worse is it for places to want everyone to wear masks than for places to want men to wear underwear and pants nowadays, or for places to want everyone to wear hats or ties at some other points in time? It'll at least be a weird fashion restriction that serves some function in lowering the overall burden of respiratory diseases.
Is the San Francisco / mask one supposed to be "greater" instead of "fewer"?
"Sorry, things may get worse before they get better."
Depends on how you define "things". I believe the only metrics we should really be looking at are hospitalizations and deaths (yes, Long COVID is a problem, but a lesser one).
34% of America's 65+ population has already been vaccinated. This number is rising quickly (I'm tracking it on a vaccination spreadsheet here and will start tracking longitudinal data as of tomorrow's update: https://docs.google.com/spreadsheets/d/1Kti2ccedNp05K_jlda-jozj55WUIv5VmNMjTDnnS6R4/edit#gid=0).
Cases will quickly start to decouple from deaths and hospitalizations. So a spike in cases is not as big of a deal, and really shouldn't result in many additional restrictions, at least not ones similar to the winter spike.
We are soon entering a period where not getting infected will become a matter of personal preference, rather than an altruistic endeavor to avoid killing someone or causing them serious illness.
I'd predict that the cultural effects will be influenced by economics, culture-war factors, or a desire for signaling. Profitable gatherings will continue to be encouraged by advertising; more spontaneous or altruistic ones will stay online. Public group activities will be tilted towards the political factions that are now more likely to engage in them. I'd give a 70% chance that at least one NFL stadium reaches 90% capacity during the 2021 season, a 90% chance that food service workers in Democrat-governed cities will remain obligated to wear masks through the end of 2021 (and beyond), and a 75% percent chance that mass transit ridership stays below 2019 levels for a decade (unfortunately for the climate).
What I have wondered about for a while, is whether a system would arise (or be proposed) for allowing people entry to mass gatherings *only if they provide proof of vaccination*. The Super Bowl this year was eager to announce that it had allowed a few thousand vaccinated health care workers to attend in person. It hasn't come up yet because vaccination rates are still low, but once they are generally available and rates get around 50%, there will be eagerness to push those rates higher, impatience to drop restrictions, and concern about the portion of the population that refuses and the resulting persistence of the virus. Someone will get the bright idea to offer incentives, and someone else will get the bright idea to claim it's the Mark of the Beast; it would be challenging to avoid economic or racial disparities in such a system as well.
Back in March or April, I counted the "it's an annual disease like the flu" as the awful-tier outcome. I stand by that, but it's looking even more likely now than it was then, and I have less hope that I was wrong in my estimates.
New reader here. Excellent article, thanks Scott. Very curious about your predictions on the long term impacts on young children. Like your WW2 analogy, will there will be a whole generation of kids who are fearful of any human contact even after things return to normal(ish)? Or will the extreme social distancing learned behavior fade as we become more integrated into the world again?
This reads like a very Bay Area view of Covid and response - or a very myopic view of a particular part of the culture.
I've flown at least 12 times since the pandemic started. We've hosted many visitors from out of state, we go to restaurants multiple times a week - they are fairly full - and most of our friends do the same. We gather indoors and out for various occasions. Family is the same.
We're fragmenting into two covid subcultures, as others have noted. But the simple fact is that the data does not justify anyone who is under 70 and healthy changing the way they live now vs. how they lived in 2019 and prior.
I would really like to know about the likelihood of long-term effects from mild covid cases in people who have been vaccinated. I am okay with taking the reduced risk of catching mild covid post-vaccination if we're just talking about the ~two weeks of acute symptoms, but if "long covid" is still occurring frequently (I recall seeing estimates of 10-50% in current inoculated cases) then I'm back to dreading the virus, even post-vaccine.
In the SUPER long-term, does COVID seem like something we could completely eradicate, like we did with smallpox? Or is it more similar to the seasonal flu?
And why? Why is that smallpox never mutated beyond control? Why is COVID more or less prone to mutation?
Note that the annual flu vaccine update does not go through the entire FDA approval process that a brand-new vaccine does. I'm not sure what happens, I'd think there is some FDA involvement, but we manage to get out an updated vaccine for new flu strains every year without the FDA getting in the way of that. So I'm relatively confident we can do the same for coronavirus. (Also, my understanding is that coronaviruses don't mutate as fast as influenza.)
A very California-centric post, short on data and attitudes from other states much less other countries. There are already indications that severe lockdowns cause more harm than good. The fact that vaccinations are touted as no guarantee of anything or a return to maskless lifestyle, the 99.5+ survival rate even without the vaccines, the politicalization and incompetence of the vaccine rollout schemes mean even the most brain-dead CNN watcher will stop listening sooner rather than later. It may be worse than the Flu, but that's not saying much.
From: https://www.politico.com/news/2021/02/04/biden-covid-vaccine-strategy-fda-466031
"Public health experts say there are key lessons for FDA in the global, long-held approach to influenza. Shots made yearly to battle the flu rely on a foundational vaccine that is then altered to fight the particular variants that pop up each season. Drawing from flu vaccine practices would mean that manufacturers could skip monthslong trials that enroll thousands of people and instead prove safety and effectiveness in smaller studies that track a few hundred volunteers for weeks.
"Peter Marks, FDA’s top vaccine official, signaled Friday that the agency is leaning toward this approach. [...]"
Seems promising?
I just don't understand how, even after a year-long major life disruption keeping this issue constantly on everyone's mind, the number of people willing to accept a free vaccine against a deadly global plague is STILL only two-thirds.
I like the idea of having the FDA give categorical approvals for modular vaccines. What can we do as citizens to make that happen: write to Joe Biden? call our members of Congress? something else?
I feel like people aren't talking enough about rapid testing. It seems like some California schools have managed to open using testing, which seems promising:
https://www.kqed.org/news/11857118/can-rapid-covid-19-testing-for-kids-help-reopen-schools-some-california-districts-bet-yes
But tests need to be made cheaper by removing unnecessary electronics. For a nice advocacy site, see: https://rapidtests.org/
Michael Mina (https://twitter.com/michaelmina_lab) has been flogging rapid tests (https://www.rapidtests.org/) for a year and I am among those baffled by the lack of take-up. I assume it should work against any variant, and by driving the incidence of cases to near zero, vastly reduce the opportunities for variants to emerge in the first place. And then, why not use it for the flu, cold, norovirus (cruise ships ahoy). Use it on pigs, use it on cows, use it on chickens, use it on anything that moves. A virtual population immune system.
