Author Topic: Superspreaders and small sample size  (Read 3939 times)

Offline Ergel

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Jun 2010
  • Posts: 12818
  • Total likes: 905
  • DansDeals.com Hat Tips 3
    • View Profile
Superspreaders and small sample size
« on: May 26, 2020, 02:34:34 PM »
Based on some of the articles posted here and in @Dan's roundup, there seems to be growing support for the fact that most infections are spread by a few hyper-infectious virus-shedding superspreaders. In Israel estimates are that 5% of the infected were responsible for at least 80% of the infections.

If these numbers are correct, and they are true elsewhere (see SK nightclub superspreader), then perhaps there is more "chance" to the virus than previously thought. Let's say we assume that the R0 number of the virus is somewhere in the 2-3 range. If there were a low standard deviation from that number (meaning most infected infect around 2 or 3 others), then given the huge sample size of infected there isn't a tremendous room for chance to effect the outcome of virus spreading.
However, if as it appears, there is a HUGE standard deviation, meaning many infected will infect 0 others and a few infected will infect many, then the sample size of infecters is smaller than we thought.
Let's take Israel as an example. 5% of the total 16K infections is only 750 people. When you are dealing with exponential growth, the beginning of the chart has much more impact than the later you get in the chart. If so, then we are looking at maybe the first quarter of those numbers as the most important - at the point we are only talking about 190 people. With those numbers, we are talking a tiny sample size. So a specific person's behavior (and chance) can have a much greater impact on the overall outcome of the pandemic.
Life isn't about checking the boxes. Nobody cares.

Offline yelped

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Mar 2015
  • Posts: 10471
  • Total likes: 3833
  • DansDeals.com Hat Tips 43
    • View Profile
Re: Superspreaders and small sample size
« Reply #1 on: May 26, 2020, 02:41:45 PM »
There's an important lesson in this....


Whatever you do counts. Don't think, nah, it's only me, and I'm only doing this.....

The same is true for the inverse.

Offline Dan

  • Administrator
  • Dansdeals Lifetime 50K Diamond Elite
  • **********
  • Join Date: May 2008
  • Posts: 67599
  • Total likes: 16912
  • DansDeals.com Hat Tips 16442
  • Gender: Male
    • View Profile
  • Location: CLE
  • Programs: UA GS, AA EXP, DL Dirt, Hyatt Glob, Fairmont Lifetime Plat, DD Diamond, Blocked By @NeriaKraus
Re: Superspreaders and small sample size
« Reply #2 on: May 26, 2020, 02:52:13 PM »
Based on some of the articles posted here and in @Dan's roundup, there seems to be growing support for the fact that most infections are spread by a few hyper-infectious virus-shedding superspreaders. In Israel estimates are that 5% of the infected were responsible for at least 80% of the infections.

If these numbers are correct, and they are true elsewhere (see SK nightclub superspreader), then perhaps there is more "chance" to the virus than previously thought. Let's say we assume that the R0 number of the virus is somewhere in the 2-3 range. If there were a low standard deviation from that number (meaning most infected infect around 2 or 3 others), then given the huge sample size of infected there isn't a tremendous room for chance to effect the outcome of virus spreading.
However, if as it appears, there is a HUGE standard deviation, meaning many infected will infect 0 others and a few infected will infect many, then the sample size of infecters is smaller than we thought.
Let's take Israel as an example. 5% of the total 16K infections is only 750 people. When you are dealing with exponential growth, the beginning of the chart has much more impact than the later you get in the chart. If so, then we are looking at maybe the first quarter of those numbers as the most important - at the point we are only talking about 190 people. With those numbers, we are talking a tiny sample size. So a specific person's behavior (and chance) can have a much greater impact on the overall outcome of the pandemic.
Well said!
Save your time, I don't answer PM. Post it in the forum and a dedicated DDF'er will get back to you as soon as possible.

Offline Lurker

  • Dansdeals Lifetime Presidential Platinum Elite
  • *********
  • Join Date: Jul 2019
  • Posts: 5054
  • Total likes: 6390
  • DansDeals.com Hat Tips 2
    • View Profile
  • Location: As always, silence is NOT an admission of agreement on DDF. It just means that people lack the stamina to keep on arguing with made up "facts", illogical arguments, deceiving statements, nasty and degrading comments, and fuzzy math. - @yelped
Re: Superspreaders and small sample size
« Reply #3 on: May 26, 2020, 03:15:16 PM »
This is also why keeping all the precautions in shuls is so important. In addition to super-spreader individuals, there have been super-spreader events, usually at loud venues where raised voices or singing were common. If you place one such individual at one such event for a significant amount of time, the risks would go up tremendously. Please continue to take as many precautions as possible over Shavuos and beyond.
Failing at maintaining Lurker status.

