The CDC is asking vaccinated people to continue to mask and social distance:

Q: Do I need to wear a mask and avoid close contact with others if I have gotten 2 doses of the vaccine?

A: Yes. Not enough information is currently available to say if or when CDC will stop recommending that people wear masks and avoid close contact with others to help prevent the spread of the virus that causes COVID-19. ... We also don't yet know whether getting a COVID-19 vaccine will prevent you from spreading the virus that causes COVID-19 to other people, even if you don't get sick yourself.

This does not make sense. They're saying we should act as if vaccinated people are no less likely to infect others, but if that were the case then vaccines would not be helping us get to herd immunity.

In the absence of specific studies showing whether these particular covid-19 vaccines reduce infectiousness, we need to make the best predictions we can based on the information we do have. A good starting point is the measured efficacy of the vaccine, which for the Pfizer and Moderna vaccines, the only ones I'll be talking about in this post, is ~95% (1/20th) two weeks after the second shot. [EDIT: this guess should likely be adjusted down, given that we already know asymptomatic spread is common.] This predicts that an interaction between two vaccinated people is ~0.25% (~1/400th) as likely to transmit covid as if they were both unvaccinated.

At this point, to figure out what activities makes sense, it's useful to bring in some numbers. We can use the idea of a "microcovid" to represent a one in a million chance of getting coronavirus:

An activity that's 20,000 microcovids is very unsafe, as you have a 2% risk of getting covid every time you do it. An activity that's 20 microcovids is relatively safe, as you could do it every week for a year and still have only about a 0.1% chance of getting covid.

Until recently, our house's typical weekly covid risk, given current covid rates in the Boston area, looked like:

  • One housemate working in person three [EDIT: two] days a week in a well-fit kn95: ~600 microcovids
  • Each of seven housemates spending about half an hour a day outside with other people around: ~100 microcovids

Comparing microcovid for individuals and groups is awkward, since you need to take into account that if one of us gets it the others might as well. A risk of 700 microcovids total, for a seven person house like ours is ~1.5% yearly individual risk.

The housemate who works in person got their first shot recently, and two weeks after their second shot I estimate we'll be at:

  • One vaccinated housemate working in person three days a week in a well-fit kn95: ~30 microcovids (1/20th of 600)
  • Each of six unvaccinated housemates spending about half an hour a day outside with other people around: ~86 microcovids.

This is much better, at a ~0.24% yearly risk.

Then imagine we agree to invite a random friend to dinner, inside. (Not something we've done since before the pandemic.) And say they're generally careful, at a 1% individual yearly risk. The dinner risks our household ~200 microcovids; too risky. If instead our guest were vaccinated, that would be ~10 microcovids; not bad.

We could do this every week and our yearly risk would rise from ~0.23% to ~0.26%. As the rest of us start getting vaccinated, the risk continues to fall.

If you are otherwise careful, limited socializing as/with a vaccinated person is likely a small enough risk that it's a very reasonable thing to spend your risk budget on.

Comment via: facebook

New Comment
12 comments, sorted by Click to highlight new comments since: Today at 4:01 PM

A good starting point is the measured efficacy of the vaccine, which for the Pfizer and Moderna vaccines, the only ones I'll be talking about in this post, is ~95% (1/20th) two weeks after the second shot. This predicts that an interaction between two vaccinated people is ~0.25% (~1/400th) as likely to transmit covid as if they were both unvaccinated. 

Efficacy at presenting symptoms is not the same thing as efficacy at preventing infection.

The FDA decided to allow vaccines to get approved without measuring infectiousness. That was a good decision because it allowed approving them faster (and not taking up important PCR capacity) but it comes at the cost of not knowing how protective it is against spreading the virus.

Given that nonsymptomatic spread is a thing and the way the underlying biology works it makes sense to expect less then 95% protection but significantly more then 0% protection for spreading the virus.

In recent news, the Oxford vaccine seems to have a 67% effect on transmission. Even if you expect the Biontech/Moderna vaccines to be better, eye-balling them for 75-80% would make more sense then 95%.

Also, all of these numbers are presenting efficacy in preventing the particular strains that were circulating in the times and places of the corresponding studies.  I’d personally discount transmission prevention a bit further due to uncertainty about whether these numbers fully reflect the virus strains that are circulating in my local community — our only data for when new strains spread to more communities are lagging indicators, and the effectiveness numbers for some newer strains are lower or less certain than the numbers we’re using in this discussion.  (But then I’m also quite far to the cautious end of the spectrum.)

| This does not make sense.

The advice makes sense from a medicalized "do no harm" perspective. Given the baseline behavior of extreme behavior modifications to prevent infection, relaxing those constraints poses a risk. Therefore the only acceptable advice to give is "continue behaving just as cautiously even after getting the vaccine."

This is clearly stupid, but medical institutions are not required to not be stupid; they are required to obey their hippocratic oath.

I think this might be missing a dimension of fairness considerations:

  1. People who were least at risk (broadly: the young) from COVID-19 were asked to give up socializing & income during the lockdowns for the people who are most at risk.
  2. People who are most at risk (broadly: the old) from COVID-19 get vaccinated first.
  3. Giving people who get vaccinated early an advantage would signal people who were least at risk that they incurred two costs (lockdown & late vaccination) and received no tangible benefits, which might damage future willingness to cooperate.

Oh and the other hand, once the people more at risk have had the opportunity to get vaccinated, then even if the vaccines didn't appreciably prevent transmission you should still be able to have public health messaging that says people who are least at risk can resume some amount of socializing/working/etc.

I think ojno has a point. Furthermore, to the best of my knowledge the protection from the vaccines takes a bit of time (10 days? 14 days?) to kick in after the vaccination. Arguably "proceed with the same caution as before" is a better message than "go nuts, dance and hug and visit all your friends" in this period, and for simplicity's sake this has become the default message.

Who am I kidding, this is of course because we don't want vaccination to be unfair. If you get social benefits from being vaccinated (by not having to abide by some of the restrictions) then the prioritisation discussion would be even fiercer than it is now. Plus, the more Sacrifices to the Gods you publicly support (h/t Svi) the more of a Serious Person you are, which the CDC tries very hard to be.

Furthermore, to the best of my knowledge the protection from the vaccines takes a bit of time (10 days? 14 days?) to kick in after the vaccination.

FDA is using 7d for Pfizer (https://www.fda.gov/media/144245/download) and 14d for Moderna (https://www.fda.gov/media/144434/download). I'm using "two weeks after the second shot" in the post to be on the safe side.

I meant after the first shot, sorry for the confusion.

10-14d after the first shot you're still not very protected. Protected enough that I think a strategy of focusing on first doses probably makes more sense, but not protected enough for most of this post to apply.

I have yet to see anything convincing me that a vaccinated person is unlikely to give me COVID, even 2 weeks after dose 2. Especially because right now, the vaccinated people are the ones most likely to be exposed to COVID (e.g. nurses, grandparents living in group facilities). I would not be comfortable being unvaccinated and unmasked around a vaccinated unmasked person for this reason. I am eagerly awaiting information to change my mind on this, but I have not seen it yet.

If I were also vaccinated, I would feel reasonably good that even if I got COVID, I would make it through OK. There's still some uncertainty about long term effects from mild cases that worry me, but I haven't seen anything on long term effects from asymptomatic cases. So, at that point I would probably be OK being unmasked around other vaccinated people who were unmasked, and possibly around unvaccinated lower-risk people who were unmasked. 

I find it to make sense, at least for public places like, say, shops, for a very simple reason: how do you tell if a random person has been vaccinated and is not merely lying to you that they have been vaccinated, as they have an incentive to do if they don't like wearing masks or whatever?

What is “it” when you say “it” makes sense?

I’ll probably wear masks in e.g. the grocery store even after being vaccinated, because I expect the law to lag the reason for the law, and because wearing a mask for ~half an hour is less costly than the risk of conflict if someone tries to make me wear a mask.

As for whether public places should enforce mask-wearing on people who claim to have been vaccinated, well, enforcement is pretty lax as it is, and I expect it’ll just get laxer as vaccination spreads. As it should.