(a) The original title of this article is 'Largest COVID-19 contact tracing study to date finds children key to spread, evidence of superspreaders' -- I edited it to fit within the HN character limit
(c) This is a large study on 500k people in India. The researchers found that 71% of infected individuals did not infect any of their contacts, while a mere 8% of infected individuals accounted for 60% of new infections -- providing evidence for so-called 'superspreaders'. Regarding children, the researchers found that 'These patterns of enhanced transmission risk in similar-age pairs were strongest among children ages 0-14 years and among adults ages ≥65 years',
and that they 'identif[ied] high prevalence of infection among children who were contacts of cases around their own age' -- something that was also true in adults.
The overall case-to-fatality ratio was between 0.05% at ages 5-17 years to 16.6% at ages ≥85 years.
The paper says that children are more efficient in transmitting to people of their same age range, so perhaps the title as you edited it can be misleading.
Some key limitations to keep in mind from the study:
> The contact tracing data analyzed included only 20% of all reported cases as index cases and represented only 19% of all contacts traced
So the observations, while important, are probably just part of a more complex picture.
> Another limitation was the lack of data on timing of exposure and symptoms onset in relation to testing dates; this necessitated assumptions about identification of true index cases.
This is important: some cases may not be "true positives" in the sense that the PCR test alone does not tell if you have the disease, or had the disease if you did not develop symptoms. Given the size of the population tested, admittedly doing that is difficult.
> Finally, while comorbidities data collected as part of COVID-19 mortality surveillance revealed clinical and epidemiological attributes of fatal cases, the fact that such data were not collected for all diagnosed cases prevented inference of the contribution of comorbidities to fatal outcomes.
Note that it's easy to read the Princeton press release as "children are spreading the disease to each other", but I don't think the Science paper is very good evidence for that hypothesis. Sadly the paper is making it pretty hard to get an answer to that question - the least they could have done was to make it obvious over what time their study data was extracted (answer, I think: for all time up to 1st Aug 2020), and also when schools were shut down in India (answer according to UNESCO: partial closure from 2nd March, still fully closed). Anyway, it seems perfectly consistent to me that the superspreaders are adults, and that clusters of children who interact with each other also interact with the same adults.
If schools are closed and children have to stay at home they don't spread the disease. They just get it from their parents and don't have any further contacts to whom they could spread it. The statistics during lockdown are pretty solid: Children don't spread it! (As long as you lock them inside with their adult parents - but the statistics may overlook that.)
Therefore we must always keep a close eye on the circumstances where the data was taken and what we can and cannot conclude from that.
Not an expert by any means, but I think that was (is) also true for the common flue, so not that surprising that a flue-like virus has similar spread-patterns like the common flue.
It may also be that they have a lot more close contacts at school than most adults at their workplace. Children usually have 30 in their class and often classes are mixed which potentially gives them >100 close contacts during a normal school week. Many adults have only a handful of close contacts at their workplace.
Add in the network effects of crosslinking different social circles (school, family, workplace) by children and adults then you get good conditions to spread disease.
It need not even be the children spreading it! Teacher standing at the front of a classroom, speaking loudly and clearly and spitting all over the entire class: how do you distinguish between that and "one child spread it to their entire class"?
Unfortunately, it's difficult to conclude that children are the "key" to superspreading, because the study suffers from the same limitation as the South Korean study the media used to make the same claim.
Which is that it's impossible to determine the direction of transmission. Just because someone (a child or young adult) developed symptoms first doesn't mean they actually got infected first, they might have actually gotten infected at the same time as their supposed contact. When the South Korean study removed cases of shared exposure, the new study found those children did not in fact transmit very much. https://twitter.com/apsmunro/status/1311616493445165058
> it's difficult to conclude that children are the "key" to superspreading
Not only does the study not show that, the press release doesn't even claim that - it contains two separate claims, that superspreading is important, and that children are involved in spread. It's only the use of a comma in the headline that - through incompetence or dishonesty - suggests that!
Mods, could we have the title changed, please?