This is a really helpful piece, but are we *sure* that this approach to vaccines is a good idea? Simply "changing the password" as often as possible and speeding it past normal processes? We don't know how reinfection affects people yet, and immune systems are complicated. It seems *possible* this could increase vaccine-resistant mutations or just somehow become a problem. We're moving really fast without a lot of information, and there isn't precedent for this. There's also the issue that, if we're going to make it like the flu shot, why don't we just do that now, and let those willing to take the risk interact normally, while giving vaccines to those who want them as quickly as possible, and helping them isolate? As you say, only a minority is likely to keep up with yearly injections. It scares me that our public decision-making basically operates, on the surface, according to a zero-risk logic. This is why I'm paranoid as you describe--if we continue validating this logic, it's not that hard to imagine things not going back to normal for decades, especially given the damage to the economy and social trust. I agree that, if I was thinking of it from a purely rational perspective, this would be like the flu in a few years. That's what I thought we'd decide back in March. But we didn't. And maybe that was reasonable given the unknowns early on, but it doesn't seem reasonable to me now. We're not operating like treating it as we do the flu is a comprehensible option, so my purely rational assessment is obviously false. This could change, but it worries me. From my perspective, the practical limitations on our ability to control this long-term, and the extraordinary accumulating costs that would come with trying, were very obvious from the beginning, and definitely not worth it. (Other than sensible, targeted mitigation measures, of course--I'm not saying do nothing, but anything close to eradication is not going to happen.) In the short-term, they seem worth it, but when you think about what would have to happen to keep it going long-term, the game is just unwinnable, and trying to win it just eats up the resiliency and ability to plan we could have used to cope with it and make necessary transitions. Society is a complicated system to be messing with---we don't know what we're disrupting in the long run. That's when you end up with no concerts forever---people may adjust to the risk and treat it like a flu, but there's no money left to sustain the concert industry, and a lot of people have grown to see it as an unnecessarily risk indulgence. If you don't come to a relatively clear-eyed acceptance of the situation, you can just get stuck drifting along as though it is a permanent emergency, even if most people want to live normally. They won't speak up against the others if we're operating under a logic that ignores the question of what the endgame is, and whether any of this is worth it long-term. I'm very uncomfortable with how this has proceeded and is proceeding. I appreciate that covid poses a real danger to some people, but I can't imagine any situation in which I would advise the response we've chosen. If I was high risk, I would want to shelter and be given the resources to do so, but I'd be terrified to have all of society "hibernate" like this. I would not see that as likely to protect my future, but likely to jeopardize it, by disrupting the available resources, conveniences, and stable society that make sheltering possible.
Re: "yearly COVID shot". I just got my second shot out of the two-component vaccine. That means me potentially being immune to two different vaccine vectors, hopefully for less than how long the COVID immunity lasts, but possibly longer. Is there a technology that would help us avoid running out of vectors for the foreseeable future?
This article does not make me feel good about the wedding I have scheduled for September.
We've already rescheduled twice...
What do you think about "open source vaccines" like Radvac or the one from winiftred Stoecker ?
We need to consider just how different humanity is now compared to when Spanish flu arrived. It was allowed to take its natural course because we didn’t know any different. Today if we were told that allowing a virus to run its natural course would likely result in less deaths than fighting it (allowing for more mutations to proliferate), we couldn’t make the call. International travel, coupled with the fact that we are so easily triggered in today’s online world, the ‘panic and fear’ factor could make things worse.
Based on tele-work, tele-medicine and tele-everything else: the current nerdopolis that is the Silicon Valley will start losing its dominance, steadily and surely. Not sure how to quantify it, though.
You've described what will happen in the US which seems plausible. Here's what would happen in developing countries: people collectively decide to ignore COVID as-is, it kills off 0.3-1% of the population (depending on the median age) allowing the rest to reach herd immunity. New mutations kick in but they're not perfect at beating natural immunity and in any case the most vulnerable people have already been killed at that point, so the IFR consistently hovers around 0.1% which is low enough to let people ignore COVID. So no vaccinations, no "adjusting the vaccines each year", no masks, no social distancing, nothing. Society moves on as it did with every pathogen in human history after suffering some casualties. India is already there: their case counts have been falling for 6 months now despite having very few restrictions in place, all thanks to herd immunity.
I am affraid you are far too optimistic about going roughly back to normal. Especially in Europe where lockdowns have been stronger and more general than in the state, with much less political opposition. With the economic crisis looming, governments will not revert back to full democracies naturally, there will need to be popular pressure. At this point, there is a risk to fall to real dictatures, as the power in place feels the risk about having to render counts for last year decision threaten so many political, police and media heads.
The counter power at this time is mostly judiciary, parliament is absent / aligned with executive, and only having the judiciary guard against dictature seems small...
This worry me much more than the covid itself, delation, police violence, scapegoating of tourists / lockdown parties, arbitrary fines linked to ever changing illogical regulations... All of this is up regardless of the epidemy evolution (it keep increasing even when vaccination started and the epidemy stagnate).
Prediction about the future are notoriously hard, but there are more signs about the west transitioning from democracies to fear - powered autocracies than signs that the plagues are back.
Humans don't dominate the planet. Plant life does, and it's not even close in terms of cell count. We should expect more viruses to be selected for infecting the social plants, assuming the virus's "goal" is only to increase its copy count.
How much human capital have we lost to covid fatalities. How many potential great works have died with covid patients?
I've been looking at lists of noteworthy people (e.g., people with their own Wikipedia pages) who have died of covid, such as Wikipedia's list:
https://en.wikipedia.org/wiki/List_of_deaths_due_to_COVID-19
Early victims included a couple of Broadway figures in their middle-aged primes, gifted songwriter Adam Schlesinger of Fountains of Wayne and "Crazy Ex-Girlfriend" (who had just signed a big contract to compose a Broadway show with his creative partner Rachel Bloom) and Tony-nominated actor Nick Cordero. So, eleven months ago I figured covid was going to be pretty bad for our human capital.
But, since then, most of the fatalities among semi-famous people have been concentrated in either the elderly or the long-term sickly. I periodically look at the Americans who have died in the Wikipedia list and I don't see much evidence that our culture will deprived of too many great works going forward.
That sounds brusque, but when we are making trillion dollar decisions, we really ought to look at all the evidence, even impolite evidence.
Your impressions may differ, so I encourage you to look at the Wikipedia list or the New York Times' list of people who have died.
Is it safe for people to get additional vaccinations in rapid succession to beat virus variants? I realize that there isn't enough vaccine for that to happen much now, but I wouldn't be surprised if the vaccine process gets sped up.
One more area to hope for progress is for more genome-checking to keep track of mutations.
I am skeptical about the fast track. Any time you change the antigen in the vaccine, it could, by coincidence, resemble some normal body protein and trigger autoimmunity. Or do something else unintended. Not that it is super probable.
I will be glad to be corrected, I am not super sure about this.
As an Israeli who has been vaccinated, I can confirm that they plan to vaccinate every 6 months. When you get vaccinated, they make it clear that you will have to do it again soon.
Another point: Many many health giants around the world have spent billions on new Coronavirus vaccines. They will want to put them to use. I expect that by 2022 there will be an oversupply of such vaccines covering every extant strain.
What about Covid treatments? So far we had a few proposals but the results were a bit underwhelming (chloroquine, remesdivir, favipinavir, etc), but dexomethason seems to work at least for some severe cases. Now we have the monoclonal antibodies and nano antibodies - so far the results are are still a bit disappointing but in theory antibodies should work and at some point we should be able to produce them in mass quantities cheaply.
And by the way - what do you think about amantadine? I have already asked that question here: https://astralcodexten.substack.com/p/hidden-open-thread-1595#comment-1236759 - but with no replies. I think it is an interesting case of auto-immunology disease in science - that is I think the allergic reaction of the scientific establishment to dr Bodnar is overblown.
