2019-nCoV / Corona virus Pt 2

[ceptimus]

I'm not sure what the WHO hopes to accomplish when it keeps announcing there's still time for control.

I think they're planning to let the daily death count get up to reasonable level before introducing harsher controls.

That might seem stupid at first sight - but if they believe that widespread infection is inevitable eventually, then it is logical.

Imagine a country had taken drastic measures back in early January: closed schools, banned sporting events, banned all internal travel and so on. By now the whole country would already have been in lock down for two months and for what? If you're certain that you're eventually going to suffer a months-long catastrophe, it makes no sense to delay the onset of that catastrophe for a few weeks by first enduring a different type of catastrophe.

 

[dudalb]

A number of GOP attendees to the CPAC conference have self quarantined themselves, because of the virus, including Matt Gaetz, the congressman who wore a gas mask to ridicule the diea that the Corona virus was a problem.
I love the Karma, man, love the Karma.

 

[The Atheist]

At this time it's looking like 1 in 20 for SARS2,...

Nope, you are way too high.

We know for certain that many - and almost surely a large majority - people don't have symptoms bad enough to see a doctor, far less be tested. We know that by the spread in America, where it was six weeks before community transmission was discovered.

If you want real data, look to South Korea, where they are testing vast numbers of people and accordingly find vast numbers of people not seriously affected. Their "severe case" load is not increasing in numbers that justify anything like a 5% mortality rate.

The smart money is on mortality of 0.5 - 1%.

That's still completely bad news, but nothing like the apocalyptic deaths you'd see at 5%.

Also, new case numbers daily are not rising that fast - I see yesterday's total was less than the previous day. The overall rate of increase isn't that high.

I had very similar logic in the previous thread. The only way your logic could be substantially incorrect is if they discover that most people either do not get it or get it so mildly they do not get tested.

Which we know for certain is happening.

Another example is the Diamond Princess. Yes, some of the severe cases will probably die, but their current mortality rate is 1%, and that's a very old demographic.

A number of GOP attendees to the CPAC conference have self quarantined themselves, because of the virus, including Matt Gaetz, the congressman who wore a gas mask to ridicule the diea that the Corona virus was a problem.
I love the Karma, man, love the Karma.

Haha! That is outstanding.

 

[alfaniner]

All of Italy is now on lockdown.

 

[Planigale]

Thanks for responding, blutoski and eeyore1954.

Next level OOM-ing: what do you think are reasonable upper and lower estimates?

Contagion: 100% and 10%? Something else?
Death rate: 8% and 0.3%? Something else?

The UK is modelling on 80% of the population infected.

The latest LSHTM paper
https://cmmid.github.io/topics/covid19/severity/diamond_cruise_cfr_estimates.html
From Chinese data suggests infection fatality rate (IFR) of 0.5% (95% CI: 0.2–1.2%) and case (symptomatic) fatality rate (CFR) of 1.1% (95% CI: 0.3–2.4%). From the Diamond Princess ship IFR 1.2% (0.38–2.7%) and CFR as 2.3% (0.75%–5.3%). Average age of passengers was 58.

This matches with estimates from elsewhere that half of those infected are symptomatic.

For the US assuming an adult population of 250,000,000. An 80% infection rate is 200,000,000 and a 0.5% fatality rate is 10,000,000 deaths most of which will be in the next 6 months.

This paper has some nice epidemic curves.
https://www.medrxiv.org/content/medrxiv/early/2020/02/14/2020.02.12.20022566.full.pdf

But of more interest is the seasonality modelling. For those relying on the warmer weather to save us it really depends on just how much of an impact on transmission summer has. Paradoxically a big drop in transmission may be worse than a moderate drop, in that it just pushes the epidemic in to the winter with no effect on the severity, whilst a modest impact smears the epidemic into a biphasic pattern making the peak cases much smaller and easier to manage.

 

[JeanTate]

I think they're planning to let the daily death count get up to reasonable level before introducing harsher controls.

That might seem stupid at first sight - but if they believe that widespread infection is inevitable eventually, then it is logical.

Imagine a country had taken drastic measures back in early January: closed schools, banned sporting events, banned all internal travel and so on. By now the whole country would already have been in lock down for two months and for what? If you're certain that you're eventually going to suffer a months-long catastrophe, it makes no sense to delay the onset of that catastrophe for a few weeks by first enduring a different type of catastrophe.

Perhaps we could call this the Chinese curse?

Through efforts - draconian, heroic, whatever - the spread of the coronavirus is stopped; a mere low-thousands dead.