Online avromie7

  • Dansdeals Lifetime Presidential Platinum Elite
  • *********
  • Join Date: Feb 2014
  • Posts: 8188
  • Total likes: 2713
  • DansDeals.com Hat Tips 6
    • View Profile
  • Location: Lakewood
Re: Superspreaders and small sample size
« Reply #4 on: May 26, 2020, 03:29:25 PM »
It also means relatively minimal precautions will have a huge effect. Limiting crowd sizes and limiting exposure in high traffic areas (like masks in grocery stores) will go a very long way, while limitations on small interactions won't have a big effect.
I wonder what people who type "u" instead of "you" do with all their free time.

Offline ari3

  • Dansdeals Lifetime Platinum Elite
  • *******
  • Join Date: May 2016
  • Posts: 1353
  • Total likes: 532
  • DansDeals.com Hat Tips 1
    • View Profile
  • Location: new jersey
Re: Superspreaders and small sample size
« Reply #5 on: May 26, 2020, 03:39:33 PM »
Based on some of the articles posted here and in @Dan's roundup, there seems to be growing support for the fact that most infections are spread by a few hyper-infectious virus-shedding superspreaders. In Israel estimates are that 5% of the infected were responsible for at least 80% of the infections.

If these numbers are correct, and they are true elsewhere (see SK nightclub superspreader), then perhaps there is more "chance" to the virus than previously thought. Let's say we assume that the R0 number of the virus is somewhere in the 2-3 range. If there were a low standard deviation from that number (meaning most infected infect around 2 or 3 others), then given the huge sample size of infected there isn't a tremendous room for chance to effect the outcome of virus spreading.
However, if as it appears, there is a HUGE standard deviation, meaning many infected will infect 0 others and a few infected will infect many, then the sample size of infecters is smaller than we thought.
Let's take Israel as an example. 5% of the total 16K infections is only 750 people. When you are dealing with exponential growth, the beginning of the chart has much more impact than the later you get in the chart. If so, then we are looking at maybe the first quarter of those numbers as the most important - at the point we are only talking about 190 people. With those numbers, we are talking a tiny sample size. So a specific person's behavior (and chance) can have a much greater impact on the overall outcome of the pandemic.
The question is assuming this theory is correct what can be done with this information to fight the spread.
Why are some people superspreaders and others not? Do superspreaders have severe symptoms? Can an asymptomatic individual be a superspreader? (likely not)

Offline Ergel

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Jun 2010
  • Posts: 12818
  • Total likes: 905
  • DansDeals.com Hat Tips 3
    • View Profile
Re: Superspreaders and small sample size
« Reply #6 on: May 26, 2020, 03:46:08 PM »
The question is assuming this theory is correct what can be done with this information to fight the spread.
Why are some people superspreaders and others not? Do superspreaders have severe symptoms? Can an asymptomatic individual be a superspreader? (likely not)
I was wondering this also. Do superspreaders tests look different than the standard positive test? Presumably they should, but I don't know. Wonder if they tested it with the SK nightclub superspreader
Life isn't about checking the boxes. Nobody cares.

Offline S209

  • Dansdeals Lifetime Presidential Platinum Elite
  • *********
  • Join Date: Jun 2016
  • Posts: 7469
  • Total likes: 3930
  • DansDeals.com Hat Tips 1
  • Gender: Male
    • View Profile
    • Gowns By Shevy
  • Location: Lakewood
  • Programs: Marriott Gold, Star Alliance Gold, Hyatt Explorist, Hertz PC, National EE, Rock Royalty Wild Card, Wyndham Diamond, MLife Gold, Caesars Diamond, Hilton Diamond, Uber VIP, IHG Platinum Elite, ANA Platinum, DDF Lifetime Prez Platinum Elite, AmEx Platinum
Re: Superspreaders and small sample size
« Reply #7 on: May 26, 2020, 09:34:03 PM »
Based on some of the articles posted here and in @Dan's roundup, there seems to be growing support for the fact that most infections are spread by a few hyper-infectious virus-shedding superspreaders. In Israel estimates are that 5% of the infected were responsible for at least 80% of the infections.

