PhD in math. MIRI Summer Fellow in 2016. Worked as a professor for a while, now I run my family’s business.
NormanPerlmutter(Norman Perlmutter)
Very nicely researched and explained analysis. Contains several points that I had not thought about much before if at all, including the analysis of false negatives in this particular context.
I’d be interested to hear more thoughts on your mechanistic explanation from commenters with more expertise in immunology.
Also interested to hear how you and other commenters are planning to change your own Covid risk-taking behavior after being vaccinated. I am coming up on two weeks after my second Pfizer shot and debating that for myself.
It seems the caveat regarding selection effects, that vaccinated people might be prone to more health-seeking behavior and might risk compensate, could swamp many of the other careful arguments. Is there any way we could get a rough estimate on these selection effects?
The variants caveat is scary, but at least we can monitor variants over time and try to adjust our risk behavior accordingly.
I think the points about non-obvious costs are very important to consider and nicely evaluated. Nonetheless, I think your conclusions are a bit too strong. It’s possible that I’m being too conservative and not giving adequate weight to the obvious and non-obvious costs of continued restrictions.
I think that in light of the risks of long covid from mild infections (small sample size but I haven’t seen a larger study contradicting it) and the ongoing risk of vaccine-resistant mutations, saying that young healthy people are protected from overall risk by a factor of 1000x through vaccination is overly optimistic. I’d say that a 100x reduction harm might be a reasonable estimate. This thread by Ruby which you linked to as well has lots of good analysis of vaccine efficacy for healthy young people. If I am interpreting the note at the top correctly, due to the risks of vaccine resistance, he has adjusted his harm reduction estimate from vaccination downward to 10x-100x reduction, from his initial estimate of 100x-1000x reduction. He also mentions long covid in his caveats towards the bottom. It’s not clear to me why you are still saying 1000x in spite of these factors, and I’d be very interested in understanding how and why your assessment differs from Ruby’s in this regard.
It’s also worth considering the possibility of a tipping point. If we hold off a few months longer on full reopening, we might get enough hesitant people and people under age 16 vaccinated to avoid the rise of a highly vaccine-resistant strain in the US.
Furthermore, there’s a big difference between returning mostly to normal and fully to normal. It’s possible that the most extreme potential superspreader events (examples: an indoor rock concert with thousands of people pressed shoulder to shoulder and yelling over the music; a large group indoor choir singing event; a crowded bar indoor bar in Manhattan; Burning Man) are much more risky with regard to covid spread than all the other stuff combined. And at the same time these sorts of crowded and touchy-feely mass gatherings are a very important part of the human experience, and disallowing them has significant costs.
I am now more than 2 weeks past my second Pfizer shot. Considering the factors above, I am not planning to return completely to my 2019 way of life right now, but I am moving substantially in that direction while monitoring covid prevalence, vaccination rates, and new variants for the next month or more.
My current lifestyle that I’ve experimented with for the last several days includes eating indoors at restaurants (but being mindful of how long I spend at indoor restaurants and how often I go); hugging and hanging out indoors unmasked in small groups with some consenting friends and family who are at least one of young/healthy, vaccinated, or have recovered from covid in the past. I’m probably interested in attending large events outdoors such as concerts so long as people are mostly masked and are not crammed shoulder-to-shoulder. And also small indoor parties with an adequate combination of social distancing, ventilation, and vaccine coverage.
But so far I’m avoiding such activities as going to especially crowded restaurants/bars where I would be spending a lot of time within three feet of many unmasked people, riding on a crowded subway at rush hour, attending crowded indoor concerts, attending large indoor conventions, playing Spin the Bottle, or getting anywhere near other people who are sick and who I assign a substantial probability of having covid. I think those types of activities I will continue to wait on as I monitor the prevalence of covid and spread of variants.
I think that for the next few months, we should maintain legal restrictions on very large gatherings such as the potential superspreader type events that I discuss near the top of this post, either forbidding these sorts of gatherings or requiring most or all attendees to be masked or vaccinated or test negative. Setting guidelines that we will return further to normal once enough people are vaccinated, or allowing vaccinated people different privileges from unvaccinated, may coerce people to get vaccinated, to the benefit society. This will give us a chance at actually hitting something resembling herd immunity through vaccination and preventing a new vaccine-resistant strain from spreading while the overall vaccination rate is lower. I agree that it’s problematic to have unenforced regulations. I think the optimal solution to this, both with covid and other unrelated areas of law (though it’s much easier said than done) is to reduce the number of regulations and more strenuously enforce the most critical regulations.
