(8/22/20) I know this sounds alarmist, but the U.S. for all its advancements, wealth, world class health services and other resources, will soon be the deadliest major nation per capita in the world other than Brazil.
Let me cut to the chase and point you to the tables below, which I think are an effective dashboard for comparing Covid-19 outbreaks in different populations. I believe in reporting per capita data and these are for total deaths and death rate per million. The former represents the magnitude of the problem and the latter on how it is trending. You can see that the U.S. in about a month will exceed every major European country for deadliness. That was unthinkable just 2 months ago. Within the states NY and NJ were the worst hit, but have recovered, as have all of New England. Of the more populous states, the most serious outbreaks with >30 weekly deaths per million people are FL, TX, AZ, LA (shown in table below), GA, TN, NC, SC, AK, MS, and NV. This compares to CA at 21 weekly deaths per million.
Preface
This may be my last Biweekly Update (which have been more like monthlies lately) for a while. I just don’t have enough time and the reporting in the media has gotten much better. I have to single out the NY Times as having excellent coverage. You should subscribe on-line for I think $1/week. Also, they have extensive data collection and place all their statistical files on the Johns Hopkins GitHub website, a repository for geeks sharing and doing good things for society. This has been great for me since the Administration recently yanked hospital data reporting away from the CDC and I can’t find the black hole where it got redirected.
Before we launch into our customary update, I wish to note that I have changed the plots to bar charts vs. lines through points as that is the emerging standard for reporting by government health organizations, model trackers and the media. Further because the shapes of these death rates, with so many rebound outbreaks, make a Gaussian function an ill fit, we have switched to using a 5th order polynomial to fit to the data. This is purely for visual effect and doesn’t have much forecasting value.
The Local California Scene
We will compare CA to the rest of the nation shortly (though you got a glimpse above). But first we cover the Los Angeles and Orange County statistics. Below are the hospitalizations for LA and OC for all occupancy and for ICU. Below that are the death rates for LA and OC.
Hospitalizations are trending down, but deaths are not. We expect a 1-2 week lag in deaths relative to hospitalizations, but we are now past that time so there is a little mystery here. The table below shows the percentage of hospitalizations that led to death 10 days later and you can see over the last two weeks this number has increased significantly in LA and OC. Given that this number has been relatively constant throughout the pandemic it would not appear to be a statistical quirk, but we’ll have to follow it out a couple of more weeks to be sure. Because it is hard to imagine this is due to poorer care.
Deaths will either drop very soon or we must search for another explanation. It is possible that the drop in hospitalizations may be due to shorter lengths of stay giving a lower average bed occupancy, while still attending to the sickest that are still dying. However, the ICU statistics mirror the overall occupancy suggesting the ratio of less to more sick is not changing much. The next few days will be interesting to see if the predicted drop in death rate occurs. On the whole I find the sustained reduction in hospitalizations very encouraging news for LA and OC.
Below we show the death rate plot for CA. As much as the rebound outbreak throughout CA was disheartening to see, and a result of premature social easing, it appears that this second peak is passing. As we see next, despite how the national news reports things, CA is not doing nearly as bad as many other hot spot states (also evident in the first tables above).
The Nation
The biggest hot spots of the more populous states continue to be FL, TX, AZ, GA, NC, TN, SC, AK, MS, NV and CA bringing up the rear. Here we show the daily death rates for early hotspots that recovered (NY, NJ, MI), early slighter hot spots that are having rebound outbreaks (LA, CA, WA) and then early relatively unaffected states having recent surges (FL, TX, AZ).
The major observations are:
- The hottest of the hot spots have fully recovered. This is evident in the death rate plots for NY, NJ, and MI, but also true for all of New England. This may be attributed to the fear factor of being in the middle of a firestorm and getting extremely serious about preservation. However, I wonder if this is not also due in part to some level of herd immunity. I will report on my theory on this in my next Daily Rumblings.
- The current hot spots got off relatively light during the first phase of the Covid-19 outbreak and these states may have gotten complacent. Despite seeing the carnage in the Northeast there seemed to be a “can’t happen here” mentality. Well the virus does not discriminate by state and these states suffering now, for the most part, never enacted serious social distancing measures and, in many cases, even shunned the use of masks. Humans are slow to learn.
International
Here we look at how the U.S. is doing as a nation and compare to the rest of the world.
The biggest hot spots in the world, for the more populous countries, in approximate order are Brazil, U.S., Mexico, India, Peru, South Africa, Columbia, Iran, and Argentina. Russia should be on this list, but they are under-reporting as may be other countries as well. As we did for the U.S. states above, we show the daily death rates for early hot spots that recovered (Italy, Spain, France, U.K.) and then countries that are having rebounds or first-time outbreaks (U.S., Brazil, Iran, and worldwide).
The major observations are:
- Similar to the U.S. states the hottest of the international hot spots have fully recovered. This is evident in the death rate plots above for Italy, Spain, France and U.K. A fear factor is probably the most likely explanation for the recovery, but again one has to wonder if there is some herd immunity to make recovery easier. Though that can never be enough especially to overcome excessive easing. It is worrisome that Spain has abdicated federal response and given control to regional governments, which could bite back.
- As for some U.S. states, many countries started off relatively lightly affected during the first phase of the Covid-19 outbreak only to be caught unprepared (or unwilling) to initiate an intervention when the inevitable wave hit. This is clearly the case for Brazil and arguably the U.S at least at the federal level.
So, there are many lessons to be learned; the above data shows that the variability of how outbreaks unfold has parallels in all populations, U.S. and the rest of the world a like. There are countries that got the initial full brunt of the outbreak as it hit without much warning. They developed fortitude and knocked it down. Around the rest of the world there was the delusional notion that it couldn’t affect them, but wham, it did. Surprisingly, these U.S. states and foreign countries ignored the first wave and were ill-prepared or unwilling to intervene when the inevitable hit. I raise the controversial notion that if the U.S. death toll wasn’t disproportionately afflicting minorities would our Administration take it more seriously.