A minor terminological point on "If vaccines made R0 go to 0.5 or whatever, we would loosen some restrictions until it was back at 1 again." R0 is defined as the reproduction number in a completely susceptible population; the reproduction number when some people are immune (through having had the infection or vaccination) is called R or Rt (see e.g. https://www.nature.com/articles/d41586-020-02009-w).
All I have to say is I have zero regrets for every midnight movie I've ever seen. You don't know how much I miss them, and I treasure every hour I've spent in a theater after midnight. I'm devastated that they may never come back.
YOU'RE TEARING ME APART 'RONA!!!!1111
I also miss my family too. But can quarantine for a few days and see them.
Maybe I really am that paranoid type. I promise I don't have a political axe to grind. But I am just really curious why you would put the chance of rock concerts not happening again or mask wearing staying majority social pressure at "less than 1%" chance? I think it's literally a 50/50 chance.
You even compared it to child kidnapping panics which I agree with. I personally like to compare mask wearing to no smoking in airplanes. No one can smoke in airplanes anymore. It's never coming back. Same thing with going to a rock concert without a face mask. It's never coming back (except in very exceptional one-off cases). I could be wrong obviously, but I would certainly bet money against anyone putting it at "less than 1%" odds.
Thanks for the article!
Did anyone encounter data on woman fertility and vaccinations?
The FDA report on Pfizer has very little actual data (~20 pregnancies, ~half in placebo).
WebMD and other sources just claim that the evidence suggesting a link between infertility and vaccinations is fake.
My friend found this information insufficient, so I am looking for more serious research if any exists yet.
Thanks :)
I have the feeling that a lot of peaople assume that new virus-strains generally should be less lethal than old strains, not due to previous infection but because viruses will tend to mutate towards an equilibrium were it does not kill the host (because there will be a selection-pressure to let the host live and further spread that virus-strain). Is this a valid assumtion? Couldn't it be that "equilibrium-viruses" is selected for over (long) evolutionary time, and that they also depend on the virus in question acting as a selection-pressure on the host? So that, any given mutation is random and might as well be much more (or less, or equally) lethal compared to Corona Classic. Only in the long run will this pan out to be a more flu-like situation. (Confidence in this speculatory line of thought is low.)
I find it strange that people have stopped questioning our response to COVID19 and whether or not it's been a) justified, b) effective and c) cost effective. Now that the vaccines are here, those questions have been sidelined and stigmatized. I think we should continue to interrogate every aspect of our response to the pandemic.
I would be interested in reactions to this piece. The extensively researched claim is that in UK at least, COVID has been massively systemically over-countered: https://architectsforsocialhousing.co.uk/2021/01/27/lies-damned-lies-and-statistics-manufacturing-the-crisis/
Then there's the issue of the mixed at best evidence for lockdown effectiveness. https://inproportion2.talkigy.com/do_lockdowns_work_2021-01-15.html
How should we view clearly costly policies that are not supported by the overwhelming body of empirical evidence? I can see a case being made for lockdowns being so intuitively effective and studying them so fraught that we should proceed regardless of the evidence. But I think this case needs to be made explicit so it can be interrogated. Currently that's not what's happening.
Presumably the virus is going to become and remain endemic, and we'll see perpetual annual or semi-annual vaccination for the most likely variants, just like (and along with) flu vaccines.
Question about different vaccines: To what extent do they "stack" and to what extent do they make each other less effective?
E.g., if I get two shots of AZ but J&J turns out to be more effective, is it possible the antibodies in my body will render an additional J&J shot ineffective?
If so, is it clear whether this cannot happen with mRNA vaccines?
Calling it the "China virus" makes you a really bad person, but calling it the "UK strain" is 100% fine?
I think the first prediction is not well defined. What is a "wave", i.e. what is the y-axis? Case numbers (as everyone knows well) are dependent on testing which changes over time. A more reliable metric (although it changes over time as well) is hospitalisation rates. I would predict that we will not see another wave of hospitalisation in the US similar to the previous 3: the majority of older people will have been vaccinated and the protection against severe illness is very high
For now I'm not sure what to think about all those "mutants". Because it seems researchers "threw" something in petri dishes and made some conclusions with that. But I don't really see any "real world" implications with that... Great-Britain - you know, this country, that "named" one of those highly-infections mutants - got their daily infections down from 60k/day to now around 10/k day in just a month! You could theorize if it was the "old virus" that their lockdown did that even faster. But that is basically impossible to prove in either way! So ... *meh* you mutants, I don't care!
My thinking is more along the lines of this diagram in the German wikipedia about the spanish-flu:
https://de.wikipedia.org/wiki/Spanische_Grippe
We are currently in this second, really big and deadly "wave". We will have a third one, that is smaller, but I don't think it will be the "same" months as in this diagram. Because at least here in Germany nothing happened last years summer and we basically opened up "everything": bars, restaurants, cinemas, even concerts happened (but with fixed seating). And "our" second wave started in october/november; so it seems very seasonal! And sorry, for now I don't think these mutations change anything about that! Maybe time will prove me wrong.
But important about this diagram: it ends after the third wave! And I think that is something we do too with Covid19!
There has been progress recently on making a general flu vaccine that will work against all sorts of flu: https://www.sciencemag.org/news/2020/12/innovative-universal-flu-vaccine-shows-promises-it-first-clinical-test
So, I would imagine this sort of thing bodes well for the prospect of a universal Covid vaccine as well?
I was hoping to see a discussion of the so-called "Long Covid", since as far as I can tell from picking through various studies over the past 6 months, still has limited evidence that it actually exists as a unique effect of the coronavirus.
Studies all seem to fall into two camps: either a very small study of like 50 patients who were hospitalized, are elderly & obese, and demonstrate lung damage in scans taken 1 or 2months after being discharged. Or a survey of people from the "Long-Covid Support Group" on Facebook, who self-report suffering varied symptoms several months on, such as Fatigue, Tiredness, and feeling worn out.
Neither of these are remotely what we are interested in though. There have been enough studies, even without large samples or controls, that suggest lung damage in survivors that I expect there is a real effect, but if it mostly appears in hospitalized patients, who have pre-existing conditions, are elderly, etc. then chances are it's very rare overall. And that's assuming it's actually long term and doesn't go away after 6months.
The surveys get towards what we're interested in: what are the chances that an average person who gets infected with coronavirus has to suffer long term issues, even if the disease itself isn't a danger. But self reports of very vague terms like fatigue are just way too unreliable, and again we want to know if these are long term, not just for one or two months. One particular survey did ask for effects at 12 weeks and found sufferers fell away to almost nothing. At that point you have to ask whether this isn't just standard post-viral syndrome?
Ultimately I can't spend much time tracking down medical papers so perhaps there has been more definite evidence emerging recently as to the prevalence of long term sequelae and the possible effects.
A thing I recently wondered about: Why are there no illnesses that do GOOD things to you?
Like a flu that makes you feel awesome and full of energy.
