iSteve commenter viennacapitalist writes:
Thought you and your readers might be interested in data from Austria (in lock down since March 14 – with active cases receding for five days, i.e. there have been more people recovering than newly infected) – there is too much focus on Germany here.
In Austria there have been two random studies conducted so far (sample size about 1500 each). Austria is the first country in continental Europe to come out with such a study (Iceland is the famous other country in Europe). One sample focused on professions assumed to be at high risk of infection (health care, supermarket employees) whereas the other aimed at getting a representative sample for the entire country.
First the results of the random study whose results were released yesterday:
About 0.33 percent of the population infected, or 28,500 people vs. 8,500 confirmed cases at the time (about 10 days ago), (95 percent confidence interval: 10,200 and 67,400). As of today there are slightly more than 300 fatalities which equates to a CFR of around 1 percent – Austrian hospitals are not overwhelmed with capacity similar to Germany.
One week ago, the results of another random study was published that focused on professions at risk: Result: 0.5 percent of health care workers were infected but, interestingly, out of 350 tested supermarket employees NONE were infected, suggesting the virus might be less contagious under certain circumstances than generally assumed. At the time of the study masks in supermarkets were not compulsory (they have become so since April 6th)
One week ago I read in the newspaper that in Austria there is a group who tries to trace and identify infections – similar to what Mr. Streeck is doing. They had traced about 150 cases and I remember the article said that they could not find indications of casual infections, in all cases there had been close contact for at least 15 minutes between for the infection to happen – Prof Streeck mentions similar findings. They said, that it looks the virus is less contagious than previously thought.
My preliminary conclusion:
The virus seems to have binary properties: not very contageous under most circumstances, but highly contagious in others. I am no expert on this, but believe that solving the superspreader issue is key to understanding this thing. I am surprised that no modeller has tried to come up with an “adjusted R0” , adjusted for superspreaders that is – what if we find out that eliminating superspreaders (or events) reduces R0 to close to 1?
I am also surprised that not more random studies have been conducted around the world in order to better get a final grip on the data.
But to give the Germans more attention, iSteve commenter GermanReader2_new asks some good questions:
Did they say anything about the estimated infection rate in cities vs the countryside, age groups etc.? In Munich around 0.2% of the population is already infected without extrapolating (they are doing some studies right now to estimate the true infection rate and spread in the city by randomly picking a few thousand people and testing them every 4 weeks). What I really would like to know are the infection rates of the locals in the ski towns (Ischgl) by profession. I think most of the Barkeepers in the apres ski bars are probably positive, but how about the hoteliers etc?
Right, as I keep trying to communicate, the “reopen the economy” question shouldn’t be a monolithic G0-No Go decision for everybody everywhere, but should instead optimize the tradeoff between risk and reward.
To do that, it would greatly help to know the infection / hospitalization / death rates of different occupations. For example, there are a lot of shoe salesmen in New York City. On average, have shoe store workers gotten by safely or have been they hit hard?
That may not seem like an important question, but it is to shoe store employees, shoe store owners, and to potential shoe store customers.
Let’s try to find out.
Outsiders could start by scanning obituaries for people under, say, 65 and tabulating their occupations.