I see a substantial possibility of near-eradication. Not like we eradicated smallpox, but maybe almost as well as we’ve done with measles? It’s not clear yet what percentage immunity we need need to achieve herd immunity. But given the figures from the past month or so, there’s a decent chance we’re heading in that direction. Even if we can achieve herd immunity, that may not be a good enough argument to ban superspreader events for now—we would achieve herd immunity regardless when we get R<1 with respect to normal activity. But there’s the further factor that by continuing to reduce transmission while more people are vaccinated, we reduce the chance of variants.
There’s also the possibility that we can eventually achieve near-eradication through a combination of immunity and good contact tracing, but we may only have few enough cases to do the contact tracing if we can get the case count down to a manageable level through continued social distancing for a while first.
Do those scenarios seem plausible to you? Why do you believe that the options are almost binary?
I think I see a larger possible downside than you do regarding variants. It seems possible that we could get a variant against which our current vaccines are less than 50% effective, and if we can reduce the chance of that happening through continued restrictions, I think it’s worthwhile.
Distributing more vaccines to the rest of the world would be a bigger intervention in that regard, but perhaps through continued travel restrictions combined with some vaccine protection, we can prevent especially bad variants from spreading worldwide the way the original pandemic did.
I don’t understand why you feel children clearly shouldn’t wear masks. They have a larger risk of spreading Covid than vaccinated adults.
How does take-out increase serving and cleanup costs? In my experience, take-out drastically reduces cleanup cost vs cooking. You don’t need to clean the baking dish or pan, and those are often hard to clean, whereas plates can just be thrown in the dishwasher, or you can even eat out of the disposable container provided by the restaurant and not have to wash any plates.
This is a bad analogy. A DUI or speeding could be a one-time thing. Not getting vaccinated is a continuous decision. All you have to do to reverse it is make the right choice once (or twice if you get Moderna or Pfizer).
Also, drunk driving and speeding are not contagious. A drunk driver can hurt or kill anybody they crash into, but that doesn’t make those people go on to become drunk drivers as well.
I’d be interested to see a good estimate and analysis of this multiplier. In places and times when r>1 the multiplier would be quite large indeed, whereas if r<1 then the mutiplier would be more modest. Some sort of time analysis is needed as to how long r stays greater than 1. (r here is the average number of new people infected by a person with covid.)
If I understand correctly, Zvi’s idea is that vaccine protection against infection has likely gone down, but vaccine protection against severe infection has held nearly constant, so that the vast majority of additional infections among vaccinated people will be non-severe.
There is an important practical consideration that is being left out here. Attempting to completely devote one’s life to these causes in the way described would not necessarily be effective. The best way to devote your life to a cause is not necessarily in a super-fervent way, because that’s not how humans work. We need certain types of motivation, we have physical and emotional needs, we suffer from burnout if we work too hard. So if you believe that astronomical suffering risks are the most important issue to work on, then by all means work on them, but don’t overwork yourself, it will likely result in anxiety, burnout, and unhappiness, which won’t actually help you to work more effectively. Work on these problems, make them your life’s work if you see fit to do so, but do so in the context of an overall good life.
The 97% was in case there was no other strain that comes in. So maybe the 96% is even taking into account another new strain? Or maybe it was just a typo.
Are you saying that you agree with William Eden’s claim that vaccination does not substantially prevent the spread of covid? Or is that one of the things that you would “quibble” with him on? That point seems much larger than a quibble to me, it’s a key point that’s being debated currently about public health policy. My own understanding is that while the vaccines are of course not perfect at preventing spread they do prevent spread to a significant degree, and therefore vaccination is indeed a public health issue not just an individual decision.
I have been thinking about this topic a lot on my own and with friends before finding this post and was excited to see a post so related to my recent thoughts. One idea that came up in a recent discussion with a friend was that the pitfalls of the reasonable good faith effort in connection with common communication norms, especially if somebody reveals a secret accidentally and is feeling vulnerable and then asks you to keep it secret. In that case, if you say, “I’ll make a good faith effort to keep it but I can’t promise” it may be interpreted as “I don’t care about the privacy of your secret.” What the person is actually wanting to hear is something more like “Don’t worry, your secret is safe with me.” There is a social expectation of some degree of fallibility, and depending on the social context, pointing to this fallibility may overemphasize it and be interpreted differently from how it is intended. All this is very context-dependent.
Another such case is if sharing something would embarrass somebody. They might be embarrassed in spite of others not acting adversarial towards them.
In this case, sharing it with people who don’t know her and will likely never encounter her will do minimal harm, so you might suggest that as an exception to the secret keeping.
Could you give an example of exploring un-endorsed emotional reactions? How is this related to having deeply held values?
