Science article noted by Stranger in another thread in this post that, as suggested by raspberry hunter is worth its own thread.
There’s an old rule that 80% of infections are spread by 20% those infected. Obviously it is just a rough guide but seems not far off in this case: “Probably about 10% of cases lead to 80% of the spread,”
The article is a great read. Too much good to pick quotes from! But those interested in how this disease spreads would likely benefit from reading it and discussing the implications.
Help me out here. My take is that the article presents several hypotheses but few conclusions. Being close to people for extended periods of time is risky, especially where shouting, singing, and deep breathing take place. What do you consider the main takeaway?
Since I had nothing to base it on when it started I couldn’t post what I suspected. This started out in China as a rapidly spreading respiratory disease.
If it’s primarily spread through the air then The 6 foot rule that followed might be completely pointless in an enclosed environment. Everybody is breathing the same air.
Put another way, it’s the mask that matters in an enclosed area and not the separation within the enclosed area.
And I’ll throw out another possibility. Excessive cleaning of surfaces might actually impede our ability to gain immunity by denying a much smaller dose of the virus at one time. Instead of picking up 1,000 of the viruses from physical contact we get hit with a million of them in aerosol form. The only way to look for that statistically is if any country pushed the masks but not the cleaning.
It would seem like the greatest risk is airborne transmission. Being outdoors will minimize that because the droplets will be easily dispersed. Being indoors is risky because the droplets circulate around the confined space giving many opportunities to be inhaled. This is more risky in an environment where people exhale forcefully, like from singing or yelling, since that will expel more droplets which have more chances to be inhaled. If someone happens to be in that most contagious period in an indoor space where they are yelling, projecting their voice, or breathing hard, that could make them a super spreader and lots of people may get infected.
This may mean indoor places which are naturally loud and people have to shout, like bars, may need to figure out how to change the environment so people can talk quieter. Places like gyms may need to figure something out as well, since the heavy breathing from a workout would likely release lots of droplets. I’m assuming that very quiet indoor places like a library would be relatively safe. Businesses which can transition to more outdoor spaces should do so. Indoor spaces should scrub the air with HEPA filters or exchange it with outside air as often as possible.
It seems like lots of data is pointing to the greatest risk is from breathing in the virus rather than from a contaminated surface. If stale air is a big reason for spikes in infection, focusing effort on keeping the air clean will likely mean big reductions in infections.
The main takeaway to my read is to recognize that there is no typical person infected with SARS-CoV-2 in terms of transmissibility and to then consider what the actual spread and distribution implies. A relatively very few do most of the spreading in likely a relatively few sorts of contexts.
One big issue is what it implies in terms of what we do and DO NOT relax. As the quotes Thudlow Boink highlighted express - considering this aspect allows consideration of how to transition from blunt hammer to targeting interventions more surgically for greatest impact at least costs.
We are at the point of trying to ease restrictions. The restrictions that get relaxed should be the ones with best chance to not be making the biggest difference to their costs (in terms of non-COVID-19 deaths and health impacts, quality of life, the global economy, food and housing insecurity, and so on).
Obviously identifying the events and activities most associated with that 10% leading to 80% of the spread would be useful. They are the likely “low hanging fruit” for keeping restrictions on for a long time. And identifying the events and activities least associated with such would be useful as well.
Considering how else to impact the 10% of individuals who do most of the spread is also one big implication. Yes we want to identify who they are if we can, but to the degree we cannot what low cost interventions reduces their spreading the germ the most? Does having everyone wear masks in moderate to high risk events and activities result in these individuals getting covered and prevent a large portion of their spread? Six feet distancing alone may not be enough for this minority. Not addressed in the article but that seems like a reasonable thought.
What does it imply about how else to keep rates very low? That it is going to be very hard to get to and to maintain no new outbreaks, let alone local “eradication”, because just one superspreader can spread so much before being identified.
Another factor obviously is now many people come in contact with a ‘super spreader’. WSJ article recently (paywall), showing some modeling results from Australia, that total number of cases in a given time would vary orders of magnitude just based on max size of gathering allowed. In various regions it was figured that single large events also characterized by shouting and singing (mardi gras in New Orleans, a particular fair in a hard hit region of Germany in February, etc.) were responsible for a large % of cases in those areas. Also it was believed infections contracted in large loud gatherings tended toward heavier viral load than infection in other situations.*
Large gathering is not only more contact but higher probability people very ‘good’ at spreading the disease are among the people there.
