Thread by @thomasafine

According to the CDC, there are a 170,731 excess deaths this year, and the data only goes through July 11th.
cdc.gov/nchs/nvss/vsrr…

That was calculated by comparing with expected deaths. If you calculate against their maximum feasible deaths, you still get 139,910 deaths, which is still above the 127K recorded COVID deaths through July 11th.

This is not the only CDC data that shows that we have vastly under-reported COVID-19 deaths. Using an alternative approach, we can calculate excess deaths from just pneumonia and flu (that were not counted as COVID), compared to 2018.

27,924 excess deaths over 2018.

But wait, maybe 2018 was an unusually good year for pneumonia/flu (it wasn't) or maybe this year was unusually bad (it wasn't). Fine. Let's compare to the worst pneumonia/flu year (by state) of the past 20 years of data.

We still find 16,688 excess deaths.

By comparing to the worst year for each state, that 16,688 number is unrealistically soft.

Even the higher number of 27K deaths is unrealistically low for several reasons...

We started out this year before coronavirus with some very mild flu death numbers compared to other years. This should have been a low year.

Especially if you also factor in all the disease protection from our social distancing since the pandemic became a big issue.

Also, and this is scary, we're comparing 7 months of this year's data to full previous years. These numbers should only be 58% of historical numbers at this point. There should be a deficit right now. And this number will only get worse.

Also, we're not looking at other COVID deaths that could be missed, like heart attack or stroke. Just pneumonia (the flu cases are a tiny portion of this result).

So the pneumonia-specific data and the overall CDC excess data corroborate each other. We are missing lots of deaths from COVID that have not been counted.

Let's be clear about the denier's arguments. They claim we're miscounting normal deaths as COVID. But if this were true, our all-cause mortality would not be 170K deaths higher than normal. And we wouldn't have excess pneumonia deaths, we'd have less pneumonia deaths.

It took some time sifting through all the CDC data to come up with these numbers, and I'll be appending a description of how I did these calculations later on.

But the CDC, this is their bread and butter. Why aren't they screaming this information from the rooftops?

We are six months into this mess, and I am some rando software engineer on Twitter, telling you what the CDC ought to be telling you directly. But they don't because Trump can not stop lying to protect the remnants of his flagging reputation.

Get this out there. Retweet it. Reproduce it. Steal it and call it your own. I don't care. The important thing is that we start dealing with the reality of what's happening around us, instead of Trump's delusions.

Notes:
• Data posted a couple of days ago was incorrect for New York because New York City was not included.
• the 170K total excess deaths is through july 11th. We've had roughly 10K new known COVID deaths since then.
• The pneumonia/flu data that I used was through July 18.

For the total excess deaths, go to the link in the first tweet and you can hover your mouse over each bar to see the expected number, the maximum threshold, and the actual number of deaths. The rest there was simple transcription and addition.

The pneumonia and flu excess deaths were much more complicated to calculate.

I started with Table 2 from this web page. I worked with data that I copied form this page a couple. of days ago, which went through July 18th.
cdc.gov/nchs/nvss/vsrr…

By subtracting column 2, just covid deaths, from column 8, COVID, flu, and pneumonia deaths combined, you can calculate all pneumonia and flu deaths that the CDC did not count as COVID.

To make historical comparisons you need to properly fill out the search form for CDCs Wonder Search, here:
wonder.cdc.gov/ucd-icd10.html

So follow these steps:
First Agree to the rules.
In Section 1, set "Group Results By" to "State".
set the first "And By" to "Year".
In Section 2, selecte "States", and then select "All" in the Browse tab.

(continued...)

In Section 6, select "J00-J98" and click "Open".
Now select "J09-J18 (Influenza and Pneumonia)"

Now click "Send" and you'll get results that look like this:

Note that this "J09-J18" designation is the same as used in column 8 of Table 2 of the "Daily Updates" web page where I got the pneumonia and flu deaths. (That column also includes "U07.1" the designation for COVID-19 deaths)

This is a lot of data to crunch through so I used perl scripts to do initial calculations, and Tcl/Tk scripts to make the pretty graphs. The separate scripts also helped cross check the results for me.

Note that none of this would have happened without @NoMoreBS2017 bringing the Daily Updates web page to my attention, and then when I still didn't quite see how to interpret it, by directing me to the WONDER search and historical data.

Thank YOU Ryan.

https://twitter.com/NoMoreBS2017/status/1286419562997702657