There are literally hundreds of disease strains that kill you, which is, we should not forget, a bad ending for the disease, in the coffin, it will not spread. If a disease made you feel great, you'd go out and meet people, which is great for as disease, and we would not spent billions to develop vaccinations and social distancing and such against it. From an evolutionary perspective, there should be massive advantages for a disease that makes you feel good vs. ones that make you feel bad. Yet I know of not a single one.
Now, there are a few obvious counterpoints.
1) Most of the symptoms we have when we are sick are not the disease, but the body fighting the disease. A body would fight a foreign substance no matter whether we like ot or not. But there are a lot of diseases that the body cannot successfully fight, from HIV to Herpes, that if you have them you are stuck with them. Why only those that will eventually kill you? And not all illnesses are so noticable as the flu, the body fights other sicknesses with us barely noticing. So a sickness where the signs of the body fighting it are low and the effect of the sickness making you feel great is at least thinkable.
2) If it was good, it would be a symbiote, and the body would just incorporate it, like the colon bacteria. True, but only over millions of years. Bacteria and viruses evolve much faster, so one evolving into a feelgood plague could happen spontaneously without cumbersome human evolution having had the time to give it an estabished niche.
Also, circumstances change. An illness that makes you feel awesome would have led our anchestors to be eaten by tigers or starving in winter, because they didn't build up stores in time. In our modern time, the same illness would just improve how you feel living your life.
Personally, I feel the reason for this is to show us that there are no loving gods, that, if something can be created to be good or bad, it most often is created to be as bad as possible (I kid, partially). But I really wonder. There are diseases that can manipulate an ants' brain to make the ant walk up a leaf of grasss and bite down at the end and wait for a cow to eat it. Some are that sophisticated. But none has the effect of releasing whatever hormone or neurotoxin would make you feel better, despite the huge evolutionary advantages that would have for it? It is easier for a disease to jump from bat to human than to be nice for humans?
Something is wrong here.
Teeny tiny nitpick from someone involved in SARS2-sequencing: The British lineage is actually B.1.1.7 -- not B117 -- since there is also a B.1.117 which co-circulates (at least here in Germany).
That already led to some confusion in meetings and emails, and an often repeated clarification that one-one-seven and one hundred seventeen are actually two very different things.
"There are whole countries whose cuisines are still built around weird decisions they made as part of World War II rationing."
There are whole religions whose cuisines are still built around logical decisions they made when camels were the main mode of transport!
Perhaps I have a bit of a different perspective on the political aspect as I'm from elsewhere (UK), but my biggest concern is the impact on Governance. My concern is that the total suspension of many civil liberties has now been established as acceptable - and what in the future will also be seen as an acceptable reason to implement these.
Also specifically here in the UK, its changed the dynamic of our parliamentary system to something a lot more command and control by the prime minister and his cabinet, with a reduction in public scrutiny by parliament - this concerns me, and I'm one of those actually same side of the political spectrum then them.
And for those who say these are temporary emergency measures, call me tin-foil-hat-wearer but these things have a horrible habit of becoming permanent. I'm still waiting for the 'temporary' income tax to be removed that was put in place in 1799 to cover expenditure during the Napoleonic wars. It would be interesting to predict which government tools become part of daily life, and which will go back on the shelf at least until the next pandemic.
As an expat currently living in Taiwan who was really hoping to return to the US this year: This feels seriously depressing. It sounds like there is a very good chance I won't be able return without getting some variant of the coronavirus at some point. I'm not in any super high risk group, but I'm really not looking for long-term brain or lung damage. And I don't have quite as much faith in American institutions as does Scott.
Still, it's good to get a sobering dose of reality. Now I can consider what I would need to do in order to stay here longer.
Does the FDA approval process matter all that much (or is its slowness typical for other vaccine-producing countries)? The rest of the world also creates vaccines. If the FDA is so slow that the rest of the world has vaccinated everyone already whilst they are still playing golf, then traveling US citizens would simply get vaccinated on arrival or be sent back at the border. Assuming they wouldn't have to be quarantined and tested for new strains first, if the US remains a permanent refuge for ever-changing mutations.
So new reader here. Excellent post with which I mainly agree although not sure if we get new vaccines down to 2 months - if we bet early enough based on Southern Hemisphere spread may be okay if it’s a bit longer. But a flu vaccine like model could work. Three other points 1. Therapeutics- if we get some highly effective measures against capacities that are required for the virus that changes the equation. Some interesting studies including one reverse engineering existing drugs at UCSF but not proven yet. Second excellent review recently in Science (sorry don’t know how to do the link thing yet) on how coronaviruses become endemic - they continue to hit young children who have mild disease and are not immune while the rest of the population is as you describe. There were some reports of a tilt to this pattern in the UK and right now in Boston persons between 0-19 are the highest group with almost no hospitalizations thankfully. Is this the beginning of wave 4 or is the UK variant already more widespread than we know ?
Right now Moderna and A-Z have x capacity to make the mRNA vaccines. Thanks to the crazed DIY vaccine article from last week, it appears that making and getting the peptides for the "classic" version of the spike is not hard.
So how do we create 5x capacity? Is it the lipid sheath that holds us back? Lack of trained personnel to run the factories?
We know this works. How do we scale it?
We also need a way to push this out to the 3d world or they will always be a disease reservoir.
I like your approach of making testable predictions. Much better than just stating opinions. I am still proud that I landed a good prediction of the number of deaths at the end of March in Germany. I am a bit less proud that in summer, I considered the whole second wave of Covid infections "rather unlikely".
In the interest of keeping myself honest, here are three things that I've gotten definitely wrong in, like, April, about covid:
1) I expected that countries which did well in a first wave will also be relatively fine in second and subsequent waves. Ups. Some of them were, but many weren't.
2) I expected substantially more deaths per capita in Europe than in the US. There are huge reporting issues in parts of postcommunist Europe but nevertheless this was also clearly wrong.
3) I definitely didn't expect that countries in historically Chinese cultural orbit (to which I include also Japan and Singapore) would do as a bloc so much better than the rest of the world, despite big differences between their political systems.
I haven't heard about convalescent plasma therapy in a while. Shouldn't this be easy to scale up now, considering we have so many recovered?
Also can you transfer vaccine protection via blood plasma? Would be lovely, if I could simply let my grandma donate me some plasma, instead of waiting till September when it's my turn.
I think you are interpreting the "neverending lockdown" concerns too narrowly. Mask mandates on subways are one of the least bad scenarios that could happen. "Long Lockdown" in a wider sense can include things like:
- Independent and diverse restaurants going bankrupt and ceding space to chains. This won't be easily reverted, as most customers will be eager to eat out *somewhere* and thus end up supporting the chains.
- Public schools never going back to their pre-pandemic functionality, as parents who care about face-to-face instruction have left, and those who remain are either those without the means to find an alternative or those less concerned with the quality of instruction. Evaporative cooling, to use the language of this blog.
- Women leaving the workforce for good, partly because of the previous point. It remains to be seen whether this will be permanent or will last until the next economic upswing. I don't think men's wages will rise anywhere near enough to make up for this; I expect to see a lot more broken families and delayed pregnancies than successful returns to trad.
