Shanghai Disneyland will close in effort to contain coronavirus

"Looking better" is relative. Ohio is trending better, sure, but it's still 6x worse than it was at the start of October.

TheMillenniumRider said:

Not likely, we would have had next to zero cases, but we would have had a massive population of new hosts just waiting for infection. It would have popped back up and you would be constantly reliving those shutdowns to try and keep case count to zero. This is New Zealand’s model.

Right. But what I was saying is if the world was able to do the truly impossible and isolate and lockdown every human on the planet for 14-28 days, the virus would die out due to a lack of new hosts. Even the most draconian lockdowns don't isolate 100% of the people 100% of the time. But if that was doable across the globe (and I never said it was), we'd be dry humping on Valentine's Day.

Sounds encouraging in Israel in terms of impact of vaccines:

https://www.timesofisrael.com/israel-sees-60-drop-in-hospitalizatio...ccination/

I don't understand all this hindsight as if our regulations and such totally failed in some way. I'm absolutely convinced that if we did anything less, we would have had millions dead - exponential growth and all. The time from discovery to vaccine is static. The goal (articulated in a bunch of articles, but not from govt. officials, because nobody wanted to say it) was to keep people alive long enough for a vaccine to arrive. And the regulations did that. We had many months of less than one thousand deaths a day and we're seeing now what happens when we don't regulate ourselves well: four thousand deaths a day or more.

And I still don't think isolating the elderly is, was, or ever could have been a strategy. Where do you think the 4,000 dead per day are coming from? It's from elderly who are in heavily tested and isolated nursing homes right now. It's not like they haven't thought about trying to keep those people isolated and safe.

Edit: I suppose the goal at some very early point might have been actual containment, but both those paths: flattening the curve to wait for a vaccine and containment take very similar trajectories early on. Especially if there's any ambiguity about which trajectory you're on.

Edited again: The super neurotic internal calculus teacher in me had to clarify that it's not the area which represents saved lives, but the vertical distance between the lines. If I wanted to demonstrate saved lives as an area rather than a distance, I should have use a deaths/day rather than cumulative deaths chart.

ApolloAndy said:

And I still don't think isolating the elderly is, was, or ever could have been a strategy. Where do you think the 4,000 dead per day are coming from? It's from elderly who are in heavily tested and isolated nursing homes right now. It's not like they haven't thought about trying to keep those people isolated and safe.

I feel like your logic contradicts itself a little there.

So if the elderly who overwhelingly make up the deaths are already dying in high numbers while generally isolated and tested, why not let those are less risk (and/or willing to take it) build herd immunity?

Especially since we're vaccinating the elderly and at risk first.

We (I?) kind of touched on the idea that the mortality rate should drop pretty hard once we vaccinate those groups.

According to the CDC numbers, 1/3rd of all deaths so far have been the over 85 demographic. In theory, vaccinating just those people should reverse future deaths by around 33%. And since they have, by far, the higest mortality rate, that would drop even more.

Using round numbers, the 85+ demographic is just 6 million of 328 million people in the US. In a perfect world (yeah, I know) you reduce deaths by 1/3rd with just 6 million well-placed vaccines.

I guess where I'm headed (and how did I make this left turn?), perhaps the risk becomes "acceptable" long before widespread vaccinations?

The 65+ group still account for 81% of the deaths. That has held steady since way back when I first brought it up. Once those folks are vaccinated, the remaining risk is relatively low and the mortality rate simply plummets.

As Dave said, herd immunity isn't binary. The mortality rate among the remianing age groups is strikingly low. How long before you can get this thing to burn itself out with a one/two punch?

Lord Gonchar said:

So if the elderly who overwhelingly make up the deaths are already dying in high numbers while generally isolated and tested, why not let those are less risk (and/or willing to take it) build herd immunity?

Especially since we're vaccinating the elderly and at risk first.

Because these aren't (or haven't been) actually decoupled from each other. The rate that the elderly die, even with our best efforts at isolation and testing is incredibly dependent on the probability that people moving in and out of those nursing facilities are positive. When young people got sick in schools, elderly people died in nursing homes, even with our best efforts at isolation and testing.

If/when we actually do have the majority of the elderly vaccinated and we're pretty sure that young people dry humping doesn't lead to elderly people dying in nursing homes, then sure...hump away I guess. But I'm not sure I'm willing to roll those dice, when I can get vaccinated a few months later and be certain. I mean, I wouldn't willingly go to work if there's flu outbreak at my place of work, even though it's ~1/10 the danger. (This is all hypothetical, since I'm 1b by virtue of being a teacher. I'll see you suckers at Disney World.)

Moderna says its vaccine protects against UK and South Africa variants. Same protection against UK. Weaker protection against South Africa variant. Working on a booster shot.

https://www.usatoday.com/story/news/health/2021/01/25/covid-moderna...chromepush

From what I have read/heard, tweaks to the new vaccines can be made relatively quickly (weeks/months not years). Expect that makers will need to continue to monitor variants and effectiveness and makes tweaks as necessary.

