March 26 US Confirmed Coronavirus Cases
Milestones: yesterday I made the prediction that we (the United States) would become the leader in (known) coronavirus infections, and that did come to pass. With just over 86,000 cases, we have passed current hotspot Italy (80,569) and even the country which first dealt with the new disease, China (just under 82,000). Given that China is adding new infections at around 100/day, Italy should pass them tomorrow.
I hope you guys will allow me a bit of international divergence today, before we get to the deeper dives into what's going on in the US. A friend asked me how the United States fares in comparison to other countries, so I thought I'd take a look.
First, let's look at just the raw numbers, done per capita, of a number of countries around the world. Bear in mind that some countries have less reliable statistics than others.
This is a look at 14 countries that popped out at me. Some of these should be obvious: China & Italy, for example. Spain is also getting a lot of press right now, and rightly so. Iran and Korea were a big deal in the early news of spread outside of China. France has been getting a fair bit of news about how they're dealing with the virus, so I added them, and I noticed that Germany seemed worse off than France and added them as well. I added Austria and Switzerland because they bordered Italy, the UK out of general interest and prevalence in our news, Australia because I knew it looked like they might be one of the hardest-hit Asia/Pacific region nations, Israel because their numbers looked really high, and Russia because I'd heard that they had very low numbers but that it was starting to spread.
So looking at the list you might see some surprises here. Italy, supposedly the worst spot in the world, is only third in infection rates? The US is ninth? China 13th!?
Well, the first thing to understand is that actually national rates of infection only tell part of the story. In most countries, you need to look regionally. Take China, for example. The Covid-19 virus first transferred to humans in the city of Wuhan, the capital and largest city in Hubei province. Hubei, with a population of 58.5 million (4.7% of China's total), suffered 82.8% of all the coronavirus infections in the country, and an astounding 96.3% of all the deaths in China. So while China as a whole has a fairly low infection rate of 6 per 100,000, the rate in Hubei was 116. Much higher, but still not the "threshold" number I've been working with of 300 per 100,000 (I explained how I got to this number in my first blog post, but I'll reexamine it in a bit). Unfortunately, I have been unable to find statistics for the number of cases or fatalities in Wuhan proper, but if we extrapolated (and I know this is a bit of a stretch but it is an intellectual exercise worth visiting) a similar ratio of Wuhan:Hubei as we know of Hubei:China, then we get 56,139 cases in Wuhan city, divided by the Wuhan metro area population of 19 million gives us 295 cases per 100,000--essentially equal to my "magic number" of 300.
Now let's jump to Italy briefly. Here we again have a situation where the rates of infection are (or, disconcertingly, maybe more accurately "were") concentrated in one region: Lombardy. This is where I made my calculation to get the "magic number". Knowing that the hospitals in Italy were overwhelmed by the number of patients and starting to make decisions on who and who not to treat, I divided the number of cases in Italy (31,000 at the time) by the population of Lombardy (under the assumption that almost all the cases in Italy at the time were in Lombardy), and then rounded off to come up with 300 per 100,000 as the point at which medical facilities would be overwhelmed. And this, then, is where people start dying in larger numbers. The way I think of it, there is a certain (small) percentage of people who will die of coronavirus no matter what (current) medical treatment they are given, and a (larger) percentage of people who will not die, given medical treatment. But if that treatment doesn't exist, because so many others are already using it....
I'm aware, by the way, that there are a lot of assumptions. First, that all or nearly all of the cases in Italy were in Lombardy. Second, that the medical infrastructure is similar enough in all these countries. But so far, this number has held out fairly well, and provides a useful tool. Looking to Italy now, we see that Lombardy has 34,889 cases--now their share of cases in Italy is just 43% even though their population is only 16.7% of all Italy's. The infection rate in Lombardy is 346 cases per 100,000. Looking at the rest of Italy, the infections seem to be creeping out of Lombardy.
And what we hear from Italy seems to match the data. Lombardy and its capital Milan are in a grave medical crisis. The South of Italy, however, seems much better off, though the spread outwards from Lombardy has to be disconcerting.
Spain's overall infection rate is even slightly higher than Italy's, though the infection, while centered on Madrid, seems less starkly regional than it is in Italy. Madrid's case rate, at 289 per 100,000, certainly has to be deeply concerning, while Navarre and the Basque region are over 200. (La Rioja is a very small region south of Navarre; I have no idea why they're rate is so high).
Switzerland has the highest rate of all the countries on my list, and I have been hearing a bit about their problems. It is interesting to contrast the rates in Switzerland with those of Austria (not much more than 50% of the infection rate of Switzerland). Is this difference due to Austria's very early decisions to restrict travel from Italy, or more related to Switzerland's closer proximity to Milan/Lombardy?
I hope you guys will allow me a bit of international divergence today, before we get to the deeper dives into what's going on in the US. A friend asked me how the United States fares in comparison to other countries, so I thought I'd take a look.
First, let's look at just the raw numbers, done per capita, of a number of countries around the world. Bear in mind that some countries have less reliable statistics than others.
This is a look at 14 countries that popped out at me. Some of these should be obvious: China & Italy, for example. Spain is also getting a lot of press right now, and rightly so. Iran and Korea were a big deal in the early news of spread outside of China. France has been getting a fair bit of news about how they're dealing with the virus, so I added them, and I noticed that Germany seemed worse off than France and added them as well. I added Austria and Switzerland because they bordered Italy, the UK out of general interest and prevalence in our news, Australia because I knew it looked like they might be one of the hardest-hit Asia/Pacific region nations, Israel because their numbers looked really high, and Russia because I'd heard that they had very low numbers but that it was starting to spread.
