by Marcel Müller (M.Sc. Biological Sciences)


Epistemic Status: I am not a virologist and this is not medical advice. That said I am fairly sure about the core claims of this piece, though there may be unknown unknowns I overlooked. I think that among everything I have heard so far this is by far the most promising strategy to eradicate SARS-CoV-2 without crippling economic damage and / or millions of deaths.

You have permission to distribute this wherever you think it might do good and I actively encourage everyone to do so.

This is written form a European/German perspective since I live in Germany but should apply in most places with few changes necessary.


SARS-CoV-2 emerged as a novel pandemic virus threatening to cause a world wide health and economic crisis. While measures have been put in place to temporarily slow the spread of the virus, a workable, economically feasible, long term strategy is needed to deal with the situation, to avoid both millions of deaths and severe economic disruption.

All strategies for dealing with the pandemic need to rely on controlling the spread of the epidemic by reducing either by generating immunity or by reducing the number of contacts within the population.

Therefore it is crucial to discuss the best ways of reducing . Economic lockdowns, school closures and social distancing are measures that were easy to implement but not necessarily optimal.

I propose the widespread use of reusable high quality (European P3 standard equivalent) particle filtering masks in both healthcare and community settings to reduce the risk of infection during contacts sufficiently to allow lifting of most other restrictions while still achieving continued exponential decay of new case numbers.


In the last months of 2019, SARS-CoV-2 emerged as a novel zoonotic virus on a wet market in China. It quickly spread through the country and then through the rest of the world, threatening to overwhelm health care systems worldwide. Most nations reacted by shutting down large parts of their economy, closing borders, and mandating different curfew variations. Some countries managed to significantly reduce the effective reproduction number , which caused exponential decay of their new case numbers, though at great cost to their economy. It is unclear how long these measures can be sustained before adequate supply of the population becomes questionable. [1] Unfortunately, lifting of these measures will return close to its original value again, leading to renewed exponential growth as long as there is no significant immunity in the population.

Several strategies have been proposed to deal with this problem:

Flattening the curve: Here the idea is to let the epidemic run its course at reduced speed until enough people have been infected to attain herd immunity. [2] Proponents of this idea hope to fine tune the rate of infection to the capacity of the health care system for critical care and especially ventilation. Since according to current estimates R0 of SARS-CoV-2 is about 3, herd immunity is achieved when about 70% of the population have been infected.

This strategy has two problems:

1. Sufficient flattening would be extremely difficult to achieve, since even in Germany with its high density of critical care facilities and under very favourable assumptions (all critical care just for CoViD patients, 2% critical care patients, 10 treatment days) would have to be maintained between 1 and 1.25 for about a year to not overwhelm the healthcare system. [3,4]

Under less favourable assumptions and with less critical care density, even tighter control of would be necessary and the timeline would quickly expand to decades as discussed in [5].

2. Even if sufficient flattening is achieved, this approach would cause widespread death and disability among the affected population. With the best medical care possible CoViD 19 kills at the very least 0.4% of the infected population [6] and a few % require ventilation to survive [7]. In Germany alone this would result in 230000 dead and about 2 million on ventilation with all the morbidity and disability that follows. Applied to 5.25 billion infected (70% of the world population) this would mean 21 million dead and about 200 million on ventilation in the absolute best case. But since these numbers apply to a situation with critical care for everyone who needs it, which is, as demonstrated above, completely impossible to provide with that many cases most of the people who need ventilation would probably die.

Quick vaccination: Under this strategy, all measures are left in place until a vaccine has been developed and produced in sufficient quantity to vaccinate most of the population.

Here the problem is that even in the best case, development and safety testing of a vaccine will take at least 12 to 18 months [8] and it will likely prove to be economically infeasible to leave the current lockdowns in place until next year. So the only possibility would be to forgo all but the most rudimentary safety testing with the hope of having a vaccine available this fall, thereby risking severe adverse reactions in an appreciable fraction of the vaccinated population with somewhat unclear benefit. While the outcome would probably be a lot better than with the „Flattening the curve“ strategy, at least currently it seems no state is willing to risk this.

