• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is COVID strategy moving towards herd-immunity?!

Status
Not open for further replies.

Racheski

Major Contributor
Forum Donor
Joined
Apr 20, 2020
Messages
1,116
Likes
1,699
Location
Chicago
That's the situation today, and it certainly doesn't look great. However, all those other countries are still seeing new cases. Whenever the epidemic is over (whether or not the virus becomes endemic), we may find that Sweden simply had more cases (and deaths) early while other countries spread the impact over a longer time (at great cost to the economy). Or not. It's too early to be drawing any firm conclusions from data like that.
The only conclusion I drew was that as of 8/3 Sweden is doing bad compared to the rest of Europe, if you use case counts and deaths normalized by population as metrics.
 
OP
L

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
That depends on the metrics you use to determine what is "bad." ...
View attachment 76582
yep, short term Sweden is doing pretty bad (EU-wise). Actually, the entire top5 in your table is using a form of herd-strategy. For different reasons but practically quite similar. I just said "not that bad" in the sense of "nowhere near the doom scenarios with millions of deaths". And globally their numbers are somewhat average.
There is still a chance that they'll do much better that most in the long term. No way to prove that with the available data but the chance (still) exists.

P.S.
I do not want to give the impression that I like or advocate herd-immunity. I don't. It's just one of the strategies on the table (possibly only because "the table" is quite empty). Might be the worst, might be the best, might be just average, might be...
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
4,685
Likes
10,700
Location
Hampshire
The only conclusion I drew was that as of 8/3 Sweden is doing bad compared to the rest of Europe, if you use case counts and deaths normalized by population as metrics.
It's too early to make that assessment. They could be doing bad, or they could simply be ahead of the pack. The hospitals are coping, so nobody is dying from lack of treatment.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
That depends on the metrics you use to determine what is "bad." If you use case counts and deaths normalized by population then they are doing bad compared to the rest of Europe. Here is data comparing Sweden to all countries in Europe with a population >2M as of 8/3:
View attachment 76582
So they are the worst for total cases per million, and 4th worst for total deaths. They have over 4k cases per million more than the European average, and 300 deaths per million more than the European average.
Data was taken from https://ourworldindata.org/covid-cases
If anyone wants the underlying Excel data PM me.

Since the excess deaths and covid-19 related deaths does not add up, the list cannot be used for any valid science. A few countries, including Sweden, match fairly well and Belgium even the opposite. In addition, counting the number of reported cases are even worse. Looking at the data for e.g. invasive pneumococci, the decrease has been identical in Sweden and Finland, with 90% reduction in April and May. If the strategies with social distancing would have been vastly different, this would have shown up in other respiratory diseases. But it does not, indicating that there are other variables at play. For example the number of people that initially contracted the infection and travelled to the becoming hot spots in countries may differ by a great extent, and may play more important role in how the epidemic developed.
 

BostonJack

Active Member
Editor
Joined
Jul 2, 2019
Messages
288
Likes
350
Location
Boston area, Cambridge, MA
Just read yesterday that vaccine studies are revealing that ~50% of the subjects have some existing immune response to covid-19, based on antibody tests prior to the trial.

Current theory is that existing corona viruses have similar enough proteins that there is overlap in the T-cell responses, perhaps not enough to grant full immunity, but enough to lower the severity of the infection.

I'm on the side that herd immunity is a poor strategy, but some combination of transmission mitigation and vaccines is likely to turn this virus into a recurring endemic headache, but not as tragic as current circumstances.

Jack
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
3,221
Likes
5,494
Which is another reason why the vaccine route might be quite bumpy. The best we can do might very well be a bi-annual flu+Covid vaccine (with partial results, same as the current flu ones).

Generally looks like many/most dislike the herd-strategy and use the "we don't have enough data" argument to 'demonstrate' that it'll not work. Let's not forget that "we don't have enough data" applies the same to everything else. Any other strategy can be demolished with the exact same arguments.

It's mostly guess work at this time and that's what I wanted from this thread: guess/predict the next strategy so we can better prepare for it. It's not like anyone will 'solve' Covid today .. or in this thread.

