18 Feb, 8 tweets, 2 min read
B.1.1.7 variant first seen in UK will be dominant by end of March

If we do our job, we can vaccinate everyone over 65 by then

And every teacher

And all school staff

A pipe dream? No – doable.

But hard

Lets look at some numbers to see why I'm optimistic

Right now, we have 26M people who have gotten only 1 dose

They need the second dose

How many folks >65 still haven't received any vaccine?

35 million

Even if we do a 1M first doses per say focused on seniors, it'll take us until late march to get 1 dose into all of them

2/9
We can get 2nd dose into about half (remember, have to wait 3-4 weeks between doses)

So we need 52M doses to get all remaining 35M seniors at least one dose and half their 2nd dose

What about teachers? 3.2M teachers in America

But schools have other staff too!

3/9
Best estimates are about 2M staff -- but if we round up,

6M teachers/staff

Imagine we rush to get all teachers and staff fully vaccinated by 3/31

That’s another 12M doses needed

That’s 90M doses need

This assumes no double counting and zero hesitancy

4/9
Obviously too aggressive

Assumes no teacher has already gotten vaccinated

Assumes no teacher/staff >65 (probably 10-15% are)

And assumes no vaccine hesitancy – that everyone will want it

Realistically? need closer to 75M doses if 80% uptake among seniors, teachers

5/9
How many will we have?

We have 17M doses of Pfizer/Moderna in freezers at states now

And we are close to producing, distributing 2M doses a day

41 days until 3/31 --> 82M doses

So we should have produced, distributed 99M doses by 3/31

This assumes no J&J vaccine

6/9
So we’ll have about 100M doses to administer between now and 3/31

2.5M doses a day

We can administer that if we have supply

And will need 75M to:

1. Finish current folks
2. Give two doses to teachers/staff
3. Two doses to half remaining seniors
4. Single dose to other half
We can finish 2nd doses for of rest of seniors early April

And vaccinate other high risk folks then too

Bottom line?

We should have enough vaccines for high risk folks before UK variant dominant

Need flawless execution

But its doable

And I think we can do this

End

• • •

Missing some Tweet in this thread? You can try to force a refresh

This Thread may be Removed Anytime!

Twitter may remove this content at anytime! Save it as PDF for later use!

More from @ashishkjha

15 Feb
Seeing stories of vaccinated folks getting COVID?

Yup. Not news

How common is this? Quick math says pretty common

Probably 60-100 fully vaccinated folks getting infected daily

So infections no surprise

We should instead focus on hospitalizations & deaths

h/t @jflier
Back of envelope math:

Over past month 4.2M have received both doses while 5.1M Americans have become infected

If you subject each fully vaccinated person to that day's daily infection risk:

We'd expect 8K-10K infections in that group if vaccines didn't work at all.

2/4
If vaccines offered 95% protection

We'd expect 500+ infections

Lots of hospitalizations

May be even some deaths

Headline news of a few infections expected

So what am I look for?

How many fully vaccinated folks end up hospitalized or dead from COVID

3/5
9 Feb
About 3,000 Americans are dying every day

They reflect infections from mid-January

Given dropping infections since, we'll see dramatic drops in deaths over next few weeks

This is great

It'll also tempt policymakers to relax restrictions...just at riskiest moment

Current 7-day moving avg about 3000 deaths/day

In a week, that'll be down to 2500

2000 daily deaths a week later

And 1500 a week after that..3 weeks from now

Best guess based on infection numbers

Nearly 20K fewer deaths over next 3 wks than if infections had stayed flat

2/4
And if infections continue falling, we might even get under a 1000 deaths a day by Mid March!

Which would be amazing!

With one challenge:

Dropping deaths will put pressure on political leaders to substantially lift restrictions

Just as the variants will be taking off

3/4
7 Feb
Am optimistic about late spring and summer

But concerned about next couple of months

So let's chat why

B117 – the UK variant

Great new analysis by @k_g_andersen et al out today confirms as much

Its a problem

medrxiv.org/content/10.110…
B117 is much more contagious – so it can quickly overwhelm a nation

Great write-up about Andersen piece and broader issues by @carlzimmer in @nytimes today

If you want to understand what this variant can do to a nation, let’s look at some data

2/8

nytimes.com/2021/02/07/hea…
First, let's look at pandemic in two nations:

Ireland (blue)

Germany (red)

Until mid-December, outbreaks had looked similar

Ireland had a spike in October did a lockdown, and got it under control

Germany rising in November, December with modest restrictions in place

3/8
5 Feb
Feeling pessimistic about COVID & 2021?

Don't!

I'm not. Here's why

We'll likely have about 400M doses of Moderna/Pfizer by end of June

Enough to vaccinate 80% of adults

And that means a much better summer

Of course, we'll likely also have J&J, AZ, Novavax

So by the time summer arrives, we'll have way more vaccines than people

Including for kids -- they'll likely get vaccinated over summer if not prior

Distribution problems were real in Dec/Jan

They are largely getting fixed

2/3
So I see no difficulties with us vaccinating 2-3M / day as we get into March/April. And vaccinated 150-200M by July 1

What could go wrong with scenario?

Pfizer/Moderna both suffer big production problems

AND

J&J, AZ, and Novavax don't deliver

Seems way, way unlikely

3/4
29 Jan
Heard of SARS-CoV2 variant from Japan?

No?

How about one from South Korea?

No?

Surely variants from New Zealand & Vietnam?

Of course not

Because these places haven't given rise to scary variants

But what places have.....has big implications for ending the pandemic

So let’s talk about where variants ARE coming from and under what circumstances

Variants arise when infections run wild

And selection pressures lead to dangerous mutations that can then thrive

Remember, every infection creates opportunities for “errors” – or mutations

2/6
Most mutations are meaningless

They will have no real clinical implications

But every once in a while, a set of mutations will lead the virus to become more contagious, more lethal, or improve its ability to escape our vaccines

So where are the variants coming from?

3/6
24 Jan
Wondering whether the national COVID data is just a blip or something more substantive?

I think there is something good going on

Cases across the U.S. really are falling...and while its only been 2 weeks, it looks real

Here’s @FT graph that shows drop in new infections

You could look at this and say – hey, we’ve been in downturns before.

True

But last 2 downturns were smaller and both were largely data artifacts (under-reporting during Thanksgiving and Christmas)

This one is different

Is it a big deal? I think so

2/7
Let’s look at data: 7-day moving average from @covid19tracking

US case counts peaked 1/11/21 at new 244K cases daily

We are now down nearly 30% to 173K

Hospitalizations peaked at 131K, now down about 8% to 120K

We know deaths lag so no surprise, deaths are flat

3/7