Hmm, suppose an adult had urinary problems and wetted their bed regularly. Which category would you say that fits into? Or somebody whose parents had named them something that they didn’t like and they changed their name and didn’t want others to know their original given name due to aesthetic preferences and social implications of character traits related to that name?
There would be some social harm in sharing this either of these, but would it necessarily be adversarial? Even if others were aligned with the person with the secret, they couldn’t help but look at them a bit different knowing the secret.
Yes, I agree that this nonstandard definition is a crux for this disagreement. Good analysis.
I have not used microcovid much because I am not confident in its predictions and modeling assumptions, or I don’t feel they are clearly enough defined to make the tool useful. The change that would be valuable to me (which I have difficultly operationalizing) would be if Microcovid were improved such that I could be much more confident in its modeling assumptions and could use it without having to try to make lots of guesses about which scenarios are well modeled. Maybe it would be sufficient just to explain which types of assumptions make for robust modeling outcomes (maybe this is already somewhere in the documentation). Otherwise, I will continue not to use it.
I think that in general maybe Microvid works well in low-risk situations but breaks down in high-risk situations.
Prior to the recent Omicron surge and post-vaccination, I tended to estimate my covid risks by looking at reported covid case rates in my area, and assuming that as a fully vaccinated person, my risk of getting covid was likely lower than the average person in my area (Ohio), many of whom are not vaccinated, even if I went to restaurants and bars at about the same rate as I did in 2019.
Some examples of my confusion about microcovid’s modeling assumptions . . .
Looking at the risk profiles for hypothetical other people, for fully vaccinated people in my state (Ohio):
Average person in your area: 11,000 microcovids
Has 4 close contacts whose risk profile you don’t know, in an otherwise closed pod: 6,400
Has 10 close contacts whose risk profile you don’t know, in an otherwise closed pod: 19,000What is the definition of a close contact here? Does this mean somebody who they live with or something like that or just somebody who they regularly hang out with closer than 6 feet? It seems to me that the average person in my area (the mean-risk person since this mean is largely determined by the riskiest people, maybe not the median-risk person) has more or less gone back to normal and would have more than 10 close contacts if you’re counting the people they live with, work with, or hang out with regularly. Or at least closer to 10 close contacts than 4.
Microcovid currently predicts that a fully mRNA-vaccinated person with a cloth mask who spends 8 hours in a bar acquires 380,000 microCovids (38% chance of getting covid), assuming that the average person in the bar went to a bar within the last 10 days. (reduced to 240,000 if the average person within 15 feet 10+ feet away rather than 6+ which seems more likely. But why doesn’t the model care at all about people 20 feet away?) (As a side note, the default assumption was that most of the people in the bar had the risk of “an average person in your area” which doesn’t seem right for a typical bar.)
And furthermore, the risk after 8 hours is equal to the risk after 4 hours, huh? I’d think that in 8 hours more people would be coming in and out, you’d be exposed to more possible infected people.
If this assumption were correct, then over the next week we’d see basically all the bartenders and bar workers here in Ohio getting covid simultaneously. (Or does it just max out at 4 hours so that the covid risk of working at a bar for a week is the same as for 4 hours? That just doesn’t make sense.) Even if half or so of these cases were asymptomatic, it would probably be enough that many of the bars would shut down. Seems unlikely, but I guess we could see if it happens.
Likely one of the missing parameters here is the protection from recent infection. I could imagine that the majority of bar workers who haven’t had covid in the last 3 months will get it over the next month or so, which wouldn’t be enough to shut down many bars.
A one-night stand with somebody who has covid (modeled as kissing for 10 hours) my risk is only 100,000 microcovids. It seems bizarre to me that this risk would be about 1⁄2 to 1⁄4 the risk of going to a bar for 2 hours with 15 random people who had been in bars in the last 10 days. Maybe my intuitions are just way off. I suppose at the bar there could be multiple people near me with covid, and one of them might be much more infectious than the average person with covid. But I would think that all of them together wouldn’t transmit as many viral particles to me as a single person with covid who I am kissing for 10 hours.
This is a very helpful analysis. I was independently undertaking a similar analysis, and it’s nice to have this for comparison. I hadn’t thought to exclude pedestrians, pedecyclists, and other non-occupants, nor of excluding single-vehicle crashes.
I think a some important pieces are missing from this analysis, as follows.
1) The final number, 548, is the number of miles that I must drive to accrue one micromort for all passenger vehicle occupants. But I am more interested in how many miles I have to drive to accrue one micromort for myself. The average (mean) car has 1.5 occupants. Assume that the average (mean) crash of interest involves two cars, and deaths are distributed roughly equally between each person involved in a crash, total of 3 people on average. So you would need to drive or ride 3 times as many miles as claimed by Josh’s analysis to accrue one micromort for yourself. A micromort would accrue to any other passengers in your vehicle as well.
