The maths of it all and what it means to Ireland

Peanut Butter Jelly

That would be case fatality rate. It wouldn't account for asymptomatic people who had it and never tested positive

Hmmzis

This seems to be an interesting take on herd immunity:

https://www.medrxiv.org/content/10.1101/2020.04.27.20081893v1.full.pdf+html

It's in preprint but does reference peer reviewed sources for the super-spreader data.

If I'm reading it right, then their argument that a society would need 60-70% recovered and immune people, is not quite correct or too simplistic.
Their argument seems to based on highly connected (medical workers, cashiers, bus drivers etc) vs lesser connected members (average person) of society. The highly connected drive the Re value up while the little connected drive it down. The connecticity here is meant to be interactions with other unique number of humans.

In short, they're estimating that an infection rate of 10-20% for SARS-cov-2 might be enough to have a signifficant herd immunity effect.

To me, the frist thing that pops to mind is packed public transit. That would have to be denting that CV value to some extent.
Same thing with packed pubs and clubs.
Office populations would be next in decreasing severity.
Then come schools and childcare facilities.

Not sure where that leaves it in regards to policy making. To me it seems some form of restrictions and distancing will be needed until new case counts are down to 0 or we start getting general population vaccine deployments.

Anyone care to shoot some holes in that paper or my understanding of it?

Hmmzis

This seems to be an interesting take on herd immunity:

https://www.medrxiv.org/content/10.1101/2020.04.27.20081893v1.full.pdf+html

It's in preprint but does reference peer reviewed sources for the super-spreader data.

If I'm reading it right, then their argument that a society would need 60-70% recovered and immune people, is not quite correct or too simplistic.
Their argument seems to based on highly connected (medical workers, cashiers, bus drivers etc) vs lesser connected members (average person) of society. The highly connected drive the Re value up while the little connected drive it down. The connecticity here is meant to be interactions with other unique number of humans.

In short, they're estimating that an infection rate of 10-20% for SARS-cov-2 might be enough to have a signifficant herd immunity effect.

To me, the frist thing that pops to mind is packed public transit. That would have to be denting that CV value to some extent.
Same thing with packed pubs and clubs.
Office populations would be next in decreasing severity.
Then come schools and childcare facilities.

Not sure where that leaves it in regards to policy making. To me it seems some form of restrictions and distancing will be needed until new case counts are down to 0 or we start getting general population vaccine deployments.

Anyone care to shoot some holes in that paper or my understanding of it?

yosemitesam1

Hmmzis wrote: »

This seems to be an interesting take on herd immunity:

https://www.medrxiv.org/content/10.1101/2020.04.27.20081893v1.full.pdf+html

It's in preprint but does reference peer reviewed sources for the super-spreader data.

If I'm reading it right, then their argument that a society would need 60-70% recovered and immune people, is not quite correct or too simplistic.
Their argument seems to based on highly connected (medical workers, cashiers, bus drivers etc) vs lesser connected members (average person) of society. The highly connected drive the Re value up while the little connected drive it down. The connecticity here is meant to be interactions with other unique number of humans.

In short, they're estimating that an infection rate of 10-20% for SARS-cov-2 might be enough to have a signifficant herd immunity effect.

To me, the frist thing that pops to mind is packed public transit. That would have to be denting that CV value to some extent.
Same thing with packed pubs and clubs.
Office populations would be next in decreasing severity.
Then come schools and childcare facilities.

Not sure where that leaves it in regards to policy making. To me it seems some form of restrictions and distancing will be needed until new case counts are down to 0 or we start getting general population vaccine deployments.

Anyone care to shoot some holes in that paper or my understanding of it?

That paper makes a lot of sense and would fit what was seen in the real world much better than the modelled continuous exponential growth.
The virus was never growing at the exponential rate that was modelled in the general population. It spread slower through the general population over a longer time period until it worked it's way to the most vulnerable in nursing homes where it could grow at a much faster rate.

wakka12

yosemitesam1 wrote: »

That paper makes a lot of sense and would fit what was seen in the real world much better than the modelled continuous exponential growth.
The virus was never growing at the exponential rate that was modelled in the general population. It spread slower through the general population over a longer time period until it worked it's way to the most vulnerable in nursing homes where it could grow at a much faster rate.

That is not true, ICU and hospital admission grew exponentially in continental Europe

ShineOn7

I noticed this today https://www.bbc.com/news/entertainment-arts-52537293

The Stranglers keyboard player Dave Greenfield has died at the age of 71 after testing positive for Covid-19.


Greenfield died on Sunday having contracted the virus after a prolonged stay in hospital for heart problems.

So cause of death here will be listed as Covid when he was already in a bad way before that?



Without getting into silly conspiracy theory territory, I wonder how many of those listed as Covid victims have actually passed away moreso from something they were already in hospital with

Jurgen Klopp

France confirms covid 19 has been in the country since December

https://www.dailymail.co.uk/news/article-8285461/France-saw-coronavirus-patient-DECEMBER-doctor-claims-examining-samples.html#article-8285461

ShineOn7

Jurgen Klopp wrote: »

France confirms covid 19 has been in the country since December

https://www.dailymail.co.uk/news/article-8285461/France-saw-coronavirus-patient-DECEMBER-doctor-claims-examining-samples.html#article-8285461

Hmmm, Daily Mail though Jurgen

Jurgen Klopp

ShineOn7 wrote: »

Hmmm, Daily Mail though Jurgen

https://www.google.com/amp/s/amp.theguardian.com/world/2020/may/04/french-hospital-discovers-covid-19-case-december-retested

Plenty more to fill your guts on

wakka12

ShineOn7 wrote: »

I noticed this today https://www.bbc.com/news/entertainment-arts-52537293






So cause of death here will be listed as Covid when he was already in a bad way before that?



Without getting into silly conspiracy theory territory, I wonder how many of those listed as Covid victims have actually passed away moreso from something they were already in hospital with

About 14% of people with COVID would have died anyway from other health problems, according to a Skynews study.

Though its important to note that most certainly not every country counts this way whereby all patients who die after testing positive count as COVID deaths.

drunkmonkey

Jurgen Klopp wrote: »

France confirms covid 19 has been in the country since December

https://www.dailymail.co.uk/news/article-8285461/France-saw-coronavirus-patient-DECEMBER-doctor-claims-examining-samples.html#article-8285461

More interestingly from that article,

Video cameras will be used to spy on people to check they are wearing face masks and complying with social distancing in France, it was revealed today.

The city of Cannes on the southern Côte d'Azur has trialled the monitoring software, installed at outdoor markets and on buses.

Welcome to 1984 kids, buckle up.

ShineOn7

What are the thoughts on:

• Today's numbers. Low because of the Bank Holiday weekend or are we seeing them really drop?
• The current number of a country's population who are asymptomatic. WHO says just 5% (I've really gone off WHO though), whereas studies in New York say 20%. As high as 60% has been spoken of elsewhere

IAMAMORON

ShineOn7 wrote: »

What are the thoughts on:

• Today's numbers. Low because of the Bank Holiday weekend or are we seeing them really drop?
• The current number of a country's population who are asymptomatic. WHO says just 5% (I've really gone off WHO though), whereas studies in New York say 20%. As high as 60% has been spoken of elsewhere

I think the low numbers would indicate a reduction in cases. Factor in that there is potentially a 2 week lag and better testing and I think we are beginning to see the wood from the trees. My gut says it is over.

I would say further outbreaks are plausible but we are better equipped to deal with them now. We are certainly witnessing something between the end of the beginning and the beginning of the end.

Asymptomatic figures are going to be pie in the sky until a proper antibody assessment is instigated. But IMO it will be less than more densely populated areas. I could see Dublin being a lot higher than more rural areas. I think we will also have a much lower rate than the UK, their initial herd immunity concept will be starting to benefit them now. They will have a lot more asymptomatic sufferers. In fact any country that has had a lot of cases can probably start assuming this now. The more visible and tangible cases identified the more instance of asymptomatic cases there will be, that is my logic.

We are now into the really dreadful grey area of lockdown winddown. Yuck. Be prepared for over zealous security guards with face masks on and excessive social distancing and other awkwardness. Now is not the time to stop washing your hands either, people will still need to be vigilant. It is going to be a really strange summer.

