Big news! Our systematic review and meta-analysis of the age stratified IFR of COVID-19 with @BillHanage, Andy Levin, and others has now been published in the European Journal of Epidemiologyhttps://link.springer.com/article/10.1007%2Fs10654-020-00698-1 …
-
Show this thread
-
Replying to @GidMK @BillHanage
Peter Pham Retweeted Peter Pham
this is the type of drill down we need to see, thank you. What multiplier are you using for the denominator for US? Agree it's 1:8? https://twitter.com/peterpham/status/1333671190850805764 … and do you calculate latest ifr vs overall given treatment protocol?https://twitter.com/peterpham/status/1326319759495843841 …
Peter Pham added,
Peter PhamVerified account @peterphamOptimism. in April Case Fatality Rate for under 30 was 0.18%, Sept 0.01%, 70-79 was 23% over 80 was 41%! Sept is now 4.7% & 15.5%. This is without a 20X in estimated infections as@WHO last month estimated 760M have been infected. https://www.usnews.com/news/health-news/articles/2020-10-05/who-estimates-coronavirus-has-infected-10-of-global-population#:~:text=The%20new%20estimate%20indicates%20some,around%2035%20million%20reported%20cases …. https://twitter.com/kerpen/status/1326306959419731968 …Show this thread1 reply 0 retweets 0 likes -
Replying to @peterpham @BillHanage
We didn't infer case numbers that way, we used serological data
1 reply 0 retweets 3 likes -
Replying to @GidMK @BillHanage
what would be the US ratio? or maybe a better question would be how many do you think have now been infected in the US? currently we are at 16.5M tested +
1 reply 0 retweets 0 likes -
Replying to @peterpham @GidMK
It varies a lot by region. I’m aware of a jama paper I’ve not had the chance to review and will shortly be collapsing. But would start there.
1 reply 0 retweets 1 like -
Replying to @BillHanage @GidMK
this was the 1:8 cdc & wsj was referring tohttps://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa1780/6000389 …
2 replies 0 retweets 0 likes -
Replying to @peterpham @BillHanage
Health Nerd Retweeted Health Nerd
I had a very rough go at calculating this using our IFRs by age for the US. Based on these calculations, it was more like ~1:10 in March/Apr, but down to 1:3 by October. Somewhere around 50 million total cases by October I'd sayhttps://twitter.com/GidMK/status/1321661890301366272?s=20 …
Health Nerd added,
2 replies 1 retweet 1 like -
Replying to @GidMK @BillHanage
Peter Pham Retweeted Peter Pham
ha, you nailed it. CDC said 53M by end of sept. https://twitter.com/peterpham/status/1338694999081291777?s=20 … But
@who was higher. So then the last 2 months not 1:8? 100M too high? or more like 90M per my math.Peter Pham added,
Peter PhamVerified account @peterphamCDC estimates for every 1 positive test we missed 8. End of Sept was 53M https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burden.html … Oct was 1.9M & Nov 4M = 5.9M x8 = 47M. So 100M naturally immune now, hopefully we get the vaccine out to the rest quickly & start seeing lightShow this thread1 reply 0 retweets 0 likes -
I think the aggregate is slightly unhelpful in terms of the ratio - it was much higher earlier in the year, but by now 1:8 is almost certainly too high. I'd say more like 1:3 or 1:4
2 replies 0 retweets 2 likes -
given bayes theorem what do you think is % of false positives?
1 reply 0 retweets 0 likes
I wrote a bit about that. In short, the false positive rate is extremely lowhttps://gidmk.medium.com/most-positive-coronavirus-tests-are-true-positives-60c95fe54fec …
-
-
thoughts then on CT value? what should we be testing at? shoudn't labs publish the value?
1 reply 0 retweets 0 likes -
I think the value is helpful for a lot of reasons, and should be at least reported to improve multiple elements of situational awareness
2 replies 0 retweets 2 likes - 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.