EDIT: Mods changed the title already, thanks mods!
The one interesting point I saw that doesn't have some sort of caveat was this:
> The researchers found that the chances of a person with coronavirus, regardless of their age, passing it on to a close contact [were] 9% in the household.
This stuff has been studied over and over and over, and the conclusion is that children spread less than adults. See 'Top 10 Epidemiological Papers on Transmission' here:
At this point, a study which concludes the opposite has to bring the proverbial extraordinary evidence.
We aren't even close to that point. There's a heckuva lot that we still don't know about this virus. That paper was update, what, 30 June? They even admitted that they were using their earlier conclusions to create filters for the new literature they were reviewing- in effect building their own echo chamber.
As long a the study is competently run and has reasoned conclusions it should be considered.
It does not say that it is the opposite, see other remarks. It does say that children play a role. In my country it is claimed they don't and they have a lot of freedoms. Even though they are probably less "good" at spreading they have a lot more occasions to do so. At least here in the Netherlands.
Just a note to say that "extraordinary evidence" doesn't exist. It's just evidence. It's a circular notion, since if such a thing as "extraordinary evidence" were to exist, then proving it existed would require itself.
From the paper Epidemiology and transmission dynamics of COVID-19 in two Indian states [1]:
> Although they are not the wealthiest states in India, Andhra Pradesh and Tamil Nadu are among the states with the largest healthcare workforces and public health expenditures per capita, and are known for their effective primary healthcare delivery models. Both states initiated rigorous disease surveillance and contact tracing early in response to the pandemic. Procedures include syndromic surveillance and SARS-CoV-2 testing for all individuals seeking care for severe acute respiratory illness or influenza-like illness at healthcare facilities; delineation of 5km “containment zones” surrounding cases for daily house-to-house surveillance to identify individuals with symptoms; and daily follow-up of all contacts of laboratory-confirmed or suspect COVID-19 cases, with the aim of testing these individuals 5-14 days after their contact with a primary case, irrespective of symptoms, to identify onward transmission.
A low/medium income region with good healthcare infrastructure is a rarity and may potentially provide unique insights.
We have seen very different transmission patterns in high latitude versus subtropical regions. I wonder if a third pattern will emerge in tropical regions.
General comment: posting a random study and drawing conclusions from it seems dangerous. This kind of science is hard, so we usually rely on a spread of evidence from many sources, building literature until conclusions are more robust - often through meta studies.
I think it’s a bit dangerous for an intelligent HN reader to go looking at scientific papers without properly taking a systematic approach to cover all bases. A much sounder, yet less exciting, method is to rely on the advice of expert organisations such as the WHO.
Counterpoint: it strikes me that WHO basically lied about masks early on to avoid a panic that compromised mask acquisition for medical needs.
The motivation was moral, but it seems such organisations are 'happy' to use misinformation to the public; that is, they can't fully be trusted.
Finding trustworthy, competent voices that one can follow is v.hard.
Academic papers, and the response of academics that have some trust [with you] already, are more useful than supranational organisations whose declarations might not be entirely aligned with promulgation of truth.
> it strikes me that WHO basically lied about masks early on to avoid a panic that compromised mask acquisition for medical needs.
This thinking has circulated widely on social media, but I don't see evidence for it.
The current WHO line is pretty much the same - they do accept that it may help to wear a mask e.g. in supermarkets, but they're pretty clear that the evidence for community mask wearing being a net positive is thin.
I don't think you have to consider the WHO a "conspiracy" in order to lose trust in them.
For example, many other sources (which tend to elucidate their sources and reasoning more fully) claim, convincingly, that the evidence for community mask wearing being a net positive is quite strong. When WHO claims that it's weak, I lose trust in them.
> that the evidence for community mask wearing being a net positive is quite strong
(genuine question) please could you provide a link to that? I've seen one meta-study that people quote, but I've not seen any evidence that community mask wearing is a net positive in practice.
