There is a large number of asymptotic cases which is quite encouraging.
In their preliminary conclusion they write:
"By adhering to strict hygiene measures it is to be expected that the virus concentration of an infected individual can be reduced to the point that the illness manifests more mildly, with simultaneous development of an immunity."
I wonder how they came to that conclusion. It would be great if it turns out to be correct.
That's a theory I've heard before. I don't remember where, unfortunately. Does anyone have a source?
The basic theory is that if you catch the virus, you want to have it in your nose first so your body has time to start a response before it reaches your lungs.
If you get a large dose into your lungs at the start (perhaps because you work at a hospital with insufficient PPE) it's less likely to be mild.
> The basic theory is that if you catch the virus, you want to have it in your nose first so your body has time to start a response before it reaches your lungs.
That's a thought that has crossed my mind before: if it's better the farther up in the respiratory system the initial infection happens, then all that hand washing might be a bad idea, at least for the personal outcome. It depends a lot on wether you expect the full wave to eventually roll through until herd immunity or if you expect it to be stopped early, either by continued distancing until it's starved our by a vaccine.
>That's a thought that has crossed my mind before: if it's better the farther up in the respiratory system the initial infection happens, then all that hand washing might be a bad idea, at least for the personal outcome.
Huh? How does that figure? Hand washing will means less (or no) virii in your nose/mouth.
So, isn't it (a) good in itself, and (b) totally orthogonal to the infection happening "farther up in the respiratory system" (aside the positive, that if it happens and is further up, it will also be less viral content)
> Huh? How does that figure? Hand washing will means less (or no) virii in your nose/mouth.
If you make "first contact" by inhaling tiny droplets it's possible that they will make landfall right down in the lungs, whereas an infection from dirty fingers is guaranteed to start in the nose/mouth area and will already be an immune system project while it works it's way deeper down. Still worse than not getting infected at all, but if you assume that it will eventually hit you anyways, easing into the infection via the nose would surely be preferable to starting in the lungs where it's actually dangerous.
The reverse of this would be accumulating an infective dose from the leakage amount allowed through by PPE while working in extreme exposure. I doubt that there is empirical data regarding viruses, but everything we've learned about particulate emissions suggests that smaller particles reach deeper into the lungs. So it seems quite convincing to me that people who do fall ill despite PPE fall ill harder. Might explain the prominent medical worker deaths.
I've heard that as well. Basically, the spread of the virus inside your body is exponential too, whereas the immune system's response time will be fairly static. So if you're exposed to a little, the immune system has time to respond before the infection gets out of control; if you're exposed to a large amount, by the time the immune system starts to respond, the virus is already widespread in your system.
Why wouldn't it be the other way around? If the virus replicates very fast, people with minimum infectious doses have a couple replication generations of extra time.
If the virus replicates more slowly, a large infectious dose would have the same impact as multiple replication generations.
I'm not asserting the above is what happens, I'm asking the question.
There is an interesting accidental study about this regarding norovirus. (Back in 2014 I think, not the 2017 Tennessee thanksgiving event)
Some guy vomited in a restaurant. Then you can clearly model the time from exposure to symptoms in everyone else based on how far away they were when the guy vomited. Though in this case the viral exposure was very high for everyone in the restaurant.
If it were 'less' exponential, and each cell only infected, say, 5 others, then you get to 500x late in the 4th generation.
In a cartoon example where a low exposure leads to 1 infected cell and a high exposure leads to 500 infected cells, the low exposure matters more if the reproductive factor is lower.
Interesting - almost identical to the idea of not overwhelming relatively inflexible health services with spikes of new cases, only within a single human body?
I couldn’t get tested so not 100% that I actually had covid but as soon as I started having some symptoms I started gargling with diluted hydrogen peroxide and rinsing my sinuses with a strong saline solution. I can’t see how this would hurt and why this isn’t recommended given that the virus replicates in the nose and throat.
The viruses are replicating inside your cells. You can't just "disinfect" living tissue without killing it. There is a reason why medicine relies on evidence and does not recommend whatever random "treatment" someone came up with.
And you disinfect your mouth because disinfecting doesn't mean killing your mouth/tongue tissue. There are several mouthwashes on the market with antibacterial/antiviral properties, dentists do the same, etc...
