14 ms·

This entire study seems to be centered around a theoretical, mathematical model, and I didn't see any attempt to actually validate that model. Basically, the a

This entire study seems to be centered around a theoretical, mathematical model, and I didn't see any attempt to actually validate that model. Basically, the authors seem to assume that the virus behaves according to their formulas, and show that under their assumptions, face masks work, but don't actually prove that their assumptions match reality - or did I miss something?

FiReaNG3L 5y ago
No you didn't miss anything - most people upvoting this didn't read anything more than the title.
- Louis_Sarwal 5y ago
  So true.
omegaworks 5y ago
It invites rigor in studying mask efficacy. Some studies out there show inconclusive results of mask usage, and this article helps give us an explanation as to why we might be seeing that. The model shows that in an extremely virus-rich environment masking can be rendered ineffective, but in virus-poor environments, even surgical masks decrease the probability of infection. Researchers need to measure concentration of virus in the environment as well as mask usage to show the efficacy of masks in real world situations.
rsj_hn 5y ago
Indeed this is the money paragraph: "The effectiveness of masks, however, is still under debate. Compared to N95/FFP2 respirators which have very low particle penetration rates (around ~5%), surgical and similar masks exhibit higher and more variable penetration rates (around ~30-70%) (2, 3). Given the large number of particles emitted upon respiration and especially upon sneezing or coughing (4), the number of respiratory particles that may penetrate masks is substantial, which is one of the main reasons leading to doubts about their efficacy in preventing infections. Moreover, randomized clinical trials show inconsistent or inconclusive results, with some studies reporting only a marginal benefit or no effect of mask use (5, 6). Thus, surgical and similar masks are often considered to be ineffective. On the other hand, observational data show that regions or facilities with a higher percentage of the population wearing masks have better control of the coronavirus disease 2019 (COVID-19) (7–9). So how to explain these contrasting results and apparent inconsistencies? Here, we develop a quantitative model of airborne virus exposure that can explain these contrasting results and provide a basis for quantifying the efficacy of face masks. " So they are left with a conundrum that places which use surgical masks seem to be better off, whereas randomized control studies of surgical masks show little to no benefit. This is what they try to explain with a mathematical model.
- BugsJustFindMe 5y ago
  The mistake that many people make, and that I think you're implicitly making here too, is this notion of "controlled study". If the study denies the result seen time and again in real life, perhaps the study isn't studying correctly.
  - colonelanguz 5y ago
    How many agains makes the study wrong
  - bawolff 5y ago
    If we fully understood what was happening in real life, we wouldn't need studies.
- function_seven 5y ago
  > On the other hand, "observational data show that regions or facilities with a higher percentage of the population wearing masks have better control of the coronavirus disease 2019 (COVID-19) (7–9)." So how to explain these contrasting results and apparent inconsistencies? My hypothesis regarding surgical mask prevalence and how it correlates with better virus control: People who wear masks--however effective they may be--also behave in other ways to limit the spread. They minimize time spent in public indoor spaces, they keep distance from other people, they're more likely to self-quarantine if a family member is sick, etc. The mask-wearing prevalence is a proxy for how serious the wearer takes Covid as a threat.
  - jasonlaramburu 5y ago
    It’s worth also considering the case of ski helmets. Since these have became common on the slopes head trauma has actually increased because wearers take more risks.
    - swader999 5y ago
      Probably related to the uptake in snow boarding.
admissionsguy 5y ago
Working down from the conclusion to find the necessary assumptions is the standard approach to COVID-related questions.
- nerdponx 5y ago
  This is a valid way to do science. "What assumptions are required to make this conclusion valid?" If the assumptions turn out to be outlandish, then maybe we can't take the conclusion to be true.
  - admissionsguy 5y ago
    In principle, sure, but here, it's just grotesque. They admit RCTs show masks don't matter, so they put together a few differential equations (which most certainly could be chosen differently to support something else entirely) such that in some domains (virus-limited) masks help a lot, in some (virus-rich) not. Then they implicitly assume that RCTs didn't show anything because they were done in virus-rich environments and somehow conclude from this dubious claim that "Face masks effectively limit the probability of SARS-CoV-2 transmission". Am I not reading it correctly?