Hammer and Dance: Here the „Hammer“ of the above described lockdowns is used to drive below 1 until the number of currently infected falls below a manageable number. This is followed by the „Dance“, where regulations are relaxed far enough to allow at least basic economic activity, while at the same time maintaining slightly below 1 so renewed exponential growth of new case numbers is prevented until a vaccine can safely be deployed to most of the population. [9]

The main problems with this approach are:

1. How many countries are able to impose sufficient measures to halt the spread of the epidemic in the first place?

2. Is it possible to guarantee continued economic functioning to supply the population with basic necessities while keeping below 1 for at least 18 months?

Contain and Eradicate: This strategy looks in many ways similar to Hammer and Dance. The difference is that after the “Hammer” phase, travel restrictions are left in place and every remaining infection chain is contained through extensive tracking and testing until the virus is eradicated from the population. While not relying on vaccination, this strategy shares every problem with “Hammer and Dance” but is even more demanding regarding the comprehensiveness of the testing and tracking regime and the extensiveness of travel bans needed [9]. Even a single uncontained case could lead to a full resurgence of the epidemic.

An approach missing from these discussions is the widespread use of high quality, reusable particle filtering masks which should, as demonstrated below, be a very promising strategy to drive below 1 and ultimately eradicate SARS-CoV-2 from the population.While this idea shares many similarities with both “Hammer and Dance” and “Contain and Eradicate”, the big advantage is that the demands on testing, tracking and continued reduced economic activity are much less severe than in the above scenarios.

Current discussion on mask usage

There has been much discussion about the use of masks to prevent SARS-CoV-2 infection [11]. Many experts, including the american CDC and the WHO, regard masks in community settings as not effective at all or only marginally effective against other aerosolized viruses such as influenza [12]. They also recommend against their use for self protection in the case of SARS-CoV-2 [13] though some contradictory studies exist [14]. This stands in stark contrast to both the expectation derived from principles of biology and physics and decades of experience in both health care [15] and lab settings [16]. To explore why this difference arises, a closer look at the different types of masks available is necessary.

Different kinds of masks

There are three different types of masks that could be used for infection control purposes:

1. Surgical Masks:

Cloth and surgical masks, which are not designed to form a seal on the face, and thus are unable to filter all the air inhaled by the wearer. While larger droplets on a direct path are stopped by these masks, aerosol particles that move primarily under the influence of air currents can gain access to the wearers airways by entering the gap between face and mask. Depending on the exact material used, small aerosol particles may even penetrate the mask itself [17].

2. Disposable “Filtering Facepiece Particle” (FFP) masks:

Disposable masks, built with a face piecemade from a filtering material which is designed to form a seal on the face of the wearer. These are available in different grades, which differ on both seal quality and filter quality, with the highest quality generally deemed suitable for use against aerosolized and even airborne pathogens [18]. In Europe these grades are FFP1, FFP2 and FFP3 [18].

A big problem with these masks is that they need to be fitted to the user by bending wires in the mask and adjusting various straps, which many people, even trained medical personnel, often fail to do properly [19]. Also, not every mask can be fitted to every face and a filtering mask with a compromised seal is no better than a surgical mask since in both cases unfiltered air can enter from the sides. These are the masks almost universally used in health care for protection against infection with SARS-CoV-2.

Another downside of these masks is that the filtering element – the facepiece itself – is continuously exposed to the breath of the wearer, which soaks the filter with moisture. In the presence of liquid water it is possible for virus particles bound to filter fibers to become resuspended in this liquid. Since all particle filters used in masks are not membrane filters but adsorption filters with pore sizes much larger than the particles to be removed from the air [18], these resuspended virus particles can diffuse across the filter to the inside of the mask as soon as a continuous water column is formed across the filter material. Now they can either be re-aerosolized by the wearer's breath or taken up by lip contact to the inside of the mask. This necessitates frequent changing of damp masks. Since it is unclear if these masks can be cleaned and reused without significant drop in filter performance, the continued supply of the health care system with a sufficient number of masks is difficult and supply of the general population for everyday use is a logistical impossibility.

