Danish study shows no significant reduction of Covid-19 in wearers of surgical masks

[PeterNSteinmetz]

Danish study finally out. NO significant difference in infection rate between those wearing surgical masks and those who do not.

"Our results suggest that the recommendation to wear a surgical mask when outside the home among others did not reduce, at conventional levels of statistical significance, the incidence of SARS-CoV-2 infection in mask wearers in a setting where social distancing and other public health measures were in effect, mask recommendations were not among those measures, and community use of masks was uncommon."

This is rather consistent with the prior randomized trials on masks and the seasonal flu as well.

https://www.acpjournals.org/doi/10.7326/M20-6817

 

[PeterNSteinmetz]

I ask that you either correct your thread title and the statement, or delete the thread, as you are spreading fake news.
The context is EXTREMELY important, which is outdoors and in addition to other measures, such as social distancing.

Intervention:
"Encouragement to follow social distancing measures for coronavirus disease 2019, plus either no mask recommendation or a recommendation to wear a mask when outside the home."

Limitation:
Inconclusive results, missing data, variable adherence, patient-reported findings on home tests, no blinding, and no assessment of whether masks could decrease disease transmission from mask wearers to others.

Conclusion:
The recommendation to wear surgical masks to supplement other public health measures did not reduce the SARS-CoV-2 infection rate among wearers by more than 50% in a community with modest infection rates, some degree of social distancing, and uncommon general mask use. The data were compatible with lesser degrees of self-protection.

What do you is "fake news"? The study showed no significant effect of the intervention of people wearing surgical masks.

 

[PeterNSteinmetz]

Here is one of the main quotes from the paper, as above,

"Our results suggest that the recommendation to wear a surgical mask when outside the home among others did not reduce, at conventional levels of statistical significance, the incidence of SARS-CoV-2 infection in mask wearers in a setting where social distancing and other public health measures were in effect, mask recommendations were not among those measures, and community use of masks was uncommon"

My title was "Danish study shows no significant reduction in Covid-19 with surgical mask wearing".

The quote you cite was proceeded by "Reduction in release of virus from infected persons into the environment may be the mechanism for mitigation of transmission in communities where mask use is common or mandated, as noted in observational studies. Thus, "

I would read those statements in context as saying that the study does not really say anything about source control. Which is true. But that lack of data about source control is not evidence that source control works. I am not aware of any actual clinical data or study that says that source control works. Are you? If so, please cite. It is a theory. And one that I frankly find improbable. Here is why.

The likelihood of being infected by Covid-19 is likely related to the number of virions you are exposed to. Why should it matter whether a, for example, 90% reduction in virion load is achieved at the source or the the person exposed? And I don't think there is any evidence that the masks are somehow a rectifier. This is a popular theory, I agree, but I am not aware of any real evidence for it. And frankly, measuring an effect of source control would be rather difficult.

So I don't think that the primary result of the study should somehow be watered down in a brief title summary to reflect a theory for which there is no good evidence.

But in a spirit of improved precision I changed the title to "Danish study shows no significant reduction of Covid-19 in wearers of surgical masks".

 

[PeterNSteinmetz]

Not extrapolating a study beyond its scope is not equivalent to watering it down.

I agree. But I don't think either the original title or the original post was extrapolating in any way. Where did it extrapolate?

Again here is the original title "Danish study shows no significant reduction in Covid-19 with surgical mask wearing".

If we parse that very finely, a "reduction in Covid-19 with surgical mask wearing" would quite normally be construed to apply to the people or persons wearing the masks. Another example would be a reported "increase in lifespan with coffee drinking". I think normally one would interpret that as meaning the people who drink coffee live longer.

The only reason to think it was extrapolating to the general population would be if you thought that source control worked for masks and that somehow this argued against it. But neither my title nor my original post said anything explicitly about source control, did they?

Perhaps those who believe strongly that source control works might read that into it, but that would be reading more into it than was stated.

While I can see the point that some might read it that way, I think it is rather unfair to try and characterize the original title as "horribly misleading" and then support that assertion by quoting a phrase out of context and chopping off the transitional "Thus," which indicated that more context was required.

