But WHY do we need to flatten the curve? (anxiety warning!)

FVP3

CelticRambler wrote: »

An antibody test has become Flavour of the Week for most polticians and internet commentators. As things stand at the moment, though, it'll change nothing. There is some evidence to show that the immune response to Covid-19 is paradoxical: if you've got a strong immune system, you'll get over it, but you might test negative; but equally, the worst cases are linked to a particularly strong immune response that kills the patient along with the virus.

That evidence is pretty weak. You'd assume younger people would get it more were that the case.

There is currently no evidence to suggest that a vaccine will work, some evidence to suggest that it'll kill at least some of the people it's supposed to protect (not from an unwanted side effect but by triggering the worst aspects of Covid-19 disease), and an awful lot of "jury's still out" statistically unreliable reports of people having recovered from a first Covid-19 infection and become test-postitive again several weeks later.

There's never any evidence that a vaccine will work until it does, and killing some people is a danger in any vaccine which is why the trials are so long.

The more likely explanation for people testing positive after "recovery" is either a false positive at the beginning or the end, or a false negative in the middle.

DeVore

The graph doesnt peak twice. Its three seperate graphs showing the outcomes of three separate approaches.

One is "do nothing" which spikes immediately.

The other two are varying degrees of isolation.

My criticism of the model would be that is presumes an abrupt and total cessation of social measures on the same day. Then, unsurprsingly, the virus returns to doing exactly what it was doing before.

It also doesnt factor in growth in ICU beds over time (though it IS graphing surge capacity).

Managing a step-down (and occasional step-up!) of social measures may very well "flatten the curve" more dramatically over a longer period.

FVP3

DeVore wrote: »

The graph doesnt peak twice. Its three seperate graphs showing the outcomes of three separate approaches.

One is "do nothing" which spikes immediately.

The other two are varying degrees of isolation.
.

I get that there are three graphs, but the graph looks to me like it definitely has a time line and there's a second peak in November for the two varying degrees of isolation, but not for the do nothing, I assume because of herd immunity.

DeVore

Sorry, yes, I just didnt think of them as "peaks" but yes they are there. IMHO thats the result of the delay in doing something. Its the initial "getting this thing under control" bump before the impact of social measures kicks in. Hard to determine the cause without access to their model but thats my opinion.

Squiggle

Lodi was in lockdown on February 23. Bergamo didn't lockdown until March 8. Check out the differences in the curve, which Lodi managed to flatten.

CelticRambler

FVP3 wrote: »

That evidence is pretty weak. You'd assume younger people would get it more were that the case.

There's never any evidence that a vaccine will work until it does, and killing some people is a danger in any vaccine which is why the trials are so long.

Coronaviruses in particular are known to trigger overwhelming - and fatal - immune responses. That's one of the reasons why we don't have very many vaccines against diseases caused by coronaviruses.

And there's no reason to assume that younger people would respond with a stronger immune response; in fact, it could be argued that our older generations have a better immunological memory than the younger ones. This is all waiting to be figured out, but a plausible hypothesis is that people growing up in the 1950s (+/-10 years) were exposed to a similar virus at the time, and that exposure has made them more susceptible to the massive immune-mediated pulmonary effusion that characterises this infection in its severe form.

GreeBo

DeVore wrote: »

The graph doesnt peak twice. Its three seperate graphs showing the outcomes of three separate approaches.

One is "do nothing" which spikes immediately.

The other two are varying degrees of isolation.

My criticism of the model would be that is presumes an abrupt and total cessation of social measures on the same day. Then, unsurprsingly, the virus returns to doing exactly what it was doing before.

It also doesnt factor in growth in ICU beds over time (though it IS graphing surge capacity).

Managing a step-down (and occasional step-up!) of social measures may very well "flatten the curve" more dramatically over a longer period.

If there is an abrupt and total cessation of social interaction for say, 3 weeks... What's the origin for the virus continuing n your criticism?

DeVore

GreeBo wrote: »

If there is an abrupt and total cessation of social interaction for say, 3 weeks... What's the origin for the virus continuing n your criticism?

Several origins.

1. International travel (thats how it got here in the first place).

2. Resevoirs we may not know about (cats? longer infectiousness timeframes than we expect/know?)

3. The likelihood of quite a number of people simply not obeying 21 days of total absolute lockdown.

4. Person A and B live together... Person A gets its on Day 0 before lockdown but doesnt pass it to Person B until some time into lockdown and their infectiousness exceeds the 21 day lock down.


