Coronavirus Sars-CoV-2/Covid-19 Megathread

[Valmy]

You guys should change your motto as E pluribus unum is not relevant anymore.

Looking out for number one! would be more appropriate. Or as the great Kevin Sorbo recently put it on twitter:

"I don't know you, my lunch is more important to me than you are."

Perfetto!

Well it is the union of many states into one country, not the union of many assholes into one borg.

[Zoupa]

First published: 31 October 2020

Data set ends in February does not mean that on March 1st all researchers were on this data parsing it, analyzing it and had a report published by March 2nd.  This isn't Star Trek or some other sci-fi were a deadly plague within 1hr.

So, Zoupa (and others) point stands: IN HINDSIGHT, MISTAKES WERE COMMITTED.

I seriously doubt that a complete lockdown could have been avoided though.  Just look at Ontario's 3rd wave, they had to close outdoor spaces too at some point to preven people from traveling from area to area.

Nope: final publication through peer review was October 31, 2020.

Look at the second link: prepublication on that site was May 4, 2020. Could have been earlier elsewhere: not sure.

The data was available to the scientific community by early May 2020 at the absolute latest.

And we all know Governor Kemp scours scientific litterature before deciding public health measures. In fact he's so in-tuned to it that he presciently opened up your state in April, before that one paper was published.

[alfred russel]

First published: 31 October 2020

Data set ends in February does not mean that on March 1st all researchers were on this data parsing it, analyzing it and had a report published by March 2nd.  This isn't Star Trek or some other sci-fi were a deadly plague within 1hr.

So, Zoupa (and others) point stands: IN HINDSIGHT, MISTAKES WERE COMMITTED.

I seriously doubt that a complete lockdown could have been avoided though.  Just look at Ontario's 3rd wave, they had to close outdoor spaces too at some point to preven people from traveling from area to area.

Nope: final publication through peer review was October 31, 2020.

Look at the second link: prepublication on that site was May 4, 2020. Could have been earlier elsewhere: not sure.

The data was available to the scientific community by early May 2020 at the absolute latest.

And we all know Governor Kemp scours scientific litterature before deciding public health measures. In fact he's so in-tuned to it that he presciently opened up your state in April, before that one paper was published.

Kemp is a dumbass. He got elected governor by proving he was the most pro gun republican in the race by pointing a gun toward a 16 year old in a campaign ad.

He opened up businesses in the state on April 24 (for the most part), so I could go to indoor gyms, but most outdoor climbing in the state was still closed. Which of course was ridiculous on a lot of fronts but he was probably mostly responding to business lobbyists.

I don't know why that is really relevant to anything. By the end of 2020, before vaccines were rolled out, basically everywhere had reopened gyms and outdoor parks. If I'm evil for going to the gym, maybe there should be a roll call of people that used indoor spaces like gyms and restaurants before the vaccines were widely available. It can be our come to Jesus thread.

[DGuller]

I assure you that I haven't gone to any indoor gyms during the Covid outbreak.  I even had the foresight to avoid indoor gyms for a 20 year period immediately preceding the Covid outbreak.

[alfred russel]

For whatever it is worth, here is a version of that study that says the preprint not through peer review was posted April 7, 2020.

https://www.medrxiv.org/content/medrxiv/early/2020/04/07/2020.04.04.20053058.full.pdf

[Tamas]

Ok so if you guys can detach yourself from Dorsey for a moment, you are a well-informed lot, do you know if:

1. Can you have just a partial loss of smell from covid, meaning you can still smell some things?

2. Is there a test that can show if you had covid like a few weeks ago, even if you have been vaccinated previously? I am assuming antibody test then is useless for that purpose.


I am asking because my Dad apparently (he didn't bother to tell anyone until now) has lost most of his ability to smell weeks ago, and roughly at the same time his mild-ish coughing (related to his heart rhythm problems according to his doctor) has become noticably worse.

