Time to update my corona graphs

Well, let’s rephrase that – I’ve been updating my graphs daily but I need to do an updated post on what I’m seeing and that is basically some states are seeing increases in infection rates since re-opening the economy.

The source for these graphs and for all graphs in this post and pulled on 6/13/2020 comes from Wikipedia: https://en.wikipedia.org/wiki/Template:2019–20_coronavirus_pandemic_data/United_States_medical_cases

By the way, I think the ultimate end sources of these data may be coming from one of three typical sources: 1) the state health departments; 2) John Hopkins Hospital; and 3) the Covid-19 Tracking project which is being run by The Atlantic. John Hopkins Hospital says that their source is the Covid-19 Tracking project. And I wonder if the Covid-19 Tracking project is actually coming from the state health departments?

The graph on the left shows the entire U.S. If you include all of the states in the U.S., you would think that the cases of infections are going down (see upper left on the left graph), but if you were to exclude the Northeast, which would take out New York and New Jersey, you would see that in the rest of the U.S. infections are still rising, albeit slowly. Over the last two days, the infection rate seems to have risen a lot more dramatically. The graph on the right hand side shows that the increases in infection rates stem mostly from the South and a bit from the West.

Louisiana is not typically mentioned in the news; I added Louisiana because their cases shot up yesterday and I’m wondering what is going on there. Florida is just rising and rising, pretty precipitously so that is a concern, especially if part of the Republican convention is held there. Arizona and South Carolina, and possibly even Arkansas, are also seeing rather steep increases.

The state that is most interesting is Georgia: why isn’t it rising since it was one of the last state to order a shut down and was one of the first to re-open. The infection rate has maintained a steady beat. But…Georgia was also the state reported to have shown “deceiving” graphs on its website, so I’m wondering if Georgia’s data is valid. If the graphs on their website was deceptive, couldn’t their data be manipulated? Furthermore, this is the state that has claims of voter suppression lodged against it and whose current governor ran for governor while also being the head of some voter commission. And this week their primary voting was a disaster with more claims of voter suppression. So is it any wonder if I question the validity of their data?

But if their data is correct, I’m interested in knowing what they are doing right.

Here’s a graph depicting states with the largest increase of the weekly moving average of daily infection cases since last week. It’s the graph in the upper left showing increases greater than zero. So the top five are Florida, Arizona, Alabama, South Carolina and Texas.

One thing I’m trying to do is find data on hospitalization rates and positivity rates. I haven’t really focused on the hospitalization before because I felt that the hospitalization rates were really a lagging indicator: it takes about 5 days to 2 weeks before someone develops symptoms, if they ever develop symptoms, and then a few more weeks (one or two?) before their symptoms get so severe that they require hospitalization.

However, hospitalization rates are now becoming more important because I want to distinguish when rising infection rates is due to increased testing or due to increasing spread. If infection rates and hospitalization rates are both increasing, then we are facing increasing spread.

I’m a little more confused on the positivity rate: a high positivity rate can mean that not enough testing of the general population is being done – meaning the testing is focused on just those with symptoms rather than being widespread. Here’s where my confusion comes in: what if the higher positivity rate actually reflects a greater percentage of the population is being infected but don’t show signs of infection or are really asymptomatic? I haven’t come to terms with this question in my mind yet.