Merck abandoned its vaccine campaign. Used older vaccine technology and was unable to obtain the protection provided by other new tech vaccines.

Lord Gonchar said:

It flat out describes the deal in the article:

If anything, the 95 percent number understates the effectiveness, because it counts anyone who came down with a mild case of Covid-19 as a failure. But turning Covid into a typical flu — as the vaccines evidently did for most of the remaining 5 percent — is actually a success. Of the 32,000 people who received the Moderna or Pfizer vaccine in a research trial, do you want to guess how many contracted a severe Covid case? One.

Based on the trial of 32,000 people:

It was 95% effective at preventing infection and reduced the risk to that of the typical flu for an additonal 4.997%.

It was 99.997% effective at preventing severe cases of COVID.

So I've been thinking about and parroting this a lot, and unless I'm missing something, 99.997% is not right.

The effectiveness at preventing severe cases of COVID is a relative measure and should compare the number of severe cases of COVID in the placebo group to the severe cases in the vaccine group. That proportion is the effectiveness at preventing severe cases of COVID. The fact that 1 vaccinated person out of 30,000 had severe symptoms does not mean that the vaccination was 99.997% effective at preventing severe cases of COVID, because if 2 unvaccinated out of 30,000 had severe symptoms then it's only 50% effective. Right?

I don't at all dispute the point being made, but Coasterbuzz Semantics Game > Rip Ridin' Rocket.

I don't know, Andy. I get what you're saying, but it feels wrong.

If 1 in 32,000 people had severe symptoms then the vaccine is 99.997% effective at preventing severe cases.

If you compare to a control where only two people had severe symptoms then it reduced instances of severe symptoms by 50%.

That feels like semantics.

I found this press release.

Near the top it states:

"The first primary objective analysis is based on 170 cases of COVID-19, as specified in the study protocol, of which 162 cases of COVID-19 were observed in the placebo group versus 8 cases in the BNT162b2 group. Efficacy was consistent across age, gender, race and ethnicity demographics."

I remember them needing to get to 170 cases before they relevant data to submit. A little quick in-my-head mathing says 8 is roughly 5% of 170. Hence, the 94%-95% efficacy...at preventing infection.

A little further down it states:

"There were 10 severe cases of COVID-19 observed in the trial, with nine of the cases occurring in the placebo group and one in the BNT162b2 vaccinated group."

It's 90% effective at preventing severe infection...and here's the catch...after already being 95% effective at stopping infection at the first place. So here's where I'm not sure how the math works.

Basically, what we have is 9 severe and 153 non-severe cases among the placebo group and 1 severe and 7 non-severe cases among the BNT162b2 group.

I mean, I think I could technically argue that your chance of severe infection goes up after vaccination if I wanted.

In the placebo group, 9 of 162 cases were severe. (5.6%)
In the BNT162b2 group, 1 of 8 cases were severe. (12.5%)

Oh God! I have a better chance of severe infection if I get vaccinated!

I'm not sure how you'd want to express it and exactly how you'd want to do the math to be able to express it in that way.

Math away.

I'm pretty sure the standard way to express that is that it's 90% effective (1 out of 10) at preventing severe infection. That's not contradictory to the previous claim that it's 95% effective at preventing any infection. Imagine if you had the exact same numbers of "any infection" and had exactly one severe infection in both groups. Then you would say the vaccine is 95% effective at preventing any infection but has no effect (0% effective) at preventing severe infections.

Which obviously points to the very small sample size which is why they probably haven't been emphasizing that result too much.

I feel like "semantics" has gotten to a weird place in this disucssion. Like it's a label of some peripheral or borderline or "well, actually..." kind of nitpicking. But what we mean when we say 0%, 50%, or 99.997% effective is super important, even if it is semantics. Like, the fact that 1 out of 32,000 people who were vaccinated had a severe case is totally meaningless unless compared against unvaccinated people, and whether we call that semantics or not (itself, a semantic argument), it is the entirety of whether the vaccine does anything. There's a standard way to express that and whether or not "it feels right" to common English parlance (i.e. it could be that it's 95% effective at preventing any infection but 90% effective at preventing severe infection) is, to me, a matter of explaining and not a matter of changing terminology.

But in case you haven't noticed yet, I love semantics and I teach so maybe that's just my natural bent.

My brain breaks at the point of "severe infection" because that's not a different set, it's a subset of "any infection"

All "severe" infections are "any" infections. But I digress.

But even still, it doesn't change the fact that only 1 of 32,000 vaccinated got a "severe" infection. I guess we'd need something to compare that to. So what you're saying makes sense.

But if the placebo set was 9 times higher then the risk of severe infection goes from .003% (1-in-32,000) to .027% (9-in-32,000).

And that can't be accurate because if the risk of severe infection were 3/100ths of one percent, we wouldn't be very worried about the virus. We know that the real world mortality rate is around 1.7% (419k deaths in 25 million cases) so a severe infection rate of .003% in a trial seems like a reasonable and incredibly favorable comparison.

