Jump to content

COVID-19 #13 or: How I Learned to Stop Worrying and Love the Disease


Mr. Chatywin et al.

Recommended Posts

3 hours ago, Ran said:

Reopening primary and elementary schools is about the only thing I'd recommend for the US. There is no evidence of schools in Stockholm, our hardest-hit city, causing clusters of infection. In all this time with open schools, we've only had an issue at one school, in the small city of Skellefteå.

This is weird. Unless a large fraction of the children are somehow immune,  I would have expected that COVID19 had already burned its way through Swedish schools and kindergartens and the families and teachers of these children. Who is then getting infected in Sweden? Which segment of the populace has more social contact than children to get infected faster?

 

Link to comment
Share on other sites

9 minutes ago, The Great Unwashed said:

Reading through that thread, I see the argument against the reporting on this. However, all it seems that she's saying is that the results haven't yet been replicated in a laboratory setting, and she concedes that the strain "mutation" may "increase viral fitness or infectivity", but cautions against extrapolating changes in the R0.

I completely agree from a scientific standpoint. I'm also not a scientist, so if anyone wants to put me on blast for talking out my ass, be gentle.

However, it seems like this information could have time-sensitive policy implications if the mutation does indeed turn out to be more transmissible, especially since the earliest hit nations, like China and South Korea, are opening up and, presumably, have not yet had to deal with this new strain. 

From a decision-making perspective, I'd want this information asap, in order to prepare accordingly, even if it turns out to be a false alarm.

Wow, interesting and worrisome. Specially in the light of the recent reveal that there was community spread in France back in December, which is to some level in contradiction what we saw in February and March. However, Wuhan in January also saw a European-like surge of cases and deaths. Did it come from there in late January whilst the original strain arrived much earlier and avoided detection for too long? Or the French dude got another older strain?

Link to comment
Share on other sites

9 minutes ago, rotting sea cow said:

Wow, interesting and worrisome. Specially in the light of the recent reveal that there was community spread in France back in December, which is to some level in contradiction what we saw in February and March. However, Wuhan in January also saw a European-like surge of cases and deaths. Did it come from there in late January whilst the original strain arrived much earlier and avoided detection for too long? Or the French dude got another older strain?

It definitely has a lot of explanatory power given the cases popping up earlier and earlier in the record.

Also, has possible implications for antibody testing, no?

Link to comment
Share on other sites

11 hours ago, Ran said:

 

Australia seems to have had a similar experience to Sweden, FWIW.

 

8 hours ago, rotting sea cow said:

Which segment of the populace has more social contact than children to get infected faster?

 

Yes it is super strange. In the Australian context we're doing super aggressive track and trace... and the data seems to indicate that children aren't particularly prone to being infected by the virus, and in the rare case they are they don't seem to transmit readily. You'd expect this thing to be tearing through our schools otherwise as especially young kids can't really social distance. But no real clusters have been found yet.

ETA: Changes are to aren't. Damn early morning brain.

Link to comment
Share on other sites

9 minutes ago, The Great Unwashed said:

It definitely has a lot of explanatory power given the cases popping up earlier and earlier in the record.

I'm starting to think there might be another - milder - strain in the wild which hasn't been detected yet because all sequences are coming from hospitals and therefore very aggressive strains . The French should really sequence the genome of whatever that guy had.  Fast.

 

9 minutes ago, The Great Unwashed said:

Also, has possible implications for antibody testing, no?

I guess so? There are wild variation in the serological studies. I cannot judge them but those numbers reporting 50x the number of infected seem to me that are in contradiction with what we know about the pathology of the disease. What are they detecting? Some cold antibodies? Some previous milder wave of this virus? Some garbage?

More importantly, do these antibodies protect against the most aggressive variants? Why do we see so many asymptomatic people and even some seemingly immune?

Link to comment
Share on other sites

15 minutes ago, The Great Unwashed said:

Also, has possible implications for antibody testing, no?

Maybe this might make the cheap do-it-yourself tests pretty pointless, if you'd need multiple ones to test all the different strains. Just like there's no point to produce antibody tests for a flu virus, as there's a new version every year. :(

Fortunately there seem to be a few new machines in laboratories that might be able to detect the different strains.  (Sorry, only in German, but it says that this machine can make 100 tests per hour, and you seem to be able to use it to detect many different kinds of antibodies to different diseases).

