blonde poker forum
Welcome, Guest. Please login or register.
August 14, 2020, 09:00:41 PM

Login with username, password and session length
Search:     Advanced search
2252988 Posts in 66539 Topics by 16612 Members
Latest Member: MT’s bro
* Home Help Arcade Search Calendar Guidelines Login Register
+  blonde poker forum
|-+  Community Forums
| |-+  The Lounge
| | |-+  COVID19
0 Members and 1 Guest are viewing this topic. « previous next »
Pages: 1 ... 154 155 156 157 [158] 159 160 161 162 ... 195 Go Down Print
Author Topic: COVID19  (Read 79992 times)
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2355 on: May 21, 2020, 09:52:09 AM »


....and the best measure for the U.K.:

https://twitter.com/ons/status/1263386765882273797?s=21

Still no difference in infection rates across age groups, I think we can move on from that idea.
Logged
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2356 on: May 21, 2020, 10:26:21 AM »


....and the best measure for the U.K.:

https://twitter.com/ons/status/1263386765882273797?s=21

Still no difference in infection rates across age groups, I think we can move on from that idea.

You do realise that there is zero evidence there is no difference too?  The sample size is way too small here. There are 35 people across all age bands that tested positive.  So for the 12-19 band this likely just equates to 1 or 2 people. Even if they saw none in any band, they wouldn't be able to conclude anything on different rates of infection for that age group.  Look at figure 3 for the size of the confidence intervals.

I'd say the same on the conclusion about health care professionals not having different rates of infection.  If you look at figure 4 it is easy to think the infection rate is the same, but look at the confidence interval.  From this you can see that the number of positive health care professionals is in single figures.  So even if you saw double or half the numbers amongst health care workers I suspect you couldn't conclude that there was any difference.  

You cannot provide any evidence for the last two bullets without big enough sample sizes.   If I produced this study, and had been asked to comment on the differences, I would only have included the last two bullet points if I could have included a major caveat on sample sizes.  

So whilst the Spanish study I linked to may have provided evidence we could move on, this provides no evidence we can.
« Last Edit: May 21, 2020, 10:33:01 AM by Doobs » Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2357 on: May 21, 2020, 10:34:07 AM »


....and the best measure for the U.K.:

https://twitter.com/ons/status/1263386765882273797?s=21

Still no difference in infection rates across age groups, I think we can move on from that idea.

You do realise that there is zero evidence there is no difference too?  The sample size is way too small here. There is 35 people across all age bands that tested positive.  So for the 12-19 band this likely just equates to 1 or 2 people. Even if they saw none in any band, they wouldn't be able to conclude anything on different rates of infection for that age group.  Look at figure 3 for the size of the confidence intervals.

I'd say the same on the conclusion about health care professionals not having different rates of infection.  If you look at figure 4 it is easy to think the infection rate is the same, but look at the confidence interval.  From this you can see that the number of positive health care professionals is in single figures.  So even if you saw double or half the numbers amongst health care workers I suspect you couldn't conclude that there was any difference. 

You cannot provide any evidence for the last two bullets without big enough sample sizes.   If I produced this study, and had been asked to comment on the differences, I would only have included the last two bullet points if I could have included a major caveat on sample sizes. 

So whilst the Spanish study I linked to may have provided evidence we could move on, this provides no evidence we can.

It was meant to be taken in combination with the previous link from all the seroprevalence studies. It is the best measure we currently have in the U.K, I agree that by itself it doesn’t tell us much but combine it with the Italian, Spanish, Swedish, Norwegian and Danish seroprevalence studies and I am ready to assign it a high level of confidence.
Logged
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2358 on: May 21, 2020, 11:14:48 AM »


....and the best measure for the U.K.:

https://twitter.com/ons/status/1263386765882273797?s=21

Still no difference in infection rates across age groups, I think we can move on from that idea.

You do realise that there is zero evidence there is no difference too?  The sample size is way too small here. There is 35 people across all age bands that tested positive.  So for the 12-19 band this likely just equates to 1 or 2 people. Even if they saw none in any band, they wouldn't be able to conclude anything on different rates of infection for that age group.  Look at figure 3 for the size of the confidence intervals.

I'd say the same on the conclusion about health care professionals not having different rates of infection.  If you look at figure 4 it is easy to think the infection rate is the same, but look at the confidence interval.  From this you can see that the number of positive health care professionals is in single figures.  So even if you saw double or half the numbers amongst health care workers I suspect you couldn't conclude that there was any difference. 

