Asked by a friend to comment on the reasonableness of the IHME forecasts for Canada (30+k deaths by Feb). (…). The IHME model is impressive...
And again, based on the 2nd (winter) wave of the pandemic ahead of us, and given that we currently stand around 10k deaths, the projection of 30k deaths by February seems reasonable. Note my earlier tweet about 2:1 ratio of 2nd to 1st wave in 1918/1919.
What's impressive to me in IHME is the forecast that we would/could save 10,000 Canadian lives in the months ahead with a national mask mandate. This, again, seems reasonable, based on best available data.
Mask mandates stand as an easy win. I think we are seeing them bear fruit right now in Ontario and Quebec. While both provinces have high case counts, growth is slower than would have been expected.
IHME, as with any model (our model, for sure) is imperfect. Canadian deaths r strongly driven by outbreaks in LTC facilities and (need to dive more into their supplementary materials) I don't think that can be captured in a nuanced way in a project of that scope.
The seasonal component is also really interesting...we'll only know post-hoc if that's right. But the model stands as a monumental achievement, and I think order-of-magnitude on deaths is correct.
Again, as @enkimute has noted, the spring wave was all about the virus, and this wave is all about us, and the choices we (and our leaders) make. We understand this beast well enough to control it if we choose to.
The new IHME paper that incorporates evolving data on strength of mask effects is available here:….

h/t @AshTuite for helping me wrap my little microbrain around some of this stuff.

• • •

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

Keep Current with David Fisman

David Fisman 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! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @DFisman

25 Oct
Global R(t) has been remarkably stable since the first wave. Here it is plotted against global doubling time. Will try to unpack this further when I get a chance but it’s a good news/bad news story: many infections ahead, but this isn’t open-ended.
“Moving average” = 7 day moving average for doubling time.
Just to unpack this a bit, there's a direct relationship between R and doubling time, inasmuch as R(t) is a function of growth rate, as is doubling time. Doubling time may just be a bit more intuitive.
Read 12 tweets
24 Oct
This is amazing. We're headed into the second wave of a once in a century pandemic. Ont and Qc have been backstopped by the feds. Now here come the Cons with a motion to initiate a massive AUDIT of the public health response? Now? October 2020?…
Like, THIS is when you're going to tie up PHAC, Procurement, Health Canada, not to mention the companies the govt is working with to procure vaccines and tests, with what looks to be the audit from hell?

Because PHAC, Health Canada, and every other federal office/bureau/department isn't already hanging on by their fingernails trying to build systems, structures and responses to protect Canadians?

Now? Really?
Read 10 tweets
13 Oct
This (as with the Quebec karaoke superspreader event) is extremely difficult to explain without invoking aerosol. High viral load in resp mucus of the index case (asymptomatic) created a superspreader event.

It would be extremely interesting if the index case was the instructor: prolonged shouting/exhorting during a spin class would make sense in terms of aerosol that info out there?
Events like this are literally showing us the gateway out of this situation. It is distressing to see that institutional public health across Canada is still reluctant to acknowledge the importance of aerosol: costing time, money, lives now.
Read 4 tweets
12 Oct
I had some wonderful back-and-forths with @aatishb about this and it’s fantastic to see the finished product on “math of masks” at
There is a wonderful YouTube video here:

If you want a mathy-er version of same, we have an approach using next generation matrices which you can access here (and which we’ll cover this week in chl5425 at @UofT_dlsph)…
Read 4 tweets
10 Oct
Thanks to the friend who reached out to suggest that having my own dedicated troll army suggests that our messaging and information are effective. That's a nice shot in the arm.
With that in mind, I'll use this account to disseminate good info on what's happening with covid...and hopefully drive the trollbots crazy.

Let's start with covid in Canada today:
R(t) by province...note New Brunswick is the place with a qualitative change in status, and is few days into a LTC outbreak after a few months of elimination of local trans.
Read 8 tweets
7 Oct
@LeaferReefer_ @BogochIsaac Getting to herd immunity via disease transmission vs. getting to herd immunity via vaccination are very different.

For disease transmission, you cross the susceptibility threshold for transmission when fraction immune = 1 -(1/R0). Let's say R0 for covid is 2; that means you...
@LeaferReefer_ @BogochIsaac hit that threshold when (1-(1/2) = 50% of the population is immune. In Canada, that's at 20 million infections.

x 0.7% IFR = 140,000 deaths.

And at 50% immune, R = R0 x S = 2 x 0.5 = 1. R hits 1 at the PEAK of the epidemic you still need to come down the other side
@LeaferReefer_ @BogochIsaac of the epidemic curve. As these curves are somewhat symmetrical, once you hit that threshold, you still have 10's of thousands of deaths to go, and the final epidemic size is much greater than 50%.
Read 9 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

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

Become a Premium Member ($3/month or $30/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!

Follow Us on Twitter!