 

 

Link to comment
Share on other sites

1 hour ago, The Great Unwashed said:

Reading through that thread, I see the argument against the reporting on this. However, all it seems that she's saying is that the results haven't yet been replicated in a laboratory setting, and she concedes that the strain "mutation" may "increase viral fitness or infectivity", but cautions against extrapolating changes in the R0.

I completely agree from a scientific standpoint. I'm also not a scientist, so if anyone wants to put me on blast for talking out my ass, be gentle.

However, it seems like this information could have time-sensitive policy implications if the mutation does indeed turn out to be more transmissible, especially since the earliest hit nations, like China and South Korea, are opening up and, presumably, have not yet had to deal with this new strain. 

From a decision-making perspective, I'd want this information asap, in order to prepare accordingly, even if it turns out to be a false alarm.

I saw this twitter thread about the mutation:

I know we've just been discussing the usefulness of Twitter posts and I can't really judge the accuracy of the posts, but I thought it was interesting that at the end of March this Twitter thread was discussing the same mutation mentioned in the LA Times as being the dominant mutation. From reading the LA Times article you would think this was brand new information (the scientific paper may not be claiming this), but it seems to have been known about back in March.

Link to comment
Share on other sites

1 hour ago, The Great Unwashed said:

Reading through that thread, I see the argument against the reporting on this. However, all it seems that she's saying is that the results haven't yet been replicated in a laboratory setting, and she concedes that the strain "mutation" may "increase viral fitness or infectivity", but cautions against extrapolating changes in the R0.

I completely agree from a scientific standpoint. I'm also not a scientist, so if anyone wants to put me on blast for talking out my ass, be gentle.

However, it seems like this information could have time-sensitive policy implications if the mutation does indeed turn out to be more transmissible, especially since the earliest hit nations, like China and South Korea, are opening up and, presumably, have not yet had to deal with this new strain. 

From a decision-making perspective, I'd want this information asap, in order to prepare accordingly, even if it turns out to be a false alarm.

Trying to extrapolate what an effect a point mutation will have on protein function, and then infectivity of the virus in the absence of any experimental data is extremely shaky ground. You just can't make any kind of good conclusions based soley on sequence data. You use this kind of bioinformatic analysis to give an indication of what to look at, but a good portion of the time it turns out to be nothing. So yeah I agree that sure this needs to be looked at, but at the same time just seeing the headline and start of the first paragraph (can't see the article as it's behind a paywall) that reporting is pretty sensationalist and irresponsible.

1 hour ago, The Great Unwashed said:

It definitely has a lot of explanatory power given the cases popping up earlier and earlier in the record.

Also, has possible implications for antibody testing, no?

So a new strain (strain is a very fuzzy and not particularly scientific concept based largely around consensus of what is / isn't a significant shift) doesn't necessarily mean that the antibodies will differ in any significant way. In the case of a single mutation I doubt it'd make much difference if any at all. I'd think a good antibody test will be going for antibodies hitting some pretty well conserved antigen.

 

Link to comment
Share on other sites

10 hours ago, Which Tyler said:

UK's excess deaths figures - updated today, and as of 10 days ago. Also included is FT's comparison to the same figures for other countries - last updated on 01/05/20
 

 

Is there any convincing explanation yet for why the non-Covid death rate is so high? I hoped that there might be a silver lining here, surely vehicle accidents must be down? Maybe work place related accidents? It seems strange that it’s pushing up the figure up so much, unless of course they later transpire to be Covid after all.

Link to comment
Share on other sites

3 hours ago, Kalbear said:

There's some of it here - the Puyallup schools closed and Bothell schools closed for a bit after staff and a kid tested positive. This was super early in though, and we closed shortly afterwards. 

https://www.kiro7.com/news/local/coronavirus-washington-state-timeline-outbreak/IM65JK66N5BYTIAPZ3FUZSKMUE/

Thanks. The reports of these basically seem to be of the "some person tested positive, maybe" (the Bothell one was "a relative of someone who works here is maybe positive") and I can't find any further reporting providing evidence of any kind of clusters at these schools. Bothell's also a high school, and for Science Reasons that I am only vaguely familiar with it's primary and elementary schools that are recommended to be open, not high schools.

