The Brazilian state of Maranhão (as large as Italy) reported a seroprevalence of COVID-19 of 40.4%
Obvious findings:
• «herd immunity threshold is not as low as 20%»
• «infection rates were significantly lower among mask wearers»
h/t @CovidSerologyhttps://www.medrxiv.org/content/10.1101/2020.08.28.20180463v1 …
-
-
.
@GidMK I see you retweeted, looks hard to draw many conclusions from this - what is your take on methodology, unclear recruitment method, IgM and IgG survey validity, very low IFR, low asymptomatic rate, no male predominance ?2 replies 0 retweets 1 like -
The sampling method was pretty robust - household random blocks. Some issues with non-response bias, which is unlikely to be undifferentiated, but impossible to know precisely the impact that this would've had
1 reply 1 retweet 5 likes -
The asymptomatic rate is very much in line with other estimates (most of the time, very high asymptomatic rates are due to different definitions of 'symptoms'). The low IFR seems to me likely to be explained by right-censoring
2 replies 0 retweets 4 likes -
It appears that the authors took the death numbers up until 15/08 despite the survey finishing 08/08. You'd expect more people to die from the infected population over the next few weeks/months, pushing the IFR up
1 reply 0 retweets 3 likes -
Given that the region was still experiencing 100 deaths per week at the censoring date, this difference is likely to be fairly substantial!
1 reply 0 retweets 2 likes -
Agreed, also excess death analysis suggest underreporting may be a much bigger issue than the 15% adjustment they authors made. Excess death is double reported COVID-19 fatalities. https://www.conass.org.br/indicadores-de-obitos-por-causas-naturais/ …pic.twitter.com/exsjQLA0uC
1 reply 1 retweet 7 likes
Yes, this is a fairly big problem in a number of countries with large COVID-19 outbreaks but relatively sparse infrastructure (Brazil, India etc). It would not surprise me if the true death count was substantially higher
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.