It may be "correct" to compare in the way you're describing, but it misses the mark to me. We know what is actually happening. 1.7% of the people infected die. In the trial, only .003% got anywhere near that point.

I think this is exactly where a disconnect happens. It may not be "correct" to compare this way, but to ignore it is missing what's right in front of your face. It just seems like getting hung up on "correct" instead of "actual" (or something, I'm really struggling to express this, but I think the point is there) to me.

I dunno. Welcome to my thought process.

Well, it may be because of the design of the study. Maybe they weren’t testing on nonagenarians and diabetics, so all the numbers are a little lower than national or global figures? I have no idea, but it makes sense to me to compare to the control group rather than to other numbers collected with different methodology on samples selected differently.

To be severely infected you would have to be infected first. So wouldn’t you be in the 10% of the 5%? If so it is a dependent event.

If so then that would be a conditional probability and thus 1/20 * 1/10 is .005.

So half of a percent chance of severe infection. Did I math correctly?

I don't think it works exactly like that. The 95% doesn't actually mean that you're 95% less likely to get it or that 95% of the people become immune. It's not really a probability. It just means that in the trial, 95% of sick people had a placebo and not a vaccine. Similarly, in the trial 90% of the severely sick people had a placebo and not a vaccine. It could be the case that, for whatever reason, the vaccine does a very good job at preventing most infections, but doesn't do as good a job at preventing severe infections. That may be because of strain or entry method or some other invisible variable.

Just as a random example:
Strain A: Causes mild infections.
Strain B: Causes severe infections.
Both strains are equally present.

If the vaccine stops 90% of strain B and 100% of strain A, you would see 95% success in preventing all infection and 90% success in preventing severe infections. I'm not saying AT ALL that this is what's happening because I have no idea, but this is the kind of situation that could lead to a greater protection against all infection than against severe infection.

Pfizer will be able to deliver 200 million doses (enough to vaccinate 100 million people) in the US by the end of May rather than the end of July. Also working on a booster with respect to the variants.

https://www.bloomberg.com/news/articles/2021-01-26/pfizer-to-delive...executive.

If you can reduce the number of people who get the virus, seems tough to imagine you are not also reducing the number of serious cases of illness. Even if being vaccinated doesn't mean that if you get the virus, your illness is likely to be shorter and less serious (from what I understand flu vaccines and shingles vaccines do both) just having significantly fewer numbers of people getting the virus means there will be significantly fewer people getting seriously ill from the virus. What the percentages are of each isn't known right now from what I have read. Effectiveness in the real world tends to be lower than it is in the trials for various reasons.

ApolloAndy said:

Like, the fact that 1 out of 32,000 people who were vaccinated had a severe case is totally meaningless unless compared against unvaccinated people, and whether we call that semantics or not (itself, a semantic argument), it is the entirety of whether the vaccine does anything. .

The fact that 1 out of 32,000 people who were vaccinated (0.003%) had a severe case of Covid is meaningless unless compared against unvaccinated people??? I disagree with that - but sure, take what is happening all around us with unvaccinated people. We're suffering through an approximately 1.6% death rate.

So unvaccinated population has a 1.6% death rate. I can't find a cumulative number of hospitalizations in US as many states aren't reporting that but a conservative estimate would be 10% hospitalization.

And the vaccinated population has a 0% death rate and a .003% severe case rate.

Sometimes things aren't that complicated. And that's really, really good news we should all be celebrating.

Did you read the last, like, 10 posts? We’ve been discussing and I would argue mostly disproving everything you just said (except the celebrating part).

You have to compare to the control group and not to "at large" numbers. It would be almost impossible to not introduce some bias in the set of people willing to volunteer for a vaccine trial. Not only that but the testing schedule would be different, the diagnosis criteria would be much more strict, incentives to fudge numbers one way or another would be eliminated, asymptomatic infections would still be caught, etc. etc. Comparing "real world" numbers to numbers from a selected and controlled study is apples and oranges.

I am hesitant to post some more good news but I was reading the daily morning blast from the NYT this AM and there really is some seriously encouraging news coming out...

Cases in the US have dropped 35% over the past 3 weeks which is the first time the US has seen such a dramatic drop in cases since the pandemic began. Attributed to what some suggest as a very early sign of herd immunity - (at least according to this article) appx 100M confirmed and unconfirmed cases (using the theory for every tested and confirmed case, there are at least 3 untested infections) plus 24M in the vaccination process - leaves us with over 1/3 of all Americans having some level of immunity against this thing. Secondly, attributed to a renewed focus on distancing, staying home, masking with the holidays behind, people seeing light at the end of the tunnel and a new tone out of the White House.

There's no evidence that it has anything to do with immunity, not when 3% of the population has been vaccinated and x% has been infected. But yeah, it's largely the expected curve of people doing what they should be doing after the holidays.

It is good to see the new cases going in the other direction but it's still so much worse than it was during the summer.

Covid 19 Act Now has started integrating vaccine data.