Louis_Sarwal 5y ago
Probably not. Most things like this are like this. There may indeed be some underlying proof, but they either can't or won't show it to you-if it exists. It is more likely that it does not. They start with their conclusion. They rarely commence an empirical study, because that might prove their prepared conclusion wrong! A model, which can be used in place of reality, is not reality, and never will be. GIGO. The model was tailored to match their wanted results, or more likely they made a bad model and tweaked it to gain desired results. This does not mean that masks are bad-it just means that the methods used are generally false. Properly gained empirical evidence should be used, and it even should not be trusted if it points to you doing something stupid-like wearing a tissue on your face all day.
neolog 5y ago
The paper is fine, the title is clickbait.
nimish 5y ago
Nope. It's circular. We need actual tests in the wild, not models.
- simpleguitar 5y ago
  During an interview with some of the doctors sent into Wuhan in their wave, they claimed none of the 10 thousand or so doctors sent in became infected. They used strict PPE protocol, including N95 masks. Given what we know about the covid transmission now, it was absolutely the N95s that prevented these doctors from getting infected.
  - NeutronStar 5y ago
    Absolutely not. You can't claim absolutely.
  - ganoushoreilly 5y ago
    Alternatively it could also be flawed data on infections, which we know is already highly likely.
  - nimish 5y ago
    Then run a proper randomized controlled trial with comparisons to cloth and surgical masks so we have actual data, not 10000 anecdotes. This was the biggest event of the last 50 years, we can throw a few billion at a real trial.
    - simpleguitar 5y ago
      One problem--ethics. Can't knowingly give humans less protection. It's not a money problem.
      - nimish 5y ago
        That is what you are testing! Equipoise demands you don't assume one outcome. Otherwise every single pharmaceutical trial would be impossible since "we can't knowingly give humans a placebo" If the effect is large and significant we will see it quickly and can halt asap.
gonational 5y ago
interesting; the same kind of erroneous supposition modeling MO is used to pretend that humans are "cooking the planet" (or is it an ice age? I guess it depends on the decade...).
- AshWolfy 5y ago
  this is wrong, unrelated, and clear flamebaiting
  - nfvme 5y ago
    This entire thread is flamebaiting, and might I add, COMPLETELY OFF TOPIC!
  - gonational 5y ago
    Should I have preceded my comment with "trigger warning: real information ahead"? I mean, it's now considered "flame be" to state your opinion? Not everybody gets their science from CNN, so not everybody is going to have the same opinion as you. People with different opinions are not trying to troll you by stating their opinions.
nfvme 5y ago
You must be a science denier! Burn the witch!
apatters 5y ago
My thinking has always been that of all the possible interventions to reduce viral spread, something like making people wear masks while grocery shopping has almost no negative side effects. Whereas closing the grocery store altogether has major negative side effects, like lost jobs and people not being able to buy groceries. So I think the evidential bar for requiring masks should be pretty low (unless it is a special case, like making people wear masks when they're exercising likely has larger negative side effects). Who cares if it may not do much? The cost of doing it is so low. Whereas the evidential bar for policies that do more collateral damage ought to be higher.
- lixtra 5y ago
  If masks turn out to be ineffective, the cost of having them recommended is reduced reputation and authority. Would you take security advice from Bruce Schneier after he proposed you to cover your keyboard in snakeoil because it might protect against computer viruses (and if not, there is not much damage)?
  - chimprich 5y ago
    That's an absurd comparison. Regardless of how well they work in practice, there's a clear theoretical reason why masks could/should work: they catch salivary droplets containing the virus. Covering your keyboard in oil is not even theoretically going to affect your computer's susceptibility to computer viruses.
    - lixtra 5y ago
      That’s the difference between theory and practice. In practice the snakeoil may prevent you from using your computer more frequently and therefore result in less infections. In practice I observe a lot of people fumbling out unwashed pieces of cloth from their pockets and placing them in their face, hoping it would preserve their respiratory health.
BurnGpuBurn 5y ago
You are correct, although I wouldn't call this a study. It's a science paper, yes, but it doesn't contain any new data that has been studied, just some fancy formulas and graphs and conjecture by a couple of people with a science degree.
collyw 5y ago
This has been the case with all of the non-pharmaceutical measures imposed. The models say they work, but real world data doesn't show much correlation.

