"better data along does not guarantee better outcomes"

Especially when results coming from such data are actively dismissed and purposefully obfuscated.

On Feb 12 AB had enough data to be *certain* that w/o extra measures it will produce the 3rd wave.

1/
On Feb 12 AB had enough data to be *certain* that w/o extra measures it'll produce the 3rd wave.
It was *clear*.

Thank you everyone who contributed to the 3rd wave by saying things like: "models are wrong".
Even w/ limited data, but knowing the properties of the phenomenon, models predict the future darn well.

On Feb 01 I knew, that if we won't implement measures, there'll be a 3rd wave.

On Mar 17 I could bet my soul, left kidney, and a collection of stuffies on that.
Yet, in Mar I could hear such statements: "If I'd have to guess, I'd say that there is 70:30 chance that there won't be a 3rd wave".
It is not a benign statement. It implies that pandemic evaluation is in a domain of foggy guessing and not of a sharp scientific prediction.
..
4/
It'd be silly to say "If I'd have to guess, the next eclipse will be on Sunday", bc there are astrophysicists who know very well when the solar eclipse will be. No need for random guessing.

Likewise, modellers do *know*. Respecting their knowledge and expertise would save lives.
Among others @CarolineColijn , @COVIDSciOntario , PHAC modellers, predicted the 3rd wave.

Those predictions had been dismissed, ignored, sometimes ridiculed.
Canadians and Albertans are paying for this dismissal with their health, lives and economical well-being.

6/
We cannot change the past.
But pandemic is not over.
Dear decision makers and decision makers' advisors, please start listening to modellers and mathematicians. And please do not mislead the public about the role and function of scientific method.

7/
And excellent simulations by @TheMemeticist predicted the 3rd wave as well.
Try running the simulation several times. Very educational about stochasticity.



trinket.io/python/a2c9985…

• • •

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

Keep Current with Malgorzata (Gosia) Gasperowicz

Malgorzata (Gosia) Gasperowicz 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!

PDF

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 @GosiaGasperoPhD

1 May
Alberta: Current measures made the growth rate less extreme.
B117 is doubling now every 20 days instead of every 7 days.
It's still very fast - similar to doubling time in the 2nd wave (18 days)

Current measures are insufficient to bend the curve downwards.

Ln scale graph 👇
At this rate we can expect:

3,000 daily new cases on May 12
4,000 daily new cases on May 22

2/
This wave is worse than the fall one.

2nd wave: at 1877 daily cases we had 100 ICUs
3rd wave: at 1860 daily cases we have 146 ICUs, and *we are still growing*

That's *despite* many people being already vaccinated. W/o vaccines it would be even worse.

3/
Read 7 tweets
27 Mar
Alberta could see the same number of new COVID-19 cases by mid-April as there were at the height of the second wave in December.

~2000 by April 19,
~4000 by April 29

Thank you @laurby for interviewing me for this article.

1/
edmontonjournal.com/news/researche…
Projection from 2 days ago. Turns out conservative.

Today's actual total cases: 717 (projected: 563)
Most recent B117 doubling time: 7.3 days (faster then on this graph)

2/
W/ update: calculated today, the B117 doubling time is 7.3
R value: 1.46

3/
Read 11 tweets
19 Mar
B117 grows exponentially in AB *since 3 months*.
Doubling time as of Mar 14: 11 days, R=1.29

That’s fast. During AB fall surge cases were doubling every 18 days.

At this rate AB may have:

Apr 19, 1000 daily new cases,
May 03, 2000 daily new cases

1/
Future effects of reopenings and vaccinations are not included in this projection.
Reopenings will speed the growth up.

Vaccines…

2/
We don’t know exactly how effective the vaccines are against transmission.
Let’s assume 60% effectiveness:
Then, in order to break the exponential growth of B117, i.e. to reduce its R from current 1.29 to 1.00, 38% of people would need to be vaccinated.
We are not there yet.

3/
Read 8 tweets
13 Mar
Alberta:
Announced* new B117 cases seem to be doubling every 2 weeks.

At this rate AB could have:
1000+ daily cases around May 10,
2000+ daily cases in late May

*Please read footnotes under the graph.
1/
It is not doubling as fast as it could, e.g. not every 7 days.

But doubling every 2 weeks is still *very fast*.

It's *faster* than Alberta's November surge, when cases were doubling every 2.5 weeks.

2/
In theory vaccine could flatten such growth, if 40% of Albertans would get vaccinated and if the vaccine would be 60% effective against transmission.

As of today, 5.5% of ppl in AB received at least one dose.

Meanwhile, reopenings will likely increase the rate of spread.

3/
Read 4 tweets
13 Mar
A great simulation comparing the spread of variants over ~6 weeks, starting from 100 cases.
Wild Type R = 0.7
Assuming B117: 50% more transmissible, P1: 70% more transmissible than the Wild Type.
R=0.7 is what AB had at the end of the shut-down last spring. Wild Type daily case count was halving every 8 days.
Under that conditions P1 would still spread like a wildfire.
Please play/run the simulation few times.
Stochasticity (randomness): Each time the result is different, but Wild Type always goes down, while P1 in vast majority of runs shoots up super fast.

Read 9 tweets
12 Mar
“Public Health Director Horacio Arruda has said that vaccination will make COVID-19 less severe and that we could tolerate up to four to six times more cases if they were less severe. Yet if there is more community transmission, there will be more variants, more long-term COVID..
“...and the possibility of more long-term effects among the younger population.
...
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!

:(