 

[PeterNSteinmetz]

This is a good review of the subject:
https://www.nytimes.com/article/coronavirus-masks.html

Jon

I think it best to stick to the scientific reviews, rather than these secondary or tertiary media reports.

I have listed 4 of the recent ones on my medical interest page at http://steinmetz.org/peter/Medical.

Overall, prior to this study Inwould say it was fair to say they were evenly divided on the question of whether masks mandates were a good idea. With this study, I think the evidence now weighs on the side of NOT wise.

 

[PeterNSteinmetz]

So I decided to sit down and parse out further why some posters here would think the original title on this thread was misleading. Many of you may not be interested in such fine linguistic parsing, so may not want to read the rest of this. But I hope it might shed some light on the subject. And I hope my interest in this is understandable given that I am the son of a linguist (a complete aside, but it still brings a smile to me to think of how my late father delighted in asking young children what the 3 or more meanings of the sentence "I saw a man eating shark" are

So the original title of this thread was "Danish study shows no significant reduction in Covid-19 with surgical mask wearing". I think whether this is misleading can be better understood by substituting some more neutral details about what the claim that the study said is. Consider the following possible titles, both of which are literally true statements about the study by Bundgaard et al:

1. Danish study shows no significant reduction in headaches with aspirin use.

2. Danish study shows no significant reduction in earthquakes with peanut butter consumption.

#1 makes a true statement, the study has nothing to do with headaches or aspirin us and so shows no reduction in same. At the same time, it is well established that aspirin use does decrease headaches. So someone reading #1 might assume that the author was trying to imply that the statement that aspirin use reduces headaches is false, even though that is not what was literally said.

#2 also makes a true statement, the study has nothing to do with earthquakes or peanut butter consumption. Since this latter assertion is fairly clearly false, most people would not assume the author of the title was trying to mislead about the contents of the study. They might think it was some kind of headline from Mad magazine.

In the case of the actual original title, the whole point here is that the evidence for the claim that masks prevent the spread of Covid-19 is not terribly convincing. It is likely best described as mixed prior to this paper. So from that perspective, there is no attempt to mislead at all. Rather the title makes a true statement which points to an important set of considerations about the evidence.

If one assumes it is well established that mask wearing by the general public slows the spread of Covid-19, then one might be inclined to make the assumption that the title is somehow misleading, as per example #1. But that is an assumption and then begs the real underlying question about the strength of the evidence for the efficacy of such mask wearing.

As to the 2nd and current title, I think it is correct and much less subject to such misinterpretation. The study says that there was no significant effect of reducing infections in those wearing surgical masks.

 

[PeterNSteinmetz]

You are trying to make people believe that masks do not do anything under all circumstances, which is FALSE.

As to this assertion that I am trying to "make people believe that masks do not do anything under all circumstances" -- I am sorry that is a complete straw-man and I never stated that. To expand a bit on the actual evidence regarding the efficacy of mask wearing in preventing people from being infected by Covid-19:

The results of the WHO metaanalysis are fairly clear that healthcare workers wearing N95s are about 95% less likely to become infected. It also reported that healthcare workers wearing 12-16 layer cloth masks may have had an 78% reduction (though these were in Chinese and not presently available online).

But these results are a far cry from the situation with the general public wearing cloth masks, where the evidence is much more mixed because there are a lot of other factors in play. Please do critically read the reviews posted on my medical interest page at http://steinmetz.org/peter/Medical/. I think you will see the evidence was fairly evenly divided - prior to this recent study. If after doing so, you want to parse in detail the meaning of the studies underlying those reviews, I am happy to do so. There is evidence on both sides, though I think with this more recent study it weighs toward mask mandates not being advisable.

 

[PeterNSteinmetz]

You stated that this study provided evidence that mask mandates are unwise:



Using this study to support your conclusion that mandatory mask wearing is unwise, again, is a perversion of logic and an irrational extrapolation of the data.