All of these will restart the infection and remember, it only takes one really...

Polar101

I'm interested in seeing new estimates on what percentage of populations will eventually get infected. We keep seeing 40% / 60% / 80% and the likes. And that's kind of discouraging, because it would mean millions of cases in most countries. Or maybe more than a billion in China or India. You would think percentages like that would take a long time with action taken, so wouldn't we get to the stage where a vaccine is available, first?

Tokyo

One of my favourite explanations (complete with simulations) about the effectiveness of flattening the curve can be found here:
https://www.washingtonpost.com/graphics/2020/world/corona-simulator/

DeVore

DeVore wrote: »

.............

I'm going to write something tomorrow to explain why we arent doomed but here's the spoilers. ..................

Was part two released on this channel?

DeVore

Polar101 wrote: »

I'm interested in seeing new estimates on what percentage of populations will eventually get infected. We keep seeing 40% / 60% / 80% and the likes. And that's kind of discouraging, because it would mean millions of cases in most countries. Or maybe more than a billion in China or India. You would think percentages like that would take a long time with action taken, so wouldn't we get to the stage where a vaccine is available, first?

That sort of thing is so sensitive to things like
1. Vaccines
2. Effective Treatment
3. Continued Social Measures etc.

R0 is the rate at which one person infects X more people. R0 of 2 means one person infects two more. COVID's R0 is between 2.2 and 2.5

With an R0 of 2.2 COVID wouldnt take long to infect a lot of people, but like I said in my "hope" thread, its not an exponential its a logistic curve because as a sizeable number of the pop get and recover from the virus, it starts to bump into itself or an immunity system that can kill it stone dead. Ever met someone who has never had the flu? Same thing. Flu (R0 of 1.3) doesnt always reach everyone because of Herd Immunity. That happens at about 60% for Covid.

Unfortunately with an R0 of 2.2, it does spread and multiply quick enough. So if nothing else changes the picture, a large number of people will get this, hopefully slowly. A vaccine changes the rate of spreading. A treatment changes the mortality.
Measles is a very very nasty virus with an R0 of between 15 and 18 (yep!) but very few people get measles (or at least USED TO) because we have the MMR vaccine.

I hope that answers your question

DeVore

Augeo wrote: »

Was part two released on this channel?

Yes it was: https://www.boards.ie/vbulletin/showthread.php?t=2058066626

DeVore

DeVore wrote: »

Yes it was: https://www.boards.ie/vbulletin/showthread.php?t=2058066626

Cheers, I thought it would be in here, hence me asking. Apologies.

CelticRambler

DeVore wrote: »

With an R0 of 2.2 COVID wouldnt take long to infect a lot of people, but like I said in my "hope" thread, its not an exponential its a logistic curve because as a sizeable number of the pop get and recover from the virus, it starts to bump into itself or an immunity system that can kill it stone dead. Ever met someone who has never had the flu? Same thing. Flu (R0 of 1.3) doesnt always reach everyone because of Herd Immunity. That happens at about 60% for Covid.

Don't pin your hopes on this. To re-phrase your question: ever met a person who's had the common cold more than once? Across all species, coronaviruses have a reputation for not provoking long-lasting protective immunity. As I mentioned earlier, that's one of the reasons why you won't find coronavirus vaccines on many pharmacy shelves, human or veterinary.

At this stage in the epidemic, all we've got to go on is
(a) some people's immune systems get rid of the virus;
(b) some peoples immune systems kill them because of virus;
(c) some people's post-infection antibodies can be harvested to help someone else recover from the disease.

There is, at the moment, no evidence to suggest that SARS-CoV-2 provokes a long-lasting protective immunity, and any model that includes this as an assumption risks being quite (or very) inaccurate.

DeVore

In the "Hope" thread I linked some studies that show the COVID mutation rate is about 1/4 of that of the 'flu per month. This is not proof of long term immunity but its definitely a Very Good Thing.

CelticRambler

DeVore wrote: »

In the "Hope" thread I linked some studies that show the COVID mutation rate is about 1/4 of that of the 'flu per month. This is not proof of long term immunity but its definitely a Very Good Thing.

Mutation rate doesn't really matter. I've been working with coronaviruses for thirty years, and even when they're relatively stable, the (very rare) vaccines that make it onto the market are effectively useless. A peculiar feature of the coronavirus family is that many of its members mess with your immune system and use the body's immune response to kill you. A vaccine risks doing the same.

