Whatever precautions you choose to take to protect yourself and reduce the risk you have covid with which you could infect your family members, which I'm glad to see the post trying to calculate out, I think people tend to ignore or at least underrate the agency of our elderly relatives in deciding their own risk tolerances. As an example, my great aunt is in her late 80s, she knows she isn't doing to live forever, and she's hard of hearing. If (and based on recent experiences at a family wedding, this seems to be the case) she wants to be able to hug all her nieces, nephews, and grandchildren, and if she'd rather have her relatives go maskless so she can read lips and actually talk to people, it's not clear to me that she's making a mistake. And if the cost to me for accommodating such a preference is a few hours of expected life (realistically, less than the duration of the holiday gathering itself), then that is well within my personal risk tolerance as well. Obviously individual preferences and tolerances vary.
Oh no, I only meant to recommend masks in the lead-up to the gathering, not the actual gathering itself. You are absolutely right and I've edited to make this clear.
Curated
Disclosure: I arranged for the Lightcone team to pay Connor to research how covid risk could affect holidays, and write this post.
I know a lot of people have been anxious about how to reason about omicron, covid and the holidays. I think this post does a good job getting the major considerations across and reasoning about them sanely.
If anyone has strong disagreements about how to model covid risk, I encourage them to speak up.
I strongly disagree with this post and think we should be much more conservative until more is known re: omicron. I don't have a writeup in great detail but thought it would be worthwhile at least to speak up here.
Omicron will make up at least 1% of cases in the US by Dec 31. Which means it could make up substantially more than that. However, in mid-December when you’re traveling and going to solstice, it probably won’t be that high—and even if it’s 5 or 10% at that point, that’s not going to have a major effect on the state of COVID.
Not sure when this post was written, but I think this is an extreme underestimate at this point. For instance, my own current median guess for Omicron overtaking Delta in the Bay Area specifically is early this coming week. This is based on eyeballing doubling rate estimates in Zvi's posts, and guessing how our initial conditions could plausibly compare to London or Denmark. (This wasn't a careful calculation, but the exact initial conditions don't actually matter that much because the doubling rate is so fast, so a factor-of-two in initial conditions only changes the takeover date by 2-3 days.)
I expect that using the Polymarket predictions as a proxy here will severely underestimate Omicron's timeline for two reasons. First, those predictions are for the whole US; we'd expect it to hit first and fastest in major urban centers with a lot of international travel (as we saw in e.g. the UK). Second, those predictions are about what the CDC data says on Jan 4, which means there's a ton of lag built in - both from the usual lag on data, and from the holidays slowing things down.
Even given all that, Polymarket currently gives an 83% chance that Omicron will be >50% in the US as a whole on Jan 1, based on the CDC's data of Jan 4.
Good point re the urban centers, was pretty dumb of me to forget to adjust for that. I've added two ETAs to the post to account for this.
An average vaccinated 30yo now loses about 6 weeks of expected life from contracting COVID instead of 2 weeks, because of waning vaccine efficacy.
The counterfactual here is "you never get covid", so I'd take this number with a large grain of salt. If, on the other hand, the counterfactual is "you get covid a few years later", then the loss of expected life does not occur. Additionally, if you do get covid, you're (probably) super immune for a while, which presumably increases your quality of life.
Using a counterfactual of "getting COVID a few years later and you balance out" is certainly tempting, but I don't think that's really how it would go down. Based on how vaccine efficacy wanes, reinfections occur, and new variants are introduced, my guess is that you lose all your immunity and more within 2 years, plus in the next decade we probably will develop increasingly effective drugs against it. Hard to sum everything up but my guess is that getting COVID causes a benefit that is less than half the badness. Probably I should make a best guess here and add it at some point in time but this is the type of factor of <2 that occasionally pop up on either side that I typically ignore.
If, on the other hand, the counterfactual is "you get covid a few years later", then the loss of expected life does not occur.
