March 29 US Confirmed Coronavirus Cases

If you look at the statistics for Sunday, it seems like it was a very good day. The number of reported new cases (19, 306) was lower than the new cases for Saturday (19,818). This is the first day that the number of new cases declined over the previous day's in two weeks. As a percentage of the previous total number of cases Sunday's new cases represented an increase of 15.5%, a significant improvement over Saturday's 18.9% and continuing a trend of declining numbers for five days now. The question that we hope to be able to answer in the next two days is whether this is a weekend slowdown (in testing, reporting, collating data) or a real sign that social distancing is working.

Let's take a look at the chart of coronavirus case rates. As always this chart shows the number of states falling within certain ranges of infection rates (cases per 100,000 people), over the last 7 days.


And now let's look at the list of states:


Notes:

• Washington's new infection rate continues to dwindle (assuming no "weekend blip") at 13.6%, but one assumes that with rates as low as 5% last week this is still disappointing to health officials there.
• New York's new infection rate is also slowing, to 13.3%, while New Jersey is growing much faster, at 20.3%.
• Connecticut grew by 30.8%, as the infections spread to the more distant suburbs of NYC.

Let's zoom in closer, to the county level. The image I posted yesterday of spring breakers heading home from Florida beaches convinced me to add some counties in Indiana and Ohio to the list of counties I'm watching.

Notes:

• Adding Indiana was a good idea, as we see that Marion County (Indianapolis) has a higher infection rate than Cook County IL (Chicago).

My list of the ten highest infection rate areas:

1. New York City. Specifically, the NYC-White Plains-Jersey City area.
2. Ski resorts.
3. Long Island.
4. New Orleans.
5. Albany, GA.
6. Bridgeport, CT.
7. Poughkeepsie, NY.
8. Seattle.
9. Bartow County, GA.
10. Shreveport, LA.

Next on the list would be Detroit, Boston, Indianapolis, Chicago, and Albany, NY. Miami would be about 20th, and San Francisco about 22nd.

Rant of the day:

Well, it's less rant than interesting note. I've been using the census definitions of Metropolitan Statistical Area (MSA), Micropolitan Statistical Area (µSA), and Combined Statistical Area, using wonderful maps such as these (1, 2--don't be confused, while similar they are different maps). I'll leave it to the truly nerdy among you to go in-depth on how these different areas are determined (Wikipedia does a good job), but the short answer is that the census studies how interconnected various communities are in terms of jobs/residences etc.

I think that an enterprising researcher could make use of the spread of infections to make some alterations to the definitions of these communities. I've noticed a number of areas in the NYC CSA area where currently-defined MSAs may (or may not) need to be altered. For example, in the counties immediately north of NYC, Westchester and Rockland Counties are clearly pretty heavily tied to NYC and part of its MSA. The county immediately north of Westchester, Putnam, is also in the NYC MSA. But Orange County, to the west of Rockland and Putnam, and Dutchess County, to the north of Putnam, while part of the greater NYC CSA, are in the Poughkeepsie -Newberg-Middletown MSA. However, looking at the counties' infection rates, Putnam County looks a lot more like Dutchess, while Orange County looks a lot more like Westchester and Rockland. New Jersey is even more mixed up. Sussex and Hunterdon Counties have much lower rates of infection than other counties in the Newark MSA, lower even than Dutchess County. Other counties in New Jersey much more distantly connected to NYC by both geography and census definitions have much higher infection rates. This could be due to greater connections than previously understood, or perhaps other reasons (Ocean County could be high because of beach visits, for example).

The point of this is to note that these patterns of infections could provide a lot of data for the ways in which people interact with the geography.