If these numbers are correct, and they are true elsewhere (see SK nightclub superspreader), then perhaps there is more "chance" to the virus than previously thought. Let's say we assume that the R0 number of the virus is somewhere in the 2-3 range. If there were a low standard deviation from that number (meaning most infected infect around 2 or 3 others), then given the huge sample size of infected there isn't a tremendous room for chance to effect the outcome of virus spreading.
However, if as it appears, there is a HUGE standard deviation, meaning many infected will infect 0 others and a few infected will infect many, then the sample size of infecters is smaller than we thought.
Let's take Israel as an example. 5% of the total 16K infections is only 750 people. When you are dealing with exponential growth, the beginning of the chart has much more impact than the later you get in the chart. If so, then we are looking at maybe the first quarter of those numbers as the most important - at the point we are only talking about 190 people. With those numbers, we are talking a tiny sample size. So a specific person's behavior (and chance) can have a much greater impact on the overall outcome of the pandemic.
The Pareto Principle in action
Quote from: YitzyS
Quotes in a signature is annoying, as it comes across as an independent post.

Offline junion

  • Dansdeals Silver Elite
  • **
  • Join Date: Mar 2020
  • Posts: 97
  • Total likes: 52
  • DansDeals.com Hat Tips 0
    • View Profile
  • Location: NYC
Re: Superspreaders and small sample size
« Reply #8 on: May 26, 2020, 10:03:41 PM »
Based on some of the articles posted here and in @Dan's roundup, there seems to be growing support for the fact that most infections are spread by a few hyper-infectious virus-shedding superspreaders. In Israel estimates are that 5% of the infected were responsible for at least 80% of the infections.

If these numbers are correct, and they are true elsewhere (see SK nightclub superspreader), then perhaps there is more "chance" to the virus than previously thought. Let's say we assume that the R0 number of the virus is somewhere in the 2-3 range. If there were a low standard deviation from that number (meaning most infected infect around 2 or 3 others), then given the huge sample size of infected there isn't a tremendous room for chance to effect the outcome of virus spreading.
However, if as it appears, there is a HUGE standard deviation, meaning many infected will infect 0 others and a few infected will infect many, then the sample size of infecters is smaller than we thought.


Let's take Israel as an example. 5% of the total 16K infections is only 750 people. When you are dealing with exponential growth, the beginning of the chart has much more impact than the later you get in the chart. If so, then we are looking at maybe the first quarter of those numbers as the most important - at the point we are only talking about 190 people. With those numbers, we are talking a tiny sample size. So a specific person's behavior (and chance) can have a much greater impact on the overall outcome of the pandemic.

This fact gets me nervous, especially for Yomim Noraim. All you need is one superspreader, and even if there is %50 immunity in the community, maybe half to a quarter of people in the room can get it. especially since people are there for many hours and davening loudly. Add that to the posbility that immunity might wear off by that time. OUCH. :o

Online etech0

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Dec 2013
  • Posts: 12862
  • Total likes: 3317
  • DansDeals.com Hat Tips 1
    • View Profile
  • Location: not lakewood
  • Programs: DDF
Re: Superspreaders and small sample size
« Reply #9 on: May 26, 2020, 10:12:29 PM »
This fact gets me nervous, especially for Yomim Noraim. All you need is one superspreader, and even if there is %50 immunity in the community, maybe half to a quarter of people in the room can get it. especially since people are there for many hours and davening loudly. Add that to the posbility that immunity might wear off by that time. OUCH. :o
We need a way to identify these super spreaders and have them quarantine until they stop
Workflowy. You won't know what you're missing until you try it.

Offline PlatinumGuy

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Apr 2011
  • Posts: 14958
  • Total likes: 2421
  • DansDeals.com Hat Tips 11
    • View Profile
Re: Superspreaders and small sample size
« Reply #10 on: May 27, 2020, 05:49:42 AM »
I was wondering this also. Do superspreaders tests look different than the standard positive test? Presumably they should, but I don't know. Wonder if they tested it with the SK nightclub superspreader
I don't think it has to do with the actual viral genome so much as other circumstances causing transmission. My guess is a brief interval of time where the virus is multiplying enough to be contagious yet the immune system hasn't neutralized it sufficiently. That period may not occur in most people, or be too short to infect others.
 