What's the intuition here? If we believe that infection confers less immunity than immunization, naively the counterfactual looks more like "get covid N-1 times" vs "get covid N times." Rather than "get covid once now" vs "get covid once some time in the future"
Average masks are probably surgical or cloth, depending on your location; KN-95s are a slight upgrade from average, maybe 1.5x, but fitted N-95s are significantly better, probably 4x. P100s would still be great if our society did not treat them like a slap in the face.
https://smile.amazon.de/-/en/Aura-Respirator-9330-Folding-Unvalved/dp/B00VAT74NG/ref=sr_1_3?crid=2FHXM7CKD66WS&keywords=aura+3m+9330&qid=1639930225&s=industrial&sprefix=3m+aura+9330%2Cindustrial%2C73&sr=1-3 is equivalent to N99 and designed to actually fit. If you don't like wearing P100s they are likely your best bet.
An average vaccinated 30yo now loses about 6 weeks of expected life from contracting COVID instead of 2 weeks, because of waning vaccine efficacy.
I think this is incorrect if you believe that vaccine efficacy against severe disease decreases more slowly than the efficacy against infection. This seems to be borne out by observational data and makes sense theoretically as neutralising antibodies don't stick around as long as B and T cells.
Therefore, conditional on infection, your protection against severe disease is actually increasing even as overall effectiveness wanes (i.e. each infection has a lower chance of leading to severe disease).
I think the big remaining question for me (as someone who got a booster a while ago) is whether it's worth it to wait for the next RaDVaC vaccine. They claim that the current version should work against Omicron, but that they plan to release an updated version 'soon' anyway:
Current RaDVaC peptide vaccine designs are largely resistant to the mutations of Omicron. This is in keeping with the strategy of RaDVaC vaccines to be more robust against variants. [...] Recent discoveries about SARS-CoV-2, in addition to Omicron, have suggested minor refinements to RaDVaC Gen 11. Therefore, RaDVaC will soon release the Gen 12 vaccine design.
(I could make the current version and then make another batch with whatever the new peptides are, but it's fairly expensive to get ingredients for).
Also, related to the expense, I wonder if we could setup some sort of buying group since I think the ingredients are mostly expensive because the minimum quantities are enough for hundreds of doses. Not sure if there's any legal complexity there though.
"6 months ago I wrote about how 30-year-olds should basically go back to normal and no longer take many COVID precautions."
Will the hospital system stress out again in many states because people did not control the transmission? We will see soon. I just did not understand that why so many people did not understand the power exponential functions.
Currently, the omicron doubles 3-4 days (Germany and British data). Let's assume the vaccines reduce the severity into the swan flu level. Now, what will the swan flu that doubles in each 4 days will lead? Simple math will tell us, it is UNACCEPTABLE.
Yes, it definitely will, and yes that will be unacceptable. Will that be because of vaccinated scrupulous LessWrong-reading mask-wearing 30yos during the holidays? No. That will contribute much less harm-to-benefit than many, many other actions.
If 30s live as normal, the transmission will not be controlled and the health system will stress out even further.
https://www.news5cleveland.com/news/continuing-coverage/coronavirus/local-coronavirus-news/gov-dewine-orders-ohio-national-guard-to-help-understaffed-hospitals-with-rising-covid-cases
First, I think we are all still pretty far from living as normal. Many things in our past lives would have been more than 1k microcovids.
Second, even the most informal versions of test and trace (telling your friends if you develop any symptoms, so they can tell their friends) can significantly reduce transmission rate.
Third, all this is in the context of the holidays. Fourth, 30yos are not the only segment of society. Fifth, the health care system is not yet close to capacity in almost all places (if your local hospitals are overwhelmed, obviously do not act normally). Etc
Simple calculation suggests the transmission rate contribute much more to the life of loss than the mortality rate. Any measures improve the transmission will cancel the vaccines' linear contribution to the death rate. The first priority of the vaccine should be prevent transmission, not mortality rate.
Thanks for this post, it helped clarify some of my concerns about the upcoming holidays.
I'm surprised you don't mention testing (PCR or antigen).
What are your views on testing before an event?
