COVID-19 (Wuhan Novel Coronavirus) Pandemic Master Thread

[PlatinumGuy]

 

Here is data suggesting somebody who gets infected after being vaccinated is much less likely to know (less symptoms + less testing motivation), and thus much more likely to transmit it (other than a difference in live viral loads)

I think there is a big mistake here. Day 7 after dose 1 will lower the probability of testing positive that happens from an infection ~5 days prior but much less impact on hospitalizations that mostly are from infections 10 days prior, before the vaccine

I'm not understanding the math. 50% get vaccinated, and 50% are still taking precautions. The ones who may be more likely to spread it are the ones who are vaccinated, who are less likely to get infected to begin with. If they aren't infected, 3x doesn't mean anything. The other 50% are still wearing masks, and their chances of coming in contact with an infected person are now much lower, so I don't see how the risk of spread would be higher. And again, we're not even going to try to quantify how much a potential lesser viral load could impact this.

You’re confusing recipients with spreaders. If your goal is to curb spread by eliminating the spreaders, and half of those you vaccinated can still get infected, and transmit it 3x as much as they would’ve done without being vaccinated (less symptoms and less tests so won’t know to isolate, there are now more spreaders than beforehand.

Obviously we don’t know it’s the case, but we know there’s a possibility it is the case.

[PlatinumGuy]

Is there even a single confirmed Covid death after 2 doses of vaccine?

The WHO chief scientist said today there isn’t one anywhere in the world.

[Lurker]

You’re confusing recipients with spreaders. If your goal is to curb spread by eliminating the spreaders, and half of those you vaccinated can still get infected, and transmit it 3x as much as they would’ve done without being vaccinated (less symptoms and less tests so won’t know to isolate, there are now more spreaders than beforehand.

Obviously we don’t know it’s the case, but we know there’s a possibility it is the case.

There's a lot of assuming going on here all around. Some points:

1) It's 40% of the vaccinated who can still get infected.
2) It's very likely that the infection will be less severe, and the viral shed will be much lower, as well.
3) The theory that they may be 3x more likely to transmit it is based on the assumption that people take precautions when they have symptoms (even though most spread happens before symptoms) and that they are less likely to get tested after an exposure. This is only relevant is you also assume that vaccinated people will take zero precautions, which is a stretch. It also doesn't factor in that X% of people they come in contact with will also be vaccinated, lowering the risk of transmission even further.

There was a Washington Post article (https://www.washingtonpost.com/graphics/2020/world/corona-simulator/) from the beginning of the pandemic that attempted to show exponential spread and explain the ever-so-popular "Flatten The Curve" policy. There were some graphics used that come to mind whenever I try to reconcile the current vaccination approach with how I think it should be done. Without revisiting the policy or anything the article says (because I haven't read it in 10 months), the image that is burned into my mind is all the balls bouncing around, changing from grey (not infected, no antibodies) to red (infected, spreading) to pink (antibodies, immunity). The fewer grey balls we have bouncing around, the lower the impact the red balls will have. And the slower we can make the spread in the time leading into the third wave, the more time we buy to vaccinate even more people and eliminate more grey balls from the equation.

[PlatinumGuy]

There's a lot of assuming going on here all around. Some points:

1) It's 40% of the vaccinated who can still get infected.

This is far from clear. It's also not clear the number is black and white that x percentage are immune under all circumstances.

3) The theory that they may be 3x more likely to transmit it is based on the assumption that people take precautions when they have symptoms (even though most spread happens before symptoms) and that they are less likely to get tested after an exposure.

The reason most of the spread happens before symptoms is because people isolate after symptoms. We don't really know that shedding itself peaks after symptoms.

There was a Washington Post article (https://www.washingtonpost.com/graphics/2020/world/corona-simulator/) from the beginning of the pandemic that attempted to show exponential spread and explain the ever-so-popular "Flatten The Curve" policy. There were some graphics used that come to mind whenever I try to reconcile the current vaccination approach with how I think it should be done. Without revisiting the policy or anything the article says (because I haven't read it in 10 months), the image that is burned into my mind is all the balls bouncing around, changing from grey (not infected, no antibodies) to red (infected, spreading) to pink (antibodies, immunity). The fewer grey balls we have bouncing around, the lower the impact the red balls will have. And the slower we can make the spread in the time leading into the third wave, the more time we buy to vaccinate even more people and eliminate more grey balls from the equation.

