COVID-19

[Riggerjack]

How sure are we that R0 is around 1.15 and how sure can we be that this would imply herd immunity at 13% of immune people in the population?

Well, that's just it. R0 of 1.15 under lockdown conditions. Open up the economy, open up the R0 rate. And then the projection is worthless...

I expect that we will open the economy, when masks are plentiful. Then we can follow a SK strategy, suppressing the transmission with social distancing and PPE, but people continuing with their lives in a semi normal sense.

If we had the PPE and masking was normalized public behavior, we wouldn't have any need for a lockdown.

This could have been "the plan" all along, expressed from the beginning. But we don't have leaders, we have shepherds. They never intended to lead, they don't know how. They don't have the tools, if they did know. (Leadership doesn't scale to that level. Despite what our history books say, it never has.) They intend to herd.

Thus the push from this direction, then that, rather than a goal and a direction.

Herding has its advantages, but none of them apply, here.

[JL13]

@bigato

Well I don't think we're be sure until it's over. But it's been almost 3 weeks since that projection was released, and it projected a doubling of cases every 6 days. DC is hitting that exactly, so it seems like a reasonable estimate so far.

[George the original one]

About herd immunity and antibodies... I'm getting mixed messages.

Up thread somewhere, we had the bit about how the antibodies found were weak and now WHO says they're unsure there will be herd immunity because antibodies may not give full protection. https://www.yahoo.com/news/unsure-antib ... 06736.html

On the other hand, UW has produced a very good antibody test that is ready for public release. https://katu.com/news/nation-world/test ... rchers-say

[chenda]

This was quite a good article about the practicalities of vaccine development:

https://www.nytimes.com/2020/04/18/heal ... pe=Article

[Bankai]

So far, like have been said many times in this thread already, the best statistical samples that we have of a population where the testing was as wide as possible are the Diamond Princess ship, the Vo Euganeo* Italian village, and South Korea. All these three point to something a little over 1% death rate among all infected, considering also the ones that didn’t show symptoms. I don’t know whether repeating it one more time will make it sink in anyone’s mind, but here here is.

I realize that it's very appealing to declare once and for all "x is definitely true, case closed". After all, we are humans and humans yearn certainty.

However, things are usually more complicated than they seem and the current situation is evolving all the time as new data becomes available. Different groups will have different death rates, depending on many factors (age structure, health care capability, general health of the population to name a few).

The 1% death rate argument is such a case as on a surface you have fairly convincing data so it's natural to assume that you can derive the one and only true answer from it. Bare in mind I have no statistics or maths background, however, when looking at the data using my "common sense" approach I see several issues. Let's look at "The Diamond Princess":

Of the 3,711 people aboard Diamond Princess, 1,045 were crew and 2,666 were passengers.The median age of the crew was 36 while the median age of the passengers was 69.The passengers were 55% female and the crew was 81% male. Of the 712 infections, 145 occurred in crew and 567 occurred in passengers

3,618 individuals were tested between 5th Feb and 5th March, 712 tested positive, 14 died. Death rate: 14/712 = 1.97%.

1) We know that up to 30% of individuals tested get "all clear" even though they actually have an active virus in their body (false negatives). If we assume a 30% false-negative rate on the ship and include it in the denominator, death rate changes to: 14/712*0.7 = 1.37%.

2) Although the whole population was tested, this was done over the course of 30 days. We know the median time from onset to recovery is only 14 days for mild cases. Therefore, it's reasonable to assume there was a fairly large population on the ship who got infected but had no symptoms or very mild symptoms and recovered before they were tested, especially considering that Japan estimated that most of the transmission occurred before the quarantine. They would show as "true negatives" as although the had a virus, it was no longer detectable by the time they were tested. If we include them in the figures, they'd increase the denominator, thus pushing the death rate lower still. I have no idea how many would there be, but let's say another 30%; death rate changes again: 14/712*0.7*0.7 = 0.96%

3) "The Diamond Princess" passengers and crew population is a representative sample of a typical cruise ship population only. Based on the numbers of passengers and crew, the blended median age would be around 60? That's way higher than any country in the world (even Japan is 'only' 47) and means that the median passenger is already in the high-risk group. Therefore, pushing the death rate higher (age is the main risk factor).

4) Around 55% of people on the ship were male. Considering that the death rate for males is around 60% vs 40% for females, this again produces an inflated death rate on the ship due to the sample not being representative.

