Mask mandate and use efficacy for Covid-19 containment in US States
escipub.com
escipub.com
> Higher self-reported mask use protected against SARS-CoV-1 in Beijing residents [30], but RCTs found no differences in PCR confirmed influenza among Hong Kong households assigned to hand hygiene with or without masks [mask use 31% and 49%, respectively] [31]. Medical and cloth masks did not reduce viral respiratory infections among clinicians in Vietnam [9] or China [10], and rhinovirus transmission increased among universally masked Hong Kong students and teachers in 2020 compared with prior years [32]. These findings are consistent with a 2020 CDC meta-analysis [33] and a 2020 Cochrane review update.
Care to cite the literature you're reading that contradicts these statements? I'm genuinely interested in learning more about the state of scientific consensus on this topic.
I was recently reading more about Ignàc Semmelweis, the doctor famous for promoting hand washing. He was finding that hand washing had almost no impact on the outcomes in a particular hospital. He eventually discovered that they needed to wash the sheets better too. Either intervention alone would have almost no impact, but together they were powerful.
Source: https://www.thelancet.com/journals/lancet/article/PIIS0140-6...
> We did not observe association between mask mandates or use and reduced COVID-19 spread in US states. COVID-19 mitigation requires further research and use of existing efficacious strategies, most notably vaccination.
Googling turns up lots of articles like this:
* https://journals.plos.org/plosone/article?id=10.1371/journal...
* https://www.cdc.gov/mmwr/volumes/70/wr/mm7010e3.htm
The statements made seem compatible, but I think reading just this article you could come away thinking mask mandates are a bad idea. They might be, but a quick google does make that seem more doubtful. As another commenter pointed out the impact is probably pretty small though and easily swamped by other factors. Who is better controlling for those factors isn't something that's clear to me.
Covid also isn't the flu so I'm not sure how relevant masks effectiveness against the flu is to covid. I think influenza has a lot more surface transmission than covid for example
I took a look at the contents of the report, and it is denser than most. I was looking for some idea of the magnitude of difference that you are seeing, and wasn't sure that I was looking in the right place.
Beyond that I wonder what your premise amounts to. Waves of covid come and go in masked and unmasked states. What does it mean to say that masks slow growth at infection minima when a wave comes along and cases grow to a maxima despite the masks?
I'm still chewing on this, but thought I'd throw this out there since you asked.
"Mask use predicted lower minimum but not lower maximum growth rates."
The explanation that I offered is that, at low infection rates, people's probability of exposure is roughly the same whether or not they wear a mask, though their chance of infection is very different. This drags down the rate of infection proportionately to the number of masked people. Once the virus makes inroads in the unmasked community, however, the chance of an unmasked person getting exposed becomes steadily greater and greater, because an unmasked person is more likely to be around other unmasked people, which means that not only do they have a higher (relatively fixed) risk of being infected when exposed, they also have an increasingly high risk of being exposed. Then the exponentially increasing infections among the unmasked swamp the lower, relatively fixed rate at which the masked get infected.
Also I wondered what is the magnitude of the different predicted lower minimum that the paper talks about is? i.e. 5% or 25% or 125%, etc.
Some of those mask studies you linked to earlier were not great studies. I didn't click through everything, but for example the hairdresser study that one of them link to had no control (no hairdresser without a mask for comparison), confounding factors (hand hygiene compared to population, ventilation), and they didn't test all the customers (self reporting, also they didn't test customers from before the stylists developed symptoms (when they may have been at their most contagious). The USS Theodore Roosevelt study also mentioned had a lot going on, I wondered how they separated the self-reported mix of precautions (masks, social distancing, and common area avoidance), given the relatively small sample size for so many variables. I didn't look through the rest.
On the other hand... Here's a randomized control trial covid mask study [0] involving 4800+ people that shows surgical mask wearing was not statistically significantly better than no mask. Here's another study [1] (pre-covid) involving 1600 people working in hospitals that shows cloth masks are penetrated by 97% of particles, compared to 44% for surgical masks. The authors of this study say that cloth masks may potentially increase infection risk, double masking making it worse (holding viruses particles near the face and moist breath). Another pre-covid paper [2] that analyzed 9 studies on medical/surgical masks (3500+ participants total) and concluded "wearing a mask may make little to no difference in how many people caught a flu‐like illness".
