Out in PNAS: "The effects of school closures on SARS-CoV-2 among parents and teachers". On March 18, 2020, upper secondary schools in Sweden moved online while schools for younger students remained open. 1/n https://www.pnas.org/content/118/9/e2020834118 …
-
Show this thread
-
Replying to @jonasvlachos
Interesting study. It seems to me that the reliance on testing data, which was I believe at that point mostly done in hospitals in Sweden, might be a pretty significant limitation but I can't see this discussed?
1 reply 2 retweets 14 likes -
Replying to @GidMK
Well, a limitation in some regard but it means that testing bias is less of an issue (some groups more prone to get tested when testing is readily availble). Alos, we make use of data on diagnonses from healthcare visits and hospitalizations. It is discussed.
2 replies 1 retweet 8 likes -
Replying to @jonasvlachos
Sure, but we know that age is a significant factor in the likelihood of an asymptomatic/very mild infection, so it's not unlikely that there is a substantial fraction of transmission that would be impossible to identify using confirmed PCR+ cases at the time
2 replies 1 retweet 9 likes -
Replying to @GidMK @jonasvlachos
I get that this is potentially non-differentiated between the comparison groups, but it is plausibly also something that could vary. Not to say that it invalidates the findings of course
1 reply 1 retweet 3 likes -
Replying to @GidMK
Basically, the only thing that differentiates those with kids in final year in lower secondary and first year in upper secondary is one year of age. We also have a host of controls (as mentioned) but dropping these does not affect the results (as expected).
1 reply 0 retweets 0 likes -
Replying to @jonasvlachos
Sure, but I'm not sure that mitigates the limitation here. The fundamental issue is that the outcome measure is a biased estimate of the true infection rate, and I suspect that the true impact of that bias on different populations is very hard to estimate
2 replies 1 retweet 5 likes -
Replying to @GidMK
A less biased estimate would be for hospitalizations. For parents, these are noisy but rule out larger effects than for tests. For teachers, tests and hospitalizations yield close to identical estimates.
1 reply 1 retweet 2 likes
Mmm potentially true. I'll have a think about it, thanks for the discussion!
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.