Now, I made some assumptions in the above, but (I hope) you can see that all four of these issues affect the calculation of death rate on the ship by inflating it substantially, at least when compared to any other population. Therefore, anchoring to the ship's death rate and extrapolating it doesn't make any sense. I'm sure there are more issues with the data than what I spotted, but the bottom line is that we should not assume that the death rate in any single population is representative of the whole world and treat it as gospel. The newest science estimates the death rate to be at a much lower level than 1% and the trend even since early Chinese data is for the death rate to only be revised down. It's not unreasonable to assume that once we are able to properly, repeatedly test large groups over time, the real death rate will be lower still.

[Ego]

Kaiser published an excellent study on their 1277 Covid-19 hospitalized patients.

PDF https://www.medrxiv.org/content/10.1101 ... 1.full.pdf

Of those hospitalized....



# of (potential) deaths was used as a justification for the lockdown. The other side of the equation, the economic costs & lives lost as a result of the lockdown was not well thought out .

Now that we have better data we should no longer be using the # of deaths as the gauge. We should be using.... well, I'll let Peter Singer tell it...

Singer: We need to think about this in the context of the well-being of the community as a whole. Even if what Anne says about the recession is right, we are currently impoverishing the economy, which means we are reducing our capacity in the long term to provide exactly those things that people are talking about that we need — better health care services, better social-security arrangements to make sure that people aren’t in poverty. There are victims in the future, after the pandemic, who will bear these costs. The economic costs we incur now will spill over, in terms of loss of lives, loss of quality of life, and loss of well-being.

I think that we’re losing sight of the extent to which that’s already happening. And we need to really consider that trade-off.

I think the assumption, and it has been an assumption in this discussion, that we have to do everything to reduce the number of deaths, is not really the right assumption. Because at some point we are willing to trade off loss of life against loss of quality of life. No government puts every dollar it spends into saving lives. And we can’t really keep everything locked down until there won’t be any more deaths. So I think that’s something that needs to come into this discussion. How do we assess the overall cost to everybody in terms of loss of quality of life, loss of well-being, as well as the fact that lives are being lost?

This is killing mostly older people. I think that's really relevant. I think we want to take into account the number of life years lost—not just the number of lives lost.

The average age of death from COVID in Italy is 79½. So you do have to ask the question: How many years of life were lost? Especially when you consider that many of the people who have died had underlying medical conditions. The economist Paul Frijters roughly estimates that Italians lost perhaps an average of three years of life. And that's very different from a younger person losing 40 years of life or 60 years of life.

Neil Ferguson, the epidemiologist who came up the original estimate that was used as justification for the lockdowns calculated between 1/2 and 2/3 of those who die from Covid-19 would have died without Covid-19 by the end of the year from old age or preexisting conditions.

Life years lost.

[jacob]

Many many pages ago, I calculated the average years of life expectancy lost to be 6 weeks for males and 8 weeks for females (women live longer). The biggest losers is the 60-69 group followed by the 70-79 group followed by the 50-59 group. This follows from a combination of remaining years left to lose, the likelihood of losing them, and the percentage of people that age.

Rehashing ... Other justifications for the lockdown was not overwhelming the hospital system and raising the death rates for a bunch of other health issues (cancer, heart attacks, diabetes, COPD, ... traffic). These numbers are not included in the above estimate. Also not included is any potential but as yet unknown long-term damage in younger patients as was the case with SARS survivors. The conclusion being, this is not just a lives vs money question but also a lives lost from not flattening the curve in the next several months vs lives lost from the economic fallout over the next few years. I think we also discussed suicides/GDP%drop. This was shown to be insignificant relative to CV19 deaths.

Add: I think a political concern is also that we'd have to take the CV19 deaths now within a few months in the present whereas the economic fallout damage can be spread over years in the future. It's similar thinking behind why politicians aren't acting on climate change (damage in the future, not now) or didn't prepare for a pandemic in the first place.

[Ego]

@jacob, yes, you have mentioned the hospital overwhelm and people not getting treatment for other life threatening issues on several occasions. What I haven't heard anyone mention here is the fact that life threatening issues are going untreated anyway. There have been several articles recently about how people are not seeking treatment for strokes, heart disease, acute appendicitis and acute gall bladder disease. One of the major drivers of the open-up movement are cancer patients who are unable to get treatment.

