Hacker News Reader

Comment by _-___________-_ | Hacker News Reader

Have surgical masks really shown to be effective in reducing the spread of disease when worn by people who are not yet infected?

I've been to mainland China, Taiwan, HK, South Korea, Japan, Vietnam, Singapore, Malaysia and Thailand in the last four weeks. China has close to 100% usage of masks in public, mostly the surgical type. The virus is spreading like wildfire there. Every other country I've visited has markedly less mask usage, in particular Vietnam where at least in the south, it's relatively rare to see people wearing them.

Yet in all the other countries it's hardly spreading (obviously I'm excluding the cruise ship off the coast of Japan here, which appears to be a special case).

View on HN

There are studies that show that masks (surgical and n95) can help prevent healthy people from getting infected [1][2]. Everyone I've seen claiming the opposite (including supposed experts being interviewed) usually do it without any significant data to back it up (just some high level theorizing). I also rarely see anyone discussing the nuances of efficacy against airborne vs droplet spread viruses. Using just a minimum amount of reasoning it would seem very plausible that masks would be effective against droplets but less effective against airborne diseases.

[1] https://www.newscientist.com/article/dn3692-face-masks-are-b... [2] https://annals.org/aim/fullarticle/744899/facemasks-hand-hyg...

Public health authorities like CDC don't recommend their use among people who are not infected:

CDC does not recommend that people who are well wear [a] facemask to protect themselves from respiratory viruses, including 2019-nCoV. - https://www.cdc.gov/coronavirus/2019-ncov/about/prevention-t...

Now, I'm not one to blindly trust a government agency, but most of the evidence seems inconclusive at this point, and one thing can't be ignored: in mainland China, use of masks in public spaces is near 100%, yet the virus is spreading rapidly. Outside mainland China, far fewer people are wearing masks, yet the virus is not spreading much at all.

> Public health authorities like CDC don't recommend their use among people who are not infected

See: https://www.cdc.gov/coronavirus/2019-ncov/hcp/index.html

Your link & the CDC don't say what you think they say.

CDC's recommendations are that members of the uninfected public should not wear N95 masks.

This should NOT be understood as: CDC says N95 masks are ineffective at preventing airborne infections of Covid-19.

It IS the CDC saying: N95 masks are non-trivial to use correctly, require decontamination protocols or single-use replacement to avoid infection and contamination from used masks themselves (potentially full of virus particles), and protect against a threat many people are unlikely to be exposed to (close proximity infected persons) -- a mix of criteria that make them inefficient to recommend for general public use.

As you see in the link above, CDC absolutely recommends N95+ respirators for EMS and healthcare workers, when working in close proximity to infected persons.

But healthcare workers have a substantially higher airborne exposure risk, as well as professional support and training for properly using N95 masks.

They do say what I think they say, but yeah, I should have clarified that I meant the public, not healthcare workers. (I addressed this more clearly in another comment on the same submission.)

Not quite sure I get the reasoning in that link.

Sure, it's not likely to decrease your direct infection risk.

But it does decrease the risk of you infecting others.

At any given time you can't really know if you're infected or not if the virus is widespread.

Also the mask would presumably be more effective than coughing into your elbow or a tissue, which they do recommend.

So why not wear them? Why recommend them for healthcare workers but not the general public?

I'm wondering if the unstated reasoning is that widespread mask usage freaks people out and increases the chance of large scale panic.

> At any given time you can't really know if you're infected or not if the virus is widespread.

Unless you're in China, the virus is not widespread. I imagine they would change their recommendation if it was.

> So why not wear them? Why recommend them for healthcare workers but not the general public?

The chance that a healthcare worker who is treating a suspected CoV case is in close contact with an infected patient is close to 100%. On the other hand, the chance that a member of the general public outside of mainland China comes into close contact with someone who is infected, whether they're in the rest of Asia or not, is within a few orders of magnitude of 0.00001%.

1-Since travel is still occuring out of mainland China, your faith in the virus being limited to the nation of China seems misplaced.

2-Your estimates of the probability are unlikely shift properly as the landscape of infection changes rapidly, and is not apparent to the naked eye.