Spreading would also be increased by a cold with sneezing, caughing, & extra droplet production early on before the immune system renders it intransmittable.
״וזה כלל גדול: שישנא אדם כל דבר שקר. וכל מה שיוסיף שנאה לדרכי השקר – יוסיף אהבה לתורה.״ - אורחות צדיקים

Online etech0

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Dec 2013
  • Posts: 12862
  • Total likes: 3317
  • DansDeals.com Hat Tips 1
    • View Profile
  • Location: not lakewood
  • Programs: DDF
Re: Superspreaders and small sample size
« Reply #11 on: May 27, 2020, 09:02:05 AM »
That period may not occur in most people, or be too short to infect others.
or if it's really that short, maybe they were home/sleeping for those few hours
Workflowy. You won't know what you're missing until you try it.

Offline PlatinumGuy

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Apr 2011
  • Posts: 14958
  • Total likes: 2421
  • DansDeals.com Hat Tips 11
    • View Profile
Re: Superspreaders and small sample size
« Reply #12 on: May 27, 2020, 10:02:05 AM »
or if it's really that short, maybe they were home/sleeping for those few hours
That's my theory
״וזה כלל גדול: שישנא אדם כל דבר שקר. וכל מה שיוסיף שנאה לדרכי השקר – יוסיף אהבה לתורה.״ - אורחות צדיקים

Offline S209

  • Dansdeals Lifetime Presidential Platinum Elite
  • *********
  • Join Date: Jun 2016
  • Posts: 7469
  • Total likes: 3930
  • DansDeals.com Hat Tips 1
  • Gender: Male
    • View Profile
    • Gowns By Shevy
  • Location: Lakewood
  • Programs: Marriott Gold, Star Alliance Gold, Hyatt Explorist, Hertz PC, National EE, Rock Royalty Wild Card, Wyndham Diamond, MLife Gold, Caesars Diamond, Hilton Diamond, Uber VIP, IHG Platinum Elite, ANA Platinum, DDF Lifetime Prez Platinum Elite, AmEx Platinum
Re: Superspreaders and small sample size
« Reply #13 on: May 27, 2020, 10:11:27 AM »
That's my theory
Very interesting. Have you seen anything to back that up?
Quote from: YitzyS
Quotes in a signature is annoying, as it comes across as an independent post.

Online etech0

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Dec 2013
  • Posts: 12862
  • Total likes: 3317
  • DansDeals.com Hat Tips 1
    • View Profile
  • Location: not lakewood
  • Programs: DDF
Re: Superspreaders and small sample size
« Reply #14 on: May 27, 2020, 10:18:24 AM »
Very interesting. Have you seen anything to back that up?
Then it wouldn't be a theory :)
Workflowy. You won't know what you're missing until you try it.

Offline PlatinumGuy

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Apr 2011
  • Posts: 14958
  • Total likes: 2421
  • DansDeals.com Hat Tips 11
    • View Profile
Re: Superspreaders and small sample size
« Reply #15 on: May 27, 2020, 10:30:37 AM »
Very interesting. Have you seen anything to back that up?
The same data that you saw. We know that at a certain point people still test positive in PCR & yet PFU assays show the virus isn't potent enough for transmission.
We also know it takes some time from exposure to becoming contagious. At least 15 minutes that it takes to circulate...

So perhaps the interval is too short?

״וזה כלל גדול: שישנא אדם כל דבר שקר. וכל מה שיוסיף שנאה לדרכי השקר – יוסיף אהבה לתורה.״ - אורחות צדיקים

Offline junion

  • Dansdeals Silver Elite
  • **
  • Join Date: Mar 2020
  • Posts: 97
  • Total likes: 52
  • DansDeals.com Hat Tips 0
    • View Profile
  • Location: NYC
Re: Superspreaders and small sample size
« Reply #16 on: May 27, 2020, 12:37:22 PM »

We also know it takes some time from exposure to becoming contagious. At least 15 minutes that it takes to circulate...

source?

