Coronavirus (COVID-19) Discussion Thread

[SvennoJ]

John2290 said:

SvennoJ said:

Why are you celebrating it being far more infectious (if the tests are accurate) than previous believed, needing a far higher percentage with enough anti bodies to reach herd immunity, which is still a big maybe how long and how much that will be effective.

If anything, it's only more dangerous if this is accurate. 0.8% of the world population is still 62 million people. Herd immunity won't be reached until 82% has enough capable anti bodies. This only points to even more strict social distancing to limit the damage.

If the 2.7 million figure based on (a little biased) anti body testing in NY is true, reported deaths in NY state divided by 2.7 million is a current death rate of 0.77%. Of course with average time of 20 days between onset of symptoms and death, that rate is still going to go up, plus uncounted / undetected deaths.


The more infectious, the harder it will be to keep the vulnerable safe. I feel more worried for my wife now and will not go to the chiropractor any time soon, not worth the risk.

I am getting my pool opened next Thursday (a bit early but my pool guy wants to escape up North). He said he'll do all the work there, no need to come see him. I guess I'll put the cash in the pool shed before he shows up and leave everything alone for 4 days (water and chemicals need to settle anyway) to be absolutely safe. I still haven't been to any store since early March, just to the trail to get some fresh air while avoiding people. It's only a few km from here so tank is still nearly full. Food still well stocked.

Meanwhile Ontario posted another new daily high reported cases:
Ontario health officials logged 54 new COVID-19 deaths and recorded a single-day high by confirming 634 more cases on Thursday, as the province continues to ramp up testing for the novel coronavirus.
One positive, no new cases reported in my local county the last 2 days. Yet if these anti body results are true then there could be a thousand undetected cases walking around...

I don't know, I didn't look into any of this yet past the surface, I'm waiting on a larger sample size but I just made the assumtion that if it's spreading since mid January the reproductive rate can't be be much of a problem? I mean if it's r6 it would have gone exponential and took nearly the entire population within a month, if already pretty close to that with r3, no? If it's double that then with these numbers the cap on the population should have been reached before lockdown or am I missing something... Huh, Idk. I'm mightly confused now, I think I'm going to take break for a day before I bother looking back into it, my head is full of assumptions. I mean, it can't be half as infectious as measles or we'd have seen everyone lining up for the ICU by now, right? If r6 is true something else might be wrong...

R0 of 5.7 is based on Chinese data from late December to Januari 23rd (before the lock down), so take a grain of salt with that. They come to a doubling rate every 2.4 days for the early growth.

March 20 is when NY became serious about the lock down, which stops the exponential growth in its tracks.
According to the NY times, NY peaked on April 10th at about 10,000 reported cases a day.
https://www.nytimes.com/interactive/2020/us/new-york-coronavirus-cases.html#county

Of course reported cases are a bit useless with these new estimates especially with the late to start testing.

But we can look at that doubling rate. To get to 2.7 million infected by April 1st (totally wild guess to avg lock down slowdown date to peak date) with unhampered spread,
50 days = 1.9 million
51 days = 2.5 million
52 days = 3.3 million
It doesn't matter that much, exponential growth goes fast on the upper end.
51 days before April 1st is start on Februari 10th.

So wild estimate and assuming NY was infected later than California (reasonable assumption) the R0 of 5.7 is somewhat possible with the 2.7 million (or 10x more infected than detected) estimate.

However people would have been aware earlier and been more cautious. Also that study that came to R0 of 5.7 warned it might be different in other countries, new year celebrations likely played a part in early transmission in Wuhan and since nobody knew of it yet, nobody was being careful yet. Plus there's the uncertainty in the early data.

The difference with the old R0 of 2.2 is huge, after 51 days you only get to 2.3K, 10 times less. Or you need 97 days to go from 1 to 2.5 million.
Also he doubling rate for reported deaths begs to differ
https://www.statista.com/statistics/1104836/days-for-covid19-deaths-to-double-select-countries-worldwide/

Russia where it recently exploded and with decent numbers to put some trust in, is at a doubling rate of 5 days, in line with the old R0 of 2.2. A lot of countries have much higher doubling rates since that stat is a real time statistic (snap shot at Apr 19th) and the lock down measures already had an effect. In Russia however measures are just now starting to affect the death rate. And those with smaller doubling rates had few deaths reported. So Russia's number is as pure as it comes from that link.

Nothing from that link supports an R0 of 5.7 and doubling rate of 2.4 days, I wouldn't put all that much faith in that. It was based on Chinese data after all.


A doubling rate of 5 days, suppose there were 20 people infected mid Januari (15th), by March 20th there would have been 163.8K people infected.
March 25th 328K
March 30th 655K
Apr 04 1.3 Million
Apr 09 2.6 Million

However since March 20th it should have slowed down to a flat increase by Apr 10th since that was the peak. (of 10K, or at most 100K if 10x under reported)
Even with 20 infectious people walking around in the population on Januari 15th and taking the fastest doubling rate observed in reported deaths, it still seems unlikely 2.7 million people are already infected in New York.



But yeah, with R0 of 5.7, spread starting in mid Januari, even if only 1 person started it on Januari 15th, 142 million people would have been infected by March 20th in NY. Everyone would have had it by now.



That's the problem with all these different biased tests, different test criteria, different ways to count data. By now you can fit or disprove anything :/ The difference between R0 of 2.2 and 5.7 is so big you can make up anything you want...

So basically we only know less now :(