Humans intuitively do not tend to comprehend exponential/quadratic growth. This video outlines it. The main point is to look how late, in this scenario you'd be saying "no problem" and how quickly it goes from fine, to effed. Note: this is not an endorsement on the other items in this channel, I just think this particular explanation is demonstrative.

https://www.youtube.com/watch?v=iIwyMif5EOg&feature=youtu.be...

> 1-Since travel is still occuring out of mainland China, your faith in the virus being limited to the nation of China seems misplaced.

I never said it was limited to the nation of China, I said it wasn't widespread outside of China and a cruise ship, which remains true despite significant travel still occurring.

Humans are also terrible at intuitively assessing risk, and the current panic is a great example of this.

> Humans are also terrible at intuitively assessing risk

Doesn't that also apply to you? Have you looked into when and where testing is performed? In the US, if you don't have a china travel history, they won't test you.

This has an incubation period that is still indeterminate, and in the US and other large countries, they are barely testing. So you literally don't have a way to be sure of what you're claiming--there's just no data, but somehow you are very confident in what you claim?

Have you read up on the Spanish Flu? Was that just a "panic"? So far, by every measure, this looks to be worse--more transmissible, harder to detect, with a higher mortality rate. The Spanish Flu killed between 1 and 5% of the world's population.

> Doesn't that also apply to you?

Yes indeed, and to you also. But in the same way that one can learn to better appreciate exponential growth, one can learn to better appreciate risk, and try to apply that knowledge rationally rather than basing decisions on fear.

> Have you looked into when and where testing is performed? In the US, if you don't have a china travel history, they won't test you.

Yes. The most reliable testing procedures are quite invasive and carry at least some exposure risk for the healthcare worker carrying them out, especially if done outside a hospital setting. They also aren't guaranteed to detect the virus in the early stages, or in minor infections. This is one reason why China has moved to include clinical diagnoses as confirmed cases, and it's also why they didn't just test every person on the Yokohama cruise ship straight away.

> This has an incubation period that is still indeterminate

Not really. There are some pretty good bounds on it, especially with the significant number of cases in Wuhan available as data.

I am confident in what I claim (given that what I claim is quite limited). I'm not making predictions here, I'm talking about the current state of things. Beyond that, let's see what happens...

Yes, totally agree.

I wonder what the rational tipping point would be for mask vs no mask.

> I'm wondering if the unstated reasoning is that widespread mask usage freaks people out and increases the chance of large scale panic.

Different cultures and situations handle this very differently. I'm hearing reports in London that people are being attacked for wearing a mask in public.

Meanwhile, if you want to scare everyone around you in Hong Kong, go out without a mask and then cough.

Surgical masks are meant to stop droplets and large particles but aren't rated to filter airborne particulates. These more or less protect people from sneezing or coughing directly onto one another.

The next step up from a surgical mask is an NIOSH rated mask / respirator. These carry a letter and number rating the letter is either N, R, or P and the number is typically 95 - 100. N95 is a NIOSH filter rating and where the term N95 mask comes from. These will filter out any viruses and other particulates in the air. There are studies showing that N95 Masks are effective.

https://www.cdc.gov/niosh/npptl/topics/respirators/disp_part...

edit: all that said there are studies that show both types have similar effectiveness at preventing infections in healthcare workers so I wouldn't go around telling people that any of them are useless.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4868605/

Yeah not only are surgical masks not even clinically effective in this situation, they’re often not worn properly. I’m in Beijing now and you see people with masks not covering their nose, or worse, pulled down onto their chin. A friend of mine tried to remind an older lady in an elevator about proper mask usage, and her response was 你们年轻人管太多了 — you young people worry too much/should mind your own business.

The people in my compound who were confirmed to have the virus (we found out on the 12th) got it at a family gathering over the new year holiday in Shanghai. So obviously they weren’t wearing a mask then.