What would be a good protocol for testing before a specific event – test on the day of, a day before?
Regarding timing, I think you want to test the day of. The day before is probably fine too. But Delta seemed like it was progressing fast enough that a 1-day lag would lose you a large chunk of the effectiveness.
Regarding mentioning testing in general—I think it helps a little but not enough to matter in most cases. I'm under the impression that PCR tests have a false negative rate of about 50% and antigen tests 70%, which basically translates into risk of .5x and .7x for an event. But if you're home for the holidays, you'd have to keep testing repeatedly if you wanted this risk multiplier to extend, otherwise you might just develop sickness later.
So a $30 test for people only matters if you are going to an event where the average person is losing more than an hour of life, which is maybe 1000+ microcovids. I don't think this comes up super often. However, I guess it's not crazy to take 7 tests over a week to save people several days of life over the holidays. Maybe I should have added this actually. I guess it just seems like it won't be cost-effective for the overcautious people (and will be overused!) and won't be attention-effective for the undercautious people. But probably I should have thought through and been clear about this.
I thought rapid tests were generally considered to have a much lower false negative rate for detecting contagiousness, though they often miss people who are infected but not yet contagious. I forget why I think this, and haven't been following possible updates on this story, but is that different from your impression? (Here's one place I think saying this, for instance: https://www.rapidtests.org/blog/antigen-tests-as-contagiousness-tests) On this story, rapid tests immediately before an event would reduce overall risk by a lot.
Ah, good point.
The main reason I don't automatically make a huge adjustment for this is that it seemed that there were still ~25% false negatives at peak contagiousness by PCR according to some studies. And 40% false negatives a few days later. All sort of things are possible, like that these studies included many asymptomatic and acontagious cases, but having seen anecdotal corroboration of this phenomenon, I'm inclined to think something weird is going on.
But I should give some weight to it though—perhaps 2x less ineffective when compared to contagiousness, so maybe 60% efficacy for rapid and 75% for PCR?
Thank you! The general reasoning makes sense to me.
This Cochrane review finds a false negative for asymptomatic individuals of 42% with antigen tests – which were not self-tests. Is your rate significantly higher because you're thinking of self-administered antigen tests?
In many European countries, you can get antigen self-tests for about $2-4 a piece, this might make a testing scheme more cost-effective.
Ah, stocking up on $2 tests would be awesome! That I would certainly endorse.
My reasoning on antigen false negatives is coming from a few lines of evidence. Perhaps I can share some later. But in short, 1) lots of studies have found much higher than average false negative rates, so results are high-variance/heterogeneous 2) my anecdotal counts of people around me concords with the above studies 3) my prior is fairly high on studies overestimating the efficacy of tests, based on BOTH lab conditions being extra controlled and on scientists being biased toward finding higher efficacy (and this affecting studies in a real way that is hard to control for). Thus my preferred resolution of the mystery between anecdotal efficacy and average study efficacy is that studies overestimate efficacy.
I apologize if this is explained somewhere, but I have a question about this statement;
The key takeaway is that a 1% chance of having COVID, which is about the base rate of COVID in the US, costs older relatives a few days of life if you pass it on to them.
Is that an average loss of life over a large population of people exposed?
So in an oversimplified example, if the only effect of the behavior is 1 in 1000 older relatives would die 1000 days earlier than they would otherwise, the average loss is one day of life?
If that is the meaning, I am not sure I find that helpful. Nobody would notice the loss of one day. But in reality, nobody is losing one day. 999 people lose nothing, and one person loses 3 years. I do not want to be the cause of that, even if the odds are low.
6 months ago I wrote about how 30-year-olds should basically go back to normal and no longer take many COVID precautions.
The holidays make this policy somewhat worse, because you can infect your family members. Further, vaccine efficacy has waned about 3x for most people since it’s been about six months since the vaccine wave. Some people have expressed concern about solstice in particular, which seems to turn out relatively safe. So let’s examine how you might navigate this.
