Main Edition
US
FR
September 3, 2021 - 3:47pm
There’s a row going on about a study into masks. The researchers took 300,000 people in Bangladeshi villages. To hugely oversimplify: in half of the villages, they promoted mask-wearing. In the other half, they didn’t. They then looked at whether the people in the mask-promotion villages were less likely to get Covid. They did: 8.6% of people in the control villages reported Covid symptoms, compared to 7.6% in the treatment villages.
But some people said that this result may not be statistically significant. It’s a bit more complicated than that, but let’s just take the claim at face value. The row tells us something interesting about science and evidence.
Here’s what “statistically significant” means. In science, there’s a thing called the “p-value”. That is: how likely you are to see a given result by fluke. Say you’re trying to find out if your dice are loaded. You roll two sixes. That could mean that the dice are loaded, or you might have just rolled two sixes. Your chance of seeing two sixes on fair dice is one in 36. P-values are written as a score out of 1, so your p-value for that result is 1/36≈0.028.
A result is “statistically significant” if your p-value is less than 0.05: if you would expect to see that result, or a more extreme one, by fluke less than one time in 20. (What it doesn’t mean is that there’s only a one in 20 chance that it’s wrong. Read this for more.)
If that sounds complicated: it is. Most psychology lecturers get it wrong, as do most psychology textbooks. This may explain quite a lot about psychology.
In science, statistical significance is often used as a cutoff: you can’t get your study published if your p-value is greater than 0.05. This system has led to people juking the stats to get their p-values below 0.05, because we say “if it lands on this side of the line it’s real”.
Join the discussion
Join like minded readers that support our journalism by becoming a paid subscriber
To join the discussion in the comments, become a paid subscriber.
Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.
SubscribeEveryone I know who has had Covid was masked up – some used to even drive around along in the car with a mask on. In fact millions of masked people have gotten Covid, ergo masks don’t work.
Israel shows that double jabbed and masked up results in the highest boom in covid infections yet!
“Israel is reporting the highest coronavirus infection rate in the world, showing that neither vaccine mandates nor “vaccine passports” are suitable means to limit or end the pandemic.
Israel has been a global pioneer in covid mass vaccinations as well as in introducing the highly controversial “vaccine passport” (Green Pass). Nevertheless, since late August 2021, Israel has been reporting the highest coronavirus infection rate in the entire world (see chart above).”
Yes and vaccinating people with a non sterilising vaccine can drive variant escape. Tom Chivers – I think this should be a topic for an article.
Bret Weinstein does great talks on that – he and his family (wife also a PhD in Biology) refused the ‘vaccines’ and use ivermectin.
But it is just intuitive – if almost all Israelis have the double dose, and yet they have the highest covid infection in the world as of right now, in the vaccinated, then all those covid (one billion in each infected person) viruses are being fought by One Spike S Protein antibody the RNA vaccine uses – and so any which get around that by a mutation will be selected for.
It would seem we are essentially creating resistant mutations – mu being the latest varriant I have herd of.
Yes I watch Bret Weinstein and Heather Heying…!
His sort of Liberal plaintive mewling grates a bit – but he is fun, and intelligent, and brave to face the Social Media Stasi as he does – and is also a voice of reason in all things Covid-wise.
I also like Jordon Peterson, but he has the same flaws, and strengths, as Bret.
My favorite of them all is George Gammon.
https://www.youtube.com/c/GeorgeGammon/videos
And wile fun, you also pick up some vital economic terms and concepts, and also learn how devious the real world is. The last two give a real example of his stuff.
The fact that you judge Tom Chivers politically is revealing. He is a science writer, and rational evidence based argument is pretty much anathema to both sides of the culture war, one side is good, the other bad, nuff said.
I think he only said that it might help prevent, to actually prevent, by trapping some droplets. Ie it might be marginally useful. Which isn’t saying much. But it’s something
NOT actually prevent. Sorry I don’t seem to be able to edit my last post. Edit button not working for me
Not necessarily! Classic example of absolutist black and white, and sorry, irrational reasoning. It entirely ignores degree, masks may reduce transmission, but not eliminate it.