- The panda in the room: A new-found trigger-happiness of Western governments to pre-emptively suspend basic civil liberties (freedom of movement, assembly and commerce) based on opinion from medical experts or -- even worse -- epidemiologists. After all, we can't wait for the virus to come to us, can we? This sort of thinking has become popular after COVID-19, to the point that experts are now berating Western politics for not taking previous epidemics (SARS-1, swine flu, avian flu, ebola) seriously enough. But what would it mean to take them seriously? Border closures during travel season? Mandatory quarantines? Countries like Italy and France have taken the axe to civil liberties in ways that no one had seriously entertained since the 1960s; inter-European borders were closed for months. Demonstrations were outlawed in several countries, and at least one (Poland) has used this occasion to ram an unpopular law through. While the most stringent measures that have been actually taken in the US were restrictions on commerce in blue states, the European model was being seriously discussed in media and often positively contraposed against the American laissez-faire one. Unsurprisingly, a general feeling of "it could happen here" has emerged. Meanwhile, no one has bothered to delineate the level of danger that warrants extreme measures like stay-at-home orders and border closures; by the standards that are emerging, we could have a pandemic scare every 3 years, not to mention that a less scrupolous government could easily pull one out of its hat whenever convenient.
All of these concerns would have looked paranoid some 10 years ago, where the typical Westerner believed liberal democracy was self-sustaining at least in the ever-growing Dar al-Democracy, and whatever mistakes politicians would make the market would fix with its gentle invisible hand. Suffice it to say, one thinks of solid ground differently after an earthquake.
You can see more covid predictions from various platforms here: https://metaforecast.org/?query=covid&starsThreshold=2&numDisplay=72&forecastsThreshold=0&forecastingPlatforms=CSET-foretell|Good%20Judgment|Good%20Judgment%20Open|Guesstimate|Hypermind|Metaculus|PolyMarket|PredictIt
I wonder what you predict/think about Covid "long-haulers" and whether that will distinguish Covid precaution from flu precaution. The anecdotal evidence about the persistence of symptoms and the unknown long term damage to lungs, heart, etc. significantly increases my concern about getting it. If vaccination prevented long-haul Covid, that would be great news.
Some random questions:
1) Genomic surveillance is so spotty in the US. What are the chances that other variants of concern with similar properties to B.1.1.7 or B.1.531 are currently circulating in the US? I read a paper (can't find it now) suggesting that B.1.1.7 wasn't increasing in relative proportion at quite the same rate in California as it was in the UK. One suggestion I think was that there are other variants that are similarly more infectious out in California. (On the whole, if this is more likely then that should decrease our expectation of a large fourth wave?)
2) To what extent is it possible to form an expectation of a fourth wave without mathematical modeling? I haven't seen a tone of modeling that attempts to take vaccination, seasonality and emerging variants into consideration. I think Youyang Gu does this, and his fourth wave is very moderate: https://covid19-projections.com/path-to-herd-immunity/
So, question: are there other attempts to model this mathematically rather than heuristically?
3) In the endgame where SARS-CoV2 becomes endemic, does it have persistently low IFR because of immunity from prior immunity? And more broadly is this what happened with other currently circulating CoVs? Suppose that you took one of the other "common cold" CoVs and removed it from human circulation for 150 years, and then reintroduced it. Would that cause a pandemic, because of weak prior immunity? And would it be similarly lethal?
It turns out antidepressants greatly mitigate Covid symptoms. Nature is an impeccable source. The magnitude of the effect reported is great. I don’t know why this is it getting more publicity
https://www.nature.com/articles/s41380-021-01021-4
I don't know if I'd trust Amazon vaccines.
Those on reddit.com/r/skincareaddiction say you shouldn't buy skincare products from them because their problem with fake products. People have said that they've bought from trusted sellers, received brand name packages, and only discovered the product was a knock off because the consistency/color was different from their last bottle (or something worse happened like they got a chemical burn).
My prediction: 100% chance that Joe Biden is going to OWN a coronavirus vaccine failure if his administration does not get its act together and create a national corona vaccine appointment scheduling system. The current every-grocery-store-for-itself mess of vaccine appointment methods is far far worse than the Obamacare rollout site ever was, and is far more infuriating than not having the vaccine at all (ala #trumpolini times)
Bioprocess engineer here, I wanted to comment on your proposal here:
"They allow anybody to manufacture vaccines and charge market price for them, subject only to usual safety restrictions. If Amazon wants to get into the vaccine distribution business, for God's sake, let them."
This seems like a really unlikely outcome to me, for a few reasons. For a bit of context on drug manufacturing, a drug approval also requires approval of the manufacturing process and the analytical methods testing the quality of the drug. If a drug manufacturer wants to make major changes to their manufacturing process, then they may need to perform new clinical trials to justify the new manufacturing process. This is based on the concept of "quality by design," usually abbreviated as QbD, which is often compared to the older, crappier paradigm of "quality by testing." Quality by testing just means that you ensure product quality by running a bunch of analytical tests on important properties. For the mRNA vaccines, that's probably going to be stuffy like number of copies per dose of the mRNA transcript by qPCR, nanoparticle size by DLS or some other sizing technique, etc. The problem with this approach is that there is always the potential that there is some super important quality attribute that you aren't aware of and don't test for. If you have some critical quality attribute you aren't testing for, then any changes in the manufacturing process that also impact our hidden quality attribute could negatively impact drug safety or drug quality. Therefore, the QbD paradigm was invented, which involves control of drug quality through control of the manufacturing process, so that even if you aren't testing every critical quality attribute, you know you are making a consistent product because your manufacturing process is consistent. All modern drug processes rely to an extent on both testing and manufacturing design for quality control.
Because of the expectations around QbD, it's going to be really hard to open up vaccine manufacturing to all potential manufacturers. You will either need to 1) force the original vaccine maker to transfer all information on their manufacturing process and analytical methods to all prospective manufacturers, 2) make new manufacturers do clinical trials on their drugs or 3) abandon the QbD paradigm and allow manufacturers to develop their own manufacturing processes to make existing vaccines, without a requirement for new clinical trials.
#1 is never going to happen, because the innovators are going to scream bloody murder about having their patents violated. Even potential manufacturers would likely avoid this, because they don't want to set a precedent that the FDA can just seize your manufacturing IP.
#2 is essentially the existing biosimilar pathway, just starting before the existing vaccine patents expire. This is probably feasible in theory if the FDA and Congress nullified all patents on the vaccine drug substance, or they bullied the manufacturers into allowing competition. Even then I can say from experience that making a biosimilar still takes a ton of time and resources. The requirement for new clinical trials will delay the release of the biosimilar vaccines significantly, probably to the point that the biosimilar vaccine is no longer effective against newly circulating strains. I don't really see anybody taking this route.
#3 is basically a supercharged biosimilar pathway where you abandon QbD principles to avoid doing new clinical trials. Without QbD control, you would almost certainly need extensive testing of the biosimilars using the analytical methods of the innovators to ensure comparability. However, you would still need to force vaccine innovators to share their analytical methods, which seems unlikely. And you are still voiding all of the initial vaccine patents to allow early biosimilar competition. And the FDA is super risk averse and is not likely to ever approve of an approach like this.