In Bavaria they are only testing people with symptoms and the positive rate is around 10% [1].
Everybody thinks they have it but the symptoms could also be down to a cold. I am not pointing this statistic out to prove you wrong -- I just find it interesting.
Another anecdote: someone online who works in a virus lab doing other stuff said that if you are ill, have a shower twice a day and rinse mouth with mouthwash to reduce viral load.
If you need to go out, when you come back, remove all clothes and wash and then have a shower.
I think it is based on the idea that staying as close as possible to the infectious dose leads to a better outcome/less severe progression. The infectious dose being "the average number of viral particles needed to establish an infection"[1].
So the studies show that there is no correlation between viral load and severity of symptoms. One study showed there was a correlation between amount of virus in the nose and severity of symptoms.
No studies have looked at various levels of initial dose (does initial dose correlates to viral load?) I imagine such a thing is very hard to measure.
Can’t remember right now where I heard, it was in a podcast. Will return with the information if/when I find it.
An epidemiologist said that there is correlation between the viral load and the severity of the resulting illness in studies on mice. He also said something that surprised me, namely that the mechanism involved is unknown. It seems fairly obvious that it has to do with the exponential growth of the pathogen inside the body.
There are also rumours (and I stress, rumours) that healthcare workers are getting more severe disease due to being infected with a higher infectious load. This would also fit with this theory.
Makes me wonder if it might be possible to deliberately infect body regions where the virus has a low chance of reaching the lung. Could the virus populate (until sufficient immune reaction) the lower intestine if put there, and not spread elsewhere due to the unambiguously directed flow down there (as opposed to bidirectional flow in the respiratory system) or would it be transmitted elsewhere nonetheless through the blood stream? Is this something that can and is checked with animal models?
He spoke more about that than was written. It seems to be the case that the viral load of the point of infection is a major factor in determining how bad it gets. Therefore low-load infections are actually desirably since those are the asymptomatic cases.
That was actually my reaction, too. But I don't think that will go through any ethics commission. Also the immunity will also last for about a year he said. Which is mind-boggling in itself.
Low viral load exposure will likely increase the chances of a milder progression. But that's a very risky gamble. Will still be way too lethal. It is still unknown if this effect is insignificantly small.
Instead we could grow lots of viral particles. Irradiate them to certain death. Then use it to repeatedly challenge the immune system over a long period. Very expensive and nowhere near as effective or reliable as a proper vaccine, but would induce some level of resistance in a fraction of the population.
Can this study really be expected to provide data on the viral load of the point of infection, and at the time of infection? (I would guess that measurements of the distribution of the viral load, after the infection has progressed for some time, would be dependent on how the person's imune system is responding, and conseqently not good evidence for this proposition.)
If not, then is this a well-established feature of respiratory virus infections?
In other words, to what extent is this claim backed by evidence? And is it evidence from studying C19 specifically?
Isn't this the report of the case study that was discussed with much controversy in German media, in particular because the researchers hired a high-profile marketing agency for promotion and held a press conference together with the prime minister of the federal state (NRW) before subjecting the findings to the scrutiny of the scientific community?
Yes, lots of controversy about this study. All in all, it’s not too representative after all because of the outbreak in the festival.
There’s currently a similar study going on in Munich, which represents a normal bigger city in Germany, which is expected to give more insights.
But the Munich study is still several weeks away. Which is frustrating. If the sky is falling like everyone is saying, intermediate results could be published daily. It's really not a difficult study to take such a test.
However, the "controversy" has been overstated by the German media as well. Drosten, virologist who has been cited as one of the critics, called the Gangelt/Heinsberg study "very solid and robust" just yesterday: https://www.zdf.de/politik/maybrit-illner/christian-drosten-...
Particulary, the death rate will be close to the estimate.
The thing is that it's simply not exactly stated what kind of population they tested. Whether they were representative of that town or the German population. Of their sample was actually representative of the German population (which I heavily assume since they talked a lot about that) then the results will be quite robust.