So looking at the list you might see some surprises here. Italy, supposedly the worst spot in the world, is only third in infection rates? The US is ninth? China 13th!?
Well, the first thing to understand is that actually national rates of infection only tell part of the story. In most countries, you need to look regionally. Take China, for example. The Covid-19 virus first transferred to humans in the city of Wuhan, the capital and largest city in Hubei province. Hubei, with a population of 58.5 million (4.7% of China's total), suffered 82.8% of all the coronavirus infections in the country, and an astounding 96.3% of all the deaths in China. So while China as a whole has a fairly low infection rate of 6 per 100,000, the rate in Hubei was 116. Much higher, but still not the "threshold" number I've been working with of 300 per 100,000 (I explained how I got to this number in my first blog post, but I'll reexamine it in a bit). Unfortunately, I have been unable to find statistics for the number of cases or fatalities in Wuhan proper, but if we extrapolated (and I know this is a bit of a stretch but it is an intellectual exercise worth visiting) a similar ratio of Wuhan:Hubei as we know of Hubei:China, then we get 56,139 cases in Wuhan city, divided by the Wuhan metro area population of 19 million gives us 295 cases per 100,000--essentially equal to my "magic number" of 300.
Now let's jump to Italy briefly. Here we again have a situation where the rates of infection are (or, disconcertingly, maybe more accurately "were") concentrated in one region: Lombardy. This is where I made my calculation to get the "magic number". Knowing that the hospitals in Italy were overwhelmed by the number of patients and starting to make decisions on who and who not to treat, I divided the number of cases in Italy (31,000 at the time) by the population of Lombardy (under the assumption that almost all the cases in Italy at the time were in Lombardy), and then rounded off to come up with 300 per 100,000 as the point at which medical facilities would be overwhelmed. And this, then, is where people start dying in larger numbers. The way I think of it, there is a certain (small) percentage of people who will die of coronavirus no matter what (current) medical treatment they are given, and a (larger) percentage of people who will not die, given medical treatment. But if that treatment doesn't exist, because so many others are already using it....
I'm aware, by the way, that there are a lot of assumptions. First, that all or nearly all of the cases in Italy were in Lombardy. Second, that the medical infrastructure is similar enough in all these countries. But so far, this number has held out fairly well, and provides a useful tool. Looking to Italy now, we see that Lombardy has 34,889 cases--now their share of cases in Italy is just 43% even though their population is only 16.7% of all Italy's. The infection rate in Lombardy is 346 cases per 100,000. Looking at the rest of Italy, the infections seem to be creeping out of Lombardy.
And what we hear from Italy seems to match the data. Lombardy and its capital Milan are in a grave medical crisis. The South of Italy, however, seems much better off, though the spread outwards from Lombardy has to be disconcerting.
Spain's overall infection rate is even slightly higher than Italy's, though the infection, while centered on Madrid, seems less starkly regional than it is in Italy. Madrid's case rate, at 289 per 100,000, certainly has to be deeply concerning, while Navarre and the Basque region are over 200. (La Rioja is a very small region south of Navarre; I have no idea why they're rate is so high).
Switzerland has the highest rate of all the countries on my list, and I have been hearing a bit about their problems. It is interesting to contrast the rates in Switzerland with those of Austria (not much more than 50% of the infection rate of Switzerland). Is this difference due to Austria's very early decisions to restrict travel from Italy, or more related to Switzerland's closer proximity to Milan/Lombardy?
I think it also worthwhile to watch Israel closely over the next week, with a higher infection rate than the US but much lower fatality rate; Australia, and Russia. Russia sealed its borders very early (leading to real problems of migrant workers being trapped in Russian airports, not allowed to leave and with no flights to return home on), and seemed to be doing well, but now the numbers are starting to climb.
OK, let's look at the US, starting with the chart I use to show the growth in infection rates by showing the number of states with various infection rates. As always, infection rates are measured as cases per 100,000, and each column represents a doubling of the rate of the column to the left.
And now let's look at the list of states:
Notes for the states:
- New Jersey continues to grow dramatically, 56% over Wednesday's numbers.
- Washington state saw a big surge in cases, 23% over Wednesday.
Now let's look at some counties:
Notes:
- Westchester is now at a level twice the "magic number".
- Not much else has changed in terms of "rank", though the numbers continue to grow everywhere.
Why hasn't Westchester's medical infrastructure collapsed yet? Well, I've mentioned before that Westchester shouldn't be considered alone, but as part of the NYC area. Going by the Census Bureau's definition of the main metro area described as New York-Jersey City-White Plains, I've compiled the numbers for NYC proper and the six counties that together form this statistical area. I've created two charts, in the second one I've increased the New Jersey counties' rates by 20% as the state of New Jersey is still not identifying the source county for 20% of their cases. In the end, it didn't make a huge difference.
So the whole metro area is now very close to that "magic number" of 300 cases per 100,000, and we are starting to hear of a serious crisis in ability to deliver care, sharing ventilators, lack of space for those who've died, etc.
We've got a ways to go....
Comments
Post a Comment