Yeah but, I present a conditional, I'm not making any categorical claims. If it is the case C19 behaves in the same manner with respect to immune responses in the same way one would have a response to the seasonal flu, then it stands to reason heard immunity's effects on the flu would also be similar to the effects it would have on C19. Maybe I'm misunderstanding what people say when they mean heard immunity in this context. The extrapolation is based on the current understanding of precedent (that being, there is no long-immunity timeline for C viruses, they're all within a few months, to two years at best basically), those sorts of timelines make the C19 virus seem like (for all intents and purposes) like a seasonal flu, just without mutation requirements to persist with repeated patterns of infection/rates.
 

Racheski

Major Contributor
Forum Donor
Joined
Apr 20, 2020
Messages
1,116
Likes
1,699
Location
Chicago
Since the excess deaths and covid-19 related deaths does not add up, the list cannot be used for any valid science. A few countries, including Sweden, match fairly well and Belgium even the opposite. In addition, counting the number of reported cases are even worse. Looking at the data for e.g. invasive pneumococci, the decrease has been identical in Sweden and Finland, with 90% reduction in April and May. If the strategies with social distancing would have been vastly different, this would have shown up in other respiratory diseases. But it does not, indicating that there are other variables at play. For example the number of people that initially contracted the infection and travelled to the becoming hot spots in countries may differ by a great extent, and may play more important role in how the epidemic developed.
This is not an excessive deaths due to COVID calculation. It is simply the countries COVID deaths per million minus the European average per million to give a sense of how each country’s rate compares to the European average.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
This is not an excessive deaths due to COVID calculation. It is simply the countries COVID deaths per million minus the European average per million to give a sense of how each country’s rate compares to the European average.

Yes, and since covid-19 deaths are registered very different between countries, it cannot be used as true covid-19-related deaths. It must be quite clear by now, given all reports on the subject.

https://www.economist.com/graphic-detail/2020/07/15/tracking-covid-19-excess-deaths-across-countries
 

Racheski

Major Contributor
Forum Donor
Joined
Apr 20, 2020
Messages
1,116
Likes
1,699
Location
Chicago
Yes, and since covid-19 deaths are registered very different between countries, it cannot be used as true covid-19-related deaths. It must be quite clear by now, given all reports on the subject.

https://www.economist.com/graphic-detail/2020/07/15/tracking-covid-19-excess-deaths-across-countries
Which is why I filtered the data to only European countries: first-world nations, government-sponsored healthcare, vetted by the WHO & ECDC. Now there are going to be outliers, like Belgium, where some of argued their official numbers are inflated compared to other countries because of less stringent requirements for what is considered a COVID19 death. When you start to deep dive on individual countries there are indeed nuances that may bias the data.
The main issue I have with the economist article is they are racking and stacking countries by excess deaths over different time periods, which is a bad methodology. They need to keep the time periods the same, or normalize the numbers over a static period of time otherwise it is not apples to apples. Instead the article arbitrarily starts counting after the first 50 COVID deaths.
We need to be very careful about trusting epidemiological analyses from those without expertise in the subject matter.
 

Vasr

Major Contributor
Joined
Jun 27, 2020
Messages
1,409
Likes
1,922
The main issue I have with the economist article is they are racking and stacking countries by excess deaths over different time periods, which is a bad methodology. They need to keep the time periods the same, or normalize the numbers over a static period of time otherwise it is not apples to apples. Instead the article arbitrarily starts counting after the first 50 COVID deaths.
Not necessarily. But it does matter what inferences you draw from it. So, I would question that approach specific to the inference being made, not a blanket statement. It is a valid statistical method, nothing to do with epidemiological factors.

Economist does better vetting of its content than most magazines (whether I agree with their conclusions or not).

There is a good reason to start tracking once the pandemic takes hold in a country as the spread took some time for different geographical areas. It is a statistical way of normalizing for different starting periods of an event so that comparisons can be made for the same stage of the epidemic in the country.