But then infections took off, in Iran, Italy, South Korea, Japan, ... the US, North Korea. Whose containment efforts were not as, um, successful.

So, a few months down the track, an infected visitor from one of those places, perhaps one who, it turns out, was highly infectious but totally asymptotic, ignites a community cluster, in Qinghai, say, or Ningxia (both provinces barely affected to date). Which is not detected for a month (easy to invent plausible scenarios).

And so begins draconian/heroic part 2.

Or: draconian/heroic Italy succeeds, though the total numbers of deaths ends up being ~10k. Then an infected visitor from the US, say, perhaps someone who went to that CPAC meeting but was very rare in having been asymptotic and infectious for far longer than normal (or the infection bounced around between her family members, pets, etc), turns up in Genoa, say.

And so begins Italy draconian/heroic part 2.

Lather, wash, rinse, repeat ...

 

[Andy_Ross]

All of Italy is now on lockdown.

Italy is extending its strict coronavirus quarantine measures, which include a ban on public gatherings, to the entire country.

Prime Minister Giuseppe Conte said that people would only be permitted to travel for work or family emergencies.

He said the measures, which come into effect on Tuesday, were to defend the most fragile members of society.

https://www.bbc.co.uk/news/world-europe-51810673

 

[ceptimus]

A report from Italy on UK Channel 4 news, had a section where a medical expert admitted there weren't enough ventilation machines for every patient so the machines were being used on the younger fitter patients who were "more likely to recover."

"So you're leaving the old people to die?" asked the interviewer - which was a harsh way of saying it, but obviously the truth.

 

[marting]

A report from Italy on UK Channel 4 news, had a section where a medical expert admitted there weren't enough ventilation machines for every patient so the machines were being used on the younger fitter patients who were "more likely to recover."

"So you're leaving the old people to die?" asked the interviewer - which was a harsh way of saying it, but obviously the truth.

Triage. One does what one must.

 

[Andy_Ross]

Boris says it might be best if we just 'grit our teeth' and bear it. Let it run it's course unchecked, get it over quick and do the least economic damage.

 

[Andy_Ross]

Two more deaths this afternoon in the UK

 

[dudalb]

Boris says it might be best if we just 'grit our teeth' and bear it. Let it run it's course unchecked, get it over quick and do the least economic damage.

What next? Boris recommending that buildings with infected people be marked a big red cross painted on them?

 

[JeanTate]

Boris says it might be best if we just 'grit our teeth' and bear it. Let it run it's course unchecked, get it over quick and do the least economic damage.

Perhaps he could lead by example?

Get himself deliberately infected, then spread it to his Cabinet, etc.

In fact, didn’t we recently have a anti-vaxxer, here, who suggested pretty much just that (for measles)?

 

[marting]

An interesting detail from the Harvard study that hasn't been much discussed is that while China was able to get a reduction in R0 to well under 1, transmission from infected children did not decrease. I'm speculating this may be due to the difficulty of getting kids to understand hygiene and social distancing.

While kids have very little risk from the disease, if they become a significant infection vector to others then this will need to be addressed.

 

[Planigale]

An interesting detail from the Harvard study that hasn't been much discussed is that while China was able to get a reduction in R0 to well under 1, transmission from infected children did not decrease. I'm speculating this may be due to the difficulty of getting kids to understand hygiene and social distancing.

While kids have very little risk from the disease, if they become a significant infection vector to others then this will need to be addressed.

Certainly there was some flu modelling that suggested if you could effectively vaccinate children against flu then flu cases would dramatically fall.

 

[Meridian]

For the US assuming an adult population of 250,000,000. An 80% infection rate is 200,000,000 and a 0.5% fatality rate is 10,000,000 deaths most of which will be in the next 6 months.

You've got an extra 0 there. Half a percent of 200M is 1M, which is still a very large number!

 

[The Atheist]

All of Italy is now on lockdown.

Soon to be joined by Spain, by the looks of it.

But of more interest is the seasonality modelling.

NZ, Australia & South Africa still not showing it breaking out everywhere in the late summer, so it's on the table. NZ & Aus were both early finds of the disease as well, so it might be the case.

It's going to make for an ugly winter.

Boris says it might be best if we just 'grit our teeth' and bear it. Let it run it's course unchecked, get it over quick and do the least economic damage.

Hey, he's been listening to Jacinda! That's our plan.

An interesting detail from the Harvard study that hasn't been much discussed is that while China was able to get a reduction in R0 to well under 1, transmission from infected children did not decrease. I'm speculating this may be due to the difficulty of getting kids to understand hygiene and social distancing.