I would bet most of my pension plan on no vaccine ever being commercialised against Covid-19 ; if I was an investor, I wouldn't put a single centime into it - I'd put all my money into pursuing drug treatment options and ICU equipment.

Ethereal Cereal

CelticRambler wrote: »

Mutation rate doesn't really matter. I've been working with coronaviruses for thirty years, and even when they're relatively stable, the (very rare) vaccines that make it onto the market are effectively useless. A peculiar feature of the coronavirus family is that many of its members mess with your immune system and use the body's immune response to kill you. A vaccine risks doing the same.

I would bet most of my pension plan on no vaccine ever being commercialised against Covid-19 ; if I was an investor, I wouldn't put a single centime into it - I'd put all my money into pursuing drug treatment options and ICU equipment.

Since this is considerably different narrative than I am hearing from media, can I ask what your expertise in this area is?

You said you have worked with it for 30 years, in what capacity? A WHO anti-virologist, a general practitioner, a chemical engineer formulating effective cleaning compounds, a micro-biologist, etc. etc.

Hmmzis

CelticRambler wrote: »

Mutation rate doesn't really matter. I've been working with coronaviruses for thirty years, and even when they're relatively stable, the (very rare) vaccines that make it onto the market are effectively useless. A peculiar feature of the coronavirus family is that many of its members mess with your immune system and use the body's immune response to kill you. A vaccine risks doing the same.

I would bet most of my pension plan on no vaccine ever being commercialised against Covid-19 ; if I was an investor, I wouldn't put a single centime into it - I'd put all my money into pursuing drug treatment options and ICU equipment.

Long lasting immunity is certainly quite possible with this one as it's a relative of the first SARS-cov. The survivors of that one still had antibodies in their blood after 12 years.
https://www.medrxiv.org/content/10.1101/2020.02.12.20021386v1.full.pdf

Hmmzis

CelticRambler wrote: »

Don't pin your hopes on this. To re-phrase your question: ever met a person who's had the common cold more than once? Across all species, coronaviruses have a reputation for not provoking long-lasting protective immunity. As I mentioned earlier, that's one of the reasons why you won't find coronavirus vaccines on many pharmacy shelves, human or veterinary.

At this stage in the epidemic, all we've got to go on is
(a) some people's immune systems get rid of the virus;
(b) some peoples immune systems kill them because of virus;
(c) some people's post-infection antibodies can be harvested to help someone else recover from the disease.

There is, at the moment, no evidence to suggest that SARS-CoV-2 provokes a long-lasting protective immunity, and any model that includes this as an assumption risks being quite (or very) inaccurate.

Bolded section: the Common Cold is caused by hundreds of viruses, most commonly the Rhinovirus group, near a 100 in that one alone.
It's not a good comparison. You'd have to be catching multiple types of them a year to have some immunity against all of them by the end of the average human lifespan.
SARS-cov-2 is a single slow mutating virus, people will and are getting immunity against it and a there is a good chance it will be long lasting as it was with the first version of it.

hmmm

CelticRambler wrote: »

Mutation rate doesn't really matter. I've been working with coronaviruses for thirty years, and even when they're relatively stable, the (very rare) vaccines that make it onto the market are effectively useless. A peculiar feature of the coronavirus family is that many of its members mess with your immune system and use the body's immune response to kill you. A vaccine risks doing the same.

Didn't that happen with the first SARS vaccine? It ended up causing a worse response to the infection.

The Moderna approach is an interesting non-traditional approach, albeit unproven.

There was an interesting discussion with the J&J CEO yesterday. They have selected one vaccine candidate which he was confident about in terms of efficiency and safety, but had two other backup candidates if required.

DeVore

There is a sizeable cash incentive for companies to work on a vaccine. That doesnt make the vaccine any more *possible*, but it does improve the odds of finding one if one IS possible.

One of the new approaches is to target the RNA in the virus itself which is a novel if unproven tack.

CelticRambler, I'm very interested to hear more of you experience with the conoronavirus family. Can you give us an idea of your background?

AllForIt

CelticRambler wrote: »

Mutation rate doesn't really matter. I've been working with coronaviruses for thirty years, and even when they're relatively stable, the (very rare) vaccines that make it onto the market are effectively useless. A peculiar feature of the coronavirus family is that many of its members mess with your immune system and use the body's immune response to kill you. A vaccine risks doing the same.

I would bet most of my pension plan on no vaccine ever being commercialised against Covid-19 ; if I was an investor, I wouldn't put a single centime into it - I'd put all my money into pursuing drug treatment options and ICU equipment.

