My Authors
Read all threads
Update COVID April 9: The big conversation is around the 60,000 death toll model so I had confidential conversations with the modelers. Can share without attribution.

Also a palliative care doc walked me through a day with a COVID patient day.

1 more Senate call. Soon.1/
Trapped on a call (about nursing home safety so I need to stick on it). Will be tweeting in about 25 minutes if interested, feel free to follow.
There is no topic more sensitive than what the death toll will be from COVID-19. And we feel like we get bounced around w new projections.

And the different numbers cause people a variety of factions from “it’s working” to “see this is overblown” to “big numbers are alarmist.”3/
So I want to give my view before I reveal my conversation: the numbers in the model are important in helping us understand what is working. They should not be seen as projections.

They also make assumptions about human behavior that can’t be known (as you will see). 4/
State of things— Trump had said previously that the death toll was headed to 100k - 200k. Scientists in the WH are still holding that. But a new model the WH and many people pay attention to are now saying 60k.

There is also a secret “Grand Princess” model I will touch on.5/
Trump has said the numbers “could have been” 2 million. And the Imperial College had put out higher numbers.

First question— if it comes out at 60 or 100-200, how could ppl have been so wrong? Or if it comes out higher, how could ppl be so wrong? 6/
The answer? That is exactly how exponential math works.

If 1 person infects 2.3 people on average, after 10 cycles 4100 are infected. But if one person infects only 1.3 people on average, only 14 people are infected. 7/
This is why we architected they #StayHome campaign. Because the difference is dramatic.

It’s why people look foolish to 2nd guessers. When Cuomo said NY could need 30,000 hospital beds, he said “I hope they call me alarmist.”

NOBODY predicting higher number wants to be right.8/
People like this in this thread. Very well meaning individuals. 9/

Here’s what I can tell you about the 60,000 projection.
-The first projection was on 1 city. Now it includes 16, including European cities. More data w sustained social distance.
-Assumes states that have not done social distancing don’t have outbreaks.

10/
-Assumes no cross state travel
-Assumes R0 goes to below 1. Meaning instead of 2.3 or 1.3, it is less than 1. That would crush the virus.
-Assumes we don’t let up social distancing. Critical assumption
-Assumes states don’t open until <10 cases 11/
-Assumes all this is over in August.
-Assumes no international travel
-Assumes no bed overruns 11/
There is a lot to criticize in any model. And while a lot of the President’s supporters view this as a sign that we’ve won, others can find a lot to critique.

Before I get back to the convo, I will tell you some of the critiques I heard. 12/
To be fair, you have to assume something in your model. Criticisms that I heard:
-Model not dynamic. Just runs multiple cases
-Already proving wrong in NY
-Steady state assumption on length of #StayHome not viable. 13/
-Assumes no 2nd wave comes before a virus. (When I asked what he thinks herd immunity is modelers think 5% so this is tough to imagine)
-Assumes a lot of testing is ready to go that isn’t

So those are the criticisms. And if the criticisms are right, then 60k will be higher.14/
If the model shows numbers that go down, that is a good signal that our actions are working. I think the right view is that if we sustain current activities we are saving lives. 16/
The other consideration in the model is that people are highly compliant. Separate person told me this:
-60% of people in Queens are compliant (that concerned me)
-33% in Montgomery, AL are compliant (that frightened me)

17/
Must pause for 15 minutes. Sorry. BRB.
*Zach points out that tweet 14 should say “vaccine” not “virus.”

Where was he 10 minutes ago?
Other models with different assumptions— one says 90,000. Some 100-200k. The Diamond Princess model (which is a straight line assumption of what happened on the ship says 400k). 17/
Again, slight changes in our behavior or relaxation of laws could make the number much higher.

Being 10% off could make you 1000% wrong. People might want to refrain from too much criticism.

Never mind, this is Twitter! Who am I kidding? 🤣 18/
Other topics today if you want to keep going. A national security conversation, conversation w a hospice nurse, a foos bank, how the projectn95.com work is going, NY Comfort/Javits.

Drop at any time, but I might as well get this out tonight. 19/
National security outlook was more an assessment of other factors & long run implications. Three interesting topics:
1- Debt & spending
2- China
3- Regional recovery

20/
1. Debt & spending: During the peak of WWII, our government spent 40% of GDP. We’ve been much lower ever since & and are now back up to 35%.

For reference GDP last year was about $20 trillion. Between what we will spend & how much GDP will drop, we will break that record. 21/
What’s a foos bank? Zach?
2. China. Many national security folks are upset with China. Republicans more hawkishly but Democrats as well (former diplomats & natl security folks). I’m not smart enough to evaluate the ins & outs but the forecast for the world is...21/
National security implications of CV:
-More nationalism (not just from China, but from us)
-Less trade more in-country manufacturing
-Much tighter borders
-Continued tariffs
-Regional hegemony strategies
-Weakening if WHO, UN, global institutions 22/
On that last point Dems felt there would be more cooperation but acknowledged the other points. The Rs had a stronger reaction. One said “There is no prize for membership.” 23/
3. Regional recovery: We are dividing East v West instead of North v South was the thesis.

West Coast handled this much better than East. Easier to contain, easier recovery. Southeast expected to be quite bad. 24/
My own observations from MN. We are partially surrounded by the Dakotas & Iowa. I heard today that entire meatpacking plants have 100% infection rate. That hospitals are seeing high infection rates among staff. And no social distancing standards. 25/
There are a handful of stubborn governors still in these states. And if it gets bad, unlike TX which has said they want to stop Louisiana people at the border, I hope MN helps by inviting folks in if they need help & we can do it safely. 26/
Speaking of parts of the country, someone sent me regional riffs on the #StayHome effort.

What would your state’s motto be? @somuchweirdness says Minnesota’s would be “Just stay where you are. Why would we get closer than 6 feet?” (not a warm people) 27/
(Focus, Andy, focus)

Talked to a hospice doc today (if I said nurse earlier, it was Zach’s fault). She is doing all her conversations via iPad. Talks to patients. Talks to families. Via iPad.

I asked her how to prevent ppl from feeling alone as they die wih no family there? 27/
It’s hard. There was no sugarcoating. For the family, no mourning. No confirmation of death by seeing the body which is part of the grief process. No visitation from neighbors.

And the family is often in isolation themselves As I heard from someone today. 28/
I will cover the food bank, the Javits Center situation, and project95 tomorrow since this is quite long. I will end with one thing. 30/
Earlier in the thread I had to pop off for 15 minutes. Got this from one of my oldest & closest friends. All our parents were like parents to all of us.
Kids live in different states. No way to mourn.

One day we must remember all of these people properly. end/
New ending: I don’t want to end on a sad note but a resolute one.

Whatever the “number” is, it’s far too many and we can beat it. /realend
Missing some Tweet in this thread? You can try to force a refresh.

Enjoying this thread?

Keep Current with Andy Slavitt @ 🏡

Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!