Have a question - It seems the longer COVID19 spreads in the community equals a greater virulence and strength in its symptoms . Why is this the case.?
Actually, it's usually the opposite. The virus becomes less deadly. From the point of view of the virus, the ideal situation is when no one dies, and no one gets very sick. That way, the time in which people can interact with others, spreading the virus, is maximized. So mutations that result in less sick people and lower mortality, tend to out-compete earlier, more deadly, versions of the virus.
If there appears to be greater virulence, it may be because as more people become infected, the average viral load someone is infected with is likely to increase. This may result in worse symptoms, though the jury is still out on that for this virus.
With regard to the Austrian study. Currently, the naive or nominal mortality rate in Austria is 2.44%. As I've explained before, this is an under-estimate, because many cases are still active, and will result in more deaths. We can also determine a ceiling on the mortality rate, by dividing deaths, by closed cases: 4.86%. This is an over-estimate, because deaths tend to happen more quickly than recoveries.
The infection rate the study reported, 0.33%, compares to a rate of about 0.10% at the time of the study. The difference reflects asymptomatic individuals. So we can multiply the mortality rates by .10/.33, and get a more accurate rate of 0.73 - 1.46%. The lower number is likely to be closer to the true rate than the higher rate. I say this just from observing the way the two rates converged in China. The only other qualification is that that 0.33% rate has a very high degree of uncertainty, so the mortality rate could be considerably higher or lower.
My guess is that when Austria planned this study, they anticipated a much higher number of negatives. Even 5% of the population would have given about 75 negatives, which would be a much more reliable number. The study was designed like a political poll, and keep in mind that in a political poll, a sample size of just 500 results in an uncertainty of just 3-4%. This is because in such a poll, there are usually only two choices, so either choice is going to yield several hundred answers. Given that the actual infection rate is so low, far more individuals need to be tested, at least ten times the number actually tested.
Despite the large uncertainty of the infection rate, the study blows a huge hole in theories/models that propose a large number of silent or asymptomatic individuals. Even at the limits of uncertainty, we can conclude that < 2 % of the population is infected. Given that the sample approximated randomness, and that the situation in Austria is not wildly different from other countries, this should be pretty much the case everywhere else. The German study did find a much higher % of people with antibodies, but that was in one town known to have a high infection rate.
Finally, another point to keep in mind. There are four types of individuals a study like this wants to identify: 1) positives with symptoms; 2) positives with no symptoms; 3) former positives with symptoms who have recovered; 4) former positives with symptoms who have recovered. This study only identified members of the first two groups. We can accurately estimate the number in the third group from the actual data of active vs. recovereds. They're about equal in number. I assume that because of this, the number of individuals in group 4), are probably about equal to the number in group 2). From this, we can estimate the total number of infected as about twice that actually found, i.e., 0.66%.