"Thus, these findings do not provide data on the effectiveness of widespread mask wearing in the community in reducing SARS-CoV-2 infections."

I disagree. Please see my response to MuseChaser’s query.

 

[PeterNSteinmetz]

Peter,

Most people can accept the fact that there's little scientific data showing that wearing a typical cloth or disposable mask protects the wearer, and the studies you cited in this thread support that. Can you address the efficacy of masks in terms of preventing the spread of Covid FROM the wearer to the people with whom they have contact inside of social distancing guidelines (or should I say "outside of social distancing guidelines?" ... your linguistic background has me all paranoid now..... grin... )? That seems to be the source of antipathy towards your original post.

Thanks for your thoughts.

Good question and I think goes to the heart of the scientific matter as well as why some posters have misinterpreted the original title to be misleading.

So the question is essentially - is there good scientific evidence to support the use of masks by the general public to prevent others from being infected?

There are no good randomized controlled studies of this at this time. They would be difficult to successfully implement.

There are in-vitro reasons to believe it should work, such as masks decreasing expelled droplets.

There is one observational study, with some serious issues in my opinion, on the effect of mask mandates on Covid-19 case rates. It showed a small effect in decreasing cases rates.

There are the international studies, which are really subject to a lot of cautions in interpretation.

The rest of the data is less direct. For example, there is data that when you put a mask mandate in place, people spend more time outside the home, often in risky areas.

Put all of this together, as the scientific reviews listed on my page attempt to do, and you end up with a pretty mixed set of evidence, about 50/50.

As to the meaning of the Bundgaard study. I do not think it is completely irrelevant to this question, though did not directly address it, as noted by the authors.

Here is why. For the results of Bundgaard et al to be true and for there simultaneously to be a strong source control effect of mask wearing would require the masks to someone have an asymmetric effect. But we know that the likelihood of people being infected is related to the total number of virions they are inhaling or ingesting. And there is no evidence that the masks are somehow rectifiers, blocking virions in only one direction.

So how could that be one way? Just imagine two people in a room, one infected and the other not. Why would it matter who wore the mask?

This seems physically improbable to me so I think that indirectly implies that if Bundgaard’s results are true, they argue against the source control effect. That is indirect evidence, but I certainly don’t think that Bundgaard’s results argue FOR source control.

In terms of whether mask mandates are wise, this involves a variety of other political and sociological considerations, but again, I don’t think the Bundgaard results can be said to argue IN FAVOR of mask mandates. Rather they add to a growing set of evidence to think they are unwise.

 

[PeterNSteinmetz]

So I flipped a coin once and it landed on heads.
PeterNSteinmetz's conclusion: It must be a two-headed coin.

I don’t know I would draw that conclusion. But I would say that it is far more likely to be a two-headed coin than a two-tailed one and the probability of it being two-headed, assuming some reasonable prior probabilities of it being two-sided, two-tailed, or fair, just went up!

 

[PeterNSteinmetz]

Peter, not only are you spreading false information, you are defending your actions.
Since science does not work for you, here is a metaphor:
Let's say a person named "Joe" is given a drug, as part of a trial.
Under this drug, Joe is given an intelligence test and his score is "imbecile".
Do you think Joe would like to see a report that says "Joe is an imbecile", without the context?
Maybe now you get it?

Sorry that metaphor is inaccurate. You keep repeating that I am spreading false information. Please quote the false statements and cite to studies showing they are false or stop simply repeating your assertion. That is also a fallacy.

 

[PeterNSteinmetz]

If both people wore masks, then you would multiply the effectiveness. If the first mask caught only 50% of the particles, the second mask also catches 50% of the particles, you only have 25% as much transmission as opposed to the 50% with one mask. If the masks each stop 80%, than two masks result in 5 times lower particle transmission.

That is true and a good observation. But what I am interested in in this discussion with MuseChaser is whether there is any evidence that the effect is asymmetric. I think that is interesting because that would need to be true to some large extent for the results in the Danish study to be true and for there to be a significant source protection effect in the general population.

