Fitted Filtration Efficiency of Double Masking During the COVID-19 Pandemic | Hacker News Reader

Fitted Filtration Efficiency of Double Masking During the COVID-19 Pandemic

jamanetwork.com

If you take doubling to mean "doubled FFE" score, this only occurs when you wear a procedure mask under a cloth mask (this combination is in general the most effective as described in the paper).

If you take doubling to mean reduce (1-FFE)(since it is a percentage) by 50%, this only happens when doubling up on one brand of procedure mask, or when wearing a procedure mask under a cloth mask.

Procedure mask over a cloth mask is pretty much only as effective as procedure mask alone, and doubling of cloth masks did not double FFE by any measure.

This headline is the not the best summary of the paper, its abstract, or its title.

We've changed the title to that of the article now. Submitted title was "Wearing two face masks can nearly double the filtering out SARS-CoV2 particles", which broke the site rules: "Please use the original title, unless it is misleading or linkbait; don't editorialize."

https://news.ycombinator.com/newsguidelines.html

Submitters: please don't do that. If you want to say what you think is important about an article, that's fine, but do it by adding a comment to the thread. Then your view will be on a level playing field with everyone else's: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...

I think that a more interesting approach is enhancing the fit of the mask to the face. Apart from improved protection, a better fit means that water vapor from one's body doesn't fog up one's glasses!

One of the references of this paper is exactly about that, with photographs of example setups ("Figure 2"): https://jamanetwork.com/journals/jamainternalmedicine/fullar...

The "mask fitter"/"badger seal" project already received some attention on HN:

https://making.engr.wisc.edu/mask-fitter/

https://docs.google.com/presentation/d/1bTUmxEiZ_EMo9DYjHWl1...

Personally, I used a bit of polycaprolactone to form a hard plastic part that is formed on the nose and on each cheek and which:

1. can push the mask against my face, while its shape conforms to the face, so that it pushes on the surface of the face relatively evenly

2. features protrusions (on the cheek-parts of the plastic) onto which something like a string can be put so that it (the string) goes behind the head and beneath the chin. This string then fixes the plastic piece (and the mask) in place.

Polycaprolactone is a plastic which softens at around 60 degrees Celsius, at which point it is possible to mold/form it by hand (skin contact is safe, and it's safe to let it harden on one's skin). Also, it's biodegradable and relatively cheap: https://en.wikipedia.org/wiki/Polycaprolactone

The main problem with designing something like this is how not to make yourself look absurdly ugly while wearing it on top of the mask ;)

If you compare a P100 to a P95 mask, the first thing you'll notice is the fitment. P100 are designed to actually fit your face, and fitment tests are a part of proper deployment.

I've been wearing a 3M 7503 the whole pandemic, modded to remove the check valves. It fits much more comfortably than any N95 junk (no crap behind my ears, no tight elastic on my head, etc), which is why I had originally gotten it for work around the house. I can easily tell if it's sitting a bit weird on my face and is leaking because it feels different on my skin.

In addition to being P100, I like it because it doesn't fog up my glasses. The biggest issue is that I'll tend to get condensation on the inside of the mask, and wearing it for long periods it will start to pool. I'll get condensation on an unmodded 7503 as well though (although it will be outside the outflow valve and drip on the ground), which is probably a consequence of having a higher thermal mass than mere filter material like an N95.

It's a tragedy that we're still stuck in the PPE dark ages over a year into the pandemic, with "insights" like stacking two junk masks to increase their effectiveness. Herds of people are never intelligent, but if one political team hadn't hitched their wagon to personal irresponsbility, I think we could have been having much more productive conversations.

Ditto on the masks situation. We somehow stagnated at “wear any mask made out of anything”, without incremental improvements as supply has improved.

At the other extreme, I’m surprised I haven’t seen anyone wearing scuba gear yet.

> I’m surprised I haven’t seen anyone wearing scuba gear yet.

Oh man, we researched that early on in the Pandemic. I was part of a team with a local hospital that was looking for any and all solutions. The big issue was weight, followed by the time you could be tanked, and then the cleaning issue. The overall conclusion was that they weren't worth it and just sterilizing masks again was better.

I wish I could blame the lack of reasonable mask guidance on the previous administration, but last I checked the official CDC guidance was that they'd rather have the general public tie the tattered remnants of their Def Leopard Hystaria tour t-shirt over their face than an FDA approved K/N95 mask. We've had a good supply of masks available for many months (Although you have to be careful about your supply chain). Of the many, many unforced errors of this pandemic, this may be the most baffling to me.

On the one hand, its great news that a simple hack can increase the effectiveness of face masks.

On the other, I'm really disappointed that we haven't done a lot more work during this pandemic to develop comfortable masks that are inexpensive and effective for both the wearer and as source control.