The results are not statistically significant. That is a fact.
But even if they were, this would not prove anything either way. There is an almost infinite array of other reasons why the numbers of people reporting Covid symptoms were different in the two groups.
There has only been one scientifically relevant study done on mask wearing and that was done in Denmark.
It showed that mask wearing has no effect on the wearer’s chances of contracting Covid. What it was not able to test (for ethical reasons) was the effect of mask wearing on third parties.
Absence of evidence is not evidence of absence. In the real world, doctors working with Covid patients are now starting to use FFP3 masks because there is crude but compelling data that they prevent inhalation of aerosols (as well as blocking droplet contamination, which any cheaper mask can do).
No question that an FFP3 mask can prevent inhalation of aerosols to some extent. But recall that an FFP3 is even higher quality than a N95 mask which in EU ratings corresponds to an FFP2. The Chinese KN95 corresponds to an EU FFP1. It is perfectly reasonable to wear a FFP3 mask for a short period of time while examining a patient and then throwing it away. But try and wear a FFP3 for any significant amount of time. Somehow I don’t think you’ll appreciate doing so.
That is exactly it. So often we see these correlationstudies, but there is no investigation into what actually explains the relationship (i.e. mediators and moderators). For example, could the difference between the masked and unmasked in this study be explained by the increased education and wariness of Covid resulting in behavioural changes rather than by a protective effect of masks? It is really a poorly designed study, because it does not control for all those factors. Why would they not administer the same sort of Covid education to both groups? Why did they not measure other behaviours such as time spent with large groups, time in community, and other such factors which seem to indicate risk?
A great example of someone not understanding the argument. The Danish study was very similar to the Indian one. The results did show a reduction in being infected with Covid for people wearing masks, but it was a small effect and the 0.05 p value was exceeded. So the results were considered to be not statistically significant. This is not the same as proving that masks ‘have no effect’.
Even assuming it is a “real” result, a 1% disparity between the two groups hardly suggests much impact of a mask-wearing policy. It would be interesting to know what the compliance rate was with the request.
Exactly. This kind of study would be laughed out of court by the people who insist that very large and rigorous blind RCT with placebos are the only trials that have validity.
Love the idea of the double blind where half are given masks made out of chicken wire as a placebo, and half get N95 masks.
Surely a Niel Ferguson/Fauchi study could not beat that for craziness.
It is not a 1% disparity. It is a 15% disparity (give or take) or a 1 percentage point.
One other comment for Chivers. I appreciate he likes to show off how clever he is by continually mentioning Bayesian statistics. But as was once said (whether by Dilke, Disraeli or Twain or all three of them): there are lies, damned lies and statistics. If an effect is obvious you don’t need to resort to any form of statistics. You only need to rely on statistics when the result isn’t obvious from the data.
And incidentally, Chivers’ estimation of prior probability (80%) for Bayesian statistical analysis is way off because there is actually ZERO evidence that surgical or cloth masks actually work in the real world. A valid Bayesian probability prior has to be based on actual data, not on wishful thinking. There have been innumerable randomized controlled studies prior to COVID with influenza and none of them were able to demonstrate any significant impact of masks. Given that all respiratory viruses (which are all more or less the same size) are transmitted in the same way both by aerosols and droplets, it follows that the same is true of COVID. Indeed, the Danish mask study, which was well done and sufficiently high powered, demonstrated that surgical masks offer NO significant protection to catching COVID. Now, there still remains the possibility that surgical and cloth masks could act as a source control to prevent an infected person from infecting a non-infected person. But the current Bangladesh study, irrespective of any nuance or potential flaws, clearly shows that the reduction in COVID infection obtained by surgical and cloth mask wearing is negligible (9%) in the real world even if highly statistically significant because of their massive sample size.
