'...seems like quite a stretch': Nate Silver breaks down the CDC data from Provincetown, MA that's being used to justify the new mask guidance

The CDC finally released some of the data behind its new mask guidance and as many suspected, it’s a study based on a cluster of infections that happened in Provincetown, Mass. during July.

First up, the study “includes several major caveats, most of which are almost completely missing from the hyperbolic news coverage around this”:

Here, finally, is the Provincetown study, which includes several major caveats, most of which are almost completely missing from the hyperbolic news coverage around this.https://t.co/2psyPdVob6 pic.twitter.com/VngcGG8HrP

— Nate Silver (@NateSilver538) July 30, 2021

And for some reason, the authors left out this tiny detail:

There is also one major caveat that the study authors don't mention: they are only looking at *people who chose to be tested*, which is a lot different than *all breakthrough infections*. Presumably people with symptoms are much more likely to be tested.

— Nate Silver (@NateSilver538) July 30, 2021

From Silver on this:

*Symptomatic* breakthrough infections having similar viral loads to *symptomatic* unvaccinated infections would be much less of a problem, both because symptomatic breakthroughs are rare and because people can learn to be more careful (and get tested) when they have symptoms.

— Nate Silver (@NateSilver538) July 30, 2021

The sample sizes are small, too:

Finally I'd note that the sample sizes here are tiny, leading to very wide confidence intervals. And the "real life" confidence intervals are likely even wider given that (as the study authors say) the sample isn't very diverse. (P-town = mostly affluent middle-aged gay men.) pic.twitter.com/llLfxC1jhK

— Nate Silver (@NateSilver538) July 30, 2021

Recommended

Scientific American Editor in Chief, Laura Helmuth Resigns - Science Saved

Gordon K

Advertisement

In summary, “To take a self-selected, not-statistically-significant sample of ~200 nondiverse people during a party weekend that was an outlier in many respects, and use it to conclude that breakthrough infections are just as likely to transmit the virus, seems like quite the stretch”:

To take a self-selected, not-statistically-significant sample of ~200 nondiverse people during a party weekend that was an outlier in many respects, and use it to conclude that breakthrough infections are just as likely to transmit the virus, seems like quite the stretch.

— Nate Silver (@NateSilver538) July 30, 2021

Oof.

***