This is not the speak of a professional in virology.

Particularly this part:

A peculiar feature of the coronavirus family is that many of its members mess with your immune system and use the body's immune response to kill you. A vaccine risks doing the same.

No virologist worth their salt would say the virus is 'trying to kill you'. All it's doing is trying to propagate and if anything killing the host doesn't help that end.

CelticRambler

Sorry for the delayed follow-up: got landed with a load of "oh, before you go ..." requests right at the end of an already extended contract in Alsace, then came home to intensive gardening obligations. :pac:

Ethereal Cereal wrote: »

Since this is considerably different narrative than I am hearing from media, can I ask what your expertise in this area is?

You said you have worked with it for 30 years, in what capacity? A WHO anti-virologist, a general practitioner, a chemical engineer formulating effective cleaning compounds, a micro-biologist, etc. etc.

I wear many hats, but in this context I'm a frontline clinician working in the private sector, while also having an "activate as required" role in the delightfully communist sounding Direction Départementale de la Cohésion Sociale et de la Protection des Populations

AllForIt wrote: »

This is not the speak of a professional in virology.

No virologist worth their salt would say the virus is 'trying to kill you'. All it's doing is trying to propagate and if anything killing the host doesn't help that end.

Unless it's mutated since I first registered, boards.ie is not a professional environment; even so, I'm not strictly speaking a virologist - I just deal with the critters and their annoying behaviour, so inevitably get to know some of them quite intimately.

But you're right: the virus isn't "trying to kill you" - and it doesn't; and that's why the quest for a vaccine (IMO) is probably both futile and a waste of time. The particularly interesting characteristing of coronavirus is that it has an as-yet unexplained talent for not provoking any immune response while it is in the process of infecting its host, replicating and being expelled in large infectious quantities. The disease that's making people sick is the body's reaction to it's own processes that have been compromised by the replication process, the "cytokine storm" that you'll read about in some places.

There's no reliable evidence that people who've suffered a mild dose of this or any other coronavirus infection develop any protective antibodies. There is evidence that people who have recovered from a clinical infection do have and maintain antibodies for "some time" afterwards - but there's no reliable evidence that they're protective. This stands to reason, as these antibodies, and the memory cells that trigger their increased production, are present in the blood, but not in the lining of the respiratory tract, and in any case, the memory cells are unlikely to be triggered when the virus is renowned for not its ability to not trigger an immune reaction.

What does seem to work, though, is giving antibody-rich plasma from recovered patients to those still suffering from the severe form of the disease. That's where the R&D dollars should be concentrated. Having an effective antiserum available to treat the relatively small percentage of people who get moderate-to-severe illness would change everything.

A few links for those who want to delve deeper into the subject:


2012 summary of coronavirus infections in general (good for playing "spot the difference" between the previous SARS epidemic and this one)

2008 Good (very technical) description of SARS-CoV's interaction with the immune system (part of it)

2012 research paper on another aspect of the immune reaction, with this pertinent observation at the very end:

Finally, it is important to note that T cells can play a protective and/or pathological role during an infection.

Hmmzis

CelticRambler wrote: »

...

There's no reliable evidence that people who've suffered a mild dose of this or any other coronavirus infection develop any protective antibodies. There is evidence that people who have recovered from a clinical infection do have and maintain antibodies for "some time" afterwards - but there's no reliable evidence that they're protective. This stands to reason, as these antibodies, and the memory cells that trigger their increased production, are present in the blood, but not in the lining of the respiratory tract, and in any case, the memory cells are unlikely to be triggered when the virus is renowned for not its ability to not trigger an immune reaction.

What does seem to work, though, is giving antibody-rich plasma from recovered patients to those still suffering from the severe form of the disease. That's where the R&D dollars should be concentrated. Having an effective antiserum available to treat the relatively small percentage of people who get moderate-to-severe illness would change everything.

...

Please correct me if I'm understanding this wrong, but if giving a sick patient plasma with antibodies helps, it seems it's at odds with the first paragraph above that states:

"There is evidence that people who have recovered from a clinical infection do have and maintain antibodies for "some time" afterwards - but there's no reliable evidence that they're protective."

The study about SARS-cov I linked above would also indicate that it's at odds with that statement.

If antibodies from others blood plasma are effective then they are effective from your own immune system as well (or ones grown in plant cells).

CelticRambler

Antibodies can be "curative" without being protective. SARS is a two-stage process. In the first stage (i.e. asymptomatic/mild infection) there's no benefit to having antibodies, and no significant evidence that the body produces any. In the second stage, triggered in some individuals for reasons no-one can yet explain, the body lobs everything it can at whatever virus remains in the lung tissue, antibodies included.