To put it in terms that Chivers might be able to understand given his obvious confirmation bias, let me provide a simple example. Let us say you are forced to play Russian roulette with a 6 shooter and one chamber loaded. So you have a 1 in 6 chance of killing yourself. Now let us say we can reduce that risk by 9%, your risk of killing yourself only goes down to 1 in 6.5. From my perspective, that reduction in risk is meaningless and it wouldn’t make me feel anymore confidant of coming out alive when forced to play Russian roulette. Not just that, I wouldn’t be tempted to play Russian roulette with a 100 shooter and a 1 in 100 chance of killing myself.
Here’s another example. Let us say that my risk of catching COVID when going to the supermarket is 1 in 10 (of course the real risk is 10-100 fold lower but 1 in 10 is a good number for illustrative purposes). Now let us say I wear a mask which reduces the risk of catching COVID when going to the supermarket to 1 in 11 (i.e. a 10% effect). In the real world what this effectively means is that in the former case I’m highly likely to catch COVID after 10 visits and in the latter after 11 visits. Bottom line: is such a reduction in risk useful and should it be mandated. I would say not because the effect is too small to consider meaningful, irrespective of statistical significance (even if the p value was less than 0.0001 and no matter what one’s prior Bayesian probability estimate was, especially when that prior is based on make-belief and wishful thinking and not hard data),
“…the Danish mask study, which was well done and sufficiently high powered, demonstrated that surgical masks offer NO significant protection to catching COVID.”
Do you mean Bundgaard et al ?
If “Yes” then as a study this was a dog’s breakfast.
The concerns about the methodology of the study are many. The main ones are:
1. For their sample size / statistical power calculation they assumed that mask wearing would reduce the risk of infection, to the mask wearer by 50%. Would you really expect this amount of reduction with a single NPI ?
It was underpowered. The prevalence was about 2% at the time this study was done (during Danish lockdown so social distancing etc. also in operation). In this situation you wouldn’t expect much disease transmission. They were setting themselves up to show “no difference” from the get go.
2. Participants tested their antibody levels at home and at this level of population prevalence (2%) there would have been a big problem with false positives. These would have been randomly distributed between the two groups and therefore would have biased the results towards showing no difference between the groups.
3. There wasn’t much difference between the 2 groups in terms of antibody prevalence (1.56% masks and 2.09% non-masks after correction for test characteristics). To find a real difference this small, they would have needed to recruit about 24,000 people, or 12,000 in each group.
4. Participants were asked to wear the mask outside the home. Apparently no instructions regarding mask wearing within a confined household (!?!) where there is now good evidence that plenty of transmission goes on.
Only 46% in the intervention group were fully adherent to the mask wearing protocol
What the study was designed to show, specifically, is that when a
country is in lockdown and/or under significant social distancing
restrictions, advising people to wear masks and providing them with
masks does not reduce THEIR risk of getting COVID-19 by more than half.
Some things that the study didn’t, and couldn’t, tell us:
– Whether mask wearing protects other people
– Whether masks prevent less than 50% of infections in the wearer
– Whether masks could be effective outside of lockdown situations
– Whether public health policy should include masks
A waste of time and money. Really bad science.
Lots of criticism of the Danish study. Not a word about this ‘study’?
Because I haven’t read it yet.
Primed however by Rex Douglass’s observations in three long Twitter threads looking at :
1) Estimands
2) Domain of those estimands
3) Strength of evidence (p-values, etc)
https://www.youtube.com/watch?v=AenwWw9jBCQ
You are aware, I hope, that the authors had to alter the wording of their conclusions in their preprint before they could get this paper published ?
In the end :
Some things that the study didn’t, and couldn’t, tell us:
– Whether mask wearing protects other people
– Whether masks prevent less than 50% of infections in the wearer
– Whether masks could be effective outside of lockdown situations
– Whether public health policy should include masks
“Yet, the findings were inconclusive and cannot definitively exclude a 46% reduction to a 23% increase in infection of mask wearers in such a setting.”
With a range like this you can safely assume that nothing meaningful can be gleaned from this study.