It would resolve a lot of issues with anti-mask folks if the masks themselves were simply more effective.

> It would resolve a lot of issues with anti-mask folks if the masks themselves were simply more effective.

Citation needed.

Not really something that needs a citation, IMO. These masks are source control, so their effectiveness is dependent upon everyone wearing them.

If they were able to more effectively protect the wearer, a few folks not wearing them has less impact.

Ditching cloth masks for low costs medical masks was a revelation for me. I can finally hold a conversation with someone or walk a few flights of steps without it getting suction-cupped to my face. I even leave them in place during workouts because I simply forget they are there (rather than pulling them off for my set).

This depends on the type of the cloth mask.

My mother has made many for different organizations. After a bit of "what fits me" the fitted/shaped oval doesn't fit me well. The pleated design, however, when made sufficiently large, fits reasonably.

The ability to talk was the problem that I had with the fitted design - moving the jaw up and down would cause either the top or the bottom of the fitted mask to move out of place. The pleated mask, however, is a much larger piece of fabric and has the slack in it to allow it to move with my jaw.

The authors write: it is notable that enhancements that improve the seal between the mask and the facial skin dramatically improve FFE performance, suggesting that fit, not material, is the intrinsic limiting factor for procedure masks.

3M VFlex has been around since before the start of the pandemic and has a great seal and is comfortable.

I actually believe the contrary to be true: if most people were given well-fitting masks that effectively filtered the air, mask adoption wold be much lower.

Because wearing a properly fitted mask that actually filters the air you breath in and out is extremely discomforting after about half an hour. I've used well-fitted 3M 6500 masks a lot for working with nasty glues etc. Those only filter the air you inhale, and have a very broad silicone lip to distribute the sealing pressure. Still it's a lot more uncomfortable than any of the usual Covid face masks I've tried.

I believe the two largest benefits of mask wearing as currently implemented is catchment of ballistically travelling saliva from sick individuals, and reduction of hand-face contact in all individuals. And for those two, filtration efficiency is irrelevant.

IMO the 7500's are more comfortable than the simple medical masks that pull on your ears, or simple N95's that squish your head. But I wouldn't want to wear mine for a long shift, because it gets a bit stuffy having removed the check valves.

Also, half face respirators get much lighter and easier to breathe when you swap out the chemical cartridges for P100-only puffball filters (eg 2297).

I'm looking forward to seeing how the sleek fabric ones feel, after the pandemic is over and I just need to thwart surveillance cameras.

I don't think it's a development issue, it's an education issue. There are plenty of masks out there that fit well and are meant to be worn for extended periods of time, be it in a medical environment or a trades environment.

The problem is people who didn't wear the masks pre pandemic don't necessarily know how to properly fit a mask or how to identify product lines that are of higher quality and better for long term wear.

OSHA has a lot of great videos on this topic and has regularly published on it for decades, but we've done a poor job disseminating it to the general public.

Or due to a complete lack of standards, if you have been fitted for a particular mask pre-pandemic, you’re basically screwed on buying that exact model and nobody has a list of “99% of people fitted for this mask also fit this other mask”

Where does the "doubling" in the title come from? I see a 10-15% increase from the article. And only when the masks are used in a specific ordering: procedure mask on face, and cloth mask over procedure mask.

> wearing a procedure mask under the cloth face covering produced marked improvements

It should also be noted that there were 3 volunteers.

> Limitations of this study are that we tested only 1 type of procedure mask and that 3 volunteers participated in the doubling evaluations

Wearing a helmet while riding a car will probably decrease your chance of dying if involved in an accident. It is also ridiculous.

Not when you factor in whiplash. You aren’t buckled in to motorcycles

It would probably benefit pedestrians as well.

WARNING: a lot of people wear ill-fitting masks where air escapes on the top and on the sides.

When the mask does not fit well adding a second mask can increase friction and encourage MORE air to go AROUND the mask instead of through it.

different for me. I have a construction mask that has a great filter but it is worn out and doesnt fit tightly anomre. a second whatever mask keeps that first mask close up and creates a better seal

That's write I wrote "can".

That could be a big issue with large particles.

But the air going through the mask is doubly filtered, that could make up for the increase in unfiltered air.

With very low viscosity, air is very good at sneaking out of a mask making it ineffective.

Luckily you can give a good puff and feel with your fingers any air coming through your eyebrows or around the cheeks.

Also covid tends to spread via ballistic spray more than aereosol so blocking the spraying is still useful.

Spend time adjusting the metal piece over the nose. Stay safe!

I'm well into the N95 and equivalents where my test is "does it slightly implode on inhalation" because that means the path of least resistance is still the filter itself.