The key takeaway is that a 1% chance of having COVID, which is about the base rate of COVID in the US, costs older relatives a few days of life if you pass it on to them. Thus, you should probably take a few easy actions beforehand to bring this down to a small number of hours. As a non-doctor who is just trying to make their own determination of what's healthy, I recommend being safe enough beforehand that you can then skip masking and distancing with your family and still be relatively safe.
(Sorry that this will be a US-centric post. I’ll try to show enough reasoning the determined reader can extrapolate. I’ve elided many small adjustments and hypotheticals that don’t make big differences for the sake of clarity and sanity, though at the expense of showing all my actual reasoning.)
[ETA: omicron was a much bigger deal than I expected, significantly because I forgot to focus on urban centers where it would advance much faster as johnswentworth points out. Also the "massive immune erosion" possibility turned out to be true, which I think was still agreed to be somewhat unlikely at time of writing. Sorry about this. I think this probably makes EV about 2x worse for people leaving from urban centers (40% omicron at 3-5x the badness for vaccinated people). Probably multiply everything in this post by 2x, and multiply all your standard risk estimates by 3-5x starting around NYE, maybe even more depending on how virulent we find out that omicron is.]
Older family members aren’t much worse off
An average vaccinated 30yo now loses about 6 weeks of expected life from contracting COVID instead of 2 weeks, because of waning vaccine efficacy. Boosted individuals would be back down to about 1 week.
The US currently has about .5% prevalence, which will probably be up to 1% in the next few weeks. (California is at .1%.) If you have a background 1% chance of getting COVID, and you’re traveling for about 1 serial interval—about 10 days, though the calculation is weird here—that means that you lose about 10 hours of expected life from your holiday vacations. This is small enough that you can be fairly comfortable just paying it, but admittedly you probably want to get a booster and may want to take small extra precautions like better masks on flights. Those with comorbidities may want to take extra precautions. I’ll circle back to these later.
You might think that older family members would lose far more expected life, since e.g. 80yos have about 500x the mortality rate. However, two things cause this to unintuitively fail! First, older people have much less life left to live that they are losing out on. Second, long COVID dominates the calculation for everyone under 60, and doesn’t obviously scale strongly with age.
Ultimately I calculate that 80yos lose about 4x the expected life you would, 70yos 2x, and 60 and younger are back down around your harm levels.
To briefly explain the calculation, my previous estimates of harm to older people from definitely catching Delta:
These numbers are calculated by taking early mortality estimates that we’re quite confident in, adjusting upward by 2-2.5x for Delta, and applying a ~10x vaccine effect on mortality that varies some with age. This is about 2x weaker than vaccine efficacy on earlier strains, coming from data showing weakening against Delta.
These are probably no longer correct in the face of data, but I couldn’t find good newer numbers with a bit of searching. I’ll lower the estimates for older people a bit because I think we haven’t seen mortality quite that high—but mostly I will proceed with these because I don’t imagine factors of 2 would change the outcomes overly much.
However, they are all about 3x higher now that vaccine efficacy has waned: maybe .2%, .8%, 2%, 6%. Divide these numbers by 6ish if they’ve had a booster.
This translates into rough expected loss of life from definitely catching COVID after vaccination of:
So 50yos have similar risk as 30yos! At least until we have better data on long COVID some day, since I secretly suspect it scales more with age than they’re letting on, similar to early COVID mortality estimates.
Anyways, these harms aren’t super high—the base rate 1% chance of catching COVID is about 2.5 days of expected life lost for 80yos, down to a half day for 50yos. This is within what many grandparents are willing to pay to see you, so canceling trips is probably not in order unless you have an abnormally high chance of being infectious or your family members have serious comorbidities. But you do want to do some things about it. Which, again, I’ll circle back to.
Omicron interlude
[ETA: Repeated from the intro: omicron was a much bigger deal than I expected, significantly because I forgot to focus on urban centers where it would advance much faster as johnswentworth points out. Also the "massive immune erosion" possibility turned out to be true, which I think was still agreed to be somewhat unlikely at time of writing. Sorry about this. I think this probably makes EV about 2-3x worse for people in urban centers (50% omicron at 3-5x the badness for vaccinated people). Probably multiply everything in this post by 2-3x, and multiply all your standard risk estimates by 3-5x starting around NYE.]