It's possible (but not yet proven) that these CoV-specific antibodies mimic the virus' own immunosuppressive action, so giving one person's antibodies to another calms down the recipient's immune system in a targeted fashion (as opposed to drugs, that suppress it indiscriminately)

For the purposes of prediction (and/or issuing "antibody positive" passports, so that people can go back to work), it's known that individuals can be antibody negative and be shedding virus, and also be antibody positive without any known exposure to the virus. They can also be antibody negative with known, resolved infection, and antibody positive and excreting the virus. Which makes for a socio-political nightmare if any government is daft enough to go down that route, and even more so if different governments decide to use different tests/thresholds/interpretations ...

McGiver

CelticRambler wrote: »

There's no reliable evidence that people who've suffered a mild dose of this or any other coronavirus infection develop any protective antibodies. There is evidence that people who have recovered from a clinical infection do have and maintain antibodies for "some time" afterwards - but there's no reliable evidence that they're protective. This stands to reason, as these antibodies, and the memory cells that trigger their increased production, are present in the blood, but not in the lining of the respiratory tract, and in any case, the memory cells are unlikely to be triggered when the virus is renowned for not its ability to not trigger an immune reaction.

How is this possible if other viruses do generate antibodies?

EDIT: influenza strains few years, measles decades, chickenpox decades etc.

EDIT2: https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(20)30118-3/fulltext - this is a Covid-19 vaccine working in mice.

CelticRambler

McGiver wrote: »

How is this possible if other viruses do generate antibodies?

EDIT: influenza strains few years, measles decades, chickenpox decades etc.

Because none of those are coronaviruses, and don't have the same cloaking mechanisms.

McGiver wrote: »

EDIT2: https://www.thelancet.com/journals/ebiom/article/PIIS2352-3964(20)30118-3/fulltext - this is a Covid-19 vaccine working in mice.

No it's not. It's an "interesting project" but (a) statistically weak (I have more mice running around my house than they used in the experiment); and (b) totally detached from the reality of CoV infection, specifically in the following respects:
- they injected the vaccine intracutaneously, so it will inevitably trigger an immune response, but not necessarily the right immune response for a viral attack coming via the respiratory tract;
- they looked for, and found, serum antibodies after the injection; they did not expose the mice to a respiratory challenge afterwards, so there is nothing to demonstrate that the injection acted as a vaccination.
- they used genetically modified immunodeficient mice which is standard practice, but means they've deliberately manipulated the experiment to avoid having to deal with the most problematic aspect of any coronavirus, i.e. it's tendency to ignore the host's immune system.
- following from the above, they didn't test anywhere near enough individuals of other species to confirm that the injection/vaccine itself doesn't trigger Covid19 in susceptible individuals.

So it's one experiment amongst thousands that will be carried out, added to the thousands already done, all of which (to date, over several decades) have yielded a grand total of no useful vaccines against any coronavirus infection in any species.

What we really need is for someone to identify the genetic factors that make some humans (and other species) more susceptible to the severe disease, to develop a test for that so that vulnerable populations can be screened, and in parallel to find an effective treatment protocol. In the meantime, we're just playing roulette (some governments more than others).

Hmmzis

Well, this is both terifying and somewhat reassuring (warning, German laguage incoming):

https://www.land.nrw/sites/default/files/asset/document/zwischenergebnis_covid19_case_study_gangelt_0.pdf

Lifted from the main spam thread, shamelesly.

In short, what they did was take about a 1000 people at random from 400 households and screened them for SARS-cov2 antibodies and active infections. Antibody test acuracy was stated at >99% but sensitivity was not stated.

Result was: 2% had an active infection, 14% had antibodies and no infection. From the 2% half had both an active infection and antibodies present. So in total, 15% of the sample population had antibodies present. From that the CFR they calculated was 0.37% and a total population mortality rate of 0.06%.

So 15%+ of the population infected after only 2 months or so. They did not present a new estimate for R0, but I reckon it might not be 2 or 2.5 anymore.

At the end they propse a 4 step action plan of how to get out of this mess. Spoiler, we're still at step 1.

In my mind these sort of results would explain the anomaly that is Sweden and some former Soviet countries where the CFR values are much lower than almost everywhere else.

DeVore

I'm not sure how you see this explaining the anomalous countries? (If they exist, Sweden is starting to look bad... )