Rate of New COVID-19 Infections Collapses in Italy

2020-03-24 | Dangus

[dangus]

>> No such thing with covid-19. It is more analogous to a cold in that regard. People get it over and over.<<

Nonsense. The reason people get colds over and over again is that there are more than 200 kinds of cold viruses, all of them different. In fact, you could say that a cold isn’t a virus; it’s the way your body reacts to many types of viruses. Some colds are rhinoviruses, some are coronaviruses.

Did you ever notice that kids seem to get colds a lot more often than middle-aged people? That’s because eventually, you’ve gotten most of the common colds.

[lepton]

I will note that the 3/16 figures are also lower than the previous 2. And 3/10 is completely out of whack.

Let’s be hopeful, but it’ll take a few more days for it to be clearly more than an accounting thing.

[lepton]

Politicians and media people predicting 50-80% infection rates seem to have a hard time explaining the lack of 750,000,000 to 1,200,000,000 infected Chinese

---

Not that I accepted the numbers, but you’ve got a couple of things there:

Despite how the politicians and media people described (Does anyone expect them to get details or math right? :) ), the models were 1) if NO action were taken on this, and 2) over the course of years, not next month.

Also. They were models. Lots of assumptions.

Also, at least as far as the numbers I am familiar with, you’ve overstated the population of China by at least 100 million or so.

[lepton]

Functionally, herd immunity works like picking individuals out of the transmission path here and there. As a result various strands die off as they are not replicated, which in turn causes it to act as a much less virulent disease.

It is being overused in the discussions of how to deal with this coronavirus outbreak, but as you note, it is certainly a thing.

[lepton]

The cases seem to follow the same bell shaped curve in every country over 8-10 weeks so I doubt lockdown tells the whole story

---

In the US, at least, we’ve had enough people ignoring (or even challenging) the “lockdown” that it has undoubtedly been much less efficient than it might have been.

[lepton]

And only 330,000 positive tests and 111 deaths in Korea, among a population of over 51 million.

---

huh? I don’t think they get to that even counting testing the same person multiple times.

It helps when you have a limited ingress path, and you segregate and sort them - even half-assed.

South Korea,
Confirmed: 9,037
Deaths: 120
Recovered: 3,507
Still sick: 5,410

[rdcbn]

Politicians like Gavin Newsome and Andrew Cuomo have publicly given short near term infection rates in the 50 -80 % range

Newsome predicted as many as 25 million Californians will soon get corona virus.

This is simply irresponsible insanity

The number I used for the Chinese population is 1.45 billion people - a generally accepted number

[cherry]

from my fuzzy math I figure that normally about 600000 people die each year in Italy....that would be about 12000 people each week and I would love to know what the number for all deaths is the past 4 weeks, so we could actually see how many MORE or how many LESS deaths there were.....my bet is that the normal death rate for Italy stands....

[lepton]

The head of health for Ohio extrapolated 100,000 cases from an apparent 5, based on a 2017 CDC document that said that if 2 people walk in to a hospital with community-spread cases, then assume 1% of the population has it. She apparently didn’t understand what “population” meant in context and assumed it meant population of the state, rather than population of say, the Italian Village area of Columbus.

That said, it is clear that this thing has become quite scattered and loose in a lot of communities now.

But yeah, not tens of millions yet or imminently.

[rdcbn]

From the data so far, don't see this thing breaking out in a really big way, except in highly urban places like NYC or San Francisco with high population densities

Hard to tell given that we are working on very partial info but we shall have a good read towards end of next week

Fact is, for such an infectious disease, it has made very little headway in some of the highest population density countries with populations most at risk genetically to the infection

China especially.

Claims that by the end of June or July 50% of the American population will have become ill simply does not comport with what have seen play out on the ground in hardest hit countries.

[lepton]

Not so much in a big way, but in a seeded-scattered way.

https://gisanddata.maps.arcgis.com/apps/opsdashboard/index.html#/bda7594740fd40299423467b48e9ecf6

JHU has changed their mapping in the US. Now it shows by county. Some of the nearby dots are really adjacent, but it does show that this is here and there over a very wide spread.

Still big areas with nothing reported yet.

[ManHunter]

Yeah, I screwed up in my haste; I corrected my original post in #52.

[bolobaby]

There are a few problems with the data table you posted.

First, early testing is likely to have higher day over day percentage increases as testing becomes more widely available.

Second, while the percentage from prior days is going down, 8% is still a healthy exponential growth rate, with a doubling time frame of 9 days.

Third, the rate today actually edged up slightly, and death counts were the second highest single day total. (The latter point isn’t necessarily a problem from a growth perspective. That is, you can be on the downward trajectory as but have number of fatalities due to prior infections resolving.)

Lastly, as a cumulative total, bear in mind that there were still 5,249 new cases today. To me, that is surprisingly large given that the lockdown for Italy began almost 4 weeks ago.

I’d say they aren’t out of the thick of it yet.

[dangus]

>> First, early testing is likely to have higher day over day percentage increases as testing becomes more widely available. <<

All the more reason a decline is noteworthy.

>> Second, while the percentage from prior days is going down, 8% is still a healthy exponential growth rate, with a doubling time frame of 9 days. <<

No, that is NOT an exponential growth rate. It’s sub-arithmetic. If the growth rate were STEADY what you are saying would be true.

>> Third, the rate today actually edged up slightly, and death counts were the second highest single day total. (The latter point isn’t necessarily a problem from a growth perspective. That is, you can be on the downward trajectory as but have number of fatalities due to prior infections resolving.) <<

Data will bounce. 8.2% is still WAY, WAY, WAY, WAY below the recent levels of 13-20%.

>> Lastly, as a cumulative total, bear in mind that there were still 5,249 new cases today. To me, that is surprisingly large given that the lockdown for Italy began almost 4 weeks ago. <<

The lockdowns began March 8-10. That is TWO weeks ago.

[dangus]

My guess is that the rate of NEW infections was never really 20%-27% in the first part of March, and that the absolute drop in new cases my have been from a somewhat inflated number. But given the scale of the drop in the rate (20-27% down to 8%), the curve has unquestionably flattened.