Omicron will make up at least 1% of cases in the US by Dec 31. Which means it could make up substantially more than that. However, in mid-December when you’re traveling and going to solstice, it probably won’t be that high—and even if it’s 5 or 10% at that point, that’s not going to have a major effect on the state of COVID.
However, it can still have a large effect on border closures or mandated behavior like travel quarantines. And if you’re flying places afterwards in mid-January, by that point it might be changing the landscape significantly (and be more likely to cause travel issues). And there are some other worries you could have here like hospital space, but in general since it’s not more deadly, all the previous logic about not putting your life on hold indefinitely still applies.
The only other relevant possibility here is that Omicron makes it to 20% of cases in the US by mid-late-Dec because it has massive immune erosion. (Unlikely but possible by my reading of others’ models.) This could mean that everyone’s risk goes up by 3xish for Omicron cases, but only 20% of cases are Omicron, so EV is 1.4x worse. Even this is not especially bad.
So I will continue to ignore Omicron for this analysis, except as relates to vaccines, since the booster you get now will determine your resistance to the Omicron wave coming in a month and lasting for several more.
Potential Actions
If your trip will cost you 10 hours of life on average and somewhat more for family members, taking small precautions like better masks starts looking pretty attractive. Let’s assume your family’s lost life adds up to 2 days, for the sake of calculation.
A slightly better mask might decrease your risk by a factor of 2x, saving about a day of life for much less than cost. Of course, further factors cause you to asymptote at twice this benefit, but clearly there are some things you want to do. At a safety factor of 20x, you’ve already saved about 48/50 hours available to be saved, so you’d want to have stopped paying cost for more.
That there are two types of calculation you can do: microCOVID calculations, or comparing yourself to some base rate. Comparing yourself to some base rate is the “running away from a bear” principle—just do better than the slower people. For this calculation, the first thing that helps many of you is to be in a bubble that’s safer than the average. Being in California brings you down 5x. Being surrounded by other fastidious people might bring you down a few more x, unless most of your risk is from traveling itself. That might already get you pretty close to the 20x reduction you wanted. I’ll leave you to keep track of your x-factor and ultimate harm.
(Note that a 20x reduction from a 1% baseline corresponds to having about a .05% chance of COVID, or 500 microCOVIDs.)
Vaccine boosters
As you may have noticed, the obvious thing to do is get a booster: it reduces your risk by about 6x, which is massive, for months. It isn’t actually obviously worth it merely for holiday altruistic reasons, but it’s definitely worth it for the next few months—let’s look at why.
Getting an mRNA booster before the holidays will reduce your risk by 6x. This saves almost all the 2 days of expected cost to your family at the holidays. Unfortunately, it seems to make enough people sick for a day or two that arguably it costs about 2 days of life to get the vaccine. So depending on your family size and composition, this is probably around break-even.
But the reason to still get a booster is that Omicron is coming, plus you have to get it eventually to avoid waning immunity anyways. If Omicron will cause you a 10% risk of COVID over the next year, that’s 4 days of life and worth your trip to the doctor. Even if you keep your risk super low, near 1%, you can still take 6x the risk for a number of months, which is probably worth 4 days. If you add these with the altruistic impact, it’s definitely worth it by the numbers.
If you get a booster, what booster?
Murmurings recently have been that people aren’t worried about original antigenic sin—I’ll defer to them now that the issue has been circulated. So Moderna or Pfizer is great, and it’s not that bad to get a third shot that’s the same as your first two—and since all approved US vaccines are unfortunately just identical spike protein, you’ll have to do this anyways.
… Unless you take RaDVaC too!