For inferences of the type, how do the countries compare in the trajectory at some point in time, you can normalize it to starting point and say how the trajectories have varied for similar stage in their history. You can compare without normalizing to say which are current hotspots and which aren't for which the normalization would make no sense.

There are far more variables, of course as countries later affected had better information, factors causing spread like travel had been reduced and so could clamp down faster (if they made the right decisions), so comparing across countries is at best, tenuous and not very valid whether you normalize or not. Best would be countries within a small region that is geographically and demographically close as you have done.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
Which is why I filtered the data to only European countries: first-world nations, government-sponsored healthcare, vetted by the WHO & ECDC. Now there are going to be outliers, like Belgium, where some of argued their official numbers are inflated compared to other countries because of less stringent requirements for what is considered a COVID19 death. When you start to deep dive on individual countries there are indeed nuances that may bias the data.
The main issue I have with the economist article is they are racking and stacking countries by excess deaths over different time periods, which is a bad methodology. They need to keep the time periods the same, or normalize the numbers over a static period of time otherwise it is not apples to apples. Instead the article arbitrarily starts counting after the first 50 COVID deaths.
We need to be very careful about trusting epidemiological analyses from those without expertise in the subject matter.

So you agree that you need to be careful to interpret covid-19-related deaths? Euromomo shows a couple of European countries where the z-scores differ compared to cover-reported deaths. For example, Belgium shows 25% higher z-score during the outbreak period compared to the Netherlands, and the reported deaths are a whooping 136 % higher, indicating an underreporting from the Netherlands. You can by comparing these data see indications from underreporting from several countries in Europe, which also have been confirmed in reports from Spain and UK. Sweden reports covid-19 deaths irrespectively of cause 30 days after a positive test results as well as covid-19 deaths from doctors certificate. Those deaths with a confirmed lab test and doctors certificate are almost 800 fewer deaths than the 30-day death report (which are the official numbers).

In summary, countries count differently, and numbers need to be interpreted with some caution.
 
OP
L

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
Belgium is an interesting outlier. In terms of cases/deaths per million pop they are doing very bad. Might just be "world champions".

I haven't heard mentions of an official herd-strategy but according to my contacts there, the quarantine was quite 'flexible'. Not much enforcement for masks, small bars still open, small concerts, etc. If my contacts are any indication, people had a very relaxed "don't care" attitude. All that would add up to a 'masked' form of herd-strategy.
Or maybe it's about different reporting rules like @Racheski mentioned.

Any belgians around here?
 
Last edited:

A Surfer

Major Contributor
Joined
Jul 1, 2019
Messages
1,125
Likes
1,230
I would also be curious to see if in Belgium for instance the population was on average older. Are the deaths still occurring in the age groups that elsewhere have been demonstrated to be most at risk? How about looking if there is a relationship between extended families living together. I would expect that in societies where older relatives still live with the family you see increased transmission leading to serious illness course and death. I still do not think that we are seeing a particularly lethal virus in the sense of being broadly lethal across all age ranges. While deaths do occur in younger people the numbers compared with the likely number of infections is extremely low. Low enough that of those young people who did die there could be unknown underlying conditions/factors that intersected.

We are already seeing herd immunity building as clearly more and more of the population is slowly becoming exposed and infected. It may be a slow spread, but it seems to me that it is inevitable.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,383
Likes
24,749
Location
Alfred, NY
I would also be curious to see if in Belgium for instance the population was on average older. Are the deaths still occurring in the age groups that elsewhere have been demonstrated to be most at risk?

Here, that's certainly the case (source- AZ Department of Health Services):

1596629910671.png
 

A Surfer

Major Contributor
Joined
Jul 1, 2019
Messages
1,125
Likes
1,230
It will take years yet to really understand I believe, but if I am correct as with many pathogens there will be a relationship with overall health which brings into play quite a few systems. That won't be surprising at all of course, but it is also possible that this pathogen becomes lethal when certain specific conditions are in play that we have not currently mapped out/understand.