So, my theory is that he caught it a few weeks after being double-Pfizered and thanks to that haven't had worse symptoms like that. Coincidentally, my mother has also been feeling a bit under the weather and somewhat worse with her heart rate and such (she has also been fully vaccinated), and -although this is easily just trying to find a connection- my grandma they live next door to also had a period of feeling worse than usual 2 or 3 weeks ago.

[Zanza]

China is now vaccinating at a rate of 1.4% population (or 20 million people) per day, much faster than the Western vaccination campaigns.

[alfred russel]

China is now vaccinating at a rate of 1.4% population (or 20 million people) per day, much faster than the Western vaccination campaigns.

The major mass vaccine site in Atlanta (where I got vaccinated) is shutting down because no one is showing up anymore. Enough people just don't want to be vaccinated. There should be a stick to get them into line (limited access to public facilities if not vaccinated, such as no air travel or enrollment in colleges).

[Zanza]

China is now vaccinating at a rate of 1.4% population (or 20 million people) per day, much faster than the Western vaccination campaigns.

The major mass vaccine site in Atlanta (where I got vaccinated) is shutting down because no one is showing up anymore. Enough people just don't want to be vaccinated. There should be a stick to get them into line (limited access to public facilities if not vaccinated, such as no air travel or enrollment in colleges).

Looking at the trend lines in the FT vaccination tracker, the US seems to reach this point earlier than most other countries though.

https://ig.ft.com/coronavirus-vaccine-tracker/?areas=gbr&areas=isr&areas=usa&areas=eue&cumulative=1&doses=total&populationAdjusted=1

Sheilbh posted some extremely high figures for the UK (all well north of 90% regardless of age group).

[Barrister]

Alberta is moving to more rapidly expand providing second doses to people (we prioritized giving out first doses at the expense of delaying second doses), but it seems like they're worried they might not hit their goal of 70% first dose vaccinated in order to fully open up.  Ages 12+ we're at 67.8 but numbers are definitely slowing.  If you dig into the numbers the teens and 20-somethings are only at 50-some % first dose vaccinated, whereas if you go 60+ they're at 80% or more.

So they might bring out the vaccine lottery.

[Savonarola]

So I've gotten shit for saying there is a significant component of media driven panic in the covid response, but maybe it is worth also looking at how much of that is driven by biases from entities like the CDC?

The CDC puts out projections here:

https://www.cdc.gov/coronavirus/2019-ncov/science/forecasting/forecasting-us-cases-previous.html

The last projection for a time period now concluded was on May 12 -- they were projecting cases in the US through June 5. The actual 7 day average at that point was 98,861. That is within the 2.5% lower band of their model which was 83,780 but dramatically below their midpoint and below all 23 separate model projections they ran.

Okay so they were at the low end. Shit happens.

It turns out they seem to be between the 2.5% and 25% lower bands in every projection they've put out since April 21. In the April 21 projection, they were actually below the 2.5% lower band!

I'm giving up back checking every set of projections, but decided, "okay when cases spiked at their all time max, how did the CDC do"? The US hit its all time peak in cases on January 9. The new weekly cases at that point was 1,777,615. The projection ending in that period was December 17. Unfortunately the data downloads at that point are in deaths rather than cases, but eyeballing the chart it looks like that may be just below the projection they had.

I've done predictive modeling before; when working for the phone company.  Even in that much simpler example it was hard to get right.  There's a series of assumptions that I had to make, and there was a greater penalty for being too low (loss of customers) than too high (additional cost.)  Consequently my estimates were almost always too high. 

The CDC has many more variables to work with than I did and (unlike me) either has no historic data, or historic data that may not be applicable.  So, to me, it's not surprising that their modeling was off and was too high.  This data was used to drive allocation of resources and if they had been too low the consequences would have been much worse.

[Sheilbh]

Sheilbh posted some extremely high figures for the UK (all well north of 90% regardless of age group).

