A CDC paper last month found that Kansas counties with mask mandates saw a decrease in cases in Aug, while counties without mandates saw an increase.

Since then, both groups saw a huge surge.

Counties w/mandates are doing a bit better, but it's difficult to determine causation.

Here's the original paper: https://twitter.com/CDCgov/status/1329877371613302785

The following conclusion is concerning: "Countywide mask mandates appear to have contributed to the mitigation of COVID-19 transmission in mandated counties."

But the paper does not make an attempt to show causation.

The simple reason one cannot conclude causation is that the selection of the counties is not random.

Counties with higher case growth are more likely to enact a mask mandate. So of course, cases are more likely to subsequently fall in those counties. https://twitter.com/youyanggu/status/1295411445786705920

I hope we can be especially careful about confusing correlation with causation.

For example, a state of emergency declaration is correlated with a large increase in deaths in the weeks afterwards.

Does that mean that SOE declarations contribute to an increase in deaths? (No)

When it comes to evaluating scientific research, I think the scientific community should hold the same rigorous review standards to all papers, regardless of their conclusions.

Unfortunately, I don't think that has been the case thus far, at least from an outsider's perspective.

It should be said that I'm not anti-mask by any means. We should continue to expand our efforts to educate the public about the importance of masks.

I just think the evidence for the effectiveness of mask *mandates* has not been as strong as many make it out to be.