I'm not even sure what I would do in their situation:
Do you "lie" (maybe by omission; at least going against the current {at the time} scientific consensus) by saying that 'public mask-wearing provides no benefit' [that's to my recollection their advice around February time when I was considering buying masks] in order to maintain supplies for medical personnel? Or, do you say 'masks limit transmission of ARIs' and risk leaving medical need of masks unfulfilled because mass panic has decimated the supplies?
I don't think it's necessarily wrong to lie there (akin to people who say "vaccines are perfectly safe" - the info leaflet disproves that), but it does mean if you're searching for a source of ground truth that you can't rely on them.
No conspiracy, other than that needed to protect global populations, is required. I think you'd agree with me that WHO are conspiring to protect the health of global populations. That they would hold this goal above the goal of providing information that is absolutely true is not especially surprising, I think.
FWIW: WHO updated the information later such that they offered different mask advice per country (to my recollection), which seems anomalous unless one considers that the advice is tailored according to whether countries had secured supplies. The advice for UK went from "masks do nothing" to "you should wear a mask".
> Do you "lie" (maybe by omission; at least going against the current {at the time} scientific consensus) by saying that 'public mask-wearing provides no benefit' [that's to my recollection their advice around February time when I was considering buying masks] in order to maintain supplies for medical personnel? Or, do you say 'masks limit transmission of ARIs' and risk leaving medical need of masks unfulfilled because mass panic has decimated the supplies?
Truth mixed with guilt would've probably been the best path: "Yes, masks might help reduce spread, but doctors and nurses on the front lines are the ones most in need. Please consider donating your spare supplies."
Not talking about their advice only, WHO has been actively spreading false information since the beginning of the crisis, in full accord with China's narrative
So, if this were the case, then when Denmark opened their elementary schools up in late April, one would have expected to see an explosion of cases, yes? Same with Finland a few weeks later. But, Denmark did not have a resurgence of cases until many months later, nor Finland.
The reason is buried in the article, but ought to be in the headline: children are key to spread...within the household. In other words, parents and older siblings and other household members do not socially distance from children in their own household. Well, duh.
Same in the Netherlands; if anything, cases started to increase rapidly again during the summer holidays.
Mind you, iirc only elementary school children (<12) went back to school, middle and higher education remained taught from home until after the holidays.
I believe the conclusion in Netherland is that it's mostly students, older teenagers and 20-somethings who spread the most. And mostly because they like to party and hang out together.
My son's secondary school just introduced mandatory masks this week. They should really do that in more areas.
> children are key to spread...within the household
Sorry but I don't understand the distinction. One family member has Covid-19 -> spread to children -> spread to other family members -> other family members spread it outside, so in the in the end children are key to spread. Or am I missing something?
Meanwhile, its impossible to find good information on how long can children be infectious. My daughter is 3 weeks after first symptoms and we'd like to get her back in kindergarden after 4 weeks.
The same for RT-PCR -- I tested positive one month after first symptoms, but I also have good igg antibodies.
Didn't people do this RT-PCR testing for influenza before?
Are RT-PCR tests for influenza still positive one month after first flu symptoms?
I mean, the quality of information you can google up as a COVID-19 patient is absolute shit, most of what google shows is sensationalized news articles directed at the non-positive, scared general public.
She should be fine, unless immunocompromised. RT-PCR tests do not tell, in absence of symptoms, if you are infectious. If you have symptoms, the presence of a positive test correlates with actual, infectious virus in your throat until 7 days with milder symptoms and 15 days with more severe symptoms.
Also bear in mind, in particular for your daughter, that viral RNA takes a lot to clear, even when her immune system has already defeated the virus (IOW: you're seeing the "debris after the battle" and not the virus): the milder the symptoms, the longer it takes to clear.
tl;dr: Your daughter should be fine at this point, in general terms.
Of course, as an obligatory disclaimer, IAAS (I am a scientist), IANAMD, this is not medical advice, etc etc.
There are no reliable reports of transmission past 11 days from infection. Symptoms can continue much longer due to lingering damage even when the patient is no longer infectious. Likewise PCR tests can continue to come up positive for weeks due to leftover genetic material from dead viruses.