Sorry but your talking out the back of your head. The Danish study was done as well as one could possibly do such a study. Further, it just confirmed the results from innumerable RCTs done before COVID in the context of Influenza.
But here’s the thing. Irrespective of whether the Bangladeshi study was well done or poorly done (and frankly it looks as if it was poorly done), if we take their results at face value they find that masks reduce the transmission of COVID in the community by 9% with a p value of 0.05 which is considered just statistically significant (i.e. 1 in 20 probability that the effect is real). But here’s the thing. Even if the p value were 0.0000001, a10% effect is a nothing burger and what it really shows is that the impact of surgical/cloth masks is the community negligible. Given that, is it real worth disrupting so many social interactions, including those involving children, for a 10% effect. Personally I think not and I fail to see how anybody could think otherwise if they use any sort of logical thinking.
The Danish study design, knowingly or accidentally was set up to show no difference for the reasons stated above.
Whether or not it confirms studies done looking at the influenza virus is irrelevant – totally different virus, mode of transmission not clearly understood even today, different populations, different countries. A study has to stand or fall on its own merits.
If the Bangladesh study was poorly done then you shouldn’t waste your time getting your nickers in a twist about p values – bin it and read something else.
I think you miss the point. (a) Influenza and coronavirus are not all that different. Similar size, similar composition, similar physical properties, similar location. Hence mode of transmission will be the same as indeed it has found to be. (b) Irrespective of your criticism of the Danish study saying that it was set up to show no difference between masks and no masks, the fact of the matter is that if the effect had been large it would have been observed. It wasn’t. Ergo the effect is small in terms of protection which is what they were looking at. From the perspective of physics this makes sense, as it is far harder to protect the wearer than it is to block forward emission of viral particles carried in droplets. (c) Irrespective of the flaws in the design of the Bangladesh study, the fact is that even if statistically significant, the effect of masking on COVID spread through the community was found to be minimal – 9%. i.e. something of a nothing burger.
The issue is not whether the results are statistically significant but whether there is a meaningful difference in real world terms with and without masks. A statistically significant difference (p < 0.05) can still represent a tiny effect as in the present case – i.e. 9% – if the sample sizes are large enough. But anybody who thinks that a 9% reduction in COVID transmission/protection upon mask wearing is significant in practical terms is out of their minds. What the study actually shows is that masks make next to no meaningful difference irrespective of whether that small difference is statistically significant or not. In other words, in most situations outside very specific hospital situations where a doctor or nurse is examining a COVID patient up close, masks have basically zero impact and represent nothing but pure theater. It would be great if they actually really worked but they don’t. It is possible that N95 masks would be significantly more effective, but wearing a N95 mask for any prolonged period of time is next to impossible. Further, they are not cheap and should be used as disposables, and not as something that one wears on and off for a week.
A week??? I use mine seldom and for short periods of time, but they last me WAY more than a week.
If you only use it for a very short period of time, say for going into the supermarket for 10 min, that’s fine. But let’s say you were the checkout clerk in the supermarket, then you’d be wearing one all day. Further, if you notice, they begin to smell pretty quickly as they accumulate gunk.
I agree Johann. Before I retired I was an RN in critical care. We were fitted yearly for N95 masks in case the need arose for their use. If I remember correctly they used a smell test to make sure they were fitted very tightly and each mask was sized specifically for the wearer. I would hazard a guess that 99.99% plus masks that I see are worn incorrectly. Most people are clueless about how to wear a mask correctly.
Page 28 of the link – NO Difference in covid for the 50 and under groups, just above 50. Weird, that.
Also teams were both lecturing, texting, putting up signs, handing out masks personally, and had monitors going about haranguing anyone without one – in the Mask towns. This means a lot of other things potentially at work.
Best anti covid ever was in NK, where Kim Un made it illegal to carry covid and the penalty was death by execution. (he did this at the beginning). Talk about incentivizing. Maybe a double blind trial of this method needs to be tried. I mean, it is just another coercive ‘Nudging’ method. I think Australia is also thinking of similar as they have the military out enforcing masks in Parks and on the beach.
