COVID-19

[IKE]

I posted the below on another forum after seeing a couple of guys throw out random death rate numbers. I didn't want to be like certain other people who "have been putting out daily updates on a different forum with high accuracy every day" and not share it here...so...here's some numbers I crunched this morning while procrastinating on house projects.
-----------------------------------------------
I did some simple math based on the US and Korean numbers published here. Crunching the numbers as "per captia" results in really small numbers (10^-4 type numbers), but the interwebs said it's common to multiply by 100k for diseases because of this. These are static numbers and in no way a good predictor of anything. Infection rates (Cases per 100k) are not great numbers to look at since the amount of people tested varies wildly from country to country. I'm inclined to say the death rate is a more accurate number, but that presumes each country is reporting honestly/accurately. That said, as of right now (24MAR20) and in no particular order:

Population
Korea: 51,470,000
US: 327,200,000
Italy: 60,480,000
China: 1,386,000,000
Spain: 46,660,000
France: 66,990,000
Japan: 126,800,000
Iran: 81,160,000
Germany: 82,790,000

Total Cases
Korea: 9,037
US: 46,548
Italy: 63,927
China: 81,588
Spain: 39,673
France: 20,149
Japan: 1,140
Iran: 24,811
Germany: 31,370

Total Cases per 100,000 People
Korea: 17.557
US: 14.226
Italy: 105.699
China: 5.886
Spain: 85.025
France: 30.077
Japan: 0.899
Iran: 30.570
Germany: 37.891

Total Deaths
Korea: 120
US: 592
Italy: 6,077
China: 3,281
Spain: 2,800
France: 862
Japan: 42
Iran: 1,934
Germany: 133

Total Deaths per 100,000 People
Korea: 0.233
US: 0.18
Italy: 10.047
China: 0.236
Spain: 6.000
France: 1.286
Japan: 0.331
Iran: 2.382
Germany: 0.160

Percentage of Cases Resulting in Death
Korea: 1.327%
US: 1.271%
Italy: 9.506%
China: 4.021%
Spain: 7.057%
France: 4.278%
Japan: 3.684%
Iran: 7.794%
Germany: 0.423%

Median Age
Korea: 40.8
US: 38.2
Italy: 45.4
China: 38.4
Spain: 44.9
France: 41.2
Japan: 48
Iran: 31.3
Germany: 45.9

Yes, I was bored. But also tired of looking at sensationalized numbers in the media. Feel free to check my math.

My favorite philosophical maxim is, "Check your premises." If I had a favorite statistical maxim, it might be, "Check your denominator."

I like your stats, they show that Italy really is a mess, and it seems like Japan ain't so bad (relatively). I think the media tends to provide raw numbers (total cases, total deaths, etc ) because those raw data are easier to gather, produce, present, and interpret. Unfortunately they not only miss telling the whole story, but also distort or distract from the truth.

Questions your data lead to:

• Are the U.S. & Korea testing more or healing more to get their deaths per case rate down?
• How is Japan beating Europe's death rate with such a high median age (and known large senior demographic)?
• How can we be more German?

(My family only drives Volkswagens. I also listen to copious metal, eat lots of sausage and breaded meats, drink lots of beer, and occasionally harbor feelings of racial superiority.)

 

[Pags]

Check out worldometer for some stats nerdery

 

[RedFive]

Questions your data lead to:

• Are the U.S. & Korea testing more or healing more to get their deaths per case rate down?
• How is Japan beating Europe's death rate with such a high median age (and known large senior demographic)?
• How can we be more German?

Originally I was just replying to two guys who disagreed about whether the US was following Korea or Italy...but the math is really easy and I started adding countries out of curiosity. I briefly considered adding one last table that would be some indicator of cultural compliance with government directives, but that's a hard number to quantify. I found the "Freedom Index" published by CATO, but the numbers didn't seem to correlate to anything meaningful regarding COVID. The Heritage Foundation publishes an Economic Freedom index, but I don't know that this is meaningful in any way, either.

I think it comes down to each culture and the population's relationship with their government. Shooting from the hip, I think we can all see Japan and Germany as being more likely to comply and go along with government directives for the sake of the team. On the opposite end of the spectrum, you have Brazil and the US with very free spirited "I do what I want" type cultures. Hence...

https://www.reddit.com/r/funny/comments/fn3p40

I'm not a fan of power grabs by government, but I think cultures/countries that suck it up and lock down the best will ultimately have a much shorter time in the penalty box. That's just my take. I don't know anything more than the rest of you.