3. Reusable masks with replaceable filters

This category contains masks which consist of a gas impermeable mask body, typically made from silicone or rubber and one or two separate filter cartridges which fit into filter ports in the mask body. These filters are available in the similar qualities as the above disposable masks, called P1, P2 and P3 in Europe [18].

A properly fitting mask of this type is very easy to use compared to an FFP mask, since no delicate adjustments have to be made. Also fit testing with these masks is very easy since closing the filter ports while wearing the mask makes inhaling impossible if the mask seals properly.

Another big advantage is that these masks generally have a one way valve protecting the filter from the wearer's breath, which is discharged via a separate exhaust valve. This prevents the filter material from being soaked by the wearer's breath and thus prevents the diffusion of virus particles across the filter. Since these filters are generally designed to be worn for a couple of hours in environments with very high dust load (grinding wood or stone, spray painting etc.) these filters should easily last for days, weeks or even months worn in a relatively clean health care or community setting. While this is generally not recommended in the manufacturers guidelines, as long as the filter remains dry it should be quite safe [18]. A common misunderstanding is that full dust filters begin to leak. While this is true for gas filters when their absorption capacity is exhausted, dust filters do not leak, since the free space between internal surfaces available for absorption of particles does not increase but decrease, causing them to clog, i.e. resistance to air passage starts to rise which indicates that replacement is necessary.

Effectiveness of high quality particle filtering masks against SARS-CoV-2 infection

As mentioned above, on the one hand there has been some research on efficiency of different masks against community based dissemination of typical droplet and aerosolized infections like flu, rhino viruses and SARS-CoV-1 with mixed results.

On the other hand, there is ample precedent for successful use of well fitted P3 masks even against dry airborne threats like anthrax [20]. The known properties of the SARS-CoV-2 virus suggest that it should be even less likely to penetrate such masks.

Preliminary evidence suggests that SARS-CoV-2 becomes nonviable when dried out [21]. This also fits prior expectation since SARS-CoV-2 has a lipoprotein envelope, which should be denatured by desiccation. This means that very fine particles around 2.5 µm, which are most likely to penetrate a high quality particle filtering mask, should already be dried out at the point when they might be inhaled under most plausible circumstances. And even these particles are retained to 99,95% in a P3 filter with filtering efficiency steeply rising with larger (or smaller!) particle diameters [18].

This forms a stark contrast to the mixed results found for mask use against flu transmission in community settings.

This apparent contradiction is, in the case of FFP masks, easily explained by the general observation that they are often worn in a way that compromises their seal to the wearer's face even if worn by trained professionals [19]. In the case of surgical masks, their inability to form a seal in the first place makes them incapable to reliably protect against droplets small enough to move with air currents (single to double digit µm scale) [17]. These problems are further exacerbated by compliance issues, especially when people are asked to wear masks in their own household [22]. This makes the whole body of research conducted with surgical and FFP masks in community settings highly suspect if applied to easy to use, preformed, reusable masks with P3 filters.

Infection via intact or even damaged skin is highly implausible for SARS-CoV-2 and has never been observed. Food borne spread, while not completely implausible, has not been observed to play an appreciable epidemiological role in SARS-CoV-2 [23]. This leaves direct targeting of the respiratory system via droplet and aerosolized infection and possibly smear infection to mouth and eyes as the only relevant transmission routes. Transmission via both of these should be severely curtailed by the use of a P3 standard equivalent mask. If the mask needs to be worn with spray tight eye protection or not is currently unclear, since it is currently unknown whether or not infection can occur through droplets entering the eye. A plausible mechanism for this exists, since liquids in the eye are evacuated into the nasal cavity via the nasolacrimal duct.

Therefore, until more research is conducted with well fitted P3 equivalent masks, we should go on to assume that at least P3 masks, worn properly, with appropriate eye protection while maintaining basic hand hygiene are efficient in preventing SARS-CoV-2 infection regardless of setting. The extraordinary evidence required to accept the quite extraordinary claim that a well fitted P3 mask does not prevent SARS-CoV-2 infections in community settings does not exist.

