Understanding why the CDC Gives the Guidelines that it Does
by Malte Skarupke
The CDC continues to issue very puzzling guidelines about the coronavirus. The most recent strange recommendation is that even if you already had covid, you should still get the vaccine. Which… makes no sense?
Let’s assume our goal is to reduce the number of deaths. Currently in the US, something like 26 million people have tested positive for SARS-CoV2, and roughly 440,000 people have died of COVID19. I’m using these terms to make a distinction: Not everyone who gets the virus will get sick, and not everyone who gets sick will die. Let’s be generous and assume that 100 million people got the virus so far, then mortality rate is 0.44%.
What do those same numbers look like for people who already had the virus before? Obviously the first number, the chance of getting the virus, should be roughly the same. But your body has fought this virus (or a very similar virus) before, so your chance of getting sick is much lower. For a long time it was uncertain whether you can get sick a second time at all, but now there are enough confirmed cases. How many of those have died? At least one, who was on chemotherapy at the time. Let’s estimate the number at 10, which would get us to a mortality rate of 10/26,000,000 = 0.000038%.
Should we really be giving people with that second risk the vaccine when not everyone in the first group has had a vaccine yet? Either the CDC worked with different numbers that are ten thousand times bigger, or their goal is not to reduce the number of deaths. What could that other goal be?
To figure out the other goal, lets look at some other puzzling guidance. Early on the CDC didn’t recommend wearing face masks. This really messed my mental models up and made me very much underestimate the severity of the pandemic, because I thought the CDC could be trusted. I was still going out in New York City in March, because I was convinced that the virus doesn’t spread through the air. I just thought that they wouldn’t say things like that without a good reason. If all of medical experience with similar viruses has taught us that they travel through the air, then surely they wouldn’t just toss all of that evidence out without a good reason, right? If they say that there is no evidence, then surely that must mean that they carefully looked at all the existing evidence and had a good reason for why it doesn’t apply here, right? But no, they just tossed out evidence that they didn’t like, and then said there is no evidence. I am imagining the conversation went something like this:
Fresh hire: “I looked at our experience with similar viruses, and all the doctors who treated SARS-CoV wore face masks because that virus spread through the air, and the face masks were very important for stopping the spread of that disease.”
Old hand: “Did you say SARS-CoV? No that won’t do. We’re looking for data on a virus called SARS-CoV2. We’ll just say that there is no evidence.”
What’s more illuminating for understanding the motivation is what they recommended instead: Wash your hands regularly, don’t shake hands, stay home if you’re sick. Did they have any evidence that these helped? No! At least they didn’t have any stronger evidence than they had against mask wearing. So why, given these different potential recommendations, did they decide to go with some against others?
Don’t Upset People
I think the unifying explanation is that they want to avoid upsetting people. Nobody is going to be upset if you recommend that they wash hands and stay home if they’re sick. But mask wearing was very unusual at the time. Nobody wore masks, and if anything it was associated with thugs or problematic people at protests. It would have been crazy to suggest that everyone should walk around like that in public. Lots of people would have gotten very upset at the CDC for issuing that guideline.
Instead a bunch of people died because we had stupid policies like New York bus drivers not being allowed to wear face masks, because it wasn’t part of the uniform, and the CDC said it didn’t help, and surely the CDC wouldn’t say that without a very strong reason.
The new guideline of giving the vaccine to people who already had the virus can be explained for the same reason: If you say that people who already had the virus have to go the end of the line, those people would be upset with you. They would find the above-linked news article about a woman who got the virus a second time and died.
The same reasoning also explains a lot of the other weirdness around vaccines, like how we were more worried about someone potentially skipping the line than about tossing out vaccines. It also explains why New York changed their vaccine rollout after that last-linked news story: People were getting upset about us tossing out vaccines, and we have to avoid people getting upset.
Overall I think the strategy has been successful. The CDC has minimized the number of people who are upset at it. At a certain point last spring, lots of people were getting upset that the CDC still hadn’t recommended face masks, so they changed their policy. Sure, then there were the expected large amount of people who were mad at it for issuing the new guidelines, but the CDC could see the trend lines and I think they changed recommendation at roughly the right time to minimize the total number of people who were upset at it.