Again – this is the study link Chivers gave above
https://www.poverty-action.org/sites/default/files/publications/Mask_RCT____Symptomatic_Seropositivity_083121.pdf
Scroll to page 28 and see the charts – and what is Amazing, unless it is because of the bias agenda, which is reasonable to suppose, It shows 50 years old and younger have NO SIGNIFICANT difference in infection by Mask Wearing!!!!!!!!! But then drops that topic like a hot rock.
So WTF about schools and masks? Instead of Chivers going on about p-values why did he not bring that up?
Because he is not a journalist. He is simply a opinion writer.- biased one at that.
And he knows nothing about science . I don’t understand why he at Unherd, I’ve read many of his poorly argued pieces till when I stopped reading them. He would do well in BBC.
The best way to protect yourself from Covid is to be healthy, keep your lungs in good shape (with exercise) and don’t worry too much. The rest -masks, lockdowns, vaccines are all fafs – periphery discussion that do not do much to keep you from diseases of all sorts but are copiously peddled by those who are either in the business of keeping you sick ( unwittingly) or are lazy and want the government to solve their problems, their health, their futures.
I understand that doctors are well meaning but when does one ever say – I won’t treat you till you loose 10k? V v rarely.
For all those wanting to wear masks, pls … wear them . For the others – let us be. Stop giving us the irresponsible guilt trip BS. Even if scant evidence shows masks work in a small way , pls … help your selves. I would say – pls loose that 10k too – it would be more beneficial than the mask .
Did anybody check whether the people who wore masks and got sick became sicker than the ones who didn’t? In theory, the increased viral load you get by re-breathing in your own sick air when you catch covid should make you sicker than you would have been without the increased viral load. It would be really nice to get this theory tested, because it has huge policy implications, if true.
I suspect by the time you are infectious enough to re-breathe your own viral particles that the barn-door is properly open – i.e. that isn’t a factor.
I don’t understand what you mean. I thought it was pretty well established that people are most infectious before they show symptoms, for instance with this study. https://www.nature.com/articles/s41591-020-0869-5 (Whether this finding is also true for the Delta variant, I do not know.)
I’m not sure that that is really true. Seems very doubtful that you would have no symptoms and a massive viral load. Also, it would be almost impossible to test in the real world. But you can bet that a person exhibiting full blown respiratory symptoms is highly infectious.
But I _am_ sure, or as sure as I can be be about anything printed about covid. Surprised me too, and most researchers, which is why there have been a host of studies like this one https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2783099 measuring, not the viral load (which is what the other study I posted was about) but who got infected. The last sort of measure is a bit problematic, as you have to assume that people who got sick who had a sick relative at home caught it from the relative, and not on the bus instead, but this sort of guessing is part of what makes epidemiology so messy as compared to data modelling.
We now have more than a dozen studies of this sort, all of which indicate that people are most infectious 2-3 days before they show symptoms and in the first 2-3 days after showing any. Given that there is a usual order to covid symptoms https://www.frontiersin.org/articles/10.3389/fpubh.2020.00473/full which generally begin with fever, not coughing, (and which may never proceed to anything beyond fever, at that), I’d say the evidence that people walk around in masks, shedding lots of virus before they have any symptoms that they are sick is as good as it gets.
(ooops, seem to have pasted that paragraph in twice, sorry about that. removed duplicate.)
Impressive summary of p values and even an intro to Bayesian statistics.
So, to ask the obvious question, what was the p value for the treatment arm of the Bangladesh study? I’m guessing it was slightly less than 0.05 hence the controversy.
Study after study shows that the benefits of mask wearing are either tiny or nit existing at all. Yet people like you, Tom Chivers, keep advocating them and even mandating them. On the other hand study after study shows that covid outcome is significantly tied to obesity. Yet no one seems to be advocating exercise and a healthy diet.
