The maths of it all and what it means to Ireland

IAMAMORON · 14-04-2020 1:01pm

Ignore that it was a doddle

Deaths (Number) by Month of Occurrence and Year
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
January 2,801 2,667 2,652 2,690 3,274 2,931 2,830 2,694 2,871 2,828 2,967 3,110 3,425
February 2,398 2,432 2,545 2,320 2,350 2,293 2,300 2,515 2,506 2,696 2,756 2,688 2,532
March 2,660 2,739 2,578 2,664 2,383 2,385 2,580 2,631 2,831 2,656 2,875 2,846 2,604
April 2,485 2,505 2,399 2,469 2,252 2,240 2,351 2,374 2,631 2,446 2,478 2,533 2,414
May 2,385 2,337 2,177 2,352 2,284 2,306 2,362 2,424 2,445 2,310 2,539 2,489 2,376
June 2,217 2,155 2,207 1,971 2,158 2,220 2,194 2,224 2,185 2,206 2,241 2,266 2,230
July 2,112 2,308 2,228 2,321 2,133 2,044 2,269 2,220 2,360 2,323 2,160 2,301 2,217
August 2,076 2,132 2,205 2,197 2,139 2,070 2,151 2,269 2,148 2,280 2,286 2,182 2,305
September 2,154 2,191 2,079 2,116 2,104 2,138 2,265 2,173 2,264 2,147 2,333 2,333 2,319
October 2,254 2,168 2,266 2,202 2,347 2,341 2,314 2,475 2,351 2,385 2,527 2,397 2,556
November 2,233 2,283 2,197 2,280 2,342 2,280 2,196 2,406 2,332 2,376 2,314 2,512 2,481
December 2,485 2,571 2,584 2,692 2,614 2,713 2,644 2,781 2,580 2,599 2,651 3,010 2,959
All data is final.

ShineOn7 · 14-04-2020 2:12pm

Deleted User wrote: »

I think from looking at the data, Dublin is Ireland's Lombardy

14-04-2020 2:19pm

ShineOn7 wrote: »

Eh?

You think it'll be this strong a virus for over a decade?

No chance

A virus doesn't magically "weaken" over time. If you think that infect yourself with HIV or Ebola and see how that goes.

People get it and either die or recover and are at least partially immune.

14-04-2020 2:21pm

Jurgen Klopp wrote: »

Go to world meters and the percentage of deaths recorded as having C19 against confirmed cases is a death rate of 5.3% or thereabouts

Now however factor that the Italians have said whatever your confirmed cases are you need to assume there's 5x that due to lack of testing and asymptomatic carriers

Almost no one is testing 100% of the population. Confirmed cases vs deaths doesn't give the death rate, total population vs deaths does.

14-04-2020 2:23pm

Deleted User wrote: »

I think from looking at the data, Dublin is Ireland's Lombardy, just at a less severe scale. When you look at Italy, regions such as Lazio and Sicily, which have a similar population to Ireland actually have a lower case rate and death rate than we do, and its falling rapidly. Then when we look at Ireland, the east, dominated by Dublin has the large bulk of cases and deaths. The major factor in this appears to be the development of a large outbreak prior to lockdown. This happened in Lombardy, Madrid, London, New York and to a lesser extent in Dublin. This, I believe, gives an insight into why some countries fared much better - no major outbreak occurred prior to lockdown

Considering they allowed in planeloads of people from Lombardy it's hardly surprising.

CalamariFritti · 14-04-2020 2:32pm

There was a representative study conducted recently when a German institute basically turned the worst affected area in Germany into a field science lab. Google it, the town is called Heinsberg and its quite prominent on google for obvious reasons right now.

Now the study may have had an agenda and I'm not sure it was their place to make early policy recommendations, but the numbers apparently are correct and welcomed by scientists across the world. Their interpretation may be a different matter but at the moment we all want better data.

Anyway its not claiming to be conclusive with regards to extrapolation to a larger sample size. However the mortally rate they arrived at is 0.37%, about 4 times as high as the seasonal flu.
Which is not me saying its in the same ballpark as the seasonal flu. Its may be one day but right now its not and that is because we have no treatment and this whole overwhelming the ICU thing.