The more likely scenario for expanded vaccine manufacturing is for the government to subsidize voluntary contract manufacturing relationships. This is already quite common; Lonza for instance is acting as a contract manufacturer for Moderna, for instance. In these relationships, the innovator maintains full control over their IP, and so they can share their full manufacturing process without worrying that the contract manufacturer is going to steal their IP. It's important though that this be voluntary and the innovators get to choose who to work with. While you can sign agreements saying "IP is protected," a contract manufacturer still learns a ton by making a drug. For instance, let's pretend Merck has a fledgling mRNA development program. Moderna is most definitely not going to want to use Merck as a CMO, since that would be giving Merck a huge leg up in terms of manufacturing experience for their potential new mRNA vaccines.
At the moment, with our government continuing to impose lockdown while promising that it'll all be relaxed down the line (there will be no St Patrick's Day parades this year, which doesn't have me weeping bitter tears to be honest), I have no idea. Every time something new comes along, e.g. they relaxed restrictions for Christmas, people travelled home from the UK, and we got the new mutant variant and up went the infection rates once again and we're back at Level 5 restrictions https://www.gov.ie/en/publication/2dc71-level-5/
So my natural tendency is to go "things will ALWAYS get worse" but will they get better? No idea.
On the other hand, D:Ream assure me "Things Can Only Get Better" https://www.youtube.com/watch?v=V6QhAZckY8w
I find it very hard to care whether covid is eradicated or not, compared to how much I care that the lockdowns end. I feel like people forgot that these are extreme measures that we shouldn't use unless absolutely necessary.
I remember in January when I heard about the lockdowns in China and thought "that's absolutely barbaric, I'm so glad that can't happen in America." Then in March there was talk about lockdowns, and everyone was duly skeptical, but after looking at graphs and statistics and checking the numbers and careful argument we were convinced that it was necessary. Now when I point out that the specific conditions which made the lockdowns rational (despite the incredibly high cost) might stop being true well before the virus is eradicated, I feel like I'm taking crazy-pills. "You can't stop sheltering in place, you might get covid." I'd much rather catch covid and take my chances than spend another year inside, it's only the structural issues around R0 and hospital capacity that I'm worried about.
The only question I want anyone to ask is "when will the risk of overwhelming hospitals go away, when will the death rate be less than 100k/year, even without restrictions?" If you're wondering how long lockdowns would be needed to eradicate the virus, I assume you've lost track of your priorities.
Nicely put, Scott, as usual. What's missing is the public health impact, not of the Rona, but of the response. The illness and death caused by closures and lockdowns. First order effects like suicide, alcoholism, drug abuse. Second order effects like domestic violence, child abuse (includes school closures), depression. Third order effects like unemployment, poverty, and stress. I posit that the public health impact of our response far outweighs that of the virus itself. On the whole, Rona is an opportunistic disease that attacks mostly those about to die soon of something else. All the rest of what I mention impacts healthy, middle age people and, especially, children.
Well done Scott - I like the way you share your thoughts. I would love to hear your thoughts on "What is the Goal?" or "What exactly is the problem we are trying to solve?" I think all would agree COVID has created many problems, so all the more important that we define the biggest problem clearly. I head a wise saying once that "a problem well defined is more than half solved."
I agree with Scott's sense that COVID will always be with us -- a sort of lingering flu like illness that burns hot in certain regions of the country during coronavirus season in Nov/Dec/Jan -- but I think his introductory predictions that we will have further "waves" in March and next Fall extremely pessimistic and unlikely.
First, what do we mean by wave? At this point, everyone knows that case numbers do not quite equal infections and that the scarcity of testing early on in the pandemic makes our first wave look artificially small. The best estimate for true infections that have occurred in the US come from Youyang Gu, who has run the remarkably excellent website COVID-19-Projections and essentially outperformed anyone else in the COVID modeling game. (Link: https://covid19-projections.com/).
Gu's Super Fancy Model estimates that our peak number of daily infections during wave 1 was ~300k; during wave 2 we hit ~250k; during wave 3 we hit ~660k. I will somewhat arbitrarily define a "wave" as a period in which the US exceeds 200k true infections per day. My prediction (70% confidence) is that this will never happen again.
How are flu vaccines able to get approved on a yearly basis? Do they spend 6 months out of the year proving their efficacy to the FDA?
The news cycle is currently saying California = Heavy Lock Down but Florida = No Lockdown and Both Have Same Outcomes. What are the best arguments for and against this kind of thinking? Preferably not mechanistic argument (masks and lockdowns HAVE TO do something so this CAN'T BE RIGHT).
I've been seeing articles about a treatment being developed in Israel: https://mynorthwest.com/2567116/inhalable-covid-treatment-israel/
Any chance this will be a game changer?
What I do not get in this whole Covid mess is why no country so far pushes well engineered comfortable masks with a good seal and long lasting efficient filtes with low breathing resistance in a much more aggressive way, espcially since the general effectiveness of masks is no longer really in doubt. Instead we are stuck with cheap one way products that most people either wear wrong or not at all and that get quite uncomfortable after an hour or so and essentially cannot be worn during any kind of serious physical exertion.
I have written about this at length (https://www.lesswrong.com/posts/yKYg6D7HNxLuJDcLS/hammer-and-mask-wide-spread-use-of-reusable-particle) back in April last year before wide spread mask use was a thing and think I deserve some credibility for calling the "masks don't do anything" bullshit hard and early.
It becomes increasingly clear that between mutations that evade the antibody response and the fact that only 70% of people in most countries are willing to be vaccinated vaccination will not be the silver bullet we hoped for. Thus I think implementing such a strategy, at least as backup for the next winter is a very sensible and comparatively cheap strategy that would also protect us from potential future pandemics after Covid.
Does anyoe have a good idea how to make this idea more public?
You didn't discuss the possibility that higher-contact people have gotten COVID already and the remaining vulnerable population is lower-contact.
From contact tracing we know that COVID spread is highly differential -- super spreader events and other infections going nowhere.
Maybe enough of the super spreader pathways have been exhausted.
You didn't discuss the possibility that higher contact people have gotten COVID already and the remaining vulnerable population is lower contact.
From contact tracing we know that COVID spread is highly differential -- super spreader events and other infections going nowhere.
Maybe enough of the super spreader pathways have been exhausted.
India apparently has a bunch of vaccines but can't get them deployed for various reasons.
https://www.theguardian.com/global-development/2021/feb/14/we-took-a-huge-risk-the-indian-firm-making-more-covid-jabs-than-anyone
"Instead we have a patchwork of approvals and I have 70m doses that I can’t ship because they have been purchased but not approved. They have a shelf life of six months; these expire in April."
1. Daily deaths is a far better visualization of the waves in the US. Confirmed cases make the first wave appear small, but when you look at deaths you can see the first wave was nearly double the size of the second wave. When you do check your first prediction, accuracy should be determined by checking daily deaths.
