Recent review of the studies on using masks to prevent the transmission of Covid-19.

PeterNSteinmetz

Administrator
Staff member
Here is a recent review in the New England Journal of Medicine of the studies on using masks to prevent the transmission of Covid-19. I think they did a fairly good job of summarizing the strengths and weaknesses of the evidence and the tradeoffs in terms of public policy.

https://www.acpjournals.org/doi/10.7326/M20-6625

As they note, there is only one randomized study which showed no significant decrease in infection rate from wearing surgical masks for the wearer. The evidence for source control, that is, an effect on preventing spread to others, is either based on observational studies (which can have many confounds) or in-vitro studies of droplet dynamics (which do not account for human behavior in wearing masks).

As also noted by the authors, it is almost impossible to do a proper randomized study on source control. To the extent that is true, it implies that the source control justification for masks is essentially a non-falsifiable hypothesis.

Overall, the authors believe that the potential benefits outweigh the harms in terms of their value judgments. In terms of the controversy over mask mandates, they also note that "However, mask mandates involve a tradeoff with personal freedom, so such policies should be pursued only if the threat is substantial and mitigation of spread cannot be achieved through other means."

Thus, in terms of the scientific question, mixed evidence which may not be resolvable in the near future.

The policy question of course is, are people justified in forcing others to wear masks through executive orders and laws, given such mixed evidence? Not a question I will comment on here, but that is the political or policy question.

(My understanding of the current PoA rules is that this type of scientific post on Covid-19 is permitted given the current pandemic; but if I am mistaken, I trust the MC will delete.)
 
Llewtrah381 said:
First key point from the article:

“Masks and face coverings, if widely worn, may substantially reduce the spread of COVID-19.”
Key word “may”. I think they covered the evidence for and against a source control effect in the article. Worth a read.

Another key point for those who don’t read the whole article “Randomized trials are sparse and have not addressed the question of source control.”

It is a mixed bag of evidence at this point. And since it may not be possible to do a proper randomized study, the hypothesis of source control becomes almost non-falsifiable.
 
That is their value judgement, as both the authors and I noted. Others may reasonably disagree with that value judgement. Particularly when the evidence is mixed and sparse.

Key words in that quoted statement “suggests” and “likely”. Both of these are used in scientific writing to note that this is not a clear-cut conclusion.

The important policy question is whether it is right to force people based on that kind of evidence. That is a political question likely best left out of this thread.
 
asicer said:
Was this stuff not studied 102 years ago?
The effects which are being observed are fairly small. A few percent reduction in the R0 in some observational studies for example.

While such an effect on R0 does add up over time, if true, there are 2 issues:

It is hard to detect and the statistical analyses that found that effect basically require computers.

Small effects like that can more likely be due to some uncontrolled for confound, thus making one’s theoretical interpretation of their cause more tenuous.
 
Cap'n Jack said:
It's part of a summary of the rest of the paper.
Correct. It is important to realize that when people write discussions, that is what they are, discussions.

I always find that instead of quibbling over the meaning of some word in the the summary or discussion, it is more productive to actually read the whole paper. It gives a much better sense of what the authors are really trying to say.
 
Cap'n Jack said:
Yes, for health care workers wearing N-95s, the evidence seems fairly clear there is about a 95% reduction in the likelihood of them being infected.

Big question for the public health issues, however, is how well do cloth masks work, and when worn by the general public. The Danmask-19 study shows fairly convincingly there is nothing like that large an effect for the general public wearing surgical masks. There was no significant effect for the wearer and if interpreted as an estimate there might be a 15% reduction in infection rate but also about a 25% chance of an increase in the infection rate.

So the issue of source control, as noted in this review, still up in the air scientifically.
 
Cap'n Jack said:
The second link covered both types of masks, as I understood it.
I see that the second link refers to a narrative article which appears to rely heavily on the recent WHO meta-analysis. It does not appear on first scan to refer to independent analyses of data from 1918 (but please correct me if I missed it in a brief scan).