You cannot provide any evidence for the last two bullets without big enough sample sizes.   If I produced this study, and had been asked to comment on the differences, I would only have included the last two bullet points if I could have included a major caveat on sample sizes. 

So whilst the Spanish study I linked to may have provided evidence we could move on, this provides no evidence we can.

It was meant to be taken in combination with the previous link from all the seroprevalence studies. It is the best measure we currently have in the U.K, I agree that by itself it doesn’t tell us much but combine it with the Italian, Spanish, Swedish, Norwegian and Danish seroprevalence studies and I am ready to assign it a high level of confidence.

Why are you assigning it a high level of confidence when I have shown it to be poor?   This detracts from the useful Spanish study, doesn't add to it.   We should just wait for something better from the UK. 
Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2359 on: May 21, 2020, 11:41:59 AM »


....and the best measure for the U.K.:

https://twitter.com/ons/status/1263386765882273797?s=21

Still no difference in infection rates across age groups, I think we can move on from that idea.

You do realise that there is zero evidence there is no difference too?  The sample size is way too small here. There is 35 people across all age bands that tested positive.  So for the 12-19 band this likely just equates to 1 or 2 people. Even if they saw none in any band, they wouldn't be able to conclude anything on different rates of infection for that age group.  Look at figure 3 for the size of the confidence intervals.

I'd say the same on the conclusion about health care professionals not having different rates of infection.  If you look at figure 4 it is easy to think the infection rate is the same, but look at the confidence interval.  From this you can see that the number of positive health care professionals is in single figures.  So even if you saw double or half the numbers amongst health care workers I suspect you couldn't conclude that there was any difference. 

You cannot provide any evidence for the last two bullets without big enough sample sizes.   If I produced this study, and had been asked to comment on the differences, I would only have included the last two bullet points if I could have included a major caveat on sample sizes. 

So whilst the Spanish study I linked to may have provided evidence we could move on, this provides no evidence we can.

It was meant to be taken in combination with the previous link from all the seroprevalence studies. It is the best measure we currently have in the U.K, I agree that by itself it doesn’t tell us much but combine it with the Italian, Spanish, Swedish, Norwegian and Danish seroprevalence studies and I am ready to assign it a high level of confidence.

Why are you assigning it a high level of confidence when I have shown it to be poor?   This detracts from the useful Spanish study, doesn't add to it.   We should just wait for something better from the UK. 

My reasoning would be that seroprevalence in children would be expected be consistent across all nations that have followed a broadly similar approach and so if they identify very similar patterns, they offer at least some level of corroboration to one and other. I can’t see any reason not to derive conclusions having combined all of the findings from all of the studies, especially useful in a country like the U.K. that doesn’t have enough data yet. The epidemiologists keep saying time is everything, so I chose not to wait.
Logged
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2360 on: May 21, 2020, 01:00:43 PM »


....and the best measure for the U.K.:

https://twitter.com/ons/status/1263386765882273797?s=21

Still no difference in infection rates across age groups, I think we can move on from that idea.

You do realise that there is zero evidence there is no difference too?  The sample size is way too small here. There is 35 people across all age bands that tested positive.  So for the 12-19 band this likely just equates to 1 or 2 people. Even if they saw none in any band, they wouldn't be able to conclude anything on different rates of infection for that age group.  Look at figure 3 for the size of the confidence intervals.

I'd say the same on the conclusion about health care professionals not having different rates of infection.  If you look at figure 4 it is easy to think the infection rate is the same, but look at the confidence interval.  From this you can see that the number of positive health care professionals is in single figures.  So even if you saw double or half the numbers amongst health care workers I suspect you couldn't conclude that there was any difference. 

You cannot provide any evidence for the last two bullets without big enough sample sizes.   If I produced this study, and had been asked to comment on the differences, I would only have included the last two bullet points if I could have included a major caveat on sample sizes. 

So whilst the Spanish study I linked to may have provided evidence we could move on, this provides no evidence we can.

It was meant to be taken in combination with the previous link from all the seroprevalence studies. It is the best measure we currently have in the U.K, I agree that by itself it doesn’t tell us much but combine it with the Italian, Spanish, Swedish, Norwegian and Danish seroprevalence studies and I am ready to assign it a high level of confidence.

Why are you assigning it a high level of confidence when I have shown it to be poor?   This detracts from the useful Spanish study, doesn't add to it.   We should just wait for something better from the UK. 