A policy proposal for the eradication of SARS-CoV2

This is a broad outline of a policy proposal that should be able to eradicate SARS-CoV-2 from a given population within a couple of months to a year while allowing nearly complete economic functioning over much of this period:

Over the next few months, infection numbers are maintained as envisioned under the conventional “Hammer and Dance” scenario, while at the same time public funding is used to build up a large scale production capability for 5 to 10 different mask body types to fit most faces, matching P3 equivalent filters and some form of eye protection with the aim to be able to supply most of the population with one properly fitted mask body and 3 to 5 matching P3 filters as well as spray tight eye protection.

While this is no trivial matter, it should be well within the capabilities of industrialized and most emerging economies. Prior to the Covid-19 pandemic, consumer prices for a half mask body were about € 30 to € 60, a P3 filter about € 5 to € 10 and protective goggles about € 20, forming an upper bound of € 100 per person with significant room for improvement through economies of scale. This suggests costs which are only a small fraction of the costs already incurred and still to be expected. Since mask bodies, valves and filter casings can be produced by simple injection molding, quick scale up of production capabilities should be possible by retooling of existing production lines.

Also the availability of raw materials should be no problem since a broad range of elastomeric polymers are suitable for the production of mask bodies, many of which are used in large quantities for other purposes. For example silicone (medical and industrial applications, gluing and sealing), Polyurethane (industrial and building applications, gluing and insulating) and butyl rubber (tyres and industrial applications).

As soon as masks are produced in large numbers they are issued to the population, beginning with health care workers and public facing employees with the directive to wear mask and eye protection whenever people not belonging to their household are (or have recently been) in the same room or within 5 meters.

This cuts down on wear time and thus reduces filter degradation and wearer exhaustion compared to wearing the mask at all times while outside the home without generating much risk of infection. To avoid possible infection via surfaces, everyone is instructed to carry a small bottle of 80% ethanol to disinfect hands

  • before and after putting on the mask or taking it off
  • touching the unmasked face for other reason
  • eating

Ethanol is available in sufficient quantities, since large amounts could be diverted from E5 and E10 fuel production.

Even though this is not standard practice, for the reasons described above (filters remain dry, thus virus particles bound to the filters will stay there [18] and degrade over time) filters can be used for 2 to 4 weeks unless severe contamination with liquids or large amounts of dust has occurred. After use the mask body can be cleaned with soap and water and disinfected with the same alcoholic solution used as hand rub to maintain hygienic conditions. This greatly alleviates both production and logistic demands and allows to supply most of the population in a comparatively short amount of time.

Since people following the policies outlined above are much less likely to become infected, they are also much less likely to pass the infection on to other people much in the same way an immunised person no longer acts as a host to the virus. Thus as soon as about 70 % of the population follow this protocol, an “artificial herd immunity” will be attained and the number of active infections will start to decline even in the absence of additional measures or much earlier if combined with other measures. Thus it is not necessary to provide literally everyone with a mask for this to work. People medically incapable of wearing a mask would also be protected by this effect.

This regimen could be continued with little economic damage until SARS-CoV-2 is eradicated from the population or a safe vaccine has been developed.


Many thanks to Michael Albert, Simon Fischer, David Gretzschel, Stefan Heimersheim, Ursula Korten-Schmitz, Ulrike Mazalla and Dr. Gerd Schmitz for their valuable input.