I agree with your approach if two conditions are met 1) Vaccines are as effective in curbing the spread as they are in preventing severe illness & death, 2) the size of the 'spreader' group is equal to the size of the at risk group, rendering the time it would take equal.

In reality we know vaccines are very effective against death but don't know how effective they are against transmission (if at all), and the at-risk group is 1/2 or 1/3 of the spreaders group, so one goal can be attained much quicker than the other.

[avromie7]

US deaths by occurrence date:

Jan 30-Feb 6 2528
Jan 23-30 8909
Jan 16-23 13610
Jan 9-16 16122
Jan 2-9 18584
HIGH WATER MARK EXACTLY WHEN VACCINES STARTED WORKING
Dec 26-Jan 2 21384
Dec 19-26 21493
Dec 12-19 20905
Dec 5-12 19853


https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm

Yes, there is a reporting lag impacting the numbers slightly, but US deaths peaked exactly when the vaccine came into effect, and very dramatically so. There are likely 1000-2000 US deaths avoided every day solely by vaccines.

I don't need you to agree with me. I just want it to be clear that when the numbers support 600/day in a few weeks from now, 1-2k/day right now is a very extreme position. Who knows? Maybe you'll be proven right. I don't think anyone will complain if you are. It just looks like Kool-Aid to me right now.

Here we are a couple of weeks later and the week after the "high water mark" is exactly 1 higher than the week before. As I and others have said this data is incomplete and meaningless.

Updated numbers are:
US deaths by occurrence date:

Jan 30-Feb 6 10307
Jan 23-30 13610
Jan 16-23 17969
Jan 9-16 21418
Jan 2-9 23001
HIGH WATER MARK EXACTLY WHEN VACCINES STARTED WORKING
Dec 26-Jan 2 23000
Dec 19-26 22075
Dec 12-19 21257
Dec 5-12 20047

[Lurker]

Here we are a couple of weeks later and the week after the "high water mark" is exactly 1 higher than the week before. As I and others have said this data is incomplete and meaningless.

I don't know that's it's completely meaningless, but the actual timeframe for reporting deaths needs to be acknowledged.

• As of Feb 9, almost 20% of deaths from the week of Jan 2-9 were unreported.
• Less than 8% of the deaths were not reported from the week of Dec 26 - Jan 2, and 2.5%, 1.6%, and 1% from the 3 weeks before that.
• The weeks of Jan 9-16 and Jan 16-23 saw 30%+ increases in the last 2 weeks.
• Jan 30 - Feb 6 saw the numbers 4x.

The majority of the deaths aren't reported until 3-6 weeks after occurrence. Some are reported before, and some are reported after, but you can't really begin to get a decent picture until at least 6 weeks later.

[PlatinumGuy]

Here we are a couple of weeks later and the week after the "high water mark" is exactly 1 higher than the week before.

Noted, however Covid deaths are similar - which is a stagnation and a break from the increases prior, but all-cause deaths are still 5000 lower Jan 2-9 than the week prior.

To the actual point of whether or not vaccines are lowering current deaths:

https://twitter.com/paulg/status/1364181743943753730

https://twitter.com/larry_levitt/status/1364584123348652032

As we know, as late as data is available, 25% of current US Covid deaths are in LTC facilities.

[Lurker]

Noted, however Covid deaths are similar - which is a stagnation and a break from the increases prior, but all-cause deaths are still 5000 lower Jan 2-9 than the week prior.

To the actual point of whether or not vaccines are lowering current deaths:

I'm not saying you're wrong, because vaccines most definitely are saving lives in old age homes, but that data is incomplete for the last 4-5 weeks, which is when the significant drops occur.

[AsherO]

I'm not saying you're wrong, because vaccines most definitely are saving lives in old age homes, but that data is incomplete for the last 4-5 weeks, which is when the significant drops occur.

Correct, the chart PG posted doesn’t even include data from the past two weeks, and the data they have isn’t fully up to date. If the LTC line on that chart keeps going down the way it’s trending, that’s very significant. Also keep in mind that FWIU vaccine immunity increases over time, even beyond seven-days-after-second-dose (28d for Pfizer and 35d for Moderna)...