The cluster thing is obviously hard to remove. But the population.. that should be very doable. You'll just have to adjust the people you chose according to the general population. Not sure exactly how that goes down in practice. But in the press conference he stressed multiple times how this was representative and was done in cooperation with statisticians etc.
Yeah well but is that really important? If they nail the same sample age-wise etc and use that to calculate lethality, then that number can be used on the general population, right? The only difference left is the rate of infection. But how many of those died should be the same.
I would guess that the spread of the disease and the rate of infection could be parameters to consider in planning policy for the immediate future. On the other hand, could this result be used to usefully estimate the total values in other regions from the hospitalization numbers?
As far as I've seen it, that response from him was in cases where people tried to simply use the rate of infection measured in Gangelt and apply that to the entirety of Germany. That is obviously invalid as Gangelt had a much larger outbreak than most places.
That doesn't mean that you can't learn anything from this study for the rest of the country, just that you can't extrapolate it naively.
Takeaway: death rate is lower than many initially feared. But it is not like we have already 15 % immunity. In reality it is probably around ~1-2% based on the number of deaths.
The controversy was that they presented results before even a preprint was available. Even worse, the presentation was done together with a state governor and the preliminary results were used to push a political agenda, mainly the loosening of the current COVID19 restrictions, although it is very clear that Gangelt is not representative for Germany. The dubious PR firm that heavily publicized the study on social media was just icing on the cake.
Prof. Drosten has said for quite some time that he estimates the fatality rate to be well below 1%, so the 0,37% from the study are well within his estimates. The real question is how far we are w.r.t. to "herd immunity", and if even a heavily hit town like Gangelt has only 15%, this means that we are far, far away from it.
I don't think it's fair to call that political. Everyone agrees that the current COVID19 restrictions are extraordinarily costly and should be relaxed the second we're able. If he believes he's found new evidence suggesting it's safer than commonly believed to relax the restrictions, of course he'll be desperate to tell people about it.
I might misread you here but what part of the statements of a politician are not supposed to be political there? The notion that this was based on evidence and marketing the study before it even concluded do not really seem to mix that well. And we're not talking about marketing in terms of a press release that went out to soon or an overeager university press department here, we're talking about a marketing agency campaign including singular topic social media account and sensationalization of the results.
I think it's also political, Charité vs Uni Bonn and NRW vs Berlin.
At the same time both Jens Spahn (central government) and Armin Laschet (local government) are front runners for the Chancellor's position.
Paper is short so worth reading, but some key points:
* They had an outbreak in this village as a result of a festival event in mid-February
* A random sample of 1000 from a population of 12,000 were sampled afterwards
* The immunity within the population as a whole was at about 15%.
* The fatality rate was 0.37%
It's got this statement:
> By adhering to strict hygiene measures it is to be expected that the virus concentration of an infected individual can be reduced to the point that the illness manifests more mildly, with simultaneous development of an immunity.
But it seems like the that statement is just an assertion, not something specifically supported by the data they gathered in this study, because it goes on to say:
> These
favourable conditions are not present in a superspreading event (e.g. Karneval
meeting, apres-ski bar in Ischgl, Austria).
If that mortaility rate is common, it would mean there are ~10x (or more) infected in the world...
This sort-of coincides with the large asymptomatic reports. So what do we do from here on? Such numbers would surely raise the famous r0 infection rate, and it means this is far more contagious than reported/believed.
The issue for me (and apparently also gwd) is that this does not seem to be a conclusion that could be made from the sort of evidence collected from this study, and I cannot imagine how that would result from a translation error (unless a whole sentence or more was simply dropped in translation?)
This monograph points to an IgG antibody assay with >99% specificity which means the false positive rate is tiny. Can anybody tell me if there is a commercial antibody test for Covid19 that advertises anything close to this?
My thought is that they have done this with an ELISA. Using a commercially available (or custom made) antibody instead of using one of those terribly inaccurate chromatography strips.
There are so many unknowns from this report that it is hard to make much from it. Some of the major questions.
1. How many people in this town ended up in hospital? How many are still in the ICU and likely to die?
2. Is the population sero tested representative of the town population? Was this population sampled biased by those who had obvious symptoms being more likely to participate?