But it seems that something like size of gathering is easier to fine tune as a policy, compared to the unsustainable baseline of ‘everybody stay home’, than ‘some people spread it more’. How would you predict the latter for specific people? Maybe that’s something that a genetic test would mainly nail down, or not, and anyway there are obvious issues with mandatory databases of people’s genetics and restricting people differently depending on the results.
*which could be part of the answer why socially distanced non-conversing outdoor activity might be virtually zero risk. If you also considered that theoretically aerodynamically possible virus travel among distanced people outdoors might tend to give mild infections even in unusual cases where it did cause an infection. Same or next day’s WSJ had a map of where people are calling 311 (City emergency violation hotline, at least not calling 911) about other people not wearing masks outside. The calls were heavily concentrated in upper end Manhattan neighborhoods. Support for very broad and strict restrictions has become a socio/economic/political thing. It’s pushed back on not only loudly by elements of the political right as I see on the internet and cable news, but also less loudly by non-right lower socio-econ people in NY as I see first hand walking around.
The title of the article is “Why do some COVID-19 patients infect many others, whereas most don’t spread the virus at all?” and this strikes me as one of THE most important questions that we don’t know the answer to – YET.
In the meantime, limiting crowd size IN CASE THERE IS an unidentified superspreader, is smart and making sure everyone wears masks, again, IN CASE THERE IS an unidentified superspreader in the crowd.
(My bold)
Very important and oh boy! if genetics are involved, does that ever open the door to accusations of discrimination, favoritism, violation of rights, singling out people for persecution (not to mention prosecution)! :smack: Holy shit. We can only hope that whatever genetic factor makes someone a superspreader (and that’s not to say it IS genetic-- that is unknown) does not correlate with a visible, physical characteristic like curly red hair, one blue eye and one brown eye, or an extra finger on the left hand-- :eek: Will unruly mobs force those people to wear a yellow “S” on their clothes??
*<Need more coffee…> *
ETA: Excellent discussion, BTW.
I got the impression that a super spreader was more about timing and behavior rather than something genetically particular about the individual. That is, if someone happens to be shouting in a crowded, indoor area at the time they are most infectious, they will likely be a super spreader. Their race, sex, or genetic makeup may not really make a difference.
One aspect could be that the way certain cultures communicate could play a role. If a particular culture speaks loudly and closely to other people, that could mean that culture will be more likely to have infection clusters. I wonder if that contributed to the situation in NYC, where the stereotype is that people often shout and speak loudly there. Italy also has that stereotype.
The article definitely says there are “superspreading events,” where people are crowded together and shouting or singing-- like bars, sports events, etc. And certain individuals do tend to express themselves this way.
But tantalizingly, the article also says:
Why do some people shed more virus? We’ve circled back to the title of the article again, “Why do some COVID-19 patients infect many others, whereas most don’t spread the virus at all?” which the article explores but does not answer definitively. Is it simply their style of speaking and/or breathing? Or “differences in their immune systems or distribution of virus receptors in their body”? Don’t know – YET.
Although there may be certain behavioral traits that make someone more prone to spreading contagion, there are definitely people with a (presumably genetic or epigenetic) predisposition to carry and express the pathogen by aggressive shedding. Mary Mallon, a.k.a. ‘Typhoid Mary’ is the most infamous example but it is well understood that the degree to which people can infect others is variable.
In at least a few cases, such “super spreaders” of SARS-CoV-2 have been identified as asymptomatic carriers, and in general carriers have been seen to shed high levels of active virus (as determined by cycle threshold values of RT-PCR tests, not plaque assays, but we can reasonably assume people who are in a presymptomatic infected are shedding whole active virus, not just genome fragments) in comparison to people actually suffering evident signs and symptoms of COVID-19. So it seems likely that some people are highly productive factories for this virus notwithstanding any behavioral factors for contagion.
Great link. Very clear and logically arranged. Thanks. One of the casualties (for me) of this whole lockdown is I’m not spending any time in the car, which is where I mostly listen to NPR.
From the discussion so far, it doesn’t seem to me that the answer to the key question is at all clear: are there people who spread the virus more effectively than others, or is there conduct or are there situations which more easily spread the virus?
IOW, once conduct and situations are controlled for, are there still people who clearly stand out as spreaders?
