Oh my gosh🤦♀️
When asked about mask data, @CDCDirector really said "the most comprehensive" was a TWO week MMWR study...
The one @AChandra_TO & I reanalyzed (extended in time & to a more nationally repr. sample: 565➡️1832 counties)
& the significant difference did not persist!🧵
When asked about mask data, @CDCDirector really said "the most comprehensive" was a TWO week MMWR study...
The one @AChandra_TO & I reanalyzed (extended in time & to a more nationally repr. sample: 565➡️1832 counties)
& the significant difference did not persist!🧵
As we concluded in our expanded reanalysis published in the Journal of Infection: "In a larger sample of 1832 counties, between weeks 2 and 9, cases per 100,000 fell by 38.2 and 37.9 in counties with and without mask requirements, respectively (p = 0.93)."
sciencedirect.com
sciencedirect.com
The ?s are
1. Why is she citing a 2 week NON nationally representative debunked observational study when she could cite higher quality randomized evidence?
2. Why does she ignore a peer reviewed reanalysis of the study she cites which has more robust data?
sciencedirect.com
1. Why is she citing a 2 week NON nationally representative debunked observational study when she could cite higher quality randomized evidence?
2. Why does she ignore a peer reviewed reanalysis of the study she cites which has more robust data?
sciencedirect.com
You can see the original publication here by Budzyn et al which actually states "because of the small sample size of counties selected for the analysis, the findings might not be generalizable"
cdc.gov
cdc.gov
I spoke before congress last week about why I believe @CDCMMWR rejecting our expanded analysis demonstrated clear publication bias.
But bigger ? is why is the initial small confounded obs study being cited as "the most comprehensive" masking evidence by the @CDCDirector
The end
But bigger ? is why is the initial small confounded obs study being cited as "the most comprehensive" masking evidence by the @CDCDirector
The end
Loading suggestions...