Offline PlatinumGuy

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Apr 2011
  • Posts: 14958
  • Total likes: 2421
  • DansDeals.com Hat Tips 11
    • View Profile
Re: Superspreaders and small sample size
« Reply #17 on: May 27, 2020, 01:02:36 PM »
source?

https://arxiv.org/pdf/2003.12886.pdf

Quote

Once a person is infected, it
takes a period of time known as the latent period before they are able to transmit the
virus. The current best-estimate of the median latent time is ≈3 days followed by a
≈4 day infectious period ​(Li et al. 2020​, ​He et al. 2020)​.


https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4440524/

Quote
For example, the average length of the eclipse phase has been estimated to be around 24 h for infections with the human immunodeficiency virus type 1 (HIV-1)10,11,12,13,14, and varied from 6 h to as much as 12 h for infections with different strains of influenza virus4,15,16,17,18.

https://wwwnc.cdc.gov/eid/article/26/6/20-0516_article

Quote
Finally, we evaluated the replication kinetics of SARS-CoV-2 in a multistep growth curve. In brief, we infected Vero CCL-81 and HUH7.0 cells with SARS-CoV-2 at a low multiplicity of infection (0.1) and evaluated viral replication every 6 h for 72 h postinoculation, with separate harvests in the cell-associated and supernatant compartments (Figure 4). Similar to SARS-CoV, SARS-CoV-2 replicated rapidly in Vero cells after an initial eclipse phase, achieving 105 TCID50/mL by 24 h postinfection and peaking at >106 TCID50/mL. We observed similar titers in cell-associated and supernatant compartments, which indicated efficient egress. Despite peak viral titers by 48 h postinoculation, major CPE was not observed until 60 h postinoculation and peaked at 72 h postinoculation, indicating that infected monolayers should be harvested before peak CPE is observed. Replication in HUH7.0 cells also increased quickly after an initial eclipse phase but plateaued by 24 h postinoculation in the intracellular compartment at 2 × 103 TCID50/mL and decreased after 66 h postinoculation. Virus was not detected in the supernatant of infected HUH7 cells until 36 h postinoculation and exhibited lower titers at all timepoints (Figure 4). Major CPE was never observed in HUH7.0 cells. These results are consistent with previous reports for SARS-CoV and MERS-CoV, which suggested similar replication dynamics between the zoonotic CoV strains (23,24).




See also https://www.ncbi.nlm.nih.gov/pubmed/15752773
״וזה כלל גדול: שישנא אדם כל דבר שקר. וכל מה שיוסיף שנאה לדרכי השקר – יוסיף אהבה לתורה.״ - אורחות צדיקים

Offline PlatinumGuy

  • Dansdeals Lifetime 10K Presidential Platinum Elite
  • *******
  • Join Date: Apr 2011
  • Posts: 14958
  • Total likes: 2421
  • DansDeals.com Hat Tips 11
    • View Profile
Re: Superspreaders and small sample size
« Reply #18 on: September 21, 2020, 11:24:46 PM »
I don't think it has to do with the actual viral genome so much as other circumstances causing transmission. My guess is a brief interval of time where the virus is multiplying enough to be contagious yet the immune system hasn't neutralized it sufficiently. That period may not occur in most people, or be too short to infect others.
 
Spreading would also be increased by a cold with sneezing, caughing, & extra droplet production early on before the immune system renders it intransmittable.

As to your wife, perhaps you overcame the virus fast enough that it wasn't transmittable by the time you got home from NY

Looks like my theory has been proven

https://www.vox.com/21296067/coronavirus-covid-symptoms-superspreaders-superspreading-contagious-bars-restaurants
Quote
  The research suggests that people carrying this virus are potentially infectious for about two and a half days — and rise to a “moderate probability of transmission per contact” for “less than half a day,” the researchers write


So Super-Spreaders aren't necessarily distinct, it's just per chance that their contagious period was when they were around a lot of people.
״וזה כלל גדול: שישנא אדם כל דבר שקר. וכל מה שיוסיף שנאה לדרכי השקר – יוסיף אהבה לתורה.״ - אורחות צדיקים

Offline yuneeq

  • Dansdeals Lifetime Presidential Platinum Elite
  • *********
  • Join Date: Jan 2013
  • Posts: 8611
  • Total likes: 3999
  • DansDeals.com Hat Tips 10
  • Gender: Male
    • View Profile
  • Location: NJ
Re: Superspreaders and small sample size
« Reply #19 on: September 22, 2020, 11:16:46 PM »
This fact gets me nervous, especially for Yomim Noraim. All you need is one superspreader, and even if there is %50 immunity in the community, maybe half to a quarter of people in the room can get it. especially since people are there for many hours and davening loudly. Add that to the posbility that immunity might wear off by that time. OUCH. :o

This post from May is spot on
Visibly Jewish