FiReaNG3L 5y ago

No you didn't miss anything - most people upvoting this didn't read anything more than the title.

Louis_Sarwal 5y ago
So true.

omegaworks 5y ago

It invites rigor in studying mask efficacy. Some studies out there show inconclusive results of mask usage, and this article helps give us an explanation as to why we might be seeing that. The model shows that in an extremely virus-rich environment masking can be rendered ineffective, but in virus-poor environments, even surgical masks decrease the probability of infection. Researchers need to measure concentration of virus in the environment as well as mask usage to show the efficacy of masks in real world situations.

rsj_hn 5y ago

Indeed this is the money paragraph: "The effectiveness of masks, however, is still under debate. Compared to N95/FFP2 respirators which have very low particle penetration rates (around ~5%), surgical and similar masks exhibit higher and more variable penetration rates (around ~30-70%) (2, 3). Given the large number of particles emitted upon respiration and especially upon sneezing or coughing (4), the number of respiratory particles that may penetrate masks is substantial, which is one of the main reasons leading to doubts about their efficacy in preventing infections. Moreover, randomized clinical trials show inconsistent or inconclusive results, with some studies reporting only a marginal benefit or no effect of mask use (5, 6). Thus, surgical and similar masks are often considered to be ineffective. On the other hand, observational data show that regions or facilities with a higher percentage of the population wearing masks have better control of the coronavirus disease 2019 (COVID-19) (7–9). So how to explain these contrasting results and apparent inconsistencies? Here, we develop a quantitative model of airborne virus exposure that can explain these contrasting results and provide a basis for quantifying the efficacy of face masks. " So they are left with a conundrum that places which use surgical masks seem to be better off, whereas randomized control studies of surgical masks show little to no benefit. This is what they try to explain with a mathematical model.

BugsJustFindMe 5y ago
The mistake that many people make, and that I think you're implicitly making here too, is this notion of "controlled study". If the study denies the result seen time and again in real life, perhaps the study isn't studying correctly.
- colonelanguz 5y ago
  How many agains makes the study wrong
- bawolff 5y ago
  If we fully understood what was happening in real life, we wouldn't need studies.
function_seven 5y ago
> On the other hand, "observational data show that regions or facilities with a higher percentage of the population wearing masks have better control of the coronavirus disease 2019 (COVID-19) (7–9)." So how to explain these contrasting results and apparent inconsistencies? My hypothesis regarding surgical mask prevalence and how it correlates with better virus control: People who wear masks--however effective they may be--also behave in other ways to limit the spread. They minimize time spent in public indoor spaces, they keep distance from other people, they're more likely to self-quarantine if a family member is sick, etc. The mask-wearing prevalence is a proxy for how serious the wearer takes Covid as a threat.
- jasonlaramburu 5y ago
  It’s worth also considering the case of ski helmets. Since these have became common on the slopes head trauma has actually increased because wearers take more risks.
  - swader999 5y ago
    Probably related to the uptake in snow boarding.

admissionsguy 5y ago

Working down from the conclusion to find the necessary assumptions is the standard approach to COVID-related questions.

nerdponx 5y ago
This is a valid way to do science. "What assumptions are required to make this conclusion valid?" If the assumptions turn out to be outlandish, then maybe we can't take the conclusion to be true.
- admissionsguy 5y ago
  In principle, sure, but here, it's just grotesque. They admit RCTs show masks don't matter, so they put together a few differential equations (which most certainly could be chosen differently to support something else entirely) such that in some domains (virus-limited) masks help a lot, in some (virus-rich) not. Then they implicitly assume that RCTs didn't show anything because they were done in virus-rich environments and somehow conclude from this dubious claim that "Face masks effectively limit the probability of SARS-CoV-2 transmission". Am I not reading it correctly?