My reasoning would be that seroprevalence in children would be expected be consistent across all nations that have followed a broadly similar approach and so if they identify very similar patterns, they offer at least some level of corroboration to one and other. I can’t see any reason not to derive conclusions having combined all of the findings from all of the studies, especially useful in a country like the U.K. that doesn’t have enough data yet. The epidemiologists keep saying time is everything, so I chose not to wait.

You linked to an Italian study of blood donors ffs.   How does that help your conclusion about children? The bloke you linked to is only claiming that there is low seroprevalence in the general population of these countries. 
Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2361 on: May 21, 2020, 01:12:07 PM »

Swedish study conclusion translation below.   They have actually concluded the rates are different by age group.  Don't worry, they don't look like they have enough data either.  5% of 1100 samples is only 55 people across all age groups.  So 0-19 and 65-70 seem certain to have sample sizes that are too small to make conclusions.

First results from ongoing study of antibodies for covid-19 virus PUBLISHED MAY 20, 2020 It is most common for people between the ages of 20 and 64 to have an infection with covid-19, compared to people in other age groups. It shows the first results from the Public Health Agency's investigation of the presence of antibodies against the virus in blood samples. The Public Health Authority has launched a survey to measure and estimate how many in the community have had covid-19.

Blood samples are collected from laboratories in clinical chemistry and clinical immunology in nine regions: Jämtland, Jönköping, Kalmar, Skåne, Stockholm, Uppsala, Västerbotten, Västra Götaland and Örebro. The collection takes place during eight weeks in the spring of 2020. A total of 1,200 samples are collected each week for analysis of antibodies. Antibodies show that the immune system recognizes the SARS-CoV-2 virus.

The analyzes for week 18 (a total of 1,104 analyzed samples) show, as expected, the largest proportion of positive antibody tests in Stockholm. A total of 7.3 percent of the blood samples collected from people in Stockholm were positive in the antibody study, which can be compared with a total of 4.2 percent in Skåne and 3.7 percent in Västra Götaland. The numbers reflect the state of the epidemic earlier in April, as it takes a few weeks for the body's immune system to develop antibodies.

Regarding age differences, the results show that covid-19 antibodies were most common among people between 20 and 64 years. In total, 6.7 percent of the samples in this group were positive, which can be compared with 4.7 percent in the age group 0-19 years and 2.7 percent in the age group 65-70 years.
Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2362 on: May 21, 2020, 01:15:51 PM »

Danish study is over 18s only, and the bloke who did it actually says people should use great caution due to his sample size...
Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2363 on: May 21, 2020, 01:21:55 PM »

The Norweigan link is to somebody changing their estimated population positive rate by using new French data...

Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2364 on: May 21, 2020, 01:38:58 PM »

Danish study is over 18s only, and the bloke who did it actually says people should use great caution due to his sample size...

Seems to be plenty of hostility in your response to this, forgive me if I’m wrong on that. If overall seroprevalence is the same across nations/outbreaks where similar or even almost identical approaches were taken. Why would it not follow that the mechanism of spread through the population at varying age groups wasn’t also likely to be the same? Italy and Spain are the same Norway and Denmark are the same. If the spread wasn’t occurring through the same age demographic, I would expect the numbers to be measurably different. The data is insufficient but the time for a decision on schools in the U.K. seems to be now.

Just as an aside, as you might of missed it, the Spanish study had already been linked and discussed itt before you linked it.
Logged
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2365 on: May 21, 2020, 01:45:01 PM »

The Norweigan link is to somebody changing their estimated population positive rate by using new French data...


In spite of the response above, I’ll hold my hands up to some due diligence failings on my morning Twitter check. Not an excuse, some good news actually, I had my first work meeting in months 😊, it meant research time on this topic was limited though.
Logged
Pokerpops
Hero Member
*****
Offline Offline

Posts: 1207


View Profile
« Reply #2366 on: May 21, 2020, 03:15:52 PM »

The Norweigan link is to somebody changing their estimated population positive rate by using new French data...


In spite of the response above, I’ll hold my hands up to some due diligence failings on my morning Twitter check. Not an excuse, some good news actually, I had my first work meeting in months 😊, it meant research time on this topic was limited though.

I hope your work gets really busy.
Logged

"More than at any other time in history, mankind faces a crossroads. One path leads to despair and utter hopelessness. The other, to total extinction. Let us pray we have the wisdom to choose correctly."
Doobs
Hero Member
*****
Offline Offline

Posts: 14863


View Profile
« Reply #2367 on: May 21, 2020, 03:23:32 PM »

Danish study is over 18s only, and the bloke who did it actually says people should use great caution due to his sample size...