If you read that whole WHO review they break down by mask type. The N95s resulted in about a 95% reduction. 12-16 layer cloth masks resulted in about an 85% reduction, based on some unavailable reports out of China. IIRC. The 14th root of 0.85 is around 0.988, corresponding to about a 1% reduction for a single layer. If they are assumed independent as a sort of rough estimate.

A lot of reviews have thrown together all masks types, which is unfortunate as it does not inform the public policy question of the general public wearing single layer cloth masks in many cases very well.
 
Llewtrah381 said:
One thing the article didn’t discuss but which I subjectively believe is a real possibility is that the extensive use of masks has generally reduced the number of viral particles people get when exposed, which has resulted in a decrease in the overall morbidity and mortality. I’ve seen that discussed somewhere in the medical literature but can’t find it right now.
I will have to respectfully disagree about whether the article has addressed that. What is said in this post is essentially the source control hypothesis. Namely, that wearing a mask reduces the likelihood that others will be infected.

The authors of this narrative review discuss this at some length. They conclude essentially that the evidence in this is “sparse”.

I agree with the authors on this point. I also agree that it is difficult to imagine how to ethically conduct a randomized trial on this question. To the extent that is not possible, it renders the source control hypothesis essentially non-falsifiable.

There is in-vitro evidence as well as observational evidence to suggest that the source control hypothesis might be true. But the randomized trials argue against it. So a mixed set of evidence. Which is essentially what the authors note.
 
X3 Skier said:
These type of things always have zero practical effect to me . If the rule is, wear a mask or don’t enter the location, wear it or don’t enter.

Discussions on POA about Masks can be entertaining or even sometimes illuminating. Change my actions, nope.

Cheers
For private locations, I largely agree, except when a coercive mask mandate is in place.

Then the issue becomes a more political one of whether a group of people, namely the the government, are justified in forcing people to wear a mask.

Of course the point is well taken: if you are going to violate the law, choose carefully when you do so.
 
Cap'n Jack said:
Nope, the second link was just as I stated, a review with better numbers.
Ok, I am sorry but I don’t see that. They appear to rely almost entirely on the recent WHO meta-analysis, which distinguishes in its results and tables between the health care workers wearing N-95s and cloth masks. Please cite to the page and reference in link number #2 if different.

ETA: just double-checked and I see no independent results or analysis. This appears to be nearly an opinion piece by the authors.

A note about the types scientific literature. A primary article normally presents new data, analyses, and numbers. A review normally cited the primary literature and summarizes , but does not present new analyses. A systematic review might present new numbers in the form of a meta analysis of existing literature. An opinion piece will present the authors opinions, sometimes with a small new analysis but mostly synthesizing existing literature.
 
Let'sgoflying! said:
"They might help, so I am happy to do this for you. Besides, it's not that difficult for me to wear a mask."

"Condoms are not 100% effective either, but they are still a damned good idea, right?"
That is actually an interesting example to consider. Compare the effectiveness of condoms at preventing pregnancy with the possible benefit of the general public wearing cloth masks. I think you will find there is a rather substantial difference.
 
Llewtrah381 said:
I think you miss my point in the paragraph of mine you respond to. Me wearing a mask reduces the number of virus particles I am exposed to - not others. The fewer I’m exposed to, the less severe MY disease is the hypothesis I’m subjectively pointing out. The other side of the coin of the “source control hypothesis” you mention.
Ah sorry if I misunderstood. The Danmask-19 study is the only randomized study addressing this question. It found NO significant difference in the rate of infection of the general public between the group recommended to wear masks and reporting doing so and the group not so recommended.

As a point estimate, the data showed a non significant 15% reduction in infection rate and a 25% chance of an increase in infection rate. Commentators have noted that with respect to no other treatment would medical professionals recommend a treatment to patients given such data. Perhaps have a look at the study itself - https://www.acpjournals.org/doi/10.7326/M20-6817.
 