I haven’t yet heard anyone complaining about the recommendation to give vaccines to people who already had the virus, which is why that recommendation still stands. Apparently at some point the CDC said that smoking puts you in a higher risk group, and they recommended that smokers should get the vaccine first. But then last week I saw plenty of people being upset at that, and now I can’t find that guideline on the CDC website any more. Either they removed that recommendation, or it’s well-hidden.
So… Is this a good strategy? I started the blog post with the calculation that you would do if you wanted to reduce the number of deaths. And it appears that the CDC instead wants to reduce the number of people who are upset at it. Which is a very politician-minded thing to do. But there actually is a strong correlation between these two goals, so I am actually less mad at the CDC now than I was before writing this blog post. I understand where they’re coming from, and I think in general their goal will lead them in roughly the right direction. Could they have better goals? Absolutely, but maybe this is OK? I said above that nobody gets upset at the CDC when they recommend washing hands. There was a time when that would have been controversial. Surely for the next pandemic, nobody will get upset at the CDC if they recommend wearing face masks. So… progress?
If I had known this in March of last year, I would have been much better at using the CDC guidelines, and I would have understood that them not recommending masks doesn’t say anything about the evidence. I think there are a lot of people for whom this was obvious at the time, but I just couldn’t understand what other motivation the CDC might have had. Maybe I was naive, or maybe I just couldn’t see this simple motivation. Now I do.
If you believe that this is true, please make it known in public that you’re upset at the CDC for recommending that people who already had the virus should still get the vaccine. I mean I’m not opposed to them getting the vaccine eventually, I just think for as long as the vaccine is a limited resource, it shouldn’t go to people who have a very, very low risk from this disease.
Probably Dance 
You might be on to something here. Seeing that social trust and trust in government/institutions are highly negatively correlated with deaths due to COVID (https://wzb-ipi.github.io/corona/), it makes sense to not irk the public too much.
Communication is hard. I’m living in Germany and while the situation seems less worse than in the US, I’ve experienced the same. When it first started they too stated that masks will not help and should be used by medical professionals only, then later moved on to making community masks mandatory when visiting businesses or using the public transport. And now, nearly a year later we went from community masks to medical masks as a mandatory requirement (“to protect against mutations”).
While I think that this kind of communication is actually hurtful, because it gives opponents some kind of attack surface (“You said last year that they don’t work, then that community masks work, and now you say that they too don’t work and we need medical masks!”), it might still be an OK-strategy.
When this all started and masks were deemed useless, we had just found out about the limits of (most often chinas-) production limit. Suddenly everyone wanted face masks, while china was too in need of them. We’ve seen situations were shipments of medical equiment had been rerouted on the airports multiple times because of strong demand from people waving with cash in their hands. (Like in april, when 200k medical masks were rerouted in Bangkok from Germany to the US and then disappeared ..)
I’m only speculating, but the guidelines were heavily influenced by what was possible and what not. Of course they could have said that people should use medical masks, as they are the only actual way to protect one against an infection. But the markets were empty already and things would have only gotten worse if millions of people now tried to buy masks by the means of alibaba and such.
Nowadays medical masks are cheap again and can be bought easily (in pretty much every store here for around 1.5$), it only makes sense to up the requirements. However, I would have preferred if they would have spoken clearly about it in the beginning, as it wouldn’t have sparked so much room for interpretation.
Here’s a very simple explanation to your first question: lots of people might have gotten sick or thought they had the virus when they actually don’t. I would hazard a guess that actually quite a lot of people thought they had covid when they actually didn’t (especially when there can be no symptoms).
It’s much better to over-vaccinate a few, than have a great number of people skip out on the vaccine incorrectly. It’s a much easier policy to apply and fulfill, as well, in terms of allocating shots to districts (you can just do it by population).
Why would the people think the had or did not have the virus? Here we have an official “database” of people who had it and tested. Everyone I talked with knows if they had it, that they maybe had it (but didn’t bother to check or pay $6 or so for the test) or they did not (most likely). I personally did 4 tests to make sure the results are not false.