Read the whole thread. No one is actually advocating *random* sampling... Just something smarter like SRS or active learning. I don't want to rehash the whole thread. Doctors are turning away people left and right on a hunch. Told to not come in. Why not be smart and systematic?
-
-
Replying to @sinanaral @GidMK
* This is a direct quote from your first tweet: "An argument for *random* COVID-19 testing." * This is a direct quote from your last tweet: "No one is actually advocating *random* sampling."
2 replies 0 retweets 0 likes -
"Random" sometimes denotes uniform random, other times it means non-uniform random, perhaps satisfying positivity for some population — hopefully yielding what is sometimes called a probability sample. One has to rely on context a bit.
1 reply 0 retweets 1 like -
Once they have testing resources they can do that. Until then, let the soldiers decide - it is their health and the health of other patients that is on the line.
2 replies 0 retweets 0 likes -
That seems like a different argument. And unclear that the current testing regime is one that has indeed been chosen by medical experts, epidemiologist, or front-line health care works.
1 reply 0 retweets 0 likes -
No, that was my argument to begin with (see above). My understanding is that
@GidMK has basically the same argument but I can't vouch for that.3 replies 0 retweets 0 likes -
Yes, similar. I think it's unlikely that we will be able to implement a testing regime that prioritizes perfect data, so rather it is more important to use the data we do gather with our imperfect system wisely
2 replies 0 retweets 0 likes -
Sinan Aral Retweeted Andrew Weeks 😷
Note: this suggests in an Italian town where everyone was tested repeatedly, 50% of positive cases where healthy & asymptomatic. If true & we continue doing what we're doing we will cost *millions* of lives and the pandemic will last a year not 6 months.https://twitter.com/meloncholy/status/1240080200152616967?s=19 …
Sinan Aral added,
Andrew Weeks 😷 @meloncholyReplying to @sinanaralVò in Italy is an interesting, if small scale, example where everyone in the town was tested repeatedly. Roughly half of infected had no symptoms. Aggressive testing helps Italian town cut new coronavirus cases to zero https://www.ft.com/content/0dba7ea8-6713-11ea-800d-da70cff6e4d3?shareType=nongift …1 reply 0 retweets 0 likes -
There's strong evidence from Italy and Wuhan that 50-80% of cases are mild/asymptomatic. This does indeed have implications for the health service, but random testing is not one of them
2 replies 0 retweets 1 like -
See above
you seem to not even be able to consider the possibility that this is a good idea despite the fact that some forms of it are being implemented globally. I'm simply saying can't we consider some systematic surveillance testing. Your position seems dogmatic to me.1 reply 0 retweets 1 like
Systematic does not mean random, and is in no way what Ioannidis was calling for in the article you posted
-
-
Correct! I clarified this multiple times, pointing to SRS weighted toward positives & active learning. This is akin to a smart surveillance program. We could use antibody tests & follow up in accute cases w/ clinical tests. Data suggest lots of asymptotic positives transmitting.
1 reply 0 retweets 0 likes -
Lol I feel like the argument got off the rails somewhere. My main objection was to Ioannidis ideas
2 replies 0 retweets 1 like - Show replies
New conversation -
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.