[PlatinumGuy]

I'm not saying you're wrong, because vaccines most definitely are saving lives in old age homes, but that data is incomplete for the last 4-5 weeks, which is when the significant drops occur.

Here are CDC models of current vaccine impact.




https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e2.htm#F1_down

[AsherO]

Here are CDC models of current vaccine impact.




https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e2.htm#F1_down

Thanks! Cool looking graphics for the modeling...

Doesn’t the 117 variant by definition mean a higher Rt than “current variants”? Or do they mean a blended rate?

Also, when they say with vaccination they mean at current vaccination rates, or beyond some hypothetical threshold?

ETA: The link has a trove of data, assumptions, variant breakdown data, and other gems. Very informative and addresses my questions above (but I’m too lazy to process it and summarize). Big thanks!

[PlatinumGuy]

Thanks! Cool looking graphics for the modeling...

Doesn’t the 117 variant by definition mean a higher Rt than “current variants”? Or do they mean a blended rate?

Also, when they say with vaccination they mean at current vaccination rates, or beyond some hypothetical threshold?

Current vaccinations trend and blended rates.

[AsherO]

Current vaccinations trend and blended rates.

When you look closer it’s obvious that it’s blended rate. Once the “current variants” and 117 reach the same level (with 117 becoming progressively more dominant), they converge with B117 eclipsing (replacing) “current variants”. Very cool!

[PlatinumGuy]

When you look closer it’s obvious that it’s blended rate. Once the “current variants” and 117 reach the same level (with 117 becoming progressively more dominant), they converge with B117 eclipsing (replacing) “current variants”. Very cool!

Correct, that's why I said blended.

[Lurker]

Here are CDC models of current vaccine impact.

If I'm reading this correctly, the CDC models predict no third wave with vaccinations. According to both models, cases either continue to drop or flatline between now and May. I'm not buying it.

[PlatinumGuy]

If I'm reading this correctly, the CDC models predict no third wave with vaccinations. According to both models, cases either continue to drop or flatline between now and May. I'm not buying it.

I think the question hinges around how much caution is thrown to the winds as vaccination increases (schools and theaters opening, for example), so I don't have an opinion one way or the other, but here's some discussion:

https://twitter.com/youyanggu/status/1362476073624158208
https://twitter.com/youyanggu/status/1362476078271455233
https://twitter.com/youyanggu/status/1362565923425640452

https://twitter.com/trvrb/status/1362438591406768134
https://twitter.com/trvrb/status/1362438597853413379
https://twitter.com/trvrb/status/1362438599937982464

[AsherO]

If I'm reading this correctly, the CDC models predict no third wave with vaccinations. According to both models, cases either continue to drop or flatline between now and May. I'm not buying it.

Read the link PG shared with these charts, and you’ll see how many assumptions there are and those are only some of the variables at play. Basically all we can do is model based on assumptions, predicting with any degree of certainly is impossible.

One thing that’s true is that the more and sooner we vaccinate the bigger likelihood there’ll either be no further waves or they’ll be smaller.

[Lurker]

I think the question hinges around how much caution is thrown to the winds as vaccination increases (schools and theaters opening, for example), so I don't have an opinion one way or the other, but here's some discussion:

It seems like one variable that's being ignored is the historical seasonal waves of pandemics. I'm also seeing a lot of optimism built on declining numbers, without enough acknowledgement that this is the natural cycle of pandemic waves.

[Lurker]

Read the link PG shared with these charts, and you’ll see how many assumptions there are and those are only some of the variables at play. Basically all we can do is model based on assumptions, predicting with any degree of certainly is impossible.

One thing that’s true is that the more and sooner we vaccinate the bigger likelihood there’ll either be no further waves or they’ll be smaller.

I read it, and it's the reason I'm not buying it. I completely understand that models are speculative and a number of variables can totally change everything. I just don't like the variables they're ignoring. The models assume a steady R value, when they rarely stay the same. Model A is already wrong, based on where we currently are, but Model B assumes a smaller current prevalence of the more transmittable strains, leading to a flatline where there will likely be growth.

[PlatinumGuy]

The models assume a steady R value, when they rarely stay the same.

I don't think either of them are attempting to predict the future. They are illustrating what would happen if R would be x, nobody is assuming it will be a steady x for 5 months.