Seems to be plenty of hostility in your response to this, forgive me if I’m wrong on that. If overall seroprevalence is the same across nations/outbreaks where similar or even almost identical approaches were taken. Why would it not follow that the mechanism of spread through the population at varying age groups wasn’t also likely to be the same? Italy and Spain are the same Norway and Denmark are the same. If the spread wasn’t occurring through the same age demographic, I would expect the numbers to be measurably different. The data is insufficient but the time for a decision on schools in the U.K. seems to be now.

Just as an aside, as you might of missed it, the Spanish study had already been linked and discussed itt before you linked it.

I am pretty hostile to most poor usage of stats/science.   I do make a habit of checking a lot of links, so it isn't just you.  If Toby Young posted here, I suspect I'd have much more work to do.

I suspect there is some commanality across populations, I wasn't suggesting there wasn't.   I was just getting grumpy about the claims the study claiming more than it should have, and then you doubling down and claiming lots of evidence of similar results across ages and across countries when there wasn't much.  Even now you are still saying you'd expect them to be measurably different, when I have already shown the sample sizes in many of these studies are too small to show any significant differences.  

If you look at all the 3 studies where they split by age (UK, Spain and Sweden), all show lower rates for children than adults.  Even in total, I don't think you can conclude there is ample evidence either way.  There might be a lower rate amongst children, but it isn't significantly lower and any small difference could easily disappear if they reopen schools.  

Well done on posting the Spainish study first, will try and be more diligent in wading through your links in future.  
Logged

Most of the bets placed so far seem more like hopeful punts rather than value spots
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2368 on: May 21, 2020, 03:31:33 PM »

The Norweigan link is to somebody changing their estimated population positive rate by using new French data...


In spite of the response above, I’ll hold my hands up to some due diligence failings on my morning Twitter check. Not an excuse, some good news actually, I had my first work meeting in months 😊, it meant research time on this topic was limited though.

I hope your work gets really busy.

My work won’t get busy any time soon. I imagine I’ll take another Blonde sabbatical when/if the immunotherapy restarts though, so you’ll get some respite then 😊.
Logged
kukushkin88
Hero Member
*****
Offline Offline

Posts: 3877



View Profile
« Reply #2369 on: May 21, 2020, 03:46:24 PM »

Danish study is over 18s only, and the bloke who did it actually says people should use great caution due to his sample size...

Seems to be plenty of hostility in your response to this, forgive me if I’m wrong on that. If overall seroprevalence is the same across nations/outbreaks where similar or even almost identical approaches were taken. Why would it not follow that the mechanism of spread through the population at varying age groups wasn’t also likely to be the same? Italy and Spain are the same Norway and Denmark are the same. If the spread wasn’t occurring through the same age demographic, I would expect the numbers to be measurably different. The data is insufficient but the time for a decision on schools in the U.K. seems to be now.

Just as an aside, as you might of missed it, the Spanish study had already been linked and discussed itt before you linked it.

I am pretty hostile to most poor usage of stats/science.   I do make a habit of checking a lot of links, so it isn't just you.  If Toby Young posted here, I suspect I'd have much more work to do.

I suspect there is some commanality across populations, I wasn't suggesting there wasn't.   I was just getting grumpy about the claims the study claiming more than it should have, and then you doubling down and claiming lots of evidence of similar results across ages and across countries when there wasn't much.  Even now you are still saying you'd expect them to be measurably different, when I have already shown the sample sizes in many of these studies are too small to show any significant differences.  

If you look at all the 3 studies where they split by age (UK, Spain and Sweden), all show lower rates for children than adults.  Even in total, I don't think you can conclude there is ample evidence either way.  There might be a lower rate amongst children, but it isn't significantly lower and any small difference could easily disappear if they reopen schools.  

Well done on posting the Spainish study first, will try and be more diligent in wading through your links in future.  

That’s cool, I’m always grateful when you point out where there’s (sometimes huge) room for improvement and in hindsight I can see why my approach could be irritating. One big assumption I made, that was wrong but as you suggest might ultimately be a factor is when the Italian researchers concluded ‘social distancing might be more effective in young people’, I thought they were citing school closures. The viral load study from Geneva included children, it might be worth a revisit, I’ll try and find it again.
Logged
Pages: 1 ... 154 155 156 157 [158] 159 160 161 162 ... 195 Go Up Print 
« previous next »
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!
Page created in 0.364 seconds with 20 queries.