If I have understood this correctly it is a study about promoting the wearing of masks not the incidence of the wearing of masks. The promotion helped reduce Covid, a bit, but it could have been that people did not like the promotion and stopped wearing them.
A rather more critical review
“bangladesh mask study: do not believe the hype”“this is one of the worst studies i’ve ever seen in any field. it proves nothing apart from the credulity of many mask advocates.”-el gato malo
https://boriquagato.substack.com/p/bangladesh-mask-study-do-not-believe
When our children were new-born, we put nappies on them, and the amazing thing was: Not only would they catch the poo, but the poo simply vanished! They kept their nappies on all the time, and there was never any issue.
I read those in favor of masking as having a bottom-line attitude of “Even if it only helps a tiny bit, it’s no real problem to wear a mask and my neighbor died of Covid so it’s not to much to ask so WEAR THE DAMN MASK!” This assumes that mask-wearing is no problem for adults.
We have no true back-and-forth productive discussions about these issues because those who have govt/institutional power and the media are almost never put in a situation where robust debate and accountability are present.
If we could have these vital, productive discussions, what I’d want an answer re masking is simply this:
Agree for the sake of argument that masks have some significant effect in transmission in adults or those younger people who are growing towards imminent puberty, say, over the age of 10-12. What study, what science, lead to the decision to mask toddlers? Who decided the cut-off age was TWO?? Have you ever dealt with a two-year-old? How are these tiny people with their tiny lungs any danger to the public, and how is their terrible level of mask-wearing compliance going to effectively stop anything?
Shouldn’t we begin with some baseline-sanity and at the very least start mask-wearing compliance with school-age kids who have seen their peer’s faces enough to develop crucial social skills? Can we start with the idea of not making this huge ask of children until they are in first grade?
Also important is the difference between statistical significance and practical significance. Take a large enough sample and you can usually get statistical significance but does it make a practical difference. Given the difference reported above, I would suggest not much.
The cranky Canadian Doc weighing in again.
The issue here is statistical significance vs. clinical significance.
Let’s say we accept that the answer to Question 1: “Do Masks actually work?” is yes, ie: masks actually did account for the ~10% lower rate of COVID. (To agree with this, one has to accept the study methods, that the increased social distancing was not the relevant factor, accept that the study was overall honest and fair, etc).
Even given this, there is a question 2: “… therefore should we mandate them everywhere based on this?”.
That is more complicated. A few questions spring to mind.
Is my rural area of NS with an average of 2 people per household, and a very diffuse population comparable to Bangladeshi villages? What about the area you live in? Would we not expect a lower utility to masks in less crowded/more sanitary environments?
If I told you that wearing a mask constantly would still leave you 90% as likely to catch COVID, would you wear a mask or not? I think most folks would say no. Some people don’t seem to mind them and would agree to wear them even if the risk reduction was more minimal. I personally wouldn’t.
What about environments where COVID is less rampant (either due to herd immunity, season, whatever)? Presumably in those environments the 10% relative risk reduction would be far less in terms of ABSOLUTE risk reduction (the actually relevant statistic). 10% reduction of a very small baseline risk is very small, so given that in Canada COVID numbers are very low, does it make sense for us to still mandate masks? It’s akin to mandating shark repellent in a freshwater lake.
Overall, it certainly doesn’t look like this study is a compelling argument for mask mandates given the small overall effect. It looks like a good argument for “Go ahead and wear a mask if it makes you feel safer, but leave me the hell alone please!”
isn’t Basyian thinking the problem, why cant a study be conducted to actually just show if the masks are effective or not. ? why is science so full of shit studies? to get it a way from masks, a flipped coin isn’t 50% of one face and 50% of the other. its flipped its 100% of one face and 0% the other. they work or they dont work,
Things are only that simple in extremely rare cases Most things are neither 100% nor 0% effective,but somewhere in between. Even if perfectly used, masks just reduce the likelihood of getting sick – you could get COVID from touching infected surfaces, or tiny droplets, or when you take off the mask to eat. Also, people might not get sick at all even if exposed, strong immune system, or funny air currents, or just luck. Any data set has lots of noise. Drugs, antibiotics, vaccinations, contraception, never work perfectly all the time, any more than masks do. Similarly, tobacco does cause lung cancer – but not every time. Even a bullet through the head is not 100% lethal – just close. In the real world we are unfortunately forced to work with these imperfections.