The important thing is the number. It suggests that putting confirmed cases (AKA tested positively) vs deaths and thinking the thing has a mortality rate of 4% or worse is of course nonsense. Because if you only test people with symptoms and people in hospitals you gonna miss a lot of people with no or only mild symptoms. It appears that number of people is quite large. In that Heinsberg study 15% of the people already had antibodies and they never knew they had it.

IAMAMORON · 14-04-2020 2:53pm

CalamariFritti wrote: »

There was a representative study conducted recently when a German institute basically turned the worst affected area in Germany into a field science lab. Google it, the town is called Heinsberg and its quite prominent on google for obvious reasons right now.

Now the study may have had an agenda and I'm not sure it was their place to make early policy recommendations, but the numbers apparently are correct and welcomed by scientists across the world. Their interpretation may be a different matter but at the moment we all want better data.

Anyway its not claiming to be conclusive with regards to extrapolation to a larger sample size. However the mortally rate they arrived at is 0.37%, about 4 times as high as the seasonal flu.
Which is not me saying its in the same ballpark as the seasonal flu. Its may be one day but right now its not and that is because we have no treatment and this whole overwhelming the ICU thing.

The important thing is the number. It suggests that putting confirmed cases (AKA tested positively) vs deaths and thinking the thing has a mortality rate of 4% or worse is of course nonsense. Because if you only test people with symptoms and people in hospitals you gonna miss a lot of people with no or only mild symptoms. It appears that number of people is quite large. In that Heinsberg study 15% of the people already had antibodies and they never knew they had it.

I am still not comfortable with this estimate. If you pump in Irelands' population into the .37% you are talking over 17,000 deaths in the republic, too high. If that figure is 4 times as high as the annual flu then you are talking circa 4,250 flu deaths annually in the republic, I don't think so, that is over 10% of our annual deaths, roughly speaking?

ShineOn7 · 14-04-2020 3:27pm

IAMAMORON wrote: »

I am still not comfortable with this estimate. If you pump in Irelands' population into the .37% you are talking over 17,000 deaths in the republic

Is that presuming 100% of the population gets it though?

Cordell · 14-04-2020 3:28pm

However the mortally rate they arrived at is 0.37%, about 4 times as high as the seasonal flu

From what I can see this a deceiving number, it's 0.37% of the total population of the town, whereas flu deaths are reported as a % of the ones that were diagnosed with it. If my understanding is true, then this means 18000 deaths which will be a disaster.
I hope that something was lost in translation and it's 0.37% of those 15% positive - which is still quite bad but more inline with a really really bad flu season.

CalamariFritti · 14-04-2020 3:34pm

I'm not sure its fair to call it an estimate but its not my data so I won't be looking to argue it one way or the other.

I think it does give perspective to some of the misconceptions out there. Talking to friends and family, there is a lot of people out there thinking this will kill one in twenty or one in ten. Or if you're over 60 and you get it you're a goner.

Stuff like that. People are scared.

Information is good.

Hmmzis · 14-04-2020 3:37pm

More fuel to the fire, some newer numbers for the R0 values:

https://wwwnc.cdc.gov/eid/article/26/7/20-0282_article

It is highly infectious. Right up there with rubella and polio. Without a vaccine this thing is going to haunt us (all of humanity) for a very long time. If those numbers are correct, we would need about 4 million people immune in Ireland to declare the 'all clear'. This makes me depressed.

The 'upside' to this is that the German and Iceland numbers are becoming more and more credible, so there is hope that it isn't as deadly as originally calculated. The problem with it is, that it makes lots of people very sick at the same time while the vast majority would have mild to no symptoms making tracing and isolating a herculean effort.

Godspeed to our healthcare people and everyone else on the front lines (and to the rest of us all, too).

ShineOn7 · 14-04-2020 3:38pm

Cordell wrote: »

From what I can see this a deceiving number, it's 0.37% of the total population of the town, whereas flu deaths are reported as a % of the ones that were diagnosed with it.

Mother of jaysus, can we get any survey or source or feckin' anything concrete on the mortality rate?

I'm personally still keeping it as 2% in my head to stay extra vigilant

CalamariFritti · 14-04-2020 3:39pm

Cordell wrote: »

From what I can see this a deceiving number, it's 0.37% of the total population of the town, whereas flu deaths are reported as a % of the ones that were diagnosed with it. If my understanding is true, then this means 18000 deaths which will be a disaster.
I hope that something was lost in translation and it's 0.37% of those 15% positive - which is still quite bad but more inline with a really really bad flu season.

Its not lost in translation and they went public with it. There is a fair bit of scrutiny going on at the moment. If nothing else I'd be fairly confident they didn't make elementary mistakes like that. And antibodies doesn't mean positive. Its means the opposite I thought. It means you were once positive but now you have antibodies and your body will no longer host the virus. No? So the deaths will certainly not come out of that group.

IAMAMORON · 14-04-2020 3:46pm

ShineOn7 wrote: »

Is that presuming 100% of the population gets it though?

Well yes, that is the absolute mortality rate?

Even if you apply say an 80% infection rate, you are still looking at 13,500 deaths in the republic, it looks too big if I am being honest.

If that is the case we have a long long way to go at 30 odd deaths a day.

I have said it before, I think you have to have an open mind when looking at northern Italy and Madrid etc. I reckon the virus was there in December, they just did not know it.

The reality is that it was probably also in the likes of New York of London then also. These are large cosmopolitan cities, it is entirely feasible that were suffering earlier than diagnosed.

One other thing, a lot of cranks seem to gripe on about testing etc. There are a few fundamental items which are overlooked, particularly in the west.

Language - this might sound simple but I have dealt with mainland China in the past and the language barrier is considerable. Hong Kong not as bad, but try deciphering their packaging, invoices etc, without any grasp of Mandarin and good luck to you. We are talking about an infant virus which no one really understands yet being translated literally via " Chinese Whispers " , there is lots of scope for error, especially in the transfer of medical data. Compound that with fake news or poor medical practice or sinister CCP involvement and you can imagine how mix ups arise and occur.

That being the case Europe has had to tackle the virus piecemeal. This is why Germany are trying to reclassify each case to establish mutations.

The language barrier is not such an issue for Asian countries such as Korea etc, They would have a handle on Mandarin, Cantonese and other provincial dialects. I know we take this for granted but not every international translator is fully aware of medical intricacies, just saying.

Cordell · 14-04-2020 3:51pm

CalamariFritti wrote: »

Its not lost in translation and they went public with it. There is a fair bit of scrutiny going on at the moment. If nothing else I'd be fairly confident they didn't make elementary mistakes like that. And antibodies doesn't mean positive. Its means the opposite I thought. It means you were once positive but now you have antibodies and your body will no longer host the virus. No? So the deaths will certainly not come out of that group.

I'm not debating the study as I can't find it, I'm just doubtful that the english articles I found have all the right numbers. Do you have a link for the original?

And having antibodies mean that you are either infected or you were infected and you have already recovered.

ShineOn7 · 14-04-2020 3:55pm

Hmmzis wrote: »

More fuel to the fire, some newer numbers for the R0 values:

https://wwwnc.cdc.gov/eid/article/26/7/20-0282_article

It is highly infectious. Right up there with rubella and polio. Without a vaccine this thing is going to haunt us (all of humanity) for a very long time. If those numbers are correct, we would need about 4 million people immune in Ireland to declare the 'all clear'. This makes me depressed.

The 'upside' to this is that the German and Iceland numbers are becoming more and more credible, so there is hope that it isn't as deadly as originally calculated. The problem with it is, that it makes lots of people very sick at the same time while the vast majority would have mild to no symptoms making tracing and isolating a herculean effort.

Godspeed to our healthcare people and everyone else on the front lines (and to the rest of us all, too).

at R0 = 5.7, this threshold rises to 82% (i.e., >82% of the population has to be immune, through either vaccination or prior infection, to achieve herd immunity to stop transmission)

An R0 of almost 6!

Ah lads. There's no avoiding getting this till a vaccine is here (estimated times mentioned has been everything from September to 3 years from now)

Best to just pray it's one of the milder stains we get

normanoffside · 14-04-2020 4:01pm

CalamariFritti wrote: »

There was a representative study conducted recently when a German institute basically turned the worst affected area in Germany into a field science lab. Google it, the town is called Heinsberg and its quite prominent on google for obvious reasons right now.

Now the study may have had an agenda and I'm not sure it was their place to make early policy recommendations, but the numbers apparently are correct and welcomed by scientists across the world. Their interpretation may be a different matter but at the moment we all want better data.

Anyway its not claiming to be conclusive with regards to extrapolation to a larger sample size. However the mortally rate they arrived at is 0.37%, about 4 times as high as the seasonal flu.
Which is not me saying its in the same ballpark as the seasonal flu. Its may be one day but right now its not and that is because we have no treatment and this whole overwhelming the ICU thing.

The important thing is the number. It suggests that putting confirmed cases (AKA tested positively) vs deaths and thinking the thing has a mortality rate of 4% or worse is of course nonsense. Because if you only test people with symptoms and people in hospitals you gonna miss a lot of people with no or only mild symptoms. It appears that number of people is quite large. In that Heinsberg study 15% of the people already had antibodies and they never knew they had it.

The study is here: https://reason.com/2020/04/09/preliminary-german-study-shows-a-covid-19-infection-fatality-rate-of-about-0-4-percent/

0.37% of those infected had died (the exact same % as Iceland).
While that number is low, it is still really important (obviously) to keep the infection rate as low as possible especially among the more vulnerable of the popualtion.

IAMAMORON · 14-04-2020 4:03pm

Hmmzis wrote: »

More fuel to the fire, some newer numbers for the R0 values:

https://wwwnc.cdc.gov/eid/article/26/7/20-0282_article

It is highly infectious. Right up there with rubella and polio. Without a vaccine this thing is going to haunt us (all of humanity) for a very long time. If those numbers are correct, we would need about 4 million people immune in Ireland to declare the 'all clear'. This makes me depressed.

The 'upside' to this is that the German and Iceland numbers are becoming more and more credible, so there is hope that it isn't as deadly as originally calculated. The problem with it is, that it makes lots of people very sick at the same time while the vast majority would have mild to no symptoms making tracing and isolating a herculean effort.

Godspeed to our healthcare people and everyone else on the front lines (and to the rest of us all, too).

My issue with using this date is that it is based on Chinese figures from January, they simply cannot be trusted, it is also based on date from 140 cases.... it is also examining the contagiousness of the virus under Chinese travel conditions.

Where did you get your quoted data for rubella and polio from this report? I am not seeing any such contrasts or comparisons? This report is detailing potential contagion rates based on a sample of 140 infected persons ? Given average stochasticity it could not be perfect by any means, the virus was seemingly less than 2 months old in January?

From the report -

"Data
Individual Case Reports
We collected publicly available reports of 140 confirmed COVID-19 cases (mostly outside Hubei Province). These reports were published by the Chinese Centers for Disease Control and Prevention (China CDC) and provincial health commissions; accession dates were January 15–30, 2020 (Appendix 1 Table 1)"

CalamariFritti · 14-04-2020 4:05pm

Cordell wrote: »

I'm not debating the study as I can't find it, I'm just doubtful that the english articles I found have all the right numbers. Do you have a link for the original?

https://www.theguardian.com/world/2020/mar/31/virologists-to-turn-germany-worst-hit-district-into-coronavirus-laboratory

Not sure about original but its all over the media in Germany.

https://www.tagesspiegel.de/wissen/zweifel-an-zuverlaessigkeit-ausgewerteter-tests-unplausible-zahlen-kritik-an-heinsberg-studie/25732878.html

https://www.sueddeutsche.de/wissen/heinsberg-studie-herdenimmunitaet-kritik-1.4873480

You can see from the articles its under immense scrutiny at the moment. It appears its idea was to transport a certain message. It was part funded by commercial interest. The message was we can go back to work soon. And for that its under fire. But the numbers are what they are.