  1. Wikipedia “2019-2020 Coronavirus Pandemic”–20_coronavirus_pandemic Retrieved 11.04.2020
  2. Stevens H. “Why outbreaks like coronavirus spread exponentially, and how to “flatten the curve” Washington Post Retrieved 11.04.2020
  3. Deutsche Gesellschaft für Epidemiologie (DGEpi) “Stellungnahme der Deutschen Gesellschaft für Epidemiologie (DGEpi) zur Verbreitung des neuen Coronavirus (SARS-CoV-2)” Retrieved 11.04.2020
  4. Science Media Center Germany “Auslastung der Intensivstationen: Zahlen aus Deutschland und Europa” Retrieved 11.04.2020
  5. Bach J. “Don’t “Flatten the Curve,” squash it!” Medium Retrieved 11.04.2020
  6. Streek, H. et al. “Vorläufiges Ergebnis und Schlussfolgerungen der COVID-19 Case-Cluster-Study (Gemeinde Gangelt)” Retrieved 11.04.2020
  7. Phua, J. et al. “Intensive care management of coronavirus disease 2019 (COVID-19): challenges and recommendations” The Lancet Retrieved 11.04.2020
  8. Kuznalla, R. “The timetable for a coronavirus vaccine is 18 months. Experts say that's risky” CNN Retrieved 11.04.2020
  9. Pueyo, T “Coronavirus: The Hammer and the Dance” Medium“ Retrieved 11.04.2020
  10. Chen Shen, Nassim Nicholas Taleb and Yaneer Bar-Yam, Review of Ferguson et al "Impact of non-pharmaceutical interventions...", New England Complex Systems Institute (March 17, 2020) Retrieved 11.04.2020
  11. Alexander, S “Face masks, much more than you wanted to know” Slate Star Codex Retrieved 11.04.2020
  12. Centers for disease Control and Prevention “Interim Recommendations for Facemask and Respirator Use to Reduce 2009 Influenza A (H1N1) Virus Transmission” Retrieved 11.04.2020
  13. Centers for disease Control and Prevention Retrieved 11.04.2020
  14. Zhang, L. et al “Protection by Face Masks against Influenza A(H1N1)pdm09 Virus on Trans-Pacific Passenger Aircraft, 2009” Emerging infectious diseases Retrieved 11.04.2020
  15. Centers for disease Control and Prevention “Respiratory Protection in Health-Care Settings” Retrieved 11.04.2020
  16. Wikipedia “Biosafety Level” Retrieved 11.04.2020
  17. Makison Booth, C. et al “Effectiveness of surgical masks against influenza bioaerosols” Journal of hospital infections Retrieved 11.04.2020 Retrieved 11.04.2020
  18. M3 “Respiratory Protection for Airborne Exposures to Biohazards” Technical Data Bulletin Retrieved 11.04.2020
  19. Sutton PM et al “Tuberculosis isolation: comparison of written procedures and actual practices in three California hospitals.” Infection Control and Hospital Epidemiology Retrieved 11.04.2020
  20. OHSA “Anthrax, Control and Prevention” Retrieved 11.04.2020
  21. Streeck, H. "Einzelne Übertragungen im Supermarkt sind nicht das Problem" Zeit Online Retrieved 11.04.2020
  22. MacIntyre C. “Facemasks for the prevention of infection in healthcare and community settings” BMJ 2015;350:h694 Retrieved 11.04.2020
  23. Bundesinstitut für Risikobewertung “Can the new type of coronavirus be transmitted via food and objects?” Retrieved 11.04.2020

New to LessWrong?

New Comment
8 comments, sorted by Click to highlight new comments since: Today at 3:39 PM
[...] P3 masks, worn properly, with appropriate eye protection while maintaining basic hand hygiene are efficient in preventing SARS-CoV-2 infection regardless of setting.

If this is true, then this is a great idea and it's somewhat suprising that these masks are not in widespread use already.

I suspect the plan is a bit less practical than stated, as I expect there to be problems with compliance, in particular because the mask are mildly unpleasant to wear for prolonged periods.


Thanks for pointing these things out, I probably should have adressed them more.

... suprising that these masks are not in widespread use already.

I could think of several reasons for this.

  • Many (most?) health care professionals do not know of these masks or do not think of them as "medical equipment".
  • People do not realize that filters can be used multiple times thus dismissing the idea as logistically impossible / even more expensive than FFP masks for everyone
  • People think that all masks do not work (well) to prevent transmission
  • People think that these masks are "overkill", not realizing that a well fitting!!! reusable silicone mask is actually much less unleasant to wear than FFP masks.

... problems with compliance ... unpleasant to wear for prolonged periods.

Yes, to a degree that is true. This should be addressed by...

  • Well fitting masks, at least 5 to 10 different types as discribed above with state of the art low resistance filters
  • