The CDC recently announced that people who are fully vaccinated (2 doses) do not need to quarantine after exposure to someone with COVID-19. Simultaneously, messages are coming on strong that vaccination doesn’t mean you can stop wearing masks and social distancing. We’re still being told that if you want to see vaccinated grandparents, it’s best to do it outside with masks.
This is confusing. Do vaccines work, or not? Can we return to normal after people are vaccinated? If we can’t, why are we getting vaccines? What is going on?
I’m going to try to break it down below, in three parts.
First: What are we learning about the real-world efficacy of vaccines in terms of both protecting vaccinated people and preventing them from spreading?
Second: a meditation on learning to live with COVID risk.
Third: Some numbers on that grandparent question. (I kind of feel like I have written a lot of this part before, but it remains very confusing so you get it again, slightly differently).
Vaccines: New Science
There are two big questions on the vaccines. One is how well they will protect people from serious illness in the real world (i.e. will we replicate the success of the trials) and the second is whether they protect against infection and transmission.
On the first point, the data coming out from Israel (where they have done the most vaccination and monitoring) is very encouraging. Preliminary data (not yet peer reviewed) is showing 94% reduction in symptomatic COVID with two vaccine doses. We will learn more about this protection in data from the UK and parts of the US, likely in the next month or two. But it’s looking very reassuring that those who are fully vaccinated are much, much, much, much less likely to be seriously ill.
Less certain is the question of infection and transmission. The Phase 3 vaccine trials weren’t designed to detect infection so we are still unsure the extent to which people who are vaccinated can be infected and spread disease, even while not being sick. We are starting to get some picture, though, so I wanted to run through it (and thanks here due to A. Marm Kilpatrick for a great Tweet storm on this).
Question 1: Are vaccinated people less likely to be infected? To learn about this, we need to observe measures of infection (likely, nasal swabs) in vaccinated people. That is: if you vaccinated people and then tracked their nasal swabs every day, alongside a control group unvaccinated people, you’d pick up even asymptomatic infections and be able to tell if they were lower in the vaccine group.
We have a bit of this from the Phase 3 trials. One small piece of data from Moderna is here. At the time of the second dose, there were see 60% fewer asymptomatic infections in the vaccinated than non-vaccinated group. This sample is small, but it is encouraging. (And this is only after the first dose, also).
Preliminary data from Israel also shows a 33% reduction in the chance of infection 2 weeks after the first vaccine dose. Again, this is just one dose, and a shorter time frame than the Moderna data.
Along with our basic understanding of vaccines, this generates some confidence that infection rates will be reduced with the vaccine. Exactly how much remains uncertain, but this question isn’t that difficult to answer and we’ll know more about this in the next few weeks (I think).
Question 2: Are Vaccinated People less likely to spread the virus? This is harder to answer. If people who are vaccinated cannot be infected at all, then the answer is clear. But if they may still have asymptomatic infection, they could spread the virus.
The “ideal” way to answer this would be with detailed contact tracing data on infection spread by vaccinated and unvaccinated people, but this isn’t likely given our infrastructure. Instead, we will rely on indirect evidence. Specifically: we know that those with asymptomatic infection and those with lower viral load are less likely to spread the virus. If the vaccine lowers viral load and reduces symptomatic infection, then that would be informative.
We already know the vaccine reduces symptomatic infection, and asymptomatic infections are a lot less likely (maybe 75 to 80% less likely) to spread. We also are now seeing data from Israel showing evidence of lower viral load in more heavily vaccinated groups.
Putting this all together — reduction in infection rates, reduction in chance of spread given infection based on viral load and symptomatic infection levels — the evidence suggests vaccinated people are a lot less likely to transmit the virus. The tweet thread I mentioned above goes with a 90% reduction all in; we’ll get more precise going forward.
It’s 90%, not 100%. And my sense is that we will not get to 100% reduction in spread from vaccinated people; there will be some remaining risk as there is with nearly all vaccines for all diseases of spreading the virus. Variants play into this, too. There is much that remains uncertain about them (see a whole discussion here from COVID Act Now and COVID-Explained) but the vaccines are likely to be slightly less effective, which may lower both the disease prevention rates and the transmission reductions a bit in the long term.
What I think we can say about the science at this point is: vaccination has a huge impact on severe disease and it reduces transmission by a lot. Neither number is 100%. But both are very, very high.
So what does that mean for behavior?
A Meditation on Risk
A detour.
Back in March and April 2020, there was talk of “flattening the curve”. The idea was that if we locked down, paused everything, we could lower the speed of viral spread, increase hospital capacity and get to a situation where when we opened back up, hospitals wouldn’t be over-whelmed.
Perhaps driven by the tremendous vaccine efficacy, the conversation has (to my reading) shifted away from the “flatten the curve” toward “eliminate the virus.” And with this frame, we have arrived at a place where any risk is too much. As long as there is some risk from vaccinated people to others, or some risk to them, we should maintain all of the same precautions we had before.
It may be feasible to do this in the short term, until vaccination rates are higher and disease rates go down more, but this isn’t feasible in the long term, at least not without tremendous costs to mental health, economic health and physical health. We can’t eliminate the virus completely, at least not in any reasonable time frame (smallpox took about 100 years). The next phase of this pandemic is going to be learning to live more normally with some risk of COVID-19. We do this every day with other risks, but it’s going to be hard to make this mental shift.
This doesn’t mean being cavalier about COVID, but it is going to mean consciously shifting ourselves into weighing this risk against other risks and benefits. This is especially true as we vaccinate more and more high risk people so the serious diseases risks diminish substantially.
Behavior
Back to the data and evaluating risks of some of the things you’d (possibly) like to do.
This is a parenting newsletter, so the main question I’m getting these days is: the grandparents are vaccinated, but we are not and the kids are not. Can we see them indoor without masks?
The risk evaluation frame here is the same as it was before, but with different numbers. If you get together, there is a risk that someone will arrive with the virus to your gathering. There is a risk they will spread it. Then there is a risk that someone will get seriously ill. Exactly how to combine these varies with your setting.
Example: vaccinated grandparents, parents unvaccinated but very, very isolated with 8 month old baby. The concern is grandparents bring disease on their visit.
- The US currently reports an average of 30 cases per 100,000 people per day. If we assume 7 days of infectiousness and a detection rate of 50%, this means the randomly chosen person has a viral risk of 0.42%.
- Chance of transmission. Household transmission risk from an infected person is 10 to 15%, let’s average to 12%.
- Now let’s assume a 90% reduction in risk of transmission from an infected person, combining the lower infection rate and the lower transmission rate.
- Risk of transmission from grandparent to someone else is now 0.42% X 12% X 10% = 0.0050% (about 1 in 20,000).
- Risk of serious illness for healthy adults & child are very, very small (especially for children). If this was 1%, that would put the risk of hospitalization from this activity at 1 in 2,000,000.
This is a quite low risk situation.
Example: Vaccinated grandparents, parents unvaccinated and work outside the home. Kids go to in-person school. Concern: grandparents might still get sick.
- Same prevalence estimates as above. Let’s assume independence, so between two parents and two kids there is about a 2% chance someone has COVID. (Note that these risks will increase if your area is still higher prevalence).
- Risk of transmission in the household estimated at 12%, but let’s say if grandparents are vaccinated their risk of any infection is lower by 50%. So the risk of transmitting virus to grandparents is 2% X 12% X 50% = 0.12% (about 1.2 in 1000).
- Risk of any symptomatic illness given vaccination efficacy is 5%; risk of serious illness is effectively 0%.
- Overall risk of symptomatic illness is 0.12% x 5% = 0.006% or about 1 in 16,000.
I realize there’s a lot here, but I would encourage you to think about these numbers as you contemplate this. You can look at COVID levels in your own area to think about the risk of bringing the virus in, and use your own sense of the reductions in risk. You may decide a 1 in 20,000 risk of transmission to a child is too high, but I think a reasonable person might also think this is acceptable, especially given the low risk of serious illness for most kids and healthy adults.
What you shouldn’t do — and I think this is really the key to the continued caution in messaging — is get together with all your vaccinated friends in a damp, hot basement and have a singing party. If you have close contact with 1000 people even if they are all vaccinated there is a reasonable chance someone’s carrying some virus around, and they could then carry it out to the rest of us who are waiting on vaccines. In another few months, when cases are lower, this will not be true anymore and we’ll be able to do our hot basement singing.
Part of learning to live with COVID-19 in the long haul is going to be making these kind of nuanced choices, and thinking through the numbers. Now may be the time to start.
Today’s bottom line:
- The data is increasingly encouraging on efficacy of vaccines and on their role in reducing viral spread.
- We are not going to get to zero COVID any time soon, and we are going to need to learn to adapt to that while eventually taking steps to normalcy
- It might be safe to see grandparents if they are vaccinated, so do some multiplication.