 

[PeterNSteinmetz]

Not being advisable or just not necessarily being effective? IOW are masks actually said to be pernicious? (No, I'm too lazy to read the studies.)

I meant that a public mask mandate is likely not advisable. That is for a variety of reasons. They are discussed in further detail in the editorial which is on my medical interest page. Probably should leave it at that generally as it borders on political tradeoffs.

From a purely medical perspective, there is some evidence that health care workers wearing cloth masks are more likely to catch respiratory illnesses (MacIntyre et al 2015 doi:10.1136/ bmjopen-2014-006577).

 

[PeterNSteinmetz]

If we had enough N95 masks for everyone to wear them for just 2 weeks back in April, we may have been able to control at least the first wave.

I understand that is reasonable speculation. And it is also true that it might have just made the later number of cases larger. Hard to say. My speculation is that this will run its course in one or two seasons regardless of any interventions short of vaccination or the natural achievement of herd immunity. These viruses are highly evolved to spread and under considerable evolutionary pressure to find a way.

But the really core question regarding mask mandates is what is the evidence the a mask mandate which orders the general public to wear some kind of face covering will help slow the spread of Covid-19. If the majority will wear cloth masks, due to unavailability and comfort, and cloth masks cause an increase in respiratory infections, I imagine you can see how it might not help the situation even from an infectious disease or mortality perspective.

 

[PeterNSteinmetz]

the Oxford study concluded that where distancing wasn't possible, home-made masks reduced the risk of transmission by 4× if just the infected person was wearing a mask, and 6× if both the infected person and the uninfected person was wearing a mask.

This is one of the studies that needs to considered objectively in the context of all the other studies out there. There are over 100 now. The scientific reviews try to do this. And as summarized on my medical interest page, they are overall about 50/50 on the efficacy of the general public being told to wear masks to slow the spread of Covid-19. And that is prior to the recent Danish study which I started this thread with.

If you want to objectively read those reviews, then it might be productive to discuss and possibly look at individual articles in that context. Without doing so, this is just picking and choosing specific studies to support a particular point.

There is lots of evidence on both sides of this issue, the trick in science is objectively evaluating the totality of the evidence in light of each items strengths and weaknesses. Doing that for over 100 articles is a lot of work and not likely the sort of thing most posters here are inclined to do. That is why I suggest first objectively reading the reviews.

 

[PeterNSteinmetz]

That was a legitimate concern, but the experience in most jurisdictions has been that indoor public mask mandates have reduced community transmission significantly, and as that's become clear, more and more public health authorities have come on board.

In order to have a reasonable objective discussion about the results of any study you might cite to support this claim about causing reduced community transmission, it really is necessary to read the reviews and consider such studies in the light of all available evidence. I take it you haven't done that yet.

 

[PeterNSteinmetz]

But does this show that people in this country are unwilling to follow mask mandates, or that they don’t work? Does any of the data in the studies show that the mask mandates were actually followed by a large enough percentage of the population to be successful, or if not did that skew the results?

The reviews deal with the totality of the data, including the experiences in Hong Kong, etc. Perhaps read them objectively. As I noted above, without that, picking and choosing this study or another to support a particular point is not the way to reach valid scientific conclusions.

 

[PeterNSteinmetz]

Thanks for the reply, Peter. As a complete layperson, I'd postulate the answer to that question may lie in the amount of space the virus would have to travel from the person expelling the virus to the mask. If the infected person was wearing the mask, the virus' ability to bridge the distance to the uninfected non-mask-wearing person would be strongly compromised by the mask, no? Conversely, if the infected person was NOT wearing the mask, the virus would have a much greater chance of reaching the non-infected person, and bypassing the little-to-no protection that mask offers to the wearer in terms of self-protection, since the virus could land anywhere on that person and be transmitted internally by habitual, unintentional hand-to-eye, -nose, or -mouth contact, on top of the possibility of being drawn through the mask via inhalation. Does that make sense?

Yes, what you say makes sense to me, but to analyze this further I think we should breakdown the different routes of transmission, as they are not all equally likely.

So from the infected person (I) to the non-infected person (NI) there are at least 3 routes that I can think of right away.
1. Large drops such as those expectorated when singing loudly, yelling, etc.
2. Small respiratory droplets which tend to hang around in the air.
3. Direct contact transfer.

To get some rough idea of whether things would be asymmetric in this 2 people in a room scenario, we will need to account for how masks might interrupt each route of transmission as well as how probable those different routes are.

For each route:

1. A mask on I will definitely reduce the drops that get expelled toward NI. So how many of those would really end up affecting NI absent the mask? Presumably mostly those that land on their respiratory surfaces (RS), but some that land elsewhere on the body could be subsequently picked up by NI and then passed to their RS. The drops landing directly on RS would be interrupted by a mask on NI. It seems like there might be some asymmetry in this route of transmission, but that would depend on the relative efficacy of inhaling the drops versus touching your clothing or body where a drop landed. My understanding is that this large drop mode of transmission from person to person is considerably less likely than #2.

2. A mask on I or NI will likely decrease the amount of respiratory droplets landing directly on the RS of NI equally. The same factor of reduction from the mask would be present in either case. Some small fraction of respiratory droplets presumably land on NI and might be passed to their RS via contact. Thus this mode is likely to mostly symmetric. My understanding is that the inhalation of respiratory droplets is the primary mechanism of transmission of Covid-19.

3. Masks would have no likely effect on direct contact with other surfaces or the other party. Wearing a mask by NI might decrease their likelihood of transferring virions from their hands onto their RS, but this would depend on handling of the mask. Wearing a mask by I would not affect NI's chances of passing virions from their hands onto their RS.

So based on this simple analysis I would suggest that it is highly likely the effect of mask wearing in this 2 person room scenario would be mostly symmetric. The degree of asymmetry would be quite slight.

As you can see, it is probably difficult to predict this accurately based on in-vitro studies. Direct measurement would be best. I have been thinking a bit about how one might be able to do that with some models.

Now I have not analyzed this or the literature on this deeply, mostly because I was waiting for the results of the Danish study to spend further time on this issue. But now that that is out and makes this a more pertinent issue, I will probably dig further into this.

 

[PeterNSteinmetz]

So why did you pick and choose one specific study to start your thread with?

The ineffectiveness of masks and social distancing can be seen in the way that every country is seeing the same rates of spread of COVID...oh wait, no, the US turns out to be an enormous viral dumpster fire compared to the rest of the developed world. I wonder why.

There are several "smart on paper" scientists arguing that climate change doesn't exist, or that vaccines cause autism, etc etc. Doesn't mean we have to agree with them or ignore the overwhelming majority who argue otherwise.

Because it is a new study that is not included in those prior reviews and should be added to the overall evidence. I am happy to discuss the new Danish study for that reason, but it does have to be also integrated with the totality of the other literature.

It is truly not the way to get to the scientific truth of this to continue citing this example or that one study and ignore everything else. Read the reviews if you actually want to understand. Otherwise it is just a form of entertainment for people here on PoA.

It is also not the way to get to the truth to treat scientific questions like some sort of opinion poll and worry about what “the majority” of scientists think. That is just the argument from authority. One needs to objectively evaluate the evidence and data, considered in its totality, especially with these types of questions where there is a lot of evidence on both sides of the issue.

 

[PeterNSteinmetz]

So now that there is that new study I posted from the weekly CDC reviews (https://www.cdc.gov/mmwr/volumes/69/wr/mm6947e2.htm), that is the newest study and it states that mask mandates are effective control measures. Shouldn't you change the title to reflect the latest study? Just like you titled this thread - not to reflect the 50/50 split that you have seen in studies, but to spin the topic to one side vs. the other. How about "Countywide mask mandates appear to have contributed to the mitigation of COVID-19 transmission in mandated counties". That is only fair. Change the thread title.

Is that a peer-reviewed publication? No.

After you have read the reviews and objectively considered them it might be interesting to discuss new studies, and perhaps even some items like this that are from reputable sources, but not peer-reviewed. Until then, not.