[jacob]

Yes, but the absolute numbers of the various health problems are important. CV19 is currently the second leading cause of death in the US now right on the heels of cardiovascular deaths. That's on average so it includes as yet little-affected areas in the denominator.

[CS]

Welp, those cancer patients are making a big assumption that cancer workers are willing to risk their lives working in the hospitals. After the crap I've seen from hospital admins during this pandemic, I sure as heck am not.

[J_]

What a debate; economic loss: versus let vulnerable people die.
Is it the lack of a social safety net as most European countries have but not the USA, which make it more obvious to American people to choose to avoid further economic loss? (and let the vulnerable die?)

[Bankai]

What a debate; economic loss: versus let vulnerable people die.

It's not - it's lives vs lives. Economic loss results in loss of life as well, just spread over time and not as visible:

Meanwhile, University of Bristol researchers say the benefit of a long-term lockdown in reducing premature deaths could be outweighed by the lost life expectancy from a prolonged economic dip.

And the tipping point, they say, is a 6.4% decline in the size of the economy - on a par with what happened following the 2008 financial crash.

It would see a loss of three months of life on average across the population because of factors such declining living standards and poorer health care.

I calculated the average years of life expectancy lost to be 6 weeks for males and 8 weeks for females (women live longer).

In that case, we might be better off doing nothing since the loss of life due to economic damage is very likely to be substancially larger than 6/8 weeks? (see citation above re: 3 months loss of life across the population at -6.5% GDP which by this point a lower estimate of true damage).

Also, your calculation was based on a 1% death rate IIRC while this is being revised down all the time.

What I haven't heard anyone mention here is the fact that life threatening issues are going untreated anyway.

I actually did say several pages ago that from the perspective of other health issues current situation might be worse than just letting the virus run as those cases that aren't treated at 3000% over capacity won't be treated at 20% either. What is more, 'letting the virus run' might cause 3 months of non-treatment while the lockdown approach might result in a much longer period of non-treatment and thus higher mortality from otherwise preventable issues.

[Ego]

Yes, but the absolute numbers of the various health problems are important. CV19 is currently the second leading cause of death in the US now right on the heels of cardiovascular deaths. That's on average so it includes as yet little-affected areas in the denominator.

Right, but you could argue that had we gotten the mass infection out of the way in these few weeks (with overwhelm) then we'd have herd immunity and everyone could go back to normal treatments. As it stands, we will be in this same boat for a year or more with limited access to healthcare for many over a very long period.

ETA @Bankai, apologies! I usually read your posts closely but somehow missed that.

[Bankai]

@bigato: strawman. No one is postulating letting old people die. What's been suggested were other options than stopping the world, such as shielding the vulnerable and letting the young and healthy take the hit (i.e. continue work/school) since for overwhelming majority of them it's like cold/flu.

[Ego]

New modeling report on corona in Stockholm shows

- infections peaked a couple of days ago
- about ⅓ have been infected by now
- in a month, corona infections will be almost over and Stockholm close to herd immunity

https://www.medrxiv.org/content/10.1101 ... 1.full.pdf

[Peanut]

People over 70 with and without risk factors can choose to self-isolate at home. Society should help them do that (grocery delivery, etc.). This is what sensible family members of mine are already doing, whatever their local government's actual guidance is.

@Ego: thanks for the Singer excerpts. A utilitarian approach is very useful in these times.

There are many different vulnerable populations. A friend of mine is a doctor whose primary patients are child abuse victims. She has been furloughed for weeks. I think extending shutdowns needs to be examined very carefully.

[Ego]

@Peanut, yes, so many example of unintended or unknown consequences. Mrs. Ego and I have been showing apartments this weekend. So far, three Covidivorces.

[Peanut]

And a less academically rigorous account of the Swedish strategy:

https://www.bloomberg.com/news/articles ... -effective

Of course, like all practices worth emulating that happen in Sweden, Americans will just say that they couldn't possibly work here...

[Peanut]

@Peanut, yes, so many example of unintended or unknown consequences. Mrs. Ego and I have been showing apartments this weekend. So far, three Covidivorces.

Wow! Although I suspect there may be a baby boom too.

[Dream of Freedom]

Has anyone thought about inoculating people with the full fledged virus? We could use some of those empty resorts or cruise ships. Get some young healthy volunteers and give them the virus. They could use all of the amenities. When they test negative they can leave.