Yeah - there's no sign of demand stalling here. I wondered if the US was running out of people willing to be vaccinated when they were announcing anyong could get the jab while we were still working through people in their 40s - and I think that's probably what has happened. And we'll massively overtake Israel too because their population is so much younger - about 80% of the UK adult population have had one does (and 55% have had both doses) but that still leaves millions while Israel's basically done their adult population already.

From what I've seen take-up is really across Europe even in countries that polled more vaccine-sceptic, like France - which is great news. I do worry that at those numbers the US won't reach herd immunity through vaccinations, there will communities that are very vulnerable and there'll be more deaths. And I think it's the fault of the right-wing media and politicians casting doubt on vaccines for - I don't know - reasons.

Having said that - the UK is likely to be postponing the final easing of restrictions (basically getting rid of all restrictions) by one month. The reason is the new delta variant has become dominant in the UK. John Burn-Murdoch's done a very good thread as ever on this:
https://twitter.com/jburnmurdoch/status/1403423583641210880

Core point is that deaths are still very low - but cases are growing exponentially and hospitalisation and ventilation numbers are on their way up Hospitalisations and ventilations are growing more slowly than in autumn so the link has been weakened through vaccinations, but not broken utterly. And the vaccinations have clearly had an effect among the priority groups who are more vaccinated - most cases and hospitalisations are among the young:


And the expectation is this will have an impact on the number of deaths:


The second dose also has a bigger impact against the delta variant than the UK variant so they are being sped up. But, again, we are basically in a race of vaccines against virus. The postponement means that we should have 5 weeks before the restrictions are lifted - and schools will go on holiday which may provide an additional automatic circuit breaker. But this wave is overwhelmingly affecting the young who are far lower risk, but not totally safe. So we need to get the vaccine to everyone ASAP - I think on current trends it should be do-able by mid/end-July - but this is why I'm a little bit reluctant to see travel re-open too quickly.

As in January - for the good of the rest of the world discourage Brits from visiting

[Josquius]

Summer is going to be painful.
Everyone in Europe travelling again and enjoying the benefits of the vaccine whilst Britain plays at being East Germany.

[Sheilbh]

Summer is going to be painful.
Everyone in Europe travelling again and enjoying the benefits of the vaccine whilst Britain plays at being East Germany.

Because we weren't like East Germany quickly enough

I think the failure to red list India early (for political reasons - Johnson wanting to do a foreign trip) is up there with the autumn lockdown as a huge decision that the government got wrong and, like the second wave, this is on them. This variant appears to be more transmissible than the UK variant and more dangerously. Luckily the most vulnerable were largely vaccinated by the time it arrived here. If we'd managed to keep it out maybe we could be more open now - but the government delayed red-listing India for a month and now we have the consequences

[alfred russel]

I've done predictive modeling before; when working for the phone company.  Even in that much simpler example it was hard to get right.  There's a series of assumptions that I had to make, and there was a greater penalty for being too low (loss of customers) than too high (additional cost.)  Consequently my estimates were almost always too high. 

The CDC has many more variables to work with than I did and (unlike me) either has no historic data, or historic data that may not be applicable.  So, to me, it's not surprising that their modeling was off and was too high.  This data was used to drive allocation of resources and if they had been too low the consequences would have been much worse.

Your estimates weren't objective reports: they were biased in order to meet the needs of management.

There are a few problems with the CDC biasing its models - beyond the obvious one that overreacting is hardly without risk--shutting down schools for instance is not consequence free.

-if it has an eye to the political outcome, that should be beyond the scope of their agency (to the point you can't really say "trust the science" if the scientists are pushing an agenda),
-you can't say "trust the experts" and then see them put out projections that lay people with an hour or so to sift through data can beat. What prompted me to look at this was reading Nate Silver: he polled his audience on what cases would be vs. the cdc projection: his audience did far better than the CDC.
-There are unforeseen impacts to putting out bullshit science that may initially appear harmless.