Louis_Sarwal 5y ago

Probably not. Most things like this are like this. There may indeed be some underlying proof, but they either can't or won't show it to you-if it exists. It is more likely that it does not. They start with their conclusion. They rarely commence an empirical study, because that might prove their prepared conclusion wrong! A model, which can be used in place of reality, is not reality, and never will be. GIGO. The model was tailored to match their wanted results, or more likely they made a bad model and tweaked it to gain desired results. This does not mean that masks are bad-it just means that the methods used are generally false. Properly gained empirical evidence should be used, and it even should not be trusted if it points to you doing something stupid-like wearing a tissue on your face all day.

neolog 5y ago

The paper is fine, the title is clickbait.

nimish 5y ago

Nope. It's circular. We need actual tests in the wild, not models.

simpleguitar 5y ago
During an interview with some of the doctors sent into Wuhan in their wave, they claimed none of the 10 thousand or so doctors sent in became infected. They used strict PPE protocol, including N95 masks. Given what we know about the covid transmission now, it was absolutely the N95s that prevented these doctors from getting infected.
- NeutronStar 5y ago
  Absolutely not. You can't claim absolutely.
- ganoushoreilly 5y ago
  Alternatively it could also be flawed data on infections, which we know is already highly likely.
- nimish 5y ago
  Then run a proper randomized controlled trial with comparisons to cloth and surgical masks so we have actual data, not 10000 anecdotes. This was the biggest event of the last 50 years, we can throw a few billion at a real trial.
  - simpleguitar 5y ago
    One problem--ethics. Can't knowingly give humans less protection. It's not a money problem.
    - nimish 5y ago
      That is what you are testing! Equipoise demands you don't assume one outcome. Otherwise every single pharmaceutical trial would be impossible since "we can't knowingly give humans a placebo" If the effect is large and significant we will see it quickly and can halt asap.

gonational 5y ago

interesting; the same kind of erroneous supposition modeling MO is used to pretend that humans are "cooking the planet" (or is it an ice age? I guess it depends on the decade...).

AshWolfy 5y ago
this is wrong, unrelated, and clear flamebaiting
- nfvme 5y ago
  This entire thread is flamebaiting, and might I add, COMPLETELY OFF TOPIC!
- gonational 5y ago
  Should I have preceded my comment with "trigger warning: real information ahead"? I mean, it's now considered "flame be" to state your opinion? Not everybody gets their science from CNN, so not everybody is going to have the same opinion as you. People with different opinions are not trying to troll you by stating their opinions.

nfvme 5y ago

You must be a science denier! Burn the witch!

apatters 5y ago

My thinking has always been that of all the possible interventions to reduce viral spread, something like making people wear masks while grocery shopping has almost no negative side effects. Whereas closing the grocery store altogether has major negative side effects, like lost jobs and people not being able to buy groceries. So I think the evidential bar for requiring masks should be pretty low (unless it is a special case, like making people wear masks when they're exercising likely has larger negative side effects). Who cares if it may not do much? The cost of doing it is so low. Whereas the evidential bar for policies that do more collateral damage ought to be higher.

lixtra 5y ago
If masks turn out to be ineffective, the cost of having them recommended is reduced reputation and authority. Would you take security advice from Bruce Schneier after he proposed you to cover your keyboard in snakeoil because it might protect against computer viruses (and if not, there is not much damage)?
- chimprich 5y ago
  That's an absurd comparison. Regardless of how well they work in practice, there's a clear theoretical reason why masks could/should work: they catch salivary droplets containing the virus. Covering your keyboard in oil is not even theoretically going to affect your computer's susceptibility to computer viruses.
  - lixtra 5y ago
    That’s the difference between theory and practice. In practice the snakeoil may prevent you from using your computer more frequently and therefore result in less infections. In practice I observe a lot of people fumbling out unwashed pieces of cloth from their pockets and placing them in their face, hoping it would preserve their respiratory health.

BurnGpuBurn 5y ago

You are correct, although I wouldn't call this a study. It's a science paper, yes, but it doesn't contain any new data that has been studied, just some fancy formulas and graphs and conjecture by a couple of people with a science degree.

collyw 5y ago

This has been the case with all of the non-pharmaceutical measures imposed. The models say they work, but real world data doesn't show much correlation.