Peanut Butter Jelly

Usually we see lower numbers on Mondays after the weekends, and a rise to a more consistent number on Tuesdays. Given this is a bank holiday weekend, I would imagine it would be Wednesday before we see that rise back to "normal" figures. But if by Wednesday we're somewhere around where we are now then that can definitely be sen as a drop in numbers when that lag is accounted for

1st dalkey dalkey

ShineOn7 wrote: »

Hmmm, Daily Mail though Jurgen

It's in the Guardian as well.

Seems they back tested samples from people who had been sick in December i.e. samples taken in December, and found covid19.

Could have been some cross contamination in the interim, i suppose, but it is important to find out what the truth is on this.

A lot of stuff changes if this is true, from China's culpability to how we assess transmission.

normanoffside

ShineOn7 wrote: »

What are the thoughts on:

• Today's numbers. Low because of the Bank Holiday weekend or are we seeing them really drop?
• The current number of a country's population who are asymptomatic. WHO says just 5% (I've really gone off WHO though), whereas studies in New York say 20%. As high as 60% has been spoken of elsewhere

They are still not testing the general public, only healthcare workers and Vulnerable.

Of Today's 221 infections 130 are from care homes.

I guess new infections from peoples in those categories are dying out.

ShineOn7

Reddit/Gov.ie

_________________________


New Cases: 211
Total Cases: 21983
New Death: 23
Total Deaths: 1339


Breakdown of Deaths

• Deaths: 1339
• Hospitalised + died in hospital: 542 - 46%
• Died in ICU: 65 - 5%
• Underlying conditions: 1147 - 86%
• Median age: 84
• Mean age: 82

Of 370 in ICU

• Remain in ICU: 90
• Not everyone in ICU is ventilated. Under 5 new cases admitted on daily basis. No new ones yesterday

As of Sunday 3rd May

• Cases: 21659
• Hospitalised: 2879 - 13.30%
• Total In ICU: 369 - 1.70%
• Median Age of cases: 49
• Healthcare workers: 6293 - 29% - Increase of 82 from yesterday

Residential Settings (Includes nursing homes)

• Clusters: 394 - Increase of 3 since yesterday
• Total Cases: 5174 - Increase of 89 since yesterday
• Total Deaths: 819 - 61%
• Hospitalised: 370 - 7.20%
• 91% of all residents and staff have been tested for Covid 19

Nursing Homes

• Clusters: 229 - Increase of 2 since yesterday
• Total Cases: 4108 - Increase of 59 since yesterday
• Lab Confirmed Deaths:
• Hospitalised: 250

Testing

• Total number completed to date: 214,761
• Total completed last week: 61,707
• Positive results returned last week: 2,280 - approximately 3.7% positivity rate
• Current lab testing capacity: Approx. 84,000 tests per week

ShineOn7

TESTING

• re Hitting testing targets by May 18th: De Gascun claims we're on still track to hit 105,000 tests a week by that point. Also, the positivity rate of tests across the board is currently 3.7%. (I think that's lower..? but not 100% on that).
• 214,000 people have been tested thus far. Most are getting results within 2-4 days but there are still reports of longer waits, they want to hear about these and address them. Where do these delays happen? Sometimes contact info is missing, sometimes there are logistical issues in the lab; repeat testing can occasionally be necessary, etc.
• Will we still need the German lab to hit 100,000 per week? A lot of focus on this figure - it ultimately depends on what the demand actually ends up being. But there is an ongoing capacity there should it be needed.
• More on the 100k; ideal scenario is to be able to do a large number of tests and get a small number of positives. They might not need to do 100k a week - they won't know until we ease restrictions and see what happens. The point of 100k is purely to have leftover capacity so they can be comfortable that the figures they're returning are an accurate indication of the virus' spread.

CARE HOMES

• re Care home testing sweep; 91% (527 homes) completed (not all of those results are back yet though). Expected to reach 100% in the coming days. More thorough analyses of these results to follow.
• re staff redeployment to nursing homes - 400 staff have been redeployed thus far. Decision to redeploy depends on the decision making of 23 care teams that visit and assess these homes

PINTS AND WEDDINGS

• Re the plans provided by the pubs to reopen earlier: Holohan hasn't seen these, but talks of the issues w/ physical and social distancing in environments essentially designed to encourage the opposite. Up to these individual industries to try to adjust to NPHET advice (so maybe...?)
• Weddings and similar events don't have clear guidance on rescheduling/adjusting guest numbers etc. Holohan comments that it's difficult to have an event like that be safe given their social nature and involvement of alcohol. Doesn't want to tell people bluntly to reschedule but he seemed to err more that way anyway.