They key to the spread has been anyone who resisted a 2 year long lockdown.
These people left their homes and kept the spread of COVID going.
The problem is, I hardly blame them, this is inevitable human nature. If a few percentage of people are in public, the virus will spread. It's pandemic physics.
And children, is it this group that doesn't listen the most? I have a hard enough time preventing my kid from climbing on furniture. How did we expect 100% of teenagers to listen to their parents and government?
There was 2 options, violence enforced lockdown or the spread of coronavirus.
I think the bigger claim in this paper is that superspreaders drive infection: 80% of infections come from 5.4% of cases (and that transmission rates between housemates is as low as 7-10%).
I'm quite curious to learn what stands out about these superspreaders. Their infection? Their behavior? Their genes? The timing of their interaction with others?
Theres been 1 million deaths worldwide with a population of 7 billion. Thats 0.00014285714
The odds you die of C19 is overwhelmingly slim especially now with the therapeutics and better understanding of how to treat it. If you're old you are at higher risk (about 2% chance over 65) but thats the case of anything when you're old. Eventually something gets you as you are basically walking on a tight rope. Everybody dies.
Old scared politicians have fucked the younger generation out of self preservation and greed once again. Theres absolutely no reason to shut down society like this. Wear a mask and move on.
This makes a lot of sense. Even if kids are less likely to spread coronavirus individually, they tend to get stuck in situations which spread disease. (School classrooms and daycare) Also they tend to bridge social groups that would normally not interact. Kids create a network that touches an entire community. Your family may never interact with a family from the other side of town, but your kids/grandkids/nephews/nieces probably will in their school.
Since the early days of lockdown, my apartment complex in Bangalore has been strict about masks among the elderly but not among children. I would see children playing without masks and probably this could be a reason why children end up spreading the virus easily.
53 comments
[ 3.1 ms ] story [ 103 ms ] thread(a) The original title of this article is 'Largest COVID-19 contact tracing study to date finds children key to spread, evidence of superspreaders' -- I edited it to fit within the HN character limit
(b) The actual article is published in Science: https://science.sciencemag.org/content/early/2020/09/29/scie...
(c) This is a large study on 500k people in India. The researchers found that 71% of infected individuals did not infect any of their contacts, while a mere 8% of infected individuals accounted for 60% of new infections -- providing evidence for so-called 'superspreaders'. Regarding children, the researchers found that 'These patterns of enhanced transmission risk in similar-age pairs were strongest among children ages 0-14 years and among adults ages ≥65 years', and that they 'identif[ied] high prevalence of infection among children who were contacts of cases around their own age' -- something that was also true in adults.
The overall case-to-fatality ratio was between 0.05% at ages 5-17 years to 16.6% at ages ≥85 years.
Some key limitations to keep in mind from the study:
> The contact tracing data analyzed included only 20% of all reported cases as index cases and represented only 19% of all contacts traced
So the observations, while important, are probably just part of a more complex picture.
> Another limitation was the lack of data on timing of exposure and symptoms onset in relation to testing dates; this necessitated assumptions about identification of true index cases.
This is important: some cases may not be "true positives" in the sense that the PCR test alone does not tell if you have the disease, or had the disease if you did not develop symptoms. Given the size of the population tested, admittedly doing that is difficult.
> Finally, while comorbidities data collected as part of COVID-19 mortality surveillance revealed clinical and epidemiological attributes of fatal cases, the fact that such data were not collected for all diagnosed cases prevented inference of the contribution of comorbidities to fatal outcomes.
It is time for the government to step in and reign in the panic instead of fueling it.
Note that it's easy to read the Princeton press release as "children are spreading the disease to each other", but I don't think the Science paper is very good evidence for that hypothesis. Sadly the paper is making it pretty hard to get an answer to that question - the least they could have done was to make it obvious over what time their study data was extracted (answer, I think: for all time up to 1st Aug 2020), and also when schools were shut down in India (answer according to UNESCO: partial closure from 2nd March, still fully closed). Anyway, it seems perfectly consistent to me that the superspreaders are adults, and that clusters of children who interact with each other also interact with the same adults.
