Loading...

Forum

You need to log in to create posts and topics.

Covid random testing shows crisis is overblown?

Ok so many states have done random covid testing and the results show that large portions of the population (up to 60x official numbers) already had the virus and recovered fully while being asymptomatic.

That is, this virus is actually weaker than influenza and the death rate is also same as or lower than influenza.

So basically, lock-down does not help in reducing the load on hospitals. It only delays the formation of herd immunity. For example, see Sweden.

I think with the new data from random sampling of antibodies points to a death rate around 0.5% instead of the earlier consensus estimate of 2.0%.  However, 0.5% is still much more deadly than influenza which usually has a death rate around 0.1%.  Also covid seems to be much more infectious.

I think the confusion can come from the fact that there are three numbers: (a) the number of people that have taken tests and tested positive (official numbers), (b) the number of people that we estimate had covid based on positive test results (some multiple of (a)), (c) the number of people that we estimate had covid based on random sampling of antibodies.

So I think when you mention a 10x to 60x difference I think those numbers are the ratio between (a) and (c), but I believe death rates have always used our estimate of total infections (b) not positive tests (a).  So in NYC with the rough guess (before random sampling) that 1/4 of people that had the virus went to a hospital and got tested we would have a = 250k, b = 1M, deaths = 20k, death rate = 2%.  However, with random sampling we have c = 2.7M, deaths = 20k, death rate = 0.74%.

https://www.nytimes.com/2020/04/23/nyregion/coronavirus-antibodies-test-ny.html

 

There has been some right-wing gaslighting on this but the idea that it is weaker than the flu is not at all correct.  First the basic transmission rate is between 2.4 and 4.8 in the literature I have read in Nature and Science.  Second, the fatality rate, appears to be in the range of 0.6% and 1%.

It's not really in the same league as the flu.

Yeah, this isn't the flu. It's massively more infectious and deadly.

But there's also growing evidence that the load on the hospital system is much lower than we expected. So I am beginning to see both sides of the argument more clearly. Yes, we need to be methodical and thoughtful about reopening, but we're approaching a point where the prolonged economic decline will have an even greater impact than the virus is actually having. A slow and methodical reopening of things makes sense at this point.

"Pragmatic Capitalism is the best website on the Internet. Just trust me. Please?" - Cullen Roche

I think we are starting to understand this virus better.  One thing I have been looking at (as have many) is the infection rate may be non-linear in dose (not uncommon for viruses).  The infections in hotspots have in some cases been traced back to a single infected person who was just at the edge of symptoms.     This complicates modeling since in the SIR or SEIR models the rate of infection is I^\dot = b S I  or I^\dot = b S E, which considers the transmission as uniform over the lifetime of I.  But if a person is most infectious at the end of the pre-symptomatic phase, early symptomatic phase, then this dramatically changes the dynamics.  The non-linearity appears both in people being highly infectious and in clustering many people.

We have some good evidence that a multi-pronged approach to mitigation can be made to work.  The best example might be from Mass General:  https://www.newyorker.com/science/medical-dispatch/amid-the-coronavirus-crisis-a-regimen-for-reentry