MISC

• re reported "hostile interactions" between Holohan and government ministers: He has not come across any tensions whatsoever, says there's no pressure being applied to NPHET that affects their decision-making either.
• Advice on masks: Non-surgical grade masks can play a role in the community. Currently working on further guidance re their use, should be set out for the 18th. Public transport is one setting that was mentioned, where likely advice might be to wear them here.
• Leaving Cert: this is ultimately up to the Dept of Education; NPHET have set out requirements re social distancing for students and invigilators, the rest is up to them.
• Cocooning in Phase 5; allegedly different advice by NPHET vs government document - key point here is that there will always be different advice for cocooners vs the rest of the population. But this document is just a framework; malleable to change details and fine print can change.
• Are post-mortems being performed on those who die from Covid-19? Not really, there's no indication to do so in cases where the diagnosis has already been established. - reporter claimed that other countries are making "interesting discoveries on autopsy." They will talk to the pathologist to get more info.
• Our reopening strategy vs that of other countries; we will be conservative. Seeing the rate of spread they saw in March reappear could be disastrous - this is to be avoided as much as possible. They do not want to get this wrong and have to retread back through the phases.

ShineOn7

Interesting bit from the HSE doctor doing the AMA on here

Well it is becoming increasingly clear that there are a lot of asymptomatic cases. At present it is a safe bet that at least 50% of cases are asymptomatic, quite probably more. This is important because it does lower the case fatality rate significantly since up until recently we were only testing people with symptoms. I'm aware, over the last two weeks, of many colleagues who were asymptomatic testing positive.

One problem with this is that we don't yet know whether an asymptomatic infection results in a high titre of neutralising antibodies which may lead to immunity for a prolonged period of time - 1 year+. Or it may not lead to immunity but the likelihood is a high titre of neutralising antibodies would lead to immunity TO THE STRAIN YOU GOT but not other strains.

If asymptomatic infection doesn't lead to a high titre of neutralising antibodies then even though those individuals technically had the virus they won't be immune and so can get re-infected by the same strain again.

Generally speaking the more severe an infection you get the more likely you are to come out the other end with a high titre of neutralising antibodies which should confer some level of immunity for some period of time.

SARS-CoV2 hasn't been around long enough for us to know for certain but if it behaves like other viruses then there is a decent chance that those who had signficant symptoms will mostly have high titres of neutralising antibodies ( and some level of immunity for some period of time to be determined) while those who had asymptomatic infections will have mounted a lesser response and have lower titres of neutralisiing antibodies and a lesser level of immunity to no immunity for a lesser period of time. This is the way I'd bet it'll turn out BUT I must stress that we do not know this for certain yet. It won't become clear for several more months.

For example, I'm fairly sure I had COVID-19 back in late February after exposure to a probably case who didn't meet the testing criteria at the time. I had what we now know to be a number of symptoms of COVID-19 but which we didn't realise were related to it at the time (rigors x 2/7, no fever, no dyspnoea, abdominal pain x 2/7, intermittent chest pain x 2/52, dry cough once or twice a day on most days for 2 weeks, massive fatigue worse now at 2 months remove than in the immediate aftermath) and so I never met the testing criteria even though I asked to be tested. I'm assuming I have no immunity because that's the safe way to play it... although when a neutralising antibody titre test comes along I'll be first in line to do it.

cavemeister

Can someone help me work something out? My brain just won't fire today.

I'll give you the numbers first.

USA:
Population - 328.2 Million
Covid-19 confirmed cases - 1,237,761
Total Tests per Million Population - 23,347

MEXICO:
Population - 126.2 Million
Covid-19 confirmed cases - 26,025
Total Tests per Million Population - 776

My (very simple, I'm sure) math question is, if Mexico were to do to the same tests per million as the USA is doing, what would Mexico's confirmed cases be (assuming infection rates are the same as USA)

Thanks

SteelyDanJalapeno

416,767?

cavemeister

Thanks. Could I ask how you calculated it?