So you are saying Schrodinger cat may come out of the box live, but deathly sick, as being one of the 70% – 30% chances? I guess I see that.
The analogy isn’t really correct. Sure a given antibiotic doesn’t work 100% of the time but then one switches antibiotics to one that the infecting bacterium is sensitive to. Indeed, if one is a smart doctor, one would take a blood culture and throat swab before prescribing the 1st antibiotics so as to see what antibiotics the bacterium was actually sensitive to, and then change antibiotic if appropriate. The correct antibiotic will work 100% of the time or certainly very close except under very exceptional circumstances. If antibiotics only had a 10% effect, no matter how statistically significant, they would never be prescribed. Similarly, if the contraceptive pill or even the lowly condom only worked 10% of the time, nobody would bother with them. So the issue of regular surgical/cloth masks is not whether they prevent transmission of a virus-containing droplet or two but whether they have a significant impact on transmission. With hindsight it would appear, unfortunately, that the impact of masks is actually minimal and represents theater and a safety blanket.
The original idea, when it was believed that droplet transmission was the main mechanism of spread, is that masks would act as an almost 100% (or certainly greater than 80%) method of source control. Had that been the case, mandating masks in all public places would have brought the pandemic under control in the space of around 6 weeks (i.e. 2 weeks from the time of infection to development of symptoms, 2 weeks for the actual disease to either take its course or for the patient to die, and another 2 weeks to fully recover and no longer be a transmitter).
either a mask is a barrier which prevents particles of the virus passing through it or it isn’t, there’s no probabilities involved here. Can you walk through a brick wall? no because your too big to pass through any of the microscopic gaps that might be there.
whether the virus can pass through these mask is an answerable question with a definite answer.
We already know that the virus can pass through these masks. They are known to be worse than the n95 masks (because why the heck spend all the money on n95 masks if you could get the same result with a normal paper mask or a cloth one?) The n95 masks are better. What is an N-95 mask? Masks can be N (not resistant to oil), R (resistant to oil, can use for 8 hours) or P (oil-proof). The number means ‘what percentage of .3 micro size particles do not get through the barrier’. So an N-95 mask prevents 95% of the virus getting through (and you had better not rely on it in an oily environment). That means that 5% does get through. But how much gets through matters.
@ Glashan, Strauss.
Sure, masks do not work 100% of the time. But that is par for the course – for medicine. Chemotherapy does not work 100% of the time. Treatment or prevention for heart disease do not work 100% of the time. The various actions they have worked out for COVID patients (the best way to give oxygen, dexamethasone, …) do not work 100% of the time, and may save only some fraction of the patients. Hell, Ivermectin seems to work 0% of the time, for all we know, and people are still taking overdoses of it.
It is true that no one would bother with condoms if they were only 10% effective – but that is because you have the alternatives of abstinence or non-vaginal sex. If you have cancer, or a heart attack, or a pandemic raging around you, you cannot just choose to opt out, Here anything that gives a reliable 10% improvement will save a lot of lives.
It may be that masks will prove not to be worth the hassle, once we know enough. Plenty of good medical ideas that you would think ought to work have proved disappointing in practice, over the years. But there is no point in insisting on impossible prefection for masks any more than we do for chemotherapy.
Oh, and antibiotics would sure be prescribed even if they had only 10% effect – if the alternative was no treatment at all.
Following your last paragraph, would not all the public health authorities be recommending the use of ivermectin (coupled with Zn, Vit D, azythromycin/docycycline, and vit C) if it only had a 10% effect. Yet a large body of anectodal evidence as well as various meta-analyses indicate that ivermectin actually provides something like 80-85% protection when used as a prophylactic (as demonstrated in a recent randomized controlled study on health care workers in India).