1 hour ago, rotting sea cow said:

This is weird. Unless a large fraction of the children are somehow immune,  I would have expected that COVID19 had already burned its way through Swedish schools and kindergartens and the families and teachers of these children. Who is then getting infected in Sweden? Which segment of the populace has more social contact than children to get infected faster?

People who work with others, people who use public transport, etc. 

Besides the fact that children's immune systems may be just very good at tackling this, children also are less likely to transmit it because they displace less air with their coughs, and with less force, so droplet transmission would be reduced.  There's this whole thing in pediatrics about how children don't necessarily cough in situations that would make adults cough, as well, and so far most kids who have been tested as having it seem to not cough or barely cough and so on.

 

Link to comment
Share on other sites

8 hours ago, Raja said:

Do you have a source for that?

That's totally fine, but in my experience, you have to do work with anything you read, weather that is something published in JAMA, NEJM, The Lancet etc - I don't think that is something unique to twitter or twitter threads. In my opinion, it is intellectually lazy not to look critically analyze the things you read, no matter the source. Going back to your point about physicians making mistakes, I totally agree, and that's why a healthy dose of skepticism is warranted when reading *anything*, including the aforementioned journals given the countless examples of completely bogus peer reviewed things that are published. I just think blanket statements regarding twitter threads don't really capture the nuance. To me, and maybe it's the way I use it, it is not a replacement for primary source information, but a very useful addition to it.

Well, initially, it was just a gut feeling which is why I stated that the number "seemed" low, but here's what I've found.  From page 24 of the "Surveillance of influenza and other respiratory viruses in the UK Winter 2018 to 2019" published by Public Health England:

Quote

Through the USISS mandatory scheme, a total of 3,157 ICU/HDU admissions of confirmed influenza were reported across the UK from week 40 2018 to week 15 2019, including 312 deaths, based on combined data from England, Scotland and Northern Ireland.

From page 21 of the "Surveillance of influenza and other respiratory viruses in the  UK Winter 2017 to 2018":

Quote

Through the USISS mandatory scheme, a total of 3,454 ICU/HDU admissions of confirmed influenza were reported across the UK from week 40 2017 to week 15 2018, including 372 deaths, based on combined data from England, Scotland and Northern Ireland.

These two winters already add up to greater than the 5782 cohort.  To complete the number, we would need part of the 2016-17 winter numbers (Week 1-15 2017) and part of the 2019-2020 winter numbers (Week 40-52 2019).  Conservatively, you can probably add at least a thousand to the total from the 2017-2018 and 2018-2019 reports to get to the 2017-2019 number.  Which brings me back to my original question.  How was the 5782 cohort selected?  The two sets of numbers don't match up.

I'm a skeptic by nature, so I tend to read everything with a grain of salt.  As you note, even peer reviewed papers can be wrong (for a variety of reasons from outright fraud, to poor experimental design, to poor interpretation of the results, to bad luck), and mainstream publications from reputable organizations, like the New York Times, can also be wrong.  That said, the peer review process and the fact-checking process make it less likely that junk gets through.  No one who reads a lot has time to fact check everything.  For me to fact check everything in that thread was frankly a waste of my time, but since I'm participating in this thread, I went through with it.  Imagine trying to fact check a long New York Times investigative piece.  Not going to try.  If something sticks out to me funny, maybe I'll do a little digging, but that's about it, but a lot of the time I'll just think "I don't know about that..." and move on.

Kalbear cited another twitter thread, from another tweeter that I'm not familiar with.  Her tweets seemed generally reasonable, so I'm not going to bother fact checking everything.  The problem with this though is it risks confirmation bias if I only fact check things I disagree with or think might be wrong.

Link to comment
Share on other sites

https://www.nytimes.com/2020/05/05/us/coronavirus-deaths-cases-united-states.html?

But nevermind let us lie all over the place, including the gov of Nebraska + other rethug govs writing in the WaPo about what a great job they've done in keeping the bug away, unlike those states with Dem govs.

