Seattle-area officials report new coronavirus deaths, bringing US total to 6

Yes, the US is severely undercounting.

There are a few reasons for this, none of which are acceptable.

1. The initial guidelines for test kit requirements were needlessly strict. They sought to create tests which would not test positive for related pathogens like SARS or MERS, neither of which seem to be circulating in the wild right now.

2. The first batch of test kits issued by the CDC were faulty.

3. The CDC disallowed tests for people that did not have a link to China.

The CDC has displayed staggering incompetence at handling this. I don't know whether this is due to inexperience (the administration fired the CDC's entire pandemic response team in order to cut costs) or trying to downplay the situation in order to mitigate damage to the stock market. Or something else equally inexcusable.

It's anyone's guess how many people are actually infected by coronavirus in the US right now. But I don't think estimates less than 500 or so have any chance of being accurate.

The main seattle outbreak happened at a nursing facility and older people are at higher risk. "Four coronavirus-related deaths have been linked to the nursing facility. 50 people at the facility are being tested for the virus" [0].

[0]: https://www.kiro7.com/news/local/report-dozens-kirkland-nurs...

With all the news about how few tests have been run in the US, I'd go with 'undercounting'. Also looks like some/many of the WA cases are in a senior care place. If the initial spread there was among a group of elderly people, I'd expect it to be more lethal than among a larger more varied population.

In the press conference this morning, King County officials said they had been testing predominantly the most severe cases only, and that they were likely not testing the community of people who had flu-like but otherwise mild symptoms.

So I'd go with undercounting.

Additionally, many but not all of the cases are linked to a elderly care facility, where underlying medical issues with lungs, hearts, and diabetes caused a higher fatality rate.

They said they expected the number of cases to increase, and they are increasing testing capacity to hundreds of tests daily.

I think a lot of the US cases don't have a clear chain of who got the virus from who, so there's probably a lot of people in the public at large who don't know they're infected (hopefully they think they have a cold and are staying home). However almost all of the US deaths happened at the same nursing home, so I think it's too early to make generalizations about death rates.

More striking is that the ex-China case count appears to be doubling every three days.

https://www.worldometers.info/coronavirus/coronavirus-cases/...

Yes, effectively. The US has tested ~zero people relative to population (under a thousand according to CDC), compared to eg South Korea who has tested thousands of people.

Estimates suggest that between 150-1500 people in Seattle are infected, so the 96 number is clearly off.

I'm too lazy to do the stats, but the US sample size is too small to prove a difference in means here, even if we knew the distributions and there wasn't ferocious noise in all of this.

We have confirmed cases of people getting sick from unknown carriers... So yeah, we know, for sure, that our count of infected peoples is incorrect.

> 6/96= 6.25%

Is 96 a number of (ongoing + finalized) cases or just a number of finalized cases?

The outbreak's epicenter in Seattle is a nursing home.