I'm also well into the "it's aerosol" camp. Any mask at all will catch your outward ballistic spray.

I'm waiting for one of our resident math geniuses to come and explain why this is incorrect. It sounds impossible to me.

It doesn't help the filtration that much. Anything caught by a cloth mask would also be caught by procedure mask material. It helps the fit while people move around though, and that seems pretty plausible.

Frontend software developers and their opinions on virology, name a better combo!

Backend software developers and a superiority complex

"Don't be snarky."

"Please don't sneer, including at the rest of the community."

https://news.ycombinator.com/newsguidelines.html

And three can nearly triple it!

Just wait until you've heard my idea for quadrupling the filtering of SARS-CoV2 particles...

If you like that, wait until you see how great 10 work!

this is fluid dynamics we’re talking about here; could just as well be 16 times better.

I've started with 100 on top of each other and will gradually remove each until my test subjects stop suffocating.

That’s nothing. I’ve quadrupled the number of cigarettes I smoke a day. Granted, I can’t climb the stairs any more without having an asthma attack, but I think you’ll agree that reducing my risk of infection from hardly anything to almost impossible is surely worth it.

Based on the corpus of data accumulated from our double-blind, placebo-controlled study of marginalized populations, we can extrapolate that the COVID-19 particle-filtering behavior of face masks has an ex vivo efficacy that can be modeled by the general equation:

f_n = (n) x (f_b), where n is the number of face masks, f_b is the baseline percentage of particles filtered, and f_n is the new percentage of particles filtered.

(/satire)

If you put your head in a plastic bag and tie it off tightly at the neck, you're literally guaranteed to not die from Covid. Which highlights the idea that you really have to consider the costs and downsides of any Covid prevention measure, something that is often not addressed by many people supporting draconian anti-Covid measures.

Masks aren't draconian though, and they're cheap and easy to use. They also cause little issues for the vast majority of adults. They could be the #1 life saver out there, if people only used them.

> something that is often not addressed

I don't know a single person that likes lockdowns, say, yet the majority of people I know support lockdowns.

We are doing the whole spiel, light lockdown - hard lockdown - light lockdown, for a year now, it has become quite obvious how bad lockdowns are. How do you assume otherwise?

I think would increase by a factor of 3

of if one mask reduces particles by 50%, three would reduce by 1-1/2^3

No. Stop spreading misinformation. Wearing 3 masks is less effective because it makes reduces air flow so much it starts coming out of the side of the mask completely unfiltered.

https://www.reuters.com/article/uk-factcheck-false-post-mask...

so 2 masks is better than 1 but 3 is worse than 2... how are we sure 2 is the optimal number? surely, calculus must be done to determine the optimal fractional number of masks between 2 and 3.

(my money's on Euler's number.)

I guess a sample size of 3 is better than a sample size of 1, but how did this paper even get past peer review?

A Danish randomized controlled trial with 6000 participants, published in the Annals of Internal Medicine in November 2020, found no statistically significant effect of high-quality medical face masks against SARS-CoV-2 infection in a community setting. https://swprs.org/face-masks-evidence/

Your source is lying to you. That's not actually what that study (https://www.acpjournals.org/doi/10.7326/M20-6817) says.

> 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.

They studied the efficacy of "wear a mask" recommendations. Not mask wearing itself.

> They studied the efficacy of "wear a mask" recommendations. Not mask wearing itself.

as far as policy is concerned, these two are equivalent

Not at all. "Masks work on a functional level, but people don't wear them if asked" would be something to tackle by policy differently than "masks offer no functional benefit".

Like fines and other enforcement measures for not wearing one beyond "please?".

no idea where you live but since masks have been enforced everywhere here (including: the beach, hiking trails, walking alone at night), with the police actually checking people, we've had 3 subsequent waves of covid. i'm only talking about exteriors because i haven't seen a single person not wearing a mask inside – no mask in a shop? you don't get allowed in.

expecting more than this and discussing the result of a study that can't really be done in a lab is debating philosophy, but of course critics can go on downvoting.

Again, from the specific study in question:

> In this community-based, randomized controlled trial conducted in a setting where mask wearing was uncommon and was not among other recommended public health measures related to COVID-19...

This study did not tackle the effectiveness of masks. It attempted to determine the effectiveness of asking people to voluntarily wear one in a location where masking is not a recommendation by local health authorities.

> expecting more than this and discussing the result of a study that can't really be done in a lab is debating philosophy

Testing mask effectiveness can and has been done in a lab.

no not really. the study says that one of the two groups was instructed to wear masks. it's the people outside that were only recommended to do so. half of them reported always wearing them and only a small minority reported never doing so. the results of the test are in line with the sad reality in most european countries that have mandatory mask wearing since last summer.

nobody is questioning lab tests.

to be clear: the effectiveness of mask-wearing policies on a broad population can not be done in a lab, although it would be interesting to test "not wearing masks" in a country that does have that policy; this unfortunately can't be done either because it's currently against the law - where applicable.