What seems to be making a difference in China — it’s no longer spreading like wildfire, if you believe the official reports* — is self-quarantine. Most of my friends have barely left home since the beginning of February. For a while, some restaurants that were still open wouldn’t let people sit together in parties larger than 2. And at least in Beijing, everyone who returns to the city after travel must undergo mandatory 14 day home quarantine, with fines or punishments for those caught violating (https://www.nytimes.com/reuters/2020/02/14/world/asia/14reut... though enforcement varies from place to place).

* Wuhan/Hubei are a tragedy no matter how you count. But at least Beijing is doing OK. I mostly trust the official numbers, simply because the Party has a strong vested interest in mounting a response commensurate with the threat. They can’t let the virus spread too much and rip apart the capital; neither can they afford to tank the economy for several months longer than necessary to contain the threat. Even with all the censorship, if you find yourself fighting against reality, reality wins.

No, I believe I read that the typical N95 masks that people use is basically useless against this virus, because it's very, very tiny and can sail right through a mask. It still might reduce your exposure and is probably better than nothing, but I believe only a gas mask would totally protect you.

"...there's no evidence inexpensive surgical masks can protect against flu virus particles small enough to be inhaled into the lower respiratory tract or the lungs"

http://www.digitaljournal.com/life/health/do-face-masks-help...

N95 masks (and other types of HEPA filters) actually perform better on smaller particles.

A very common misconception is that HEPA filters only filter particles larger than 0.3 microns. In fact there's multiple different filtration mechanisms at play, with interception and impaction working better for larger particles, and diffusion being more effective for smaller particles. HEPA filters are rated for their effectiveness at 0.3 microns because that size is the size where the combination of mechanisms operate worst: smaller and larger particles are filtered even more effectively.

https://en.wikipedia.org/wiki/HEPA#Mechanism

Your own cite says that surgical masks are ineffective but properly fitted N95 masks can be effective. I believe that they filter droplets well, e.g. from sneezing.

https://www.asahq.org/about-asa/governance-and-committees/as...

I think they're better than nothing but as someone else mentioned the virus is 0.1 to 0.3 microns and the N95 only filters 95% of particles 3 microns or larger.

See https://www.cdc.gov/coronavirus/2019-ncov/about/transmission...

"Most often, spread from person-to-person happens among close contacts (about 6 feet). Person-to-person spread is thought to occur mainly via respiratory droplets produced when an infected person coughs or sneezes, similar to how influenza and other respiratory pathogens spread."

https://www.cdph.ca.gov/Programs/EPO/Pages/Wildfire%20Pages/...

"An N95 respirator is particularly effective against aerosolized droplets that are invisible to the naked eye."

These droplets can infect you through your eyes as well. Like I said, the mask helps, but no it's not virus proof as Wuhan healthcare workers can attest.

You're picturing a bare virus flying through the air and getting through a mask.

That's not a plausible route of transmission. The virus would be embedded in a much larger particle of human mucus, and an N95 mask will do a fine job of filtering that out.

The virus isn't going to be airborne naked as an individual thing in the free air - it's going to be in droplets of stuff (from sneezing, coughing, etc).

That may well be, but the predominant concern is larger droplets from a cough or sneeze. I don't recall seeing anything speaking of airborne transmission.

Smaller particles get filtered through diffusion based processes: they like landing on surfaces, and the mask is full of surfaces.

I haven't seen conclusive evidence that they're better than nothing.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4868605/

Here you go

This study only finds that N95 are not more effective than surgical masks. It does not measure the effectiveness of surgical masks vs nothing. Did you actually read the study before posting a link to it?

It's worth also noting that this refers to healthcare workers. The considerations are totally different here, because the likelihood that they'll come into close contact with an infected person is pretty close to 100%. The likelihood that a member of the public will in most Asian countries with 20 or so cases and a few tens of million people is more like 0.00001% give or take a couple of orders of magnitude.

Most people are not even wearing N95 masks, they're wearing the much more common and cheaper "surgical masks" that don't form any kind of seal around the outside.

> Have surgical masks really shown to be effective in reducing the spread of disease when worn by people who are not yet infected?

> China has close to 100% usage of masks in public

You know, 100% usage in public is an excellent way to ensure that people who are infected wear the masks.

I agree. But other countries have no significant community spread besides not much mask usage.