Could there be a relationship that among younger people who do get very sick or die that being overweight and or diabetic is a factor? What type of pre-existing health conditions create a dangerous vulnerability? Certainly we know of some, but it seems unlikely that we know all such risk factors for serious illness course and death.
 

Racheski

Major Contributor
Forum Donor
Joined
Apr 20, 2020
Messages
1,116
Likes
1,699
Location
Chicago
I would also be curious to see if in Belgium for instance the population was on average older. Are the deaths still occurring in the age groups that elsewhere have been demonstrated to be most at risk? How about looking if there is a relationship between extended families living together. I would expect that in societies where older relatives still live with the family you see increased transmission leading to serious illness course and death. I still do not think that we are seeing a particularly lethal virus in the sense of being broadly lethal across all age ranges. While deaths do occur in younger people the numbers compared with the likely number of infections is extremely low. Low enough that of those young people who did die there could be unknown underlying conditions/factors that intersected.

We are already seeing herd immunity building as clearly more and more of the population is slowly becoming exposed and infected. It may be a slow spread, but it seems to me that it is inevitable.
The dataset I referenced earlier included median age, percent of population over 65, and percent of population over 75. I didn’t include the numbers because I thought it would be too much information in one table, and they looked pretty similar among European countries, but the data is there if you want to dig in.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
Belgium is an interesting outlier. In terms of cases/deaths per million pop they are doing very bad. Might just be "world champions".
I haven't heard mentions of an official herd-strategy but according to my contacts there, the quarantine was quite 'flexible'. Not much enforcement for masks, small bars still open, small concerts, etc. If my contacts are any indication, people had a very relaxed "don't care" attitude. All that would add up to a 'masked' form of herd-strategy.
Or maybe it's about different reporting rules like @Racheski mentioned.

Any belgians around here?

It is an outlier mainly due to reporting differences and this is evident by looking at euromomo statistics. The Netherlands is as ”bad” as Belgium.
 
OP
L

lashto

Major Contributor
Forum Donor
Joined
Mar 8, 2019
Messages
1,045
Likes
535
It is an outlier mainly due to reporting differences and this is evident by looking at euromomo statistics. The Netherlands is as ”bad” as Belgium.
you're right, Euromomo shows BE and NL to be ~same and the Corona-only worldometer shows BE as >2x worse in terms of deaths/million. Too many damn stats :)
 

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,437
Likes
4,686
Belgian here ;) yeah, the stats are somewhat inflated because of the reporting standard (which essentially was "any death with any respiratory syndrome is a counted as a COVID-19 death). Due to other reporting issues (multiple reporting authorities) that even led to COVID deaths being above excess mortality for a while.

That being said, we didn't do well by far, especially in retirement/elderly care homes. We usually do have a fairly decent affordable (relatively to the rest of the world) care system for elderly people and, imho, it backfired in that particular case. Many of those hospices are fairly linear in design, with a nurse (or eventually nurses + MD, depending on the type of hospice) doing daily tours, visiting each and every resident/patient. PPE equipment shortages made it impossible to swap protective gear on a per patient basis, and the speed at which contamination spread did the rest. I had to spend a couple of mandatory months in one of those "MRS" during my studies and can very well imagine how I could have been an unwilling vector (MRS stands for "house of rest and care" and are dedicated to older people who require medical treatment/monitoring).

We had a relatively strict lockdown which was decently followed and at the beginning of July, we were having only a very low number of cases. That lead to an easing of the lockdown, and the current bounce, which seems to mostly hit a different target. I don't have the time to get into detailed numerical analysis, but current cases seem to be around 20 to 30 times **less** likely to end up in intensive care.

Now, more generally, almost everyone can be unhappy about how some aspects of the pandemic were handled in their home countries. Some people were very "proud" of their country's response or strategy at first, then it soured or backfired. What we have to keep in mind is that we are dealing with a coronavirus that (except for mortality and morbidity) behaves very much like the common cold coronaviruses. That, btw, isn't very positive in terms of immunization and eradication... Governments definitely need to plan and be ready but a lot of the rest is mostly posturing, one way or another.
 
Status
Not open for further replies.
Top Bottom