Therefore we must always keep a close eye on the circumstances where the data was taken and what we can and cannot conclude from that.
Add in the network effects of crosslinking different social circles (school, family, workplace) by children and adults then you get good conditions to spread disease.
Which is that it's impossible to determine the direction of transmission. Just because someone (a child or young adult) developed symptoms first doesn't mean they actually got infected first, they might have actually gotten infected at the same time as their supposed contact. When the South Korean study removed cases of shared exposure, the new study found those children did not in fact transmit very much. https://twitter.com/apsmunro/status/1311616493445165058
This Twitter discussion does not make me think they did anything differently than the South Korean study that was later corrected: https://twitter.com/apsmunro/status/1312301601156194304
Not only does the study not show that, the press release doesn't even claim that - it contains two separate claims, that superspreading is important, and that children are involved in spread. It's only the use of a comma in the headline that - through incompetence or dishonesty - suggests that!
Mods, could we have the title changed, please?
EDIT: Mods changed the title already, thanks mods!
> The researchers found that the chances of a person with coronavirus, regardless of their age, passing it on to a close contact [were] 9% in the household.
https://dontforgetthebubbles.com/evidence-summary-paediatric...
At this point, a study which concludes the opposite has to bring the proverbial extraordinary evidence.
We aren't even close to that point. There's a heckuva lot that we still don't know about this virus. That paper was update, what, 30 June? They even admitted that they were using their earlier conclusions to create filters for the new literature they were reviewing- in effect building their own echo chamber.
As long a the study is competently run and has reasoned conclusions it should be considered.
> Although they are not the wealthiest states in India, Andhra Pradesh and Tamil Nadu are among the states with the largest healthcare workforces and public health expenditures per capita, and are known for their effective primary healthcare delivery models. Both states initiated rigorous disease surveillance and contact tracing early in response to the pandemic. Procedures include syndromic surveillance and SARS-CoV-2 testing for all individuals seeking care for severe acute respiratory illness or influenza-like illness at healthcare facilities; delineation of 5km “containment zones” surrounding cases for daily house-to-house surveillance to identify individuals with symptoms; and daily follow-up of all contacts of laboratory-confirmed or suspect COVID-19 cases, with the aim of testing these individuals 5-14 days after their contact with a primary case, irrespective of symptoms, to identify onward transmission.
A low/medium income region with good healthcare infrastructure is a rarity and may potentially provide unique insights.
We have seen very different transmission patterns in high latitude versus subtropical regions. I wonder if a third pattern will emerge in tropical regions.
[1] https://science.sciencemag.org/content/early/2020/09/29/scie...
https://www.bbc.co.uk/news/election-us-2020-54401186
I think it’s a bit dangerous for an intelligent HN reader to go looking at scientific papers without properly taking a systematic approach to cover all bases. A much sounder, yet less exciting, method is to rely on the advice of expert organisations such as the WHO.
Oxford's CEBM is one (https://www.cebm.net) and someone else has posted another for transmission in children here.
The motivation was moral, but it seems such organisations are 'happy' to use misinformation to the public; that is, they can't fully be trusted.
Finding trustworthy, competent voices that one can follow is v.hard.
Academic papers, and the response of academics that have some trust [with you] already, are more useful than supranational organisations whose declarations might not be entirely aligned with promulgation of truth.
This thinking has circulated widely on social media, but I don't see evidence for it.
The current WHO line is pretty much the same - they do accept that it may help to wear a mask e.g. in supermarkets, but they're pretty clear that the evidence for community mask wearing being a net positive is thin.
The WHO is not a conspiracy.
For example, many other sources (which tend to elucidate their sources and reasoning more fully) claim, convincingly, that the evidence for community mask wearing being a net positive is quite strong. When WHO claims that it's weak, I lose trust in them.