3. Are the 15% positive representative of the German population or is it skewed to the young?
4. How was the accuracy of the serology test determined? >99% is impressively accurate if true.
15% positive is not representative of the total German population. This is a sample from a small town with a known cluster outbreak from a festival in mid Feb, where a significant amount of people are believed to have gotten infected.
Not everyone is convinced that this study is accurate.
"It is important to understand the sensitivity and specificity of the serological test used in the German studies to be sure the test is not picking up antibody to other corona viruses (4 different common cold causing coronaviruses have been isolated in humans during the past years and 3 more serious ones (SARSCoV1 and 2, and MERSCoV)." [1]
It would be good to see more of these studies conducted in different areas by different teams to see if they come to similar conclusions.
78 comments
[ 4.3 ms ] story [ 152 ms ] threadIn their preliminary conclusion they write:
"By adhering to strict hygiene measures it is to be expected that the virus concentration of an infected individual can be reduced to the point that the illness manifests more mildly, with simultaneous development of an immunity."
I wonder how they came to that conclusion. It would be great if it turns out to be correct.
The basic theory is that if you catch the virus, you want to have it in your nose first so your body has time to start a response before it reaches your lungs.
If you get a large dose into your lungs at the start (perhaps because you work at a hospital with insufficient PPE) it's less likely to be mild.
That's a thought that has crossed my mind before: if it's better the farther up in the respiratory system the initial infection happens, then all that hand washing might be a bad idea, at least for the personal outcome. It depends a lot on wether you expect the full wave to eventually roll through until herd immunity or if you expect it to be stopped early, either by continued distancing until it's starved our by a vaccine.
Huh? How does that figure? Hand washing will means less (or no) virii in your nose/mouth.
So, isn't it (a) good in itself, and (b) totally orthogonal to the infection happening "farther up in the respiratory system" (aside the positive, that if it happens and is further up, it will also be less viral content)
If you make "first contact" by inhaling tiny droplets it's possible that they will make landfall right down in the lungs, whereas an infection from dirty fingers is guaranteed to start in the nose/mouth area and will already be an immune system project while it works it's way deeper down. Still worse than not getting infected at all, but if you assume that it will eventually hit you anyways, easing into the infection via the nose would surely be preferable to starting in the lungs where it's actually dangerous.
The reverse of this would be accumulating an infective dose from the leakage amount allowed through by PPE while working in extreme exposure. I doubt that there is empirical data regarding viruses, but everything we've learned about particulate emissions suggests that smaller particles reach deeper into the lungs. So it seems quite convincing to me that people who do fall ill despite PPE fall ill harder. Might explain the prominent medical worker deaths.
If the virus replicates more slowly, a large infectious dose would have the same impact as multiple replication generations.
I'm not asserting the above is what happens, I'm asking the question.
The Vmin to be detected by your immune system is small. You want the most time between Vdetected and Vdanger.
Some guy vomited in a restaurant. Then you can clearly model the time from exposure to symptoms in everyone else based on how far away they were when the guy vomited. Though in this case the viral exposure was very high for everyone in the restaurant.
For influenza, each infected cell apparently infects ~22 other cells:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1563736/
So in 2 generations you have 500x the virus.
If it were 'less' exponential, and each cell only infected, say, 5 others, then you get to 500x late in the 4th generation.
In a cartoon example where a low exposure leads to 1 infected cell and a high exposure leads to 500 infected cells, the low exposure matters more if the reproductive factor is lower.
[Mumbles something about hormesis.]
Err, you can wash your hands, and it will "disinfect them".
And if the viral content is still in your nose/mouth/neck, you can dissinfect that too to some degree.
So, there's that...
In Bavaria they are only testing people with symptoms and the positive rate is around 10% [1].
Everybody thinks they have it but the symptoms could also be down to a cold. I am not pointing this statistic out to prove you wrong -- I just find it interesting.
[1] https://www.lgl.bayern.de/gesundheit/infektionsschutz/infekt... and search on the page for 'Laboruntersuchungen'. The left axis is number of tests and the right axis is percentage of positive tests.
If you need to go out, when you come back, remove all clothes and wash and then have a shower.
[1]: https://www.newscientist.com/article/2238819-does-a-high-vir...