2. I highly doubt that once we reach some sort of equilibrium with the coronavirus that we'll need yearly updates to the vaccine. Right now it is "mutating quickly" because there is a huge number of replication events combined with the fact that the virus is still getting to know us. Once infections are down and the virus has discovered most of the "low-hanging fruit" mutations that provide large jumps in fitness, the rate of mutations will drop and the availability of mutations that provide large jumps in fitness will also drop. In fact, my prediction is that the only thing needed to drive the risks of covid down to the levels of the flu is a single exposure to the original strain of the virus, whether that exposure comes from the virus itself or the vaccine. After that, exposure to the newest variants might cause symptomatic infection, but the "old password" will still give previously exposed immune systems a major head start.
2b. My point above isn't a counterargument to your prediction that public health officials will recommend a yearly vaccine for the newest covid strain. Just saying that I think a yearly update will be major overkill after ~2022.
3. Many of your predictions have some vagueness that will make them difficult to check. For example, the prediction about mask usage on an average street in SF Bay Area seems particularly difficult/expensive to check. Maybe change it to focus on something you can easily check, like the proportion of businesses that still require masks (easily checked by viewing signs on doors). Also, this prediction seems to be a prediction about wearing masks OUTSIDE, which could change if people finally get the message that covid doesn't spread well outside. So, its entirely possible that your prediction is "correct" even though we're still living with ridiculous mask requirements.
Nice info as usual! Does anyone have a prediction about when schools get back to normal or what they'll look like in the fall? My 7yo is going to in-person school with restrictions and here's what those restrictions look like - Everyone is masked, which still makes me so sad to see. He gets 15 minutes to eat lunch because that's how long they're allowed to remove the mask at all. Many of the afterschool clubs and extracurriculars have been cut. A lot of typical activities are proscribed at recess and there aren't so many fun or "special" activities to spice things up. Obviously there are no field trips They're still doing a couple of special in-class events but parents are not welcome. Logistical arrangements were required to e.g. hold more class time outdoors, which obviously isn't feasible in lousy weather. Otherwise they stay in the same classroom all day rather than traveling to different rooms for specials like music or art. It sounds like such a grind to me, for both the students and teachers. It all can't possibly be sustainable long-term, and to my perception there's a lot less joy and anticipation at present.
Frankly, a lot is obscured in looking at “the peaks” across the whole us. At geographically finer resolution, nowhere that had a bad first peak (use deaths not cases) had a bad second peak, let alone 3 peaks. What looks like new peaks is simply the virus reaching new populations. the speculation about strains causing big new peaks are a bit dumb, seeing as the UK and SA covid cases have both fallen 75% over the past month.
UK strain vaccine resistant? Omg. So what is the point in getting the current mRNA vaccines?
Quick questions:
Why is there an approval process for vaccines, at all?
Has any vaccines ever done more harm than good, historically?
Like did some vaccine actually cause autism?
Is there a thalidomide-level clusterfuck that I'm not aware of?
[or if you think those are leading questions......]
What is the argument for regulating vaccine-production?
One plausible trajectory is that of the Russian flu of 1889-90. Not only is it now suspected to have been a coronavirus, but the patterns of infection and disease were very similar - severe/deadly in older people, almost no effect on children. Speculation is that children built up immunity early on, so as they grew older they had only mild symptoms; these symptoms now present as a 'common cold'. Given that covid-19 also has very few symptoms in children, it is entirely plausible that they will develop a partial immunity that protects them from all but mild symptoms as they grow older. No vaccine necessary.
I noticed this curious pattern emerging from last week's NERVTAG updates.
1. Using the SGTF bio-marker as an identifier for the B.1.1.7 variant, the B.1.1.7 variant is now at 90% of the strains detected across the UK. Yet UK case rate continues to fall, and the hospitalizations rate continues to fall.
2. By case *percentages* B.1.1.7 seems to have a significantly higher R0 value than the "classic" D614G strain. However, some epidemiologists predicted that there would be a new case surge when B.1.1.7 reached 70 percent penetration in UK's virus pool. B.1.1.7 reached ~70 percent penetration the final week of December in most areas of the UK. By those predictions, the UK is past due for another surge.
I've been discussing this with an immunologist friend, and his only explanation is that people are masking and social distancing better now in the UK. I don't know of any way to accurately measure this variable. And I'm far from convinced.
Has there been any writing/speculation that the mutations we're seeing now represent "low hanging fruit" in the evolutionary space of the COVID spike protein? A few of the point mutations in the different new sub-strains are the same, which may lend credence to this idea but I'm not sure.
I read someone say that humans triumphed over the animal kingdom because while they adapt over evolutionary time, human ingenuity adapted at an exponentially growing rate. But it may be that the tables have turned briefly - the virus evolving new strains faster than we can develop, approve and manufacture vaccines for it.
Unrelated question: why wasn't there a Metaculus Monday post yesterday?
I've been saying all along that COVID will be with us essentially forever (for the next century at least), so I feel a bit vindicated now.
That said, COVID is more deadly than the flu, and it disproportionately affects old overweight diabetic cardiovascular-compromised people -- which is a sizeable chunk of the American population, which is getting larger every day. For this reason, I still expect mask requirements to become the norm in most public indoor spaces (e.g. grocery stores). I also expect public gatherings of any kind to become a lot less frequent, and of course the same goes for nationwide/international travel.
Alternatively (or perhaps in combination), society might slowly change, so that COVID-vulnerable people are expected to stay indoors all the time. If a person dies from COVID while being old/fat/etc., the public sentiment would be, "well, he knew the risks, so it's sad but it's kind of his own fault".
I wish people would stop posting graphs that look the first one in this post; that's the number of confirmed cases, but our ability to test in the first half of 2020 was abysmal. Almost all of the models predict that wave #2 was smaller than wave #1. https://ourworldindata.org/grapher/daily-new-estimated-infections-of-covid-19?tab=chart&stackMode=absolute&time=2020-03-01..latest®ion=World
Non-COVID question, meta for the forum:
When I switch back to an ACX tab, my computer is busy for ~10 seconds re-rendering the page. I have a super-nice high-end computer so this must be horrible for people with normal or below-average hardware.
It seems the same whether or not I am using the "ACX Tweaks" extension.
I'm no expert at JavaScript debugging, not at all, but it looks like the "visibilityChange" event is triggering some extensive UI refresh. Can we ("we" meaning the community including me; SubStack Inc could do it too but they are busy and have a long QA process) write an extension that disables that functionality, and perhaps makes refreshing the page a manual event done by a button-press?
> "When the fifth wave strikes in late spring/early summer [...] another part of the population (~25%?) will have had the disease already"
Note that more than 25% of the US population has probably been infected already. My best point estimate through 2020 was 29%. See e.g. the CDC's estimate, COVID-19 Projection's estimate, etc.
(I only skimmed the post quickly, but this jumped out at me.)
This isn't working as an open thread. Everything or just about everything is about COVID. I suggest having open threads that don't lead off with highly engaging topics.
>The second wave was in July, when we got sloppy and lifted lockdowns too soon.
When I looked into this "second wave", back in the summer, I concluded that there was no second wave, it was the first wave, just in different places. New York and Massachusetts got hit hard early on, but then it calmed down there and, after a bit, picked up in other parts of the country. If you just looked at the stats for the US (which is a BIG place) as a whole, it looks like a second wave, but that's just an artifact of lumping the whole country together. If there was no second wave, that argues against the notion that "we got sloppy and lifted lockdowns too soon".