(If excluding all accidents with one car, then the assumption of the average accident including only 2 cars is actually off, as some accidents will include more than 2 cars.)
https://css.umich.edu/factsheets/personal-transportation-factsheet (source for average passengers per vehicle)
https://www-fars.nhtsa.dot.gov/Main/index.aspx—starting point for further honing this factor
2) If a passenger car gets in a crash with a larger vehicle such as a semi truck or a bus, likely the occupant of the passenger car will be injured much worse than the occupants of the larger vehicle. (I didn’t look this up but it seems reasonable.) This would bias the results in the opposite direction from excluding motorcycles and pedestrians. More generally, we should be aware of cherry-picking which risk factors we remove from the analysis.
3) Deaths per 100M motor vehicle miles driven stayed approximately constant in a range of 1.0-1.2 per 100 million miles from 2009-2019 but went up dramatically in 2020 to 1.37 deaths per 100M miles (73 miles per micromort) and has stayed at a higher level or even increased in 2021 based on preliminary data.
https://www.usnews.com/news/health-news/articles/2021-06-04/traffic-deaths-increased-in-2020-despite-fewer-people-on-roads-during-pandemic
https://www.nhtsa.gov/press-releases/usdot-releases-new-data-showing-road-fatalities-spiked-first-half-2021
4) Taking out 50% of deaths due to alcohol impairment, so as to account only for the other driver being drunk, seems fine. But I’m less confident that the average lesswrong reader does not drive distracted or drowsy. This seems like an area where we should be careful about being too confident due to self enhancement bias. On the other hand, there are other ways to drive more safely than the average driver. One of the most important of these is maintaining a safe following distance.5) Depending on the context, we might be more interested in the micromorts per mile of interstate highway type driving in particular. This risk is about half the mortality risk per mile as compared with all driving.
http://www.bast.de/EN/Publications/Media/Unfallkarten-international-englisch.pdf?__blob=publicationFile (assuming “motorways” has a similar meaning to interstate-type highways)This figure is relevant, for instance, in answering a question such as “What is my risk of highway death from taking a 1000-mile road trip, as compared with my risk of death from covid over the course of the same vacation?”
Josh’s analysis mutiplies the initial figure of 91 miles per micromort by a factor of roughly 6.
Jost includes factors of approximately
A) 1.51 for excluding motorcycles, bikes, and pedestrians
B) 1.78 for excluding single-car crashes
C) 1.86 for seatbelt wearing
D) 1.24 for drunk, distracted, and drowsy driving. This is dominated by drunk drivingI would reduce factor A, as I think it’s balanced by cases where a passenger car hits a truck. More research on this could be helpful. For now, I’d say to get rid of half of it, make it 1.25
I would get rid of factor B, but replace it with a discretionary safe driving factor (see below)
I would keep factor C
I would keep most of factor D, maybe reduce it to 1.2.Add (or rather multiply) in the following factors
E) Factor of 3 to account for only micromorts accruing to one traveler, not to co-passengers or occupants of other vehicles. This is based on an average of 3 people per crash, a figure that would benefit from further research.
F) Factor to account for being a safer than average driver (beyond the effect of wearing a seatbelt and not driving drunk). I think that it might be reasonable for especially safe drivers to use a factor of 2 or 3 here (which would be comparable to eliminating single-car crashes plus more). A factor of 10 seems like it would be too much. For myself as driver, I think I would be conservative and keep it at 1, absent further analysis. I am safer than average in some of my driving practices but less safe than average in other practices and skills.
One way to get a start at estimating this factor might be to look at your auto insurance rate and compare it to the average rate for comparable coverage of a comparable vehicle in your state. Insurance companies are in the business of rating risk after all.
G) Factor of 2 if considering only interstate highway driving (or other divided limited-access highways).
I would start with 73 miles per micromort, using the rate from 2020. Overall this gives me
73*1.25*1.86*1.2*3 = 611611 miles per micromort for each vehicle occupant, or 1222 miles per micromort considering only interstate highway driving, to be adjusted further by a factor of up to 3 depending on one’s beliefs about one’s relative safety as a driver.
Seems to be of great practical significance to me. If there’s a decent chance that I could return completely to unlimited degrees of interpersonal close contact and have only a 1⁄20 − 1⁄100 chance of getting symptomatic Covid with unlimited amounts of exposure, even if I would get covid many times over while unvaccinated, I’d be quite tempted to do it. If returning to that level of exposure would mean that I’d almost inevitably get Covid eventually, I’d be much more likely to play it safe for at least a few months more and see where things go with infection rates and new variants.