The study here[1] was inconclusive, but their limitations seemed pretty noteworthy:

> 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.

I'm not sure what conclusions could be drawn from this study alone.

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

I worked as a cashier at one of the two Walmarts in Rapid City, SD last year, from summer to this January, between programming jobs. just before that (March—April sometime), my mom got covid, was incredibly sick for about ten days, and I took care of her during that time (while not wearing a mask except during an urgent care visit). South Dakota has never had a mask mandate. Walmart has a mask mandate for employees, but at least here in SD, they don't enforce it for customers. Rapid City is close to both Sturgis and Mt. Rushmore, and as such is a major tourism hub in the summer, especially during the Sturgis Motorcycle Rally.

despite "poor mask-wearing form" (frequently worked with the mask being below my nose, so I could breathe while working), the predicted-to-be-apocalyptic nature of just how devastating the 2020 Sturgis Rally was supposed to be, the supposedly mass-fatality-inducing lack of state mask mandate, the supposedly significant outbreak caused by President Trump giving a speech at Mt. Rushmore on July 4th (with people traveling from all around the country to come see it), and me having literally taken care of a woman with covid for ten days while she had it... I have somehow avoided catching covid. I usually get sick all the time. I was the first case of H1N1 in the state a decade ago when I was (once again) working as a cashier at Walmart (& went to work with insufficient sleep due to preparing a D&D campaign for my friends the night before).

obviously anecdotes aren't evidence but I gotta say, I spent a year waiting for the supposed covidpocalypse to ravage South Dakota, especially the tourist-friendly Black Hills... and nothing significant happened, at all. many people have been sick and of course a few people have died, provably as a result of the 'vid... but overall, somehow, the pandemic has never significantly reared its head here, seemingly against all odds, given the rhetoric about the supposed importance of (now plural) mask-wearing, and my state's refusal to comply with official federal government recommendations

EDIT: should also mention that, for what it's worth, I did one PCR test for covid after visiting my mom, who had a very paranoid out-of-state friend over, and then a couple days later I caught a common cold. despite my insistence that this was a cold-ass cold and definitely not the 'vid, the friend demanded I get tested, as I was working in a "high-risk" "essential personnel" job (Walmart cashier), before she (friend) would commit to returning home as scheduled, as opposed to self-quarantining for two weeks. the test came back negative. the cold was gone in 72 hours.

South Dakota is #3 in cases per capita and #10 in deaths per capita in the US. It's August-November stretch was just brutal, peaking with a death rate of 200/1M for the week of Nov 16.

Sure it's raw numbers may not have been as dramatic as NY, a state with over 20x as many people, but it definitely was not example setter.

Edit: My numbers are based on data I collected from NYT, but Worldometers shows SD as #2 in cases per million and #9 in deaths per million: https://www.worldometers.info/coronavirus/country/us

NY had created some of the harshest anti-covid regimes in the country, and by your own data NY was still far worse.

NY had 24% fewer cases per capita. How is that "far worse"?

> 200/1M

I understand the metric but we don't even have a million people in the state. so when you compare ratios it might look high when comparing to other populations, but in reality, it's remarkable at how low the numbers actually are, again, given the difference in mask policy.

again, there was no summer spike due to tourism. there was no spike due to almost half a million additional people riding in from out of state for the Sturgis Rally (temporarily significantly increasing our population!). iirc, two people died after returning home afterwards. very few bikers wore masks anywhere. more Rally-goers died from traffic accidents (as is usual for the Rally) than died from the 'vid.

You can change it to 20/100k if you wish, it doesn't really make a difference.

SD case counts per 1M on selected dates:

Jul 15 2020 - 463

Aug 15 2020 - 725

Sep 15 2020 - ~1800

Doubling in a month seems like a spike to me.

are you inferring that the rise in case counts in September is somehow correlated to the summer tourism season, which is pretty well over by September? I see no evidence that points to this. surely if the summer tourism season caused a spike in cases(!), you would see it during the summer tourism season, and not afterward?

You keep talking about Sturgis, which ended on Aug 16. There is known incubation time between contact and testing positive, and even by 10 days later (Aug 26) the rate was at ~1200. Also July -> Aug was a 55% increase in cases as well, even before that, which coincides more with summer.

I'm not going to say definitively what caused what here, but your anecdotal recollection of what happened in SD is not really supported by the numbers.

And wearing a plastic bag on your head cuts the risk of COVID by 100%

How does this compare to wearing N95 masks?

Reply on news.ycombinator.com