A hypothetical: if you’ve got two infected but still asymptomatic people at a church service, both enthusiastic singers during the hymns, neither wearing masks, both being very social at the fellowship hour after the service, is there a likelihood that one’s going to be responsible for the vast majority of the resulting infections, or is the expectation that it will work out more or less even?
In addition to genetic or epigenetic predisposition to aggressively shed, developmental factors (age) and/or (lack of) recent experience with the other common cold causing human coronaviruses may contribute. The biggest superspreaders likely combine factors of their biological predisposition, their behaviors, and their presence in high risk events, along with the “right” timing.
I am struck by how, in retrospect, some assumptions made by epidemiology modelers early on, seem so off now. Imperial College’s model early on got lots of attention, partly because of their excellent reputation. One of their statements?
Seems now that a single superspreader person screaming at a crowded mass event, like a sporting event, music concert, political rally (especially if they are in enclosed locations), etc., can do lots of spread with even relatively brief shared time in the space … more than if the same person was talking quietly at a table six feet away in a restaurant.
One thing I figured out about myself fairly recently (though before pandemic) is that when I’m talking to someone and get overexcited I tend to spray saliva around a bit more than others. (I couldn’t figure out how to word that without it sounding super gross – I don’t think it’s too awful, although I figured it out because I noticed people stepping back from me occasionally when we were talking. They still talk to me, though, so it can’t be that bad? And I’m really trying to watch myself and not get too enthusiastic when I talk to cut down on that.) I wonder if it’s that kind of thing that could contribute to someone being a much better spreader than others.
To expand on the “and/or (lack of) recent experience with the other common cold causing human coronaviruses” …
Better to think of this as one of the possible contributors to the second part of the question - why do some at least (not sure if it is “most”) seem to not spread it at all, to function epidemiologically as if they are in the resolved bucket even though they’ve never seen this specific germ before?
What is known is only suggestive.
Coronaviruses are known to generate antibodies that nonspecificallyprotect against infections with other coronaviruses. "SARS infection can generate neutralizing antibodies against HCoV-OC43 (15) and HCoV-OC43 infection can generate cross-reactive antibodies against SARS (16). " This much is established, but do those antibodies (that wane over time after infection) also neutralize SARS-CoV-2? If so to what degree? What is the level of cross-reactive neutralizing IgA antibodies (which are ones that get into mucosal surfaces like snot most)? Unknown. But recent and/or frequent infections with those other common cold causing coronaviruses might result in a population that epidemiologically function as being in the recovered bucket, be of the group that don’t spread it at all.
34% of adults who have never been infected with SARS-CoV-2 already have “helper T cells that recognized SARS-CoV-2”! Unknown if they protect from infection, allow infection to be less serious, cause the rare multisystem inflammatory syndrome in kids, and/or if that 34% get infected and are among those who nevertheless don’t spread it at all.
And IF some sizable fraction “don’t spread the virus at all” - because of cross-protection from past (recent and/or frequent) common cold causing infection, does that impact the fraction needed to have been infected to slow down spread? How does that even get modeled? If only we had a data scientist around …
Holy cow, DSeid, is this the pathway that explains why kids don’t get it as badly? Because they’ve probably been exposed to all N of the coronaviruses currently circulating?
The data I’d like to see is how many parents with small children get bad cases of COVID-19 compared to those in the same age/health bracket who don’t have kids. I can attest that due to the two small ones I have also been exposed to probably 100% of the coronaviruses in the world before COVID-19
Something else I wonder. In January a member of my church congregation/ward came back from China where he works. We talked at church. A couple of weeks after he returned, I had the weirdest flu I have ever had: for a week it wasn’t bad, I went around saying “well, I feel crappy but at least it’s not the flu,” and then after about a week it suddenly got a lot worse. (My previous experiences with the flu have been that I feel really awful within 24 hours of starting to feel bad.) I have been wondering about this for a while but figured it couldn’t possibly be covid-19 because I would have expected a bunch of people in our congregation to have been hospitalized or die. But we have a lot of small kids, there is a lot of mingling between adults and the small kids, every winter everyone tends to get sick, and now I wonder whether we did have it but our whole congregation was partially protected by having gotten sick so many times before
But that of course is rampant speculation Next week I may go in to get tested for antibodies, just because I am curious now. (It won’t materially change anything I’m doing or not doing.)
Next week I may go in to get tested for antibodies, just because I am curious now. (It won’t materially change anything I’m doing or not doing.)