Prediction: lockdowns will continue to have no effect on the virus (compare Florida to California), nor will mask wearing nor social distancing. Also predicted: lockdowns will continue to harm the least-prepared, least-capable, and least-off people in our society. You know ... the people we don't listen to and don't care about.
Sorry if I'm cynical, but this ENTIRE last year has been filled with nonsense.
If you study the Covid Tracking Project's regional & state charts you will see that nearly everywhere in the country there have been two waves, not three. Now that the waves are in sync, it is unlikely there will be another in the next few months. Probably not another wave until next November.
If you're holding your breath near an infected person, you won't contract their infection (until you breathe in). I've summarized this idea with other insights here: https://garethidris.medium.com/mask-of-the-mind-8147fff8ea0c
Are we doing the world a disservice by labeling the variants by country of first expression? Is this just a virus with a fast rate of evolution and multiple possible vectors to do so? Would it be better to show how it may outpace a slow vaccination campaign?
> But even if they have it tomorrow, that's...what? Another four months for studies, one month before the FDA is able to meet to discuss an approval (you can't rush meetings!), two months to ramp up production
Shouldn't the approval process and ramping up production run in parallel? If a company is fairly confident that it's vaccine is likely to be approved, it should make sense to start manufacturing it before it's approved.
Indeed, why didn't this happen with the current vaccines against the original strain? At least I assume it didn't, or the initial rollout should have been much faster.
Any reason Overcoming Bias isn't on the blogroll? Seems like a surprising omission.
"I bet we'll pick up the Asian custom of wearing masks in crowded areas if we feel sick." Depends on who "we" are. Some people already skew masks. Also, couldn't wearing a mask after the pandemics ends make some people nervous, making them think the mask-wearer is very disease (maybe carrying a new COVID strain or something like that)? To be honest, I would like to see mask-wearing when one feels sick being mainstreaned, but I do not think it will happen.
I have seen polls that suggest nearly half the U.S. population may not opt to be vaccinated. Here Germany the government has tied re-opening to an r factor of 35 or less. The current German lock-down seems to be pushing the number in that direction but new variants have the government prepping the population for continued lockdown. I wonder how the vaccinated population percentage will affect government policy?
Kind of strange reading an article like this, praising lockdowns, not mentioning the data we have on their efficacy (very nearly none as far as the virus is concerned, but with staggering negative side effects). Why is that? Are you, Scott, ignoring the data?
Well written. I feel like you just wrote all of my intuitions in a meaningful way. Thank you. Sharing.
“Classic” = Wuhan, China strain?
Like most recent waves, this one began in an animal lab or a meat market, i.e. a site where we crowd together and brutally manipulate non-humans. From that filth & degradation, "zoonotic" viruses will continue to arise; and the new global system of control that takes advantage of that spread, like any good Disaster Capitalist, will continue to amplify fear (i e. the meta-virus) to maximize its own hold over us. Prediction: humans will continue to herd and slaughter, and when disease wafts off the kill pile, we'll cage and (medically) torture more animals to "solve" it.
"Over 900 UK health and care workers have died of COVID-19, most due to occupational exposure. They continue to die because infection control in health and care settings is built around an outdated 'droplet' model which ignores that #CovidIsAirborne. 1/"
https://twitter.com/trishgreenhalgh/status/1362386480975540224
I apologize if this is a dumb question. I am not a scientist but my sister is, which makes me think I have scientific genes ...
Is it possible that some significant portion of the population is simply not susceptible to COVID? And maybe the recent decrease reflects approaching herd immunity in the susceptible population?
Technical question about B.1.1.7
How, at the human mechanical level, is it "more contagious"? What makes it spread more? Does it travel further in the air when sneezed? Does it make people more likely to cough? What's the mechanism?
Most promising predictions. But it depends, what happens in the future, of course. For example, there is no such certainty with vaccine and vaccination in general in Nordic after the situation which happened with 80 old people who vaccinated in Norway. Because of this, Sweden still disputing regarding vaccines. By the way, a strange trend happening because of this uncertainty. Swedish companies started renting additional warehouse space, such as https://www.matchoffice.se/hyra/industri-lager/malmo, in order to fill it up with products. Waiting for a more strict lockdown, I guess.
If it's a seasonal coronavirus like the flu or the common cold then you'll see another surge in late March / April because that's how seasonal coronaviruses behave. The waves we've seen so far also mimic the behavior of seasonal coronaviruses.
I thought y'all might be amused to know that Byrne Hobart (whose Substack is "The Diff") has written at least a short version of the article that Cade Metz originally claimed to be writing for the NYT. "Why Did One Internet Subculture Spot Covid-19 So Early?": https://diff.substack.com/p/why-did-one-internet-subculture-spot.
My favorite snippet: "[Rationalists are] a sort of distributed, mostly open-source monastic order, spending a lot of time contemplating the world and passing down important observations, but less time directly interacting with it."
My prediction: 80% chance that on an average day in mid-2022, inside of an average business (grocery store, hair salon, bookstore, etc.) in the SF Bay Area, more than 20% of people will be wearing face masks. (This of course assumes that the relevant business is open.)
This might or might not conflict with Scott's second-to-last prediction. I would be interested to hear what probability he would give to my statement.
Well, everyone here knows the NYT is infallible, right? /sarc
https://www.nytimes.com/2021/02/18/opinion/fake-news-media-attention.html
The issue of people believing conspiracy theories is one thing, but the NYT's solution is 'don't try to think for yourself! just trust us! And Wikipedia!'
Generally, it is well established in the scientific literature that
survival times (and therefore infection rates) for enshrouded-RNA
viruses (such as CoV-19, influenza, SARS, MERS, common cold) are
strongly influenced by ambient humidity and (to a lesser extent)
temperature. (Homeland Security's research department confirms that the
CoVID-19 virus conforms to this pattern.) This pattern has been known
for more than a decade.
One Mayo Clinic study in schools showed a factor-of-2.3 decrease in
influenza when in-school humidity was raised to 50%
(https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0204337).
An Australian study found that a reduction in relative humidity of 1%
was predicted to be associated with an increase of COVID-19 cases by
6.11% (https://onlinelibrary.wiley.com/doi/full/10.1111/tbed.13631)
You may find this US Department of Homeland Security CoVID-lifetime
calculator interesting: compare the usual (less than 20%) indoor winter
humidity with 50% --
(https://www.dhs.gov/science-and-technology/sars-airborne-calculator)
Making it a policy that public-building humidity should be set at 50%
and temperature set at 75 degrees F would substantially reduce the
infection rates for not only CoV-19 but also for influenza and common
colds. Given that tens of thousands of people die each year from
influenza, this is a simple *prevention* measure would have saved
hundreds of thousands of lives from that disease alone; if it had been done
already, it might well have prevented this epidemic, or might prevent more
"bumps" -- and much less intrusively than the currently fashionable measures.
For what it's worth...
People forget that the “waves” get bigger because more TESTING became available. Not because there were actually more cases, and not all the tests gave accurate results. This is blown way too far out of proportion.