No studies have looked at various levels of initial dose (does initial dose correlates to viral load?) I imagine such a thing is very hard to measure.
An epidemiologist said that there is correlation between the viral load and the severity of the resulting illness in studies on mice. He also said something that surprised me, namely that the mechanism involved is unknown. It seems fairly obvious that it has to do with the exponential growth of the pathogen inside the body.
If true, would it be possible to infect yourself with a tiny concentration of the virus, then isolate, and expect mild illness followed by immunity?
Insert super low load, stay home for two weeks, and you have been immunized.
Instead we could grow lots of viral particles. Irradiate them to certain death. Then use it to repeatedly challenge the immune system over a long period. Very expensive and nowhere near as effective or reliable as a proper vaccine, but would induce some level of resistance in a fraction of the population.
If not, then is this a well-established feature of respiratory virus infections?
In other words, to what extent is this claim backed by evidence? And is it evidence from studying C19 specifically?
How could you proxy for viral load? And more specifically: How reliable can this measure be?
So short answer: To a degree yes, but unlikely that it is a reliable finding.
Edit: link to newspaper article (German): https://www.sueddeutsche.de/gesundheit/heinsberg-studie-coro... Note that Gangelt is a municipality in the Heinsberg district, so the article and the report refer to the same study by different names.
Particulary, the death rate will be close to the estimate.
https://twitter.com/c_drosten/status/1251034050883657729
I can easily see this to be representative for the town in question.
But I find it quite obvious that it cannot be representative for the whole country due to the initial difference in exposition.
That doesn't mean that you can't learn anything from this study for the rest of the country, just that you can't extrapolate it naively.
Takeaway: death rate is lower than many initially feared. But it is not like we have already 15 % immunity. In reality it is probably around ~1-2% based on the number of deaths.
https://www1.wdr.de/nachrichten/landespolitik/heinsberg-prot...
It is bit sad as it seems to be powerful study and given the attention to the subject, publication of priliminary results are perfectly in order.
* They had an outbreak in this village as a result of a festival event in mid-February
* A random sample of 1000 from a population of 12,000 were sampled afterwards
* The immunity within the population as a whole was at about 15%.
* The fatality rate was 0.37%
It's got this statement:
> By adhering to strict hygiene measures it is to be expected that the virus concentration of an infected individual can be reduced to the point that the illness manifests more mildly, with simultaneous development of an immunity.
But it seems like the that statement is just an assertion, not something specifically supported by the data they gathered in this study, because it goes on to say:
> These favourable conditions are not present in a superspreading event (e.g. Karneval meeting, apres-ski bar in Ischgl, Austria).
This sort-of coincides with the large asymptomatic reports. So what do we do from here on? Such numbers would surely raise the famous r0 infection rate, and it means this is far more contagious than reported/believed.
My thought is that they have done this with an ELISA. Using a commercially available (or custom made) antibody instead of using one of those terribly inaccurate chromatography strips.
1. How many people in this town ended up in hospital? How many are still in the ICU and likely to die?
2. Is the population sero tested representative of the town population? Was this population sampled biased by those who had obvious symptoms being more likely to participate?
3. Are the 15% positive representative of the German population or is it skewed to the young?
4. How was the accuracy of the serology test determined? >99% is impressively accurate if true.
This is a misprint. They meant to say 0.99 or 99%, not .99%
That means it's very reliable that any positive signal originates from covid19 antibodies, instead of from something else.
That does not mean that it's 99% accurate with regards to sensitivity.
https://www.medrxiv.org/content/10.1101/2020.03.30.20047365v...
This would mean the antibody test alone cannot tell the full extent of community spread. There's a brief summary of antibody knowledge here.
https://www.the-scientist.com/news-opinion/what-do-antibody-...
"It is important to understand the sensitivity and specificity of the serological test used in the German studies to be sure the test is not picking up antibody to other corona viruses (4 different common cold causing coronaviruses have been isolated in humans during the past years and 3 more serious ones (SARSCoV1 and 2, and MERSCoV)." [1]
It would be good to see more of these studies conducted in different areas by different teams to see if they come to similar conclusions.
[1] https://www.sciencemediacentre.org/expert-reaction-to-unpubl...