NOTE: The analysis here is NOT applicable to COVID-19 prevention.
This is a 2018 article about masks for wildfire smoke only. I have not updated it for COVID and am not going to. Please refer to more recent and authoritative sources for COVID-19 best practices.
For several hours, a dangerously misleading article from the San Fransisco Chronicle circulated on social media with the headline “If you’re using an N95 mask for smoke, you’ll need a new one every 8 hours”. I call this “dangerous” because there is a mask shortage as well as a housing shortage, and there are lots of people sleeping outside without masks as I write this.
I’ve done a fair amount of research at this point, and everything I’ve been able to find suggests the masks can be worn for much, much, longer than 8 hours. You should not smell smoke through the mask, even after extended use. The most important mask tip is to make sure you are wearing it properly so that it creates a seal. If you smell smoke through the mask, you’re not wearing it properly – so replacing the mask won’t fix it.
What the CDC report actually said
The article has since been retitled, and the paragraph in question edited to specify that the study they cite is about pathogens and contagious disease in a healthcare setting. That’s significantly different – pathogens remain dangerous even after they’re lodged in the mask, and can multiply in that environment. Smoke particles stay put, and are very much not alive nor contagious. The article now accurately states:
there is no recommended interval for changing an N95 mask in smoke-related environments.
Even then, the study doesn’t actually say anything about switching out masks after 8 hours, pathogen-related or otherwise. Here’s where the study refers to 8 hours:
A key consideration for safe extended use is that the respirator must maintain its fit and function. Workers in other industries routinely use N95 respirators for several hours uninterrupted. Experience in these settings indicates that respirators can function within their design specifications for 8 hours of continuous or intermittent use. Some research studies (14, 15) have recruited healthcare workers as test subjects and many of those subjects have successfully worn an N95 respirator at work for several hours before they needed to remove them. Thus, the maximum length of continuous use in non-dusty healthcare workplaces is typically dictated by hygienic concerns (e.g., the respirator was discarded because it became contaminated) or practical considerations (e.g., need to use the restroom, meal breaks, etc.), rather than a pre-determined number of hours.
The reference to 8 hours is a floor, not a ceiling. They’re referring to the general common knowledge that workers in other industries routinely use the same mask for a whole shift. They say nothing about switching after 8 hours, because that is assumed to be the end of the shift. To the contrary, that section concludes with
Extended use alone is unlikely to degrade respiratory protection. However, healthcare facilities should develop clearly written procedures to advise staff to discard any respirator that is obviously damaged or becomes hard to breathe through.
The rest of the study is about pathogens.
When it’s time to switch masks, you’ll know it
Here’s what the directions in the box say:
If filter becomes damaged, soiled, or breathing becomes difficult, leave the contaminated area and dispose of the filter.
The P-series ones (P95/P99/P100) also say
If used in environments containing only oil aerosols, dispose of filter after 40 hours of use or 30 days, whichever is first.
Solid aerosols tend to build up and form a “cake” on the filter which increases both the filter’s efficiency and its breathing resistance, indicating when the filter needs to be changed. […] Atmospheres that contain both oil and non-oil aerosols will most likely result in filter caking from the non-oil aerosol. Therefore, the P-series time use limitation reverts to dispose of the filter when it becomes damaged, soiled, or difficult to breathe through if the filter is used in environments that contain no oil aerosols, or if the filter is used in environments that contain a mixture of oil and non-oil aerosols. Only if a P-series filter is used in an environment that contains only oil aerosols does the full time use limitation apply.
I have not found any clear evidence on whether the wildfire smoke contains oil aerosols in any significant quantity, but I’m not too worried. 3M’s report notes, “In most atmospheres containing oil aerosols respirators are worn for protection from contaminants other than oil” and “Neither 3M nor NIOSH has been able to locate an oil-only environment.” Meanwhile every public health agency everywhere is saying N95 is indicated. N95 isn’t supposed to be used at all in environments with oil, so I take that to mean the guidance that particulates are way more likely to be the limiting factor, is applicable here.
Bottom line
[UPDATE 11/17/2018]: No one really knows how much of the non-particulate stuff in the air right now could be considered oily gas, except that it’s greater than zero but comparatively small. The increased efficiency from trapping more particles partially offsets the degradation from small amounts of oil, but based on anecdotal evidence, it seems they do have a net decrease in efficiency over time. One person’s unofficial experiments from Beijing found that a similar disposable 3M mask decreased from blocking 99.7% of 0.01-0.1 micron particles when new, to 98.3% after 11 days of “intensive use”.
[UPDATE 9/10/2020]: Oily gas is what produces the characteristic wood-burning smell. If you can smell smoke, you should use P100 instead of N95. Replace your mask/filters after 40 hours of use or 30 days, whichever comes first. If you can’t smell smoke but the PM2.5 in the air is high, N95 might be ok and clogging with particles is more likely to be the signal that you need to change your mask.
So, they do get less effective with time, but the decrease is very slow. It should still be safe to follow the manufacturer’s instructions: when breathing through it becomes noticeably more difficult, that is a sign that it is starting to degrade and should be replaced. This happens on the scale of days, not hours.]
- Particulate respirators get more efficient (they trap more particles) as you use them. As the mask becomes clogged with particles, it gets harder to breathe through.
There’s no degradation from continuing to use them as for as long as you can physically breath through them without difficulty.If you smell smoke, it means you should adjust the mask to get a better seal. It does not mean the filter itself has worn out. - The most conservative estimate of how long it’s ok to use the same mask seems to be the 40 hour (five 8-hour shifts) or 30 days (intermittent) limit for P-series masks in a hypothetical oil-only environment. At least one friend reports being able to wear a single mask daily for 2-3 weeks in Beijing before replacing it. From what I’ve been able to piece together from studies and technical reports, it seems like one mask should last you at least 5 days, maybe even a month unless it’s clearly physically damaged. But most importantly,
- No one needs ten disposable masks when there are people spending all day outside with no masks. Instead of stockpiling enough to throw a good mask away every day, give your extras to people who outside without one, especially homeless people.
- If you’re spending more than $20 on disposable masks for yourself, you should probably just get a reusable one anyway. Make sure it’s actually NIOSH-certified instead of just claiming it “meets N95 standards”. An approved facepiece + pair of P100 filters is $25-$35 on Amazon currently (you might have to buy the filters separately). Speaking of government certifications, make sure your air purifier isn’t generating ozone and making it worse!