> I've been to mainland China, ... HK ... in the last four weeks. China has close to 100% usage of masks in public, mostly the surgical type. The virus is spreading like wildfire there. Every other country I've visited has markedly less mask usage.

I dunno where you were in HK, but the only place I haven't seen everyone wearing a mask is in Soho. Everywhere else, I'm seeing near 100% coverage.

I would have put it closer to 2/3 or 3/4 in HK, versus almost 100% in China. I'm usually around Yuen Long, Mong Kok, TST, Central and Western. Has been a week since I was there though. Compared to mainland China and HK, there's way less mask usage elsewhere. For example in Vietnam I'd say it's more like 5-10%, at least in the south.

Absence of evidence is not evidence of absence. A number of public health authorities, including the US CDC, have stated that there is likely community transmission happening.

Are you still traveling to these countries? Are you concerned?

Yes, all of them except mainland China.

I'm not the slightest bit concerned. No country except mainland China has more than 70 confirmed cases, and deaths worldwide are in the single digits outside of China. These are countries with millions, tens of millions or hundreds of millions of inhabitants. I'm heading to Singapore in a few days, which has the most cases outside of mainland China (likely because they're very good at tracking them) -- 67 out of a population of 5.7 million (0.00001%), and of course all of those people are in quarantine. I imagine some cases haven't been discovered yet, but I have a hard time imagining that changing the order of magnitude.

I'd be willing to travel to many parts of mainland China, but it causes problems for traveling to other countries, so I'm unable to for now.

I'll alter my travel according to the facts as they change, but for now there's absolutely no rational reason I can see for avoiding travel to any country other than mainland China.

Edited to add: numbers are from https://www.who.int/docs/default-source/coronaviruse/situati...

Probably the main reason to consider travel slightly hazardous is that you'll be in close proximity to other travelers and people who service travelers, which likely represents some risk.

What if a group of Chinese nationals traveling to a tour get on your plane / bus / etc, or were the previous people in your seat? You don't have that high of a risk from this scenario, but it is more exposure than you'd get at home.

Most organised outbound tourism from China is canceled right now, though there are of course still independent travelers and probably smaller group travel too.

It's also worth noting that the fact that they're Chinese nationals doesn't really move the needle that much on how likely it is that they're infected. About 0.00005% of the mainland China resident population is confirmed to be infected, and the vast majority of those (over 80%) are in Hubei, which given the lockdown makes it very unlikely that you will meet them on an aircraft. Even if you allow for under-reporting the order of magnitude is probably about right.

Additionally there are another 50 million Chinese nationals living in other countries, who I would submit are more likely than those resident in China to be traveling right now, and are also far less likely to be infected. This is one reason why the racism seen recently in Australia and UK against Chinese-looking people wearing masks is so ridiculous.

How do you know if your cough is due to smog or due to infection? Best to wear the mask anyway.

Well, the best available estimate of the chance that I'm infected is on the order of 0.00001%, give or take an order of magnitude. If I would take countermeasures to every risk of that order, I wouldn't have time to do anything else.

That number really depends on where you live right now.

Not as much as you'd think. It's basically "mainland China or a cruise ship" or "anywhere else".

Isn't mainland China the only place where people are really wearing masks because of this?

Yes, it's simultaneously the only place with close to 100% mask usage, and the only place with significant spread of it.

From this one could conclude, that masks are spreading the virus

correlation != causation

It does however provide a data point that masks are likely not very effective at preventing the spread of the virus.

That's like saying cancer treatment centers aren't very good at helping people survive because many people who visit them die.

So maybe you're right. But maybe it would have broken out 100x worse by now.

Again: there are other countries where the virus has spread in the community, yet despite virtually no mask wearing the spread has been extremely limited and easily controlled.

In single or double digit counts. It's not that much. Give it time. No place is as densely infected as China because China had a huge headstart.

The important bit is that they’re not “practically” effective.

You need to wear them properly. You need to change them regularly. You need to practice effective hygiene like not touching your face.

If the government distributed masks that could be effective to everyone, it’s unlikely they would have a noticeable effect

Reply on news.ycombinator.com