(genuine question) please could you provide a link to that? I've seen one meta-study that people quote, but I've not seen any evidence that community mask wearing is a net positive in practice.
Do you "lie" (maybe by omission; at least going against the current {at the time} scientific consensus) by saying that 'public mask-wearing provides no benefit' [that's to my recollection their advice around February time when I was considering buying masks] in order to maintain supplies for medical personnel? Or, do you say 'masks limit transmission of ARIs' and risk leaving medical need of masks unfulfilled because mass panic has decimated the supplies?
I don't think it's necessarily wrong to lie there (akin to people who say "vaccines are perfectly safe" - the info leaflet disproves that), but it does mean if you're searching for a source of ground truth that you can't rely on them.
No conspiracy, other than that needed to protect global populations, is required. I think you'd agree with me that WHO are conspiring to protect the health of global populations. That they would hold this goal above the goal of providing information that is absolutely true is not especially surprising, I think.
FWIW: WHO updated the information later such that they offered different mask advice per country (to my recollection), which seems anomalous unless one considers that the advice is tailored according to whether countries had secured supplies. The advice for UK went from "masks do nothing" to "you should wear a mask".
Truth mixed with guilt would've probably been the best path: "Yes, masks might help reduce spread, but doctors and nurses on the front lines are the ones most in need. Please consider donating your spare supplies."
https://www.businessinsider.com/who-no-transmission-coronavi...
Very careful wording here by the author, where the usage of "children" includes "adult children".
The reason is buried in the article, but ought to be in the headline: children are key to spread...within the household. In other words, parents and older siblings and other household members do not socially distance from children in their own household. Well, duh.
Mind you, iirc only elementary school children (<12) went back to school, middle and higher education remained taught from home until after the holidays.
My son's secondary school just introduced mandatory masks this week. They should really do that in more areas.
Sorry but I don't understand the distinction. One family member has Covid-19 -> spread to children -> spread to other family members -> other family members spread it outside, so in the in the end children are key to spread. Or am I missing something?
The same for RT-PCR -- I tested positive one month after first symptoms, but I also have good igg antibodies.
Didn't people do this RT-PCR testing for influenza before? Are RT-PCR tests for influenza still positive one month after first flu symptoms?
I mean, the quality of information you can google up as a COVID-19 patient is absolute shit, most of what google shows is sensationalized news articles directed at the non-positive, scared general public.
She should be fine, unless immunocompromised. RT-PCR tests do not tell, in absence of symptoms, if you are infectious. If you have symptoms, the presence of a positive test correlates with actual, infectious virus in your throat until 7 days with milder symptoms and 15 days with more severe symptoms.
See https://www.cebm.net/covid-19/infectious-positive-pcr-test-r...
Also bear in mind, in particular for your daughter, that viral RNA takes a lot to clear, even when her immune system has already defeated the virus (IOW: you're seeing the "debris after the battle" and not the virus): the milder the symptoms, the longer it takes to clear.
tl;dr: Your daughter should be fine at this point, in general terms.
Of course, as an obligatory disclaimer, IAAS (I am a scientist), IANAMD, this is not medical advice, etc etc.
These people left their homes and kept the spread of COVID going.
The problem is, I hardly blame them, this is inevitable human nature. If a few percentage of people are in public, the virus will spread. It's pandemic physics.
And children, is it this group that doesn't listen the most? I have a hard enough time preventing my kid from climbing on furniture. How did we expect 100% of teenagers to listen to their parents and government?
There was 2 options, violence enforced lockdown or the spread of coronavirus.
I'm quite curious to learn what stands out about these superspreaders. Their infection? Their behavior? Their genes? The timing of their interaction with others?
The odds you die of C19 is overwhelmingly slim especially now with the therapeutics and better understanding of how to treat it. If you're old you are at higher risk (about 2% chance over 65) but thats the case of anything when you're old. Eventually something gets you as you are basically walking on a tight rope. Everybody dies.
Old scared politicians have fucked the younger generation out of self preservation and greed once again. Theres absolutely no reason to shut down society like this. Wear a mask and move on.