A 10% effect for surgical/cloth mask wearing is basically as good as a zero % effect. It’s worthless when it comes to a community setting. Now, talk about a N95 mask (FFP2 or higher) and that’s a whole different ball game. But that is not what was advocated and what mask mandates have insisted upon is it.
I think it depends on which sources you trust. The last I read, Ivermectin had little or no evidence in favour except for the very first clinical trial – which has been retracted by the preprint server that published and is highly .likely on internal evidence to be fraudulent. I think I put that link elsewhere.
But it is an interesting point. Following the links in this article, you find the claim that the WHO etc. were against masks in the beginning because there had never been a randomised, double-blind controlled etc. study that proved they worked, and the medical profession is totally paranoid on this point, .Also such a study would be extremely hard to set up. This link: https://jamesheathers.medium.com/hurry-dont-rush-e1aee626e733 shows why – medicine has been burnt so many times by treatments that looked promising, got a lot of popular following,, and proved worthless in the end. The question is when it is justified to follow promising anecdotes and common sense and try a treatment that is still not proven. Personally I think that masks are a much better bet than ivermectin, though.
If masks are such a great bet, why is it that in every real world study whether with influenza or the current COVID pandemic they have not demonstrated any significant real world (as opposed to statistical) effect. Similarly if masks were so good why are the curves for daily cases per million and daily deaths per million identical for North and South Dakota although the former had mask mandates in place, while the later had absolutely nothing. The two states are right next door to one another and have very similar characteristics in terms of population density etc….. Likewise, if masks are so great, why wasn’t Florida a total disaster and the northeastern states a total success given that the former had no mask mandates or lockdowns, while the latter had extensive mask mandates and lockdowns.
As for medicine being burned and reliance on randomized clinical trials: You only need a randomized clinical trial to prove significance of something that has a small effect. If the effect is large, there is no need for either a RCT or statistics. Had the British and US insisted upon clinical trials for penicillin prior to deployment on the battlefield it is likely that WWII would have gone on a lot longer. Why because the effect of penicillin in the treatment of wound infections was black and white. Now if you then ask the question as to whether antibiotic A is better than antibiotic B, then you might need a RCT to prove the case.
Incidentally the same is true in science in general. One only needs to resort to statistics when the effect one is looking at is dubious.
Oh, statictical results are also part of the real world. Some of them are dubious, but some are not. OK, there is a famous dictum that “If your experiment needs statistics, you need a better experiment”. 😉 – but it is not true. There are a few effects that are so obvious that you just need to look at them, but they are a minority. Many more problems are tractable, but require careful analysis and, yes, statistics. Not least because a lot of problems have complex causes, and you do not get an obvious result unless you do something about several of them at once. Penicillin is actually an exceptional case, where the result was an immediate cure for people who were otherwise pretty sure to die. If you limited yourself to the most obvious cases you would miss a lot of improvements.
A couple of things here. I don’t know whether you are a practicing academic scientist in the hard sciences or not, but I am. And the dictum you quote that “if your experiment needs statistics, you need a better experiment” is absolutely spot on and true.
As for penicillin you might recall that the initial results were actually disappointing because the antibiotic wasn’t administered long enough and immediate relapse occurred. But the individuals involved persisted, realized what might be happening and fixed the problem by administering penicillin for 7 days.
And on another note regarding penicillin, the mechanism of action at the time was completely unknown. Today, my academic medical colleagues living in ivory towers always need a mechanism of action if they are to believe a particular drug is useful. But that’s just plain silly. If something works, go for it, and there is always plenty off time later to figure out how it works.
And by the way your statement regarding antibiotic use is not quite true either. If one developed a new antibiotic and it only was 10% better than an existing well-established antibiotic with a very well established safety profile, it is unlikely that the FDA would approve the new antibiotic unless it displayed properties that were quite distinct from the existing one (e.g. if it was active against bacteria that were resistant to the original).