Shutting down the joke they called a pandemic taskforce because, you know folks, just like the deathcultists promised, it's all over with.  Translate these things and what is really being said is that all of us should be taking a chance of dying or life diminishment forever so he can tickle the stock market and get re-elected.  That's the only value for Them -- we all die, all we useless people of color, poor, elderly, disabled in any way, etc.  Just no point in us other than to help Them get richer.  If that isn't what THEY are saying -- what are They saying because everything They say out loud and in public is a lie.  Except this time they aren't even bothering to pretend.

 

Link to comment
Share on other sites

3 hours ago, Mudguard said:

Well, initially, it was just a gut feeling which is why I stated that the number "seemed" low, but here's what I've found.  From page 24 of the "Surveillance of influenza and other respiratory viruses in the UK Winter 2018 to 2019" published by Public Health England:

From page 21 of the "Surveillance of influenza and other respiratory viruses in the  UK Winter 2017 to 2018":

These two winters already add up to greater than the 5782 cohort.  To complete the number, we would need part of the 2016-17 winter numbers (Week 1-15 2017) and part of the 2019-2020 winter numbers (Week 40-52 2019).  Conservatively, you can probably add at least a thousand to the total from the 2017-2018 and 2018-2019 reports to get to the 2017-2019 number.  Which brings me back to my original question.  How was the 5782 cohort selected?  The two sets of numbers don't match up.

Two points to note here - Firstly, I also had a look at the papers that you have linked above - the data there includes Scotland whereas the ICNARC does not ( It includes England, Wales & Northern Ireland but excludes Scotland).

We could probably figure that one out if we really wanted to, but I think his broader point re: increasing ICU capacity because COVID 19 has caused a surge in ICU admissions that is equivalent to a couple of winters worth of flu cases stands. As I said, it's the best national level ICU data ( at least the English speaking ones) I have seen so far, which in turn allows us to compare the two diseases whilst being cognizant of the fact that there are limitations ( Just like we would for articles appearing in peer reviewed work) and why I think the ICNARC data is quite useful.

The argument from me is not that twitter or twitter thread replace peer reviewed work, but that it's a platform that when used correctly is a useful adjunct. As I stated in  my original comment, twitter is good when you have experts distilling information down to a level that someone outside their profession can understand because lets face it, a lot of people have very little experience with good methodology when it comes to medical research ( It is less good when you have threads regarding 8 units of blood without much else to go off of - amplifying those leaves me uneasy)

( I also think familiarity with a person means little, at least for me, but ymmv on that)

Link to comment
Share on other sites

This is an excellent set of tools and education on how the whole modeling system works and what things like R0 and R and why we care about this means - as well as what certain plans do to impact the models:

https://ncase.me/covid-19/?fbclid=IwAR011ciQjOM6Sl2ev370_MN09b20iC8rue5SQA6u2cEb9TlGt-HioH0DsSc

Link to comment
Share on other sites

9 hours ago, Ran said:

First I've heard of this. Any thing I can Google on it? I'd be curious about primary and elementary school cases, specifically. And day cares, I suppose. I've tried searching and am coming up with nothing along those lines, except for very early cases of "We closed because one person was/maybe was sick" and no real evidence of clusters of infections associated with these situations.

Australia seems to have had a similar experience to Sweden, FWIW.

True, but New Zealand's second largest cluster was in a school.  

Link to comment
Share on other sites

9 hours ago, Ran said:

Besides the fact that children's immune systems may be just very good at tackling this, children also are less likely to transmit it because they displace less air with their coughs, and with less force, so droplet transmission would be reduced.  There's this whole thing in pediatrics about how children don't necessarily cough in situations that would make adults cough, as well, and so far most kids who have been tested as having it seem to not cough or barely cough and so on.

What I'd seen on this front is that things which impact the lungs in adults can attack the digestive tract in children instead, and that this is being seen in cases of covid amongst kids. Unfortunately I don't still have the link handy but it also talked about something like "covid toes" with kids having toes turn bright red or purple due to... Some other effect I'm forgetting.