I didn’t take Radvac in 2020 because I was worried about nebulous interactions with better future vaccines, but now that it’s time for a booster it’s looking substantially better. The bigger kicker is that it probably retains its effectiveness against Omicron, which is all that matters starting January. So if Omicron has major immune erosion then Radvac’s relative efficacy goes up substantially. To pull numbers out of my ass, maybe it gives a 50% efficacy still, equivalent to 2x reduced risk. Maybe several doses gives you up to 3x (though being only mucosal seems like it will only work on incoming boli, and not cause the additional reduction in mortality we’ve seen from other vaccines). 3x, compared to less than 2x from a fourth shot of mRNA vaccine. (Though of course mRNA boosters in the future will increase in efficacy as your immunity wanes over time, but that’s apples and oranges.)
I imagine it’s still not great to layer Radvac on another vaccine in the same week, so my very unofficial recommendation as a not-doctor would be to get a booster in the next week before the holidays really pick up and then, if you’re considering Radvac, move forward on that in early January when your mRNA vaccine has settled and Omicron is just starting to be a major concern.
Another crazy thing you could do about Omicron is become a vaccine tourist and go get Coronavac or Covaxin or Sinopharm, each an inactivated-virus vaccine containing more than just the spike protein. These will work much better against Omicron. Each still offers something like 3-6x protection depending on how you measure (maybe down to 2-4x if you get it after 2 mRNA vaccines, maybe 2-3x after a third booster). This probably isn’t worth it for all but the most immunocompromised 30yos, but for older people it might become attractive if the US drags its feet too long on Omicron boosters, which it sure will.
Masks
(ETA: I do not think it is worth it for most families to wear masks when together—thanks to AnthonyC for pointing out my ambiguity and giving good reasons for not doing so. You'll be spending lots of time together anyways, and it is highly disruptive to normal socializing to not be able to see and hear each other well. I only recommend masks in the week leading up to your family gathering! Sorry this was not clear before.)
Average masks are probably surgical or cloth, depending on your location; KN-95s are a slight upgrade from average, maybe 1.5x, but fitted N-95s are significantly better, probably 4x. P100s would still be great if our society did not treat them like a slap in the face.
Wearing masks for small interactions is probably not most of your risk—most of your risk comes from packed, long events like airport security/gates or parties.
Solstice
Solstice seems to be of low to medium risk since everyone is masked. My scenario on microcovid.org gave about 100 microcovids—about .01% chance, or an hour of risk to your hypothetical family. Even if this is a factor of a few too low, this seems probably worth paying.
The main non-obvious driver here is that the ceiling is tall, which I claim is equivalent to 1 side being open to the outdoors, or 4x reduction on microcovid.org. I claim this partly because of a calculation I've now lost. But a brief illustration is that while the aerosol concentration grows over time, it will also be diffusing upward to dilute it by around 10x from a normal room. Even if this is slow, it clearly reaches 1/4 risk.)
There are also a few small reductions from what you might otherwise use, like people having half the microcovids that the average person does, or the calculater wearing a KN-95, but I endorse these and think skipping them would cause reliable overestimates of risk. However, singing loudly could be worse than the 5x "loud" penalty on microcovid.org. That's the main way I'd expect this to be an underestimate and multiply into some number of hours that actually matters.
The strongest argument for more risk is probably that it's hard to bet on—that if this means there's only a 3% chance that one of eg 300 solstice-goers gets it, that seems low enough that it would be hard to make 30 similar bets and only be wrong once. But I think I trust the numbers here more than how intuitively confident I can be, since early in the pandemic I was always overestimating risk using the bet-methodology. You can see how the numbers might be right if you consider that actually less than 1 of us is likely to have COVID, and if they do have COVID they will be masked and symptomless, and everyone else will be masked, and most of their aerosols will diffuse upwards, etc. But I don't have a ton of clarity on how to resolve this.
Conclusion
I had hoped to do more microcovid.org calculations for random activities like planes, parties, etc, but I didn’t have time. Perhaps I or someone else can add some in the comments later. My guess is the vaccinated are still quite safe for a few weeks before Omicron arrives, and it will turn out that most activities except parties are ignorable.
If I’ve made any major errors in reasoning, let me know.