Sigma101

cavemeister wrote: »

Can someone help me work something out? My brain just won't fire today.

I'll give you the numbers first.

USA:
Population - 328.2 Million
Covid-19 confirmed cases - 1,237,761
Total Tests per Million Population - 23,347

MEXICO:
Population - 126.2 Million
Covid-19 confirmed cases - 26,025
Total Tests per Million Population - 776

My (very simple, I'm sure) math question is, if Mexico were to do to the same tests per million as the USA is doing, what would Mexico's confirmed cases be (assuming infection rates are the same as USA)

Thanks

USA already has 218 deaths per million of the population, in Mexico it's just 19 deaths per million. Mexico are clearly at an earlier stage than the USA. Numbers of infections per million are probably much lower than the USA at this stage, although the rate at which infections are growing is probably much higher. I'm not sure if comparing the two gives much insights.

tedpan

cavemeister wrote: »

Can someone help me work something out? My brain just won't fire today.

I'll give you the numbers first.

USA:
Population - 328.2 Million
Covid-19 confirmed cases - 1,237,761
Total Tests per Million Population - 23,347

MEXICO:
Population - 126.2 Million
Covid-19 confirmed cases - 26,025
Total Tests per Million Population - 776

My (very simple, I'm sure) math question is, if Mexico were to do to the same tests per million as the USA is doing, what would Mexico's confirmed cases be (assuming infection rates are the same as USA)

Thanks

Workings based on your question and numbers above..

If Mexico were doing the same tests per million as the US(23,347), multiply 23,347 x 126.2(number of million) = 2,946,391

Do the Same for the USA: 23,347 x 328.2 = 7,662,485

Then divide 2,946,391 / 7,662,485 = 0.3845216 etc..

Multiply 0.38452 x 1,237,761(the US confirmed cases) = 475,946

So around 476,000 if Mexico were doing the same level of testing per million and same infection rates..

cavemeister

tedpan wrote: »

Workings based on your question and numbers above..

If Mexico were doing the same tests per million as the US(23,347), multiply 23,347 x 126.2(number of million) = 2,946,391

Do the Same for the USA: 23,347 x 328.2 = 7,662,485

Then divide 2,946,391 / 7,662,485 = 0.3845216 etc..

Multiply 0.38452 x 1,237,761(the US confirmed cases) = 475,946

So around 476,000 if Mexico were doing the same level of testing per million and same infection rates..

Thanks a million

PMBC

I've looked but can't find a schedule and often when I watch on 5.30 news they 'cut back' to the studio. Is it broadcast live on www?
I appreciate that the start migt not be at the exact time scheduled as the team finalises the information

PMBC

Is that the plan for the future as restrictions are now being relaxed?
How achievable do people think that is and/or what are the problems in achieving such an aim?

Boggles

PMBC wrote: »

Is that the plan for the future as restrictions are now being relaxed?
How achievable do people think that is and/or what are the problems in achieving such an aim?

We don't know.

We can look at data of how other countries that are ahead of us and how they are getting on.

The one problem I have seen all ready primarily in the past couple of weeks is a certain cohort of people have completely abandoned social distancing.

It only takes 5-10% to act the bollíx and we are probably on some form of "lockdown" again, which of course will be devastating to the economy.

PMBC

Thanks Boggles.
Shortly after I posted the question, I heard a piece on radio that indicated testing and tracing is coming. There was mention of privacy issues, data storage on each phone rather than centrally etc.
For sure there are and always will be people who for whatever reaons dont obey the rules. It was more the area of staffing for both testing and tracing, lab. equipment, chemicals, swabs etc. that I was querying.

Boggles

PMBC wrote: »

Thanks Boggles.
Shortly after I posted the question, I heard a piece on radio that indicated testing and tracing is coming. There was mention of privacy issues, data storage on each phone rather than centrally etc.
For sure there are and always will be people who for whatever reaons dont obey the rules. It was more the area of staffing for both testing and tracing, lab. equipment, chemicals, swabs etc. that I was querying.

Testing and tracing have all been aspirational at this stage.

It's one of the 3 reasons restrictions were not lifted last week.