Cordell wrote: »

And having antibodies mean that you are either infected or you were infected and you have already recovered.

Yes so the deaths are not going to come out that group then.

Hmmzis · 14-04-2020 4:08pm

IAMAMORON wrote: »

My issue with using this date is that it is based on Chinese figures from January, they simply cannot be trusted, it is also based on date from 140 cases.... it is also examining the contagiousness of the virus under Chinese travel conditions.

Where did you get your quoted data for rubella and polio from this report? I am not seeing any such contrasts or comparisons? This report is detailing potential contagion rates based on a sample of 140 infected persons ? Given average stochasticity it could not be perfect by any means, the virus was seemingly less than 2 months old in January?

From the report -

"Data
Individual Case Reports
We collected publicly available reports of 140 confirmed COVID-19 cases (mostly outside Hubei Province). These reports were published by the Chinese Centers for Disease Control and Prevention (China CDC) and provincial health commissions; accession dates were January 15–30, 2020 (Appendix 1 Table 1)"

Numbers for polio and rubella:
https://en.wikipedia.org/wiki/Basic_reproduction_number

among other nasties in there.

Cordell · 14-04-2020 4:17pm

CalamariFritti wrote: »

https://www.theguardian.com/world/2020/mar/31/virologists-to-turn-germany-worst-hit-district-into-coronavirus-laboratory

Not sure about original but its all over the media in Germany.

https://www.tagesspiegel.de/wissen/zweifel-an-zuverlaessigkeit-ausgewerteter-tests-unplausible-zahlen-kritik-an-heinsberg-studie/25732878.html

https://www.sueddeutsche.de/wissen/heinsberg-studie-herdenimmunitaet-kritik-1.4873480

You can see from the articles its under immense scrutiny at the moment. It appears its idea was to transport a certain message. It was part funded by commercial interest. The message was we can go back to work soon. And for that its under fire. But the numbers are what they are.

Yes, the numbers are what they are, and that was my question: what are the numbers?

Now I have the answer, it's 0.37% out of those 15% infected. Or 0.055% of the total population - that means around 2800 deaths here.

CalamariFritti wrote: »

Yes so the deaths are not going to come out that group then.

Yes it is - anyone dying from it have the antibodies.

IAMAMORON · 14-04-2020 4:18pm

The .37% could well be accurate, if it is that means we have a considerable damage limitation exercise ahead of us nationally. It would also indicate how effective the lockdown in Ireland has been. It looks to have saved 1000's of lives which is fantastic.

The next big problem then is when are we all allowed out to play again? Surely such action would be a disaster?

1st dalkey dalkey · 14-04-2020 5:21pm

Deleted User wrote: »

I think from looking at the data, Dublin is Ireland's Lombardy, just at a less severe scale. When you look at Italy, regions such as Lazio and Sicily, which have a similar population to Ireland actually have a lower case rate and death rate than we do, and its falling rapidly. Then when we look at Ireland, the east, dominated by Dublin has the large bulk of cases and deaths. The major factor in this appears to be the development of a large outbreak prior to lockdown. This happened in Lombardy, Madrid, London, New York and to a lesser extent in Dublin. This, I believe, gives an insight into why some countries fared much better - no major outbreak occurred prior to lockdown

These places are also transport hubs and have high population densities.
Both facilitating to any viral transmission and explosion.

Makes one wonder about our push toward ever denser population numbers and housing models.

CalamariFritti · 14-04-2020 5:32pm

Cordell wrote: »

Yes, the numbers are what they are, and that was my question: what are the numbers? Now I have the answer, it's 0.37% out of those 15% infected. Or 0.055% of the total population - that means around 2800 deaths here.

Yes it is - anyone dying from it have the antibodies.

Now I'm no expert and I'm not debating this but where do you get that from? Just trying to judge how I might have misunderstood.

What I get from those very articles is the number that was published was 0.37% and it was the overall mortality rate.

Not 0.37% out of the 15% that have antibodies.

The way they say it is the 15% with antibodies are the immune ones. Antibodies means you had it and now you're immune, no? The deaths couldn't possible come out of that group then. That wouldn't make any sense.

You sure you got your info straight?

normanoffside · 14-04-2020 5:45pm

CalamariFritti wrote: »

Now I'm no expert and I'm not debating this but where do you get that from? Just trying to judge how I might have misunderstood.

What I get from those very articles is the number that was published was 0.37% and it was the overall mortality rate.

Not 0.37% out of the 15% that have antibodies.

The way they say it is the 15% with antibodies are the immune ones. Antibodies means you had it and now you're immune, no? The deaths couldn't possible come out of that group then. That wouldn't make any sense.

You sure you got your info straight?

It's the same as the Iceland Figures.

They calculate the % that have been infected and work out how many people in total have been infected.

From this they can work the death rate from the total number of people that have been infected.

For example 75 people have died from covid 19 in a town of 100,000.
You work out the % of the population infected by doing the antibody test and see it's 20%

Therefore 20,000 people (approximately) have been infected and 75 have died, giving a death rate of 0.375%

(These are not the actual numbers in the study, just an example of how they are worked out.)

Cordell · 14-04-2020 5:47pm

Yes, from here: https://reason.com/2020/04/09/preliminary-german-study-shows-a-covid-19-infection-fatality-rate-of-about-0-4-percent/

Fatality rate is calculated as a percentage of those infected, as it normally is for any other disease. Calculating it as a percentage of the total population makes no sense, and 0.37% of the total population is a huge number.
And those infected are those that have antibodies.

Jurgen Klopp · 14-04-2020 6:07pm

CalamariFritti wrote: »

I'm not sure its fair to call it an estimate but its not my data so I won't be looking to argue it one way or the other.

I think it does give perspective to some of the misconceptions out there. Talking to friends and family, there is a lot of people out there thinking this will kill one in twenty or one in ten. Or if you're over 60 and you get it you're a goner.

Stuff like that. People are scared.

Information is good.

This is exactly it, it should be given more airtime. But it wont cause the one industry who's duty to inform makes their money from people being scared and generating all those lovely clicks

If sickens my stomach. I know so many poor misfortunes that think it's the new black death. Ones with a touch of asthma, ones over 60, poor oul crathúrs of 40 who are a little overweight all thinking they are goners if they get it and these scumbags like the daily mail and their likes only loving it

Jurgen Klopp · 14-04-2020 6:56pm

The Danish Health Agency believes they have 30 to 80 times more infected than their official number

https://www.thelocal.dk/20200408/danish-health-agency-says-400000-could-have-been-infected?fbclid=IwAR0Y6YufD-pEabWUHitMf6vSpgPbX8myyEwmJsNbG7jc2TTF_lmP_OTJZvs

IAMAMORON · 14-04-2020 7:22pm

If we have tested 90,000 persons so far, how come only 11,479 have tested positive? What was wrong with the other 78,521 people?

Why is 1 in every 8 people tested turning up negative?

Finally what is the benefit of testing if 1 in 8 persons are not infected and thus are potentially capable of being infected in the future?

normanoffside · 14-04-2020 7:34pm

IAMAMORON wrote: »

If we have tested 90,000 persons so far, how come only 11,479 have tested positive? What was wrong with the other 78,521 people?

Why is 1 in every 8 people tested turning up negative?

Finally what is the benefit of testing if 1 in 8 persons are not infected and thus are potentially capable of being infected in the future?

It’s a good question, there are a couple of reasons:

1. Health care workers are being tested routinely and often more than once for screening purposes.
2. When there was a big backlog lots of people will have recovered by the time they were tested.

Otherwise the simple answer is that we have to test loads of people otherwise we can’t find the positive cases.
It’s impossible in most cases to know if they have it until they are tested as the symptoms are so similar to colds and flus.

The maths of it all and what it means to Ireland

Comments