While kids have very little risk from the disease, if they become a significant infection vector to others then this will need to be addressed.

Yet the kids didn't pass it on to many others, or the rate would have continued. I think that means asymptomatic cases don't pass it on readily.

And yes, getting kids to understand social distancing ain't gonna happen.

We have a few kids at the local schools wearing masks all day. I even saw a fat boy running in one the other day and he was struggling so much he took it off and threw it away.

 

[JeanTate]

Thanks for responding, blutoski and eeyore1954.

Next level OOM-ing: what do you think are reasonable upper and lower estimates?

Contagion: 100% and 10%? Something else?
Death rate: 8% and 0.3%? Something else?

The UK is modelling on 80% of the population infected.

The latest LSHTM paper
https://cmmid.github.io/topics/covid19/severity/diamond_cruise_cfr_estimates.html
From Chinese data suggests infection fatality rate (IFR) of 0.5% (95% CI: 0.2–1.2%) and case (symptomatic) fatality rate (CFR) of 1.1% (95% CI: 0.3–2.4%). From the Diamond Princess ship IFR 1.2% (0.38–2.7%) and CFR as 2.3% (0.75%–5.3%). Average age of passengers was 58.

This matches with estimates from elsewhere that half of those infected are symptomatic.

For the US assuming an adult population of 250,000,000. An 80% infection rate is 200,000,000 and a 0.5% fatality rate is 10,000,000 deaths most of which will be in the next 6 months.

This paper has some nice epidemic curves.
https://www.medrxiv.org/content/medrxiv/early/2020/02/14/2020.02.12.20022566.full.pdf

But of more interest is the seasonality modelling. For those relying on the warmer weather to save us it really depends on just how much of an impact on transmission summer has. Paradoxically a big drop in transmission may be worse than a moderate drop, in that it just pushes the epidemic in to the winter with no effect on the severity, whilst a modest impact smears the epidemic into a biphasic pattern making the peak cases much smaller and easier to manage.

Thanks!

So at this OOM stage:
- population 330 million
- will get infected: 40-80%; geometric mean 56.6%
- of those, will die: 0.2-5.3%; geometric mean 1.0%

Total estimated deaths:
- high: 13.99 million
- low: 264k
- mean: 1.92 million.

FWIW, in my experience, this OOM-ing is pretty good, surprisingly so. And it is very easy to generate inputs ("reasonable" ranges, or upper and lower limits). In this case, with just a handful of independent estimates (or guesses), from a half-dozen ISF members who've posted regularly here (call it wisdom of the crowd, if you like), I think you'd find the estimated mean number of deaths would vary only slightly, perhaps by only +/- 30%.

 

[blutoski]

So you think a 1% fatality rate is too conservative? We don't have that much hard data yet but we do know from the small sample on the cruise ship 700 had the virus and 6 died. I haven't seen any data on the age distribution on the ship but I would think it is skewed to older people and so far it appears older people are far more susceptible. We do know that there has been some spread in the US and the first deaths were not reported until all of the ones at the nursing home. My common sense (which isn't always correct) tells me many people have gotten milder cases of the virus which did not require doctors or hospitalization. Because of the one case in Westchester NY I think so far about 80 cases have been uncovered. Why would I think this hasn't been happening for a while already in other places across the US.

I do understand there could have been deaths that were incorrectly not linked to the virus also.

In this case, I'm not doing my own calculations from media, I'm forwarding my communications from Health Canada, which is collating their data with other countries and keeping us updated.

('us' meaning, me as an immunologist and my wife as a manager at a hospital)

6% is the rough estimate at this time based on all the data.

Regarding the cruise ship, no the disturbing part of that example is that some of the crew succumbed, and they are relatively young and healthy. My impression is that the demographics on the cruise ship for deaths skew surprisingly young considering their lower ratio in the overall contained population there.

 

[blutoski]

Wasn't there a case about a week ago of someone who gave it to their dog?

Yes but I'd like to learn more about how the infection manifest, if it's sufficiently symptomatic to propagate &c, before confirming there's an animal reservoir risk.

Coronaviruses don't seem to cross species generally.

To put this another way, dogs already have their spectrum of coronaviruses, but they're not a reservoir because they don't infect humans.

This one might cross the species barrier, which would be bad for the reasons I've outlined in an earlier post. It can contaminate the food supply, for example.



ETA: research on SARS confirmed transmission to cats, pigs, ferrets and some primates, but not dogs, so that's why I'm holding off on this one specimen, it could be very rare.