For Whom Do the Cov “Fact Checkers” Really Work?
By Vinay Prasad
I recently came across two tweets that caught my eye.
Here is the first one from the CDC director:
Masks can help reduce your chance of #COV19 infection by more than 80%. Masks also help protect from other illnesses like common cold and flu. Wearing a mask- along w/ getting vaccinated- are important steps to stay healthy.
And here is the second one, from a couple of months back:
Together they got me thinking. What do they have in common? What do they tell us about the state of the public communication of science?
Let’s start with the one by Dr. Walensky. I don’t know how to put this politely, but it is a lie, and a truly unbelievable one at that.
First, of all, if it were true, it would mean that masking was more effective that the J&J vaccine (implausible). Second of all, we have actual cluster RCT data from Bangladesh showing a 11% (relative risk reduction). This occurred in a massive trial where masks were provided for free and encouraged. Even here, only surgical masks worked, and cloth did not, and had nowhere near this effect size. The idea that masks could reduce the chance of infection by 80% is simply untrue, implausible, and cannot be supported by any reliable data.
The mathematician Wes Pegden had this to say about it, and Wes is right!
Yet, as far as I can see no organization nor twitter has fact checked this tweet and labeled it misleading. It is an untruth we are allowed to say.
Now let us turn to the AP’s fact checking claim. This is where things get interesting.
There are two types of COVID19 survivors— those who have documented recovery from sars-cov-2 (either PCR, antigen or serology + tests) or those who have self-identified recovery from sars-cov 2 (said they had it).
When it comes to the former group, we know with confidence, the chance they get re-infected and severely ill is very very low, and far lower than people who have not yet had and recovered from COVID19 (this is called natural immunity). The data in support of this is massive, and quite certain. Antibody data is beside the point— we care about the thing in itself getting sick.
So do these people (those who recovered) benefit from vaccination? Current data is solely observational— and that is a huge problem. If you compare people with recovery who chose to get the vax vs. those who chose not to get it— you are comparing very different types of people. Their behavior, and appetite to take risks (going out to crowded places) may also be different. We know both groups have very low rates of re-infection, but direct comparisons to assess vaccine efficacy after recovery are fraught.
The right answer would be to conduct an RCT of vaccination among those who recovered. It could have 3 arms. No further doses; 1 dose, or 2 doses. It could be large (after all, millions have recovered), and powered to look for rates of severe disease. In the absence of this, experts are largely speculating.
So here is what blows my mind: We are living in a world where the CDC director can say something that is false, made-up and no institution will say otherwise. At the same time, major, venerable fact checking institutions are literally asserting as fact something which is at best unproven.
No matter how you feel about these issues; these are dangerous times. Truth and falsehood is not a matter of science but cultural power— the ability to proclaim and define the truth. If this continues, dark times lie ahead. Someday soon, we may not like who defines the truth.