Sunday, March 01, 2020

Coming In From the Edges

It's interesting to see the virus come into the country via its borders first. That's what you would expect, I guess, and entry via major airline hubs like LA, Chicago and New York. The little bugger is predictable in many respects, and yet it seems there's little we can do to stop it. I guess that's a big part of why it's scary.


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

8 comments:

Entropic man said...

You have a hotspot in Washington state, perhaps 1500 cases.

https://www.bbc.co.uk/news/world-us-canada-51715245?intlink_from_url=https://www.bbc.co.uk/news/world&link_location=live-reporting-story

Layzej said...

If there are only 18 confirmed cases in the state, does it mean that the other 1482 people are asymptomatic? If so, maybe the virus isn't all that bad?

Entropic man said...

It could be lower or higher depending on a lot of assumptions.

Worldwide, death rate is settling around 1%. Then 4 deaths implies 400 cases.

If the hotspot centres around care homes, then it is a particularly vulnerable group and the total cases might be lower again.

If the care home group was infected from visitors, then it might be much larger.

The nightmare for an analyst is that at this stage the data is changing from day to day and it is hard to do the maths on the hoof. Much easier in retrospect.

Entropic man said...

Eighteen confirmed cases and four deaths is a mortality of 22%. Not credible, based on mortality elsewhere.

Either something is massively amplifying the death rate of the are a lot of unconfirmed cases in the area.

David Appell said...

L: or it could mean those 1482 people either haven't had serious symptoms, or any, or haven't been tested. I'm assuming it's a number spit out by some epidemiologist's model.

David Appell said...

EM: this page shows a latest death rate of about 3.3%:

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

But this may be an overestimate if a lot of people have little-to-no symptoms, or haven't gotten tested, or (as in China) been rejected by all the hospitals they visited and sent home to suffer and die.

David Appell said...

And perhaps the latency period of the virus -- the time between someone is exposed to when they start to show symptoms, ~ weeks(?) -- is quite a complicating factor that might be difficult to include in models -- since no one really knows what this latency period is.

David Appell said...

Some new info:

"WHO estimates coronavirus death rate at 3.4 percent — higher than earlier estimates," WaPo 3/3/20.
https://www.washingtonpost.com/world/2020/03/03/coronavirus-live-updates/#link-FZWCUE6QH5AVLCAB2UUN4BHBQM

If half the people in the US get the virus, that would mean almost 6 M dead. Impossible to fathom.