Llewtrah381 said:
RATE of infection is not the same as SEVERITY of infection. My, again subjective, premise is that being exposed to, say, 100 virus particles because a mask blocked 900 others is going to give me a less severe infection than being exposed to all 1,000 particles. I may indeed get an infection with the 100 but it will likely be less severe than if I was exposed to 1000 because my body can hopefully better manage the 100. This is all simplistic analogies to help others understand the premise (and not at all meaning to sound condescending with that). Getting exposed to only 100 sparks from a sparkler is less likely to lead to a severe fire than being exposed to 1,000 sparks. Both may cause “a fire” but one is typically easier to put out than the other.
That is an interesting hypothesis. I suppose the Danmask study data could be used to address this. I am not aware of any data to say it is true from randomized trials otherwise.

While this hypothesis has a certain plausibility, in terms of what we know from the correlation of viral load and disease severity, it strikes me that the small possible effect of reducing viral load with single or double layer cloth masks, perhaps a few percent based on the 12-16 layer mask data which the WHO cites, might be lost in the noise of other effects.

Nonetheless, I agree with the authors of the review, and I imagine your judgement as well, that N95 or better masks are well advised in situations where infection is much more likely, such as enclosed spaces or in crowds. Nowadays disposable N-95s or better are becoming more available. (And my institute has recently made available the 3D printing files to produce an adaptor which allows one to use a readily available N-95 in the 3M 6200 respirator with several layers of cloth for just such purpose. PM me for the files if interested.)
 
Cap'n Jack said:
I realize you may not be able to get a good answer to this, but why do you think everyone has been wearing masks?
I suspect that this is because people really do want to think they can do something to help. Who wants to helpless?

And many authorities have been telling them it will. It is much more comforting to think it might help than to acknowledge that this pandemic will likely run its course largely independently of what humans do.

As for me, I merely wear one in private establishments such as restaurants because I am not choosing to make a protest at the time. In public outdoor areas I invite them to try and cite me. Happy to see them in court.

In other private locations, I obey the requests of the owners and if I think it is a big risk situation, like an airplane cockpit, tend to wear an N95.
 
RyanB said:
My opinion is - I don’t understand why mask wearing has become so hostile and divisive. So many people seem to like going against the grain for the sake of making a statement and just being outright difficult. It hurts nobody and can only help. Proper mask wearing DOES reduce the spread of airborne viruses and there’s plenty of data to support it.
On the last bit I will I have to disagree and I think that was sort of the point of this review. Considered objectively overall, the data to support cloth mask wearing by the general public is rather mixed.

And indeed those authors seem to have ignored MacIntyre et al 2016 which showed that health care workers wearing masks had a higher rate of infection. I don’t view that as determinative, but it is one part of the data that says the case is not so clear.
 
Cap'n Jack said:
I don't know what you are saying with that response. People wear a mask when they have to, in a store, maybe, and beyond that, no one really knows how much the population masks up outside of legal compulsion.
There is some survey data by the New York Times on the level of compliance, if that is what you are driving at.

My non randomized observation is that there is a higher level of compliance than say back in April and that the level of compliance where legally required, such as in stores here in AZ, is quite high.
 
Llewtrah381 said:
I know I don’t need to wear a mask to protect myself or others but doing so helps others remember the communal importance of doing so, in the absence of anything better to do to stop this illness right now, until immunization is rolled out broadly.
Thanks, that is an honest take on the situation. I have a somewhat different attitude in that I think our rights in general are being grossly eroded by our present government. So in this particular case, and given the over politicization of this issue and distortion
of the facts for political purposes on all sides, I think it important for me to speak out objectively about the facts.
 
Cap'n Jack said:
Or what you believe are the facts. I disagree about the erosion of rights. People were saying the same sort of stuff in 1918.
So you don’t think there are sort of outrageous distortions on both sides of the issue?


I was referring to the erosion of our rights in general, not just WRT COVID-19. However, that general phenomenon does influence my choices WRT mask mandates for COVID-19.
 
It does depend on where you live. Some places are much more restrictive than AZ and have a bigger issue, I agree. I feel bad for them.

There is almost no good evidence that lockdowns reduce mortality overall, at least here in the US. So that seems outrageous to me given the huge costs of such things. But that is of course a whole other subject.
 
Back
Top