Also I'd be wary of drawing any conclusions from the lack of spread in Australian schools, we have had only a little sustained community spread and over 50% of parents pulled their kids out of school before the formal shut down. You need to bear in mind that a big chunk of our cases, particularly the big spike, was due to cases from the Ruby Princess cruise ship, along with a couple of others, and the first wave of infections from those on the cruise ship that were allowed back into the community.

Link to comment
Share on other sites

5 hours ago, ants said:

True, but New Zealand's second largest cluster was in a school.  

Marist College, a high school, yes. My understanding is there's an associated elementary school where there have been cases, but these seem to be household transmission (older siblings passing it to parents and younger siblings). 

 

@karaddin

Sure, but this is the experience in other places. Not just Sweden, France as well for certain, and countries in Europe that shut down schools are reopening them because the incidence in serious illness in children is so rare.

Link to comment
Share on other sites

It's definitely a claim I've seen made in a number of places, including in the media here as well - I'm just saying our experience isn't really data towards either side in this.

There's also cross cultural confusion caused by some of the ways it's talked about, we only have 2 levels of school in Aus so when someone is talking in the American context and says "primary and elementary" that sounds like you're talking about all of school, and there is data showing the risk at 16 the risk is more aligned with regular adults so there is more push back on that. In reality the two schools you were talking about when you used that phrase earlier excludes those 16+ because those are in "high school" in that context. 

Link to comment
Share on other sites

11 hours ago, DaveSumm said:

Is there any convincing explanation yet for why the non-Covid death rate is so high? I hoped that there might be a silver lining here, surely vehicle accidents must be down? Maybe work place related accidents? It seems strange that it’s pushing up the figure up so much, unless of course they later transpire to be Covid after all.

Undiagnosed / misdiagnosed COVID cases

Patients being unwilling to "burden" the NHS with their problems leading to delayed diagnosis & treatment

Patients being scared to go out, or request repeat prescriptions, so are reducing their dose without consultation

Patients requiring urgent care, but there's no capacity to take them, or care arrives too late

Patients seeing ANOther therapist, who might pick up on a warning sign for something that the patient would otherwise ignore / not know about

Rise in domestic violence and suicide due to lockdown

 

All of which are still related to Covid-19 - which is why I feel they're the most useful statistic of the lot - not to mention that the total excess deaths is the one thing we can be pretty sure of (though the excess is likely higher, as they'd been lower for the year before covid arrived, and other causes of death SHOULD be reduced).

Link to comment
Share on other sites

11 hours ago, Ran said:

 

Besides the fact that children's immune systems may be just very good at tackling this, children also are less likely to transmit it because they displace less air with their coughs, and with less force, so droplet transmission would be reduced.  There's this whole thing in pediatrics about how children don't necessarily cough in situations that would make adults cough, as well, and so far most kids who have been tested as having it seem to not cough or barely cough and so on.

 

 

In the Heinsberg/Gangelt study

(https://www.ukbonn.de/C12582D3002FD21D/vwLookupDownloads/Streeck_et_al_Infection_fatality_rate_of_SARS_CoV_2_infection2.pdf/%24FILE/Streeck_et_al_Infection_fatality_rate_of_SARS_CoV_2_infection2.pdf)

they found that household transmission was low and related to household size:

"A significant association between household cluster size and the per-person infection risk was found (Fig. 5B, p<0.001). In a two-person householdcluster, the estimated risk for the second infection increasedfrom 15.53%to43.59% [25.26%; 64.60%]; in a three-person householdclustertheestimatedrisk for the second and third personsincreasedfrom 15.53%to35.71% [19.57%; 55.60%] each, and in a four-person household cluster the estimated risk for the second, third and fourth personsincreasedfrom 15.53%to18.33% [9.67%; 28.74%]each."

 

why is not everyone infected in a household if it is so infectious? Why were less people infected if the household size was bigger (that could be related to infection risk of children)?

There was this other interesting study about background immunity, showing that antibodies for other coronaviruses react similar.

Perhaps children are more likely to have had the common cold more often in the past 1 or 2 years and are a little bit protected (this is only my reasoning no source )

 

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Guest
This topic is now closed to further replies.
×
×
  • Create New...