I find the gap between what politicians and the media have been saying and these studies to be disturbing. But maybe this is why the CDC retracted their mask recommendation for vaccinated individuals several months ago, because the data doesn't show that masks help.
[0] https://pubmed.ncbi.nlm.nih.gov/33205991/
[1] https://www.researchgate.net/publication/275360639_A_cluster...
[2] https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...
There is no such thing in this study as a "masked population". The authors categorized the states based on the percentage of mask-wearers, according to the IHME data. But the reported infection rates don't distinguish between masked and unmasked.
Your reference [0] looked at whether masks protect the wearer, and explicitly did not study whether masks on the infected protected others: "...no assessment of whether masks could decrease disease transmission from mask wearers to others." (From the Limitations section.) Your other references do not pertain to covid.
The CDC changed their recommendation regarding masks for the vaccinated originally because they were satisfied that the vaccine provided sufficient coverage against serious illness to make unmasked, vaccinated individuals a very low public health risk. That changed when the Delta variant emerged and it became clear that vaccinated and infected individuals could spread the virus as readily as the unvaccinated infected, making them a potential vector and therefore a public health threat. As has been repeatedly reported, masks are primarily to keep the infected from infecting others, and again, there is a mountain of evidence that this is the case.
As an aside, I upvoted your comment because even though I disagree, you presented an argument and provided references to support it.
I might not be seeing the same mountain of evidence that you are. The first couple studies I looked at in the metastudy you linked to in the original comment in this thread (hairdressers and navy ship) were not convincing (they were quite loose, as I commented on) and the second study you linked to seemed to assume that stopping droplets slows the spread, but doesn't offer evidence of how relevant stopping droplets actually is (there's data saying that covid spreads through aerosols just as well if not more than droplets [0], like in restaurants where people were not seated closely, etc.)
I just don't see the effectiveness of cloth masks as proven. Where am I going wrong in my thinking?
[0] https://www.sciencenews.org/article/covid-coronavirus-aeroso...
With respect to type of mask, another metastudy looks at different types of masks, and concludes: "...correctly wearing masks of all kinds, despite their different designs, functions and effectiveness, will to a large degree reduce the overall risks of COVID‐19 infection and enhance general protection from coronavirus."
Like global warming, you can always dig up a handful of studies to support a contrary conclusion, such as the originally posted article appears at first glance to be. The point of my response was to reconcile the observations made in that article with the effectiveness of masks, to show that masks could be 100% effective and still not sufficient to suppress a covid maximum, if the percentage of people wearing them falls below some threshold. So whatever you may think of masks, the posted article isn't necessarily evidence that they aren't effective.
I'm not saying that you are provably incorrect. You pointed out that the studies I referenced were for personal protection, not to deter spread. However seeing how poorly cloth masks performed in those studies doesn't impart much confidence for their ability to stop spread, either.
Looking for a critique first, I stumbled found this article listing a bunch of pro-mask studies [2], with summaries. I took a quick peek at #3, which mentioned using laser light to examine droplets being exhaled. Sounded like a neat way to quantify things. As I got into it I wondered about the sizes of viruses in relation to the particles they were measuring. But I read some more and it appears they used a slightly damp washcloth, not a mask. That's the kind of thing that I find discouraging - the article quotes this as one of many studies supporting masks, but it was a damp washcloth! I don't consider that to be valid, and it doesn't help the credibility of the headline in my mind.
Have you got that far into details on any of this? The titles of these studies aren't enough to convince me, especially when I've clicked on several definitively titled studies and they are anything but. Maybe I have bad luck choosing which studies to read.
I went back to your link in the comment above. Under the Exhalation section I looked at this [1] 2020 study, it was the newest. It places a couple hundred people with the flu in cars and bedrooms, with and without masks, and used a device to measure particles in size from 20-1000nm, and found that masks had a significant effect. I then looked to see what size particles spread covid-19 and found this study [2] that mentions 80-90% of particles from human expiratory activities were aerosols with the diameter being smaller than 1μm. I confirmed that 1μm equals 1000nm [3]. They did something else with an aerosol generator and a dryer and a neutralizer, which is kind of confusing, but I think they were just doing some bench measurements. The worst I can say about this study is that the results in Table 2 include a pretty wide variations, and I'm not sure why. For example, in round numbers bedroom background particles were 51K +/- 16k, no mask was 93k +/- 63k, medical mask was 53k +/- 18k. What does it mean that background ranged from 35-67k and no mask from 30-156k? They mentioned these were "in group comparisons", maybe that has something to do with these numbers. I have to assume those aren't error ranges? I didn't see a link to their raw data to know for sure.
So that last study is interesting! It certainly passes my initial sniff tests. I'd save this link if I were you, it's the best I've looked into. At this point I'm curious to keep an eye out for a best conflicting study. Until then I've shifted my opinion, I'm not as sceptical as I was. I'm still not a fan of mask mandates, based on other reasons, but I respect hard data and wouldn't want to argue from an inaccurate position.
[0] https://www.kxan.com/news/coronavirus/do-face-masks-work-her...
[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7883189/#mds310...
[2] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7293495/
[3] https://www.traditionaloven.com/tutorials/distance/convert-m...
It's probably not fair to phrase it as overwhelming evidence. Yes there is evidence that masks can reduce transmission, and the citations you provided included a fair amount of research supporting this idea (thanks for those references). And just thinking about it logically, it seems obvious masks would have such an effect.
However the evidence presented in OP - and the previous studies it cited - provide counter-evidence suggesting that mask usage and mandates may not have a significant beneficial impact on the overall course of a pandemic.
There are a variety of possible reasons for this, which you began to speculate on. Poor compliance, people not wearing their masks correctly, porous cloth masks, poorly ventilated indoor environments contributing to high viral load environments even with 100% mask compliance - the number of places where things can go wrong are tremendous and it begs the question of if the tradeoffs of mask mandates are worth it, or if our resources would be better spent on other viral elimination strategies.
I posted a more detailed explanation of my theory above. You don't need to account for imperfections in masks, or ventilation, or any other details. Even if you make the assumption that masks are perfectly effective, the observations seen by the researchers might be explained by exponentially increasing spread among the unmasked. To use an analogy, a 4' child can drown just as easily in a 5' deep pool as in 100' deep lake. The percentage of mask usage referenced in the paper would only dictate the depth of the pool, so it's not surprising it would have little effect once the virus takes hold in the unmasked population. Barring very high mask usage (making the pool only 2' deep, to continue the analogy), exponential infection spikes in the unmasked population are unavoidable. It doesn't mean masks don't work, it only indicates just how high usage needs to be to stop these spikes.
> we did not observe lower growth rates over a range of compliance at maximum Fall-Winter growth [45-83% between South Dakota and Massachusetts during maxima] when growth rates were high. This complements a Danish RCT from 3 April to 2 June 2020, when growth rates were low, which found no association between mask use and lower COVID-19 rates either for all participants in the masked arm [47% strong compliance] or for strongly compliant participants only.
> We only assessed one biological quantity [confirmed and probable COVID-19], but the ongoing pandemic warrants assessment of other factors such as hospitalizations and mortality. Importantly, our study does not disprove the efficacy of all masks in limited and controlled circumstances, such as properly worn N95 respirators.
It's important to be aware that mandated non-pharmaceutical interventions like lock-downs, school closures, - and to a lesser degree masking - have second order effects on seasonal endemic illnesses such as influenza and RSV. Large future outbreaks of both diseases may occur following a period of extended NPIs. These outbreaks, which may reach peak numbers in the winter, could increase the burden to healthcare systems [1].
Considering these potential consequences, and the other potential negative externalities mentioned in the paper, it's imperative that such mandates be carefully scrutinized and evaluated before being widely enforced.
[1] The impact of COVID-19 nonpharmaceutical interventions on the future dynamics of endemic infections https://www.pnas.org/content/117/48/30547
You do not have to hold that view. You are free to criticize it. But downvoting seems counterproductive to any reasonable discussion.
As one HN'er put it:
> The downvote button is not a disagree button. If you disagree with something, add your own comment. That's the point of a discussion board. Downvotes are for comments which are not worthy of a response because they are not honest attempts at discussion.