 

[taxi1]

Questions your data lead to:

• Are the U.S. & Korea testing more or healing more to get their deaths per case rate down?
• How is Japan beating Europe's death rate with such a high median age (and known large senior demographic)?

Lockdown and TESTING.

Testing is the key. They did it, we didn't.

We were late on the lock down too, but that can be traced to lack of testing and therefore lack of actionable data.

 

[cfam]

I agree that the situation in Germany is better than in the US, but I also think we’re at least a week (if not two) behind the rest of Western Europe on the infection curve. It’ll be interesting to see how the numbers here look at the beginning of April, given that it took longer than I thought for them to lock things down (still not on a complete lockdown yet).

So far the government has done a good job of striking a balance between being proactive while still avoiding knee jerk reactions. Consistent messaging across the board has definitely helped (federal, state, and local officials all seem to be on the same page).

Yes, generally speaking the stereotype of rule abiding Germans is true, but there’s been a fair share of jackassery here (Corona parties, left wing activists torching things, etc.) The crackdown on that just began with the new restrictions they put out yesterday (to include checking IDs if you’re out and about).

There are still shortages here like everywhere (not enough masks, ventilators, etc), but I haven’t heard any complaints about limitations on testing. Comparatively speaking, it’s going well, to the point that hospitals in my city just flew in patients from Italy this morning.

Personally, I’ve got enough beer from the brewery I volunteer at to last me until the zombies come, so no complaints here.

 

[scoolbubba]

Personally, I’ve got enough beer from the brewery I volunteer at to last me until the zombies come, so no complaints here.

Your civic service is to be applauded. Thank you.

 

[exNavyOffRec]

I agree that the situation in Germany is better than in the US, but I also think we’re at least a week (if not two) behind the rest of Western Europe on the infection curve. It’ll be interesting to see how the numbers here look at the beginning of April, given that it took longer than I thought for them to lock things down (still not on a complete lockdown yet).

So far the government has done a good job of striking a balance between being proactive while still avoiding knee jerk reactions. Consistent messaging across the board has definitely helped (federal, state, and local officials all seem to be on the same page).

Yes, generally speaking the stereotype of rule abiding Germans is true, but there’s been a fair share of jackassery here (Corona parties, left wing activists torching things, etc.) The crackdown on that just began with the new restrictions they put out yesterday (to include checking IDs if you’re out and about).

There are still shortages here like everywhere (not enough masks, ventilators, etc), but I haven’t heard any complaints about limitations on testing. Comparatively speaking, it’s going well, to the point that hospitals in my city just flew in patients from Italy this morning.

Personally, I’ve got enough beer from the brewery I volunteer at to last me until the zombies come, so no complaints here.

The info coming out each day is showing how jacked up the care center was when everything started, it took like a month before the state notified the CDC, that is weeks after the care center said they had an outbreak and employees were going to work sick, going home sick, shopping sick, etc....... so who knows how far behind we are?

The results of the investigation show the place basically was a mess, If the state had called the CDC earlier the data would be much better.

 

[Python]

Redirect Notice

 

[Jim123]

Like any stats and information, the source data should improve as time goes on. For now, nice work crunching the numbers that are available.

 

[wink]

Lockdown and TESTING.

Testing is the key. They did it, we didn't.

We were late on the lock down too, but that can be traced to lack of testing and therefore lack of actionable data.

Korea did little in the way of lock down. Sure they tested, and they tracked down each case following the vector, apparently in leu of extensive lock down. I wonder what the impact on their economy has been (launch the alert nerds) in light of that strategy. As to massive testing, over 97% of Korean tests were negative. And each and every one of those negative could have become infected a day later. I get that testing can provide more data, but it comes at a cost too. Health care workers getting over taxed only to prove 90+% of those they test are healthy, using consumable PPE at a much higher rate then if used solely for high risk individuals and in the treatment of known cases, and burdening labs. So far the US is only testing people that are high risk or have symptoms and have tested negative for flu. Even then we are at about 10% positive. Test every swinging dick and that number will drop dramatically to Korean numbers or lower. Tell me why testing is the key when the data is transitory. So you can isolate the infected? Shit we are isolating the entire country anyway at great expense. I'll give you massive testing if you give me my life back. Not trying to be difficult. I just don't see the value when everything is moving so fast. Simply imagine a high school chemistry experiment. You dutifully observe your data, and then the teacher comes in and adds water to your solution. He continues to come by at random times to fuck with your solution. Where is the value in taking data while the subject of your testing and observations is changing in ways not fully understood? Help me understand.

 

[IKE]

Korea did little in the way of lock down. Sure they tested, and they tracked down each case following the vector, apparently in leu of extensive lock down. I wonder what the impact on their economy has been (launch the alert nerds) in light of that strategy. As to massive testing, over 97% of Korean tests were negative. And each and every one of those negative could have become infected a day later. I get that testing can provide more data, but it comes at a cost too. Health care workers getting over taxed only to prove 90+% of those they test are healthy, using consumable PPE at a much higher rate then if used solely for high risk individuals and in the treatment of known cases, and burdening labs. So far the US is only testing people that are high risk or have symptoms and have tested negative for flu. Even then we are at about 10% positive. Test every swinging dick and that number will drop dramatically to Korean numbers or lower. Tell me why testing is the key when the data is transitory. So you can isolate the infected? Shit we are isolating the entire country anyway at great expense. I'll give you massive testing if you give me my life back. Not trying to be difficult. I just don't see the value when everything is moving so fast. Simply imagine a high school chemistry experiment. You dutifully observe your data, and then the teach come sin and adds water to your solution. He continues to come by at random times to fuck with your solution. Where is the value in taking data while the subject of your testing and observations is changing in ways not fully understood? Help me understand.

So, you're saying COVID-19 testing is like acoustic "screening" with dipping helos?

"We know the sub (virus) isn't in that water column (person) right now, but we have no idea where it might be in a few minutes (days)."

 

[RedFive]

Test every swinging dick and that number will drop dramatically to Korean numbers or lower.

I'm just going to throw in an abbreviated self-quote here to help out...

Population
Korea: 51,470,000
US: 327,200,000

Total Cases
Korea: 9,037
US: 46,548

Total Cases per 100,000 People
Korea: 17.557
US: 14.226

Total Deaths
Korea: 120
US: 592

Total Deaths per 100,000 People
Korea: 0.233
US: 0.18

Percentage of Cases Resulting in Death
Korea: 1.327%
US: 1.271%

Median Age
Korea: 40.8
US: 38.2

We're still below Korean numbers...as of this morning.

I hate to say it, but the best metric is death. There's no way you can test the entire population. You can, however, pretty easily test and keep track of the people who don't make it.

 

[OscarMyers]

https://news.usni.org/2020/03/24/3-...-calls-canceled-as-navy-investigates-outbreak

 

[IKE]

Anyone seen Type I/Type II error rates for the COVID tests? I imagine the general public doesn't realize most of these tests are far from 100% accurate.

 

[sevenhelmet]

I hate to say it, but the best metric is death. There's no way you can test the entire population. You can, however, pretty easily test and keep track of the people who don't make it.

That's a trailing indicator. By the time someone is dead, you're 3-4 weeks (at best) behind the infection. Also, you have only traced (at best) 2% of cases in a new disease that is still not fully understood. Leaving 98% of the spread on the table.

I don't think there is a "silver bullet" metric here. All the mathematical models and projections I've seen fail to account for some variable or another, which may have HUGE implications on the outcome. Better to err on the side of caution, but I'll raise my glass in the hopes of environmental relief as spring warms up, drug therapies to help those at risk, and people taking precautions seriously enough to make a difference. That's just to name a few- I'm certain there are factors in play that I know nothing about, positive and negative. It's a massive game of rates, many of which nobody knows.

In the meantime, it's better in my opinion to isolate and trace known cases, as most of the geographic country doesn't have a ton of cases in their locale, and lockdown known hotspots, such as New York City and San Jose where there ARE a large number of local cases. I hope that we can adopt a refined version of this approach in the next week or two, assuming we really do see a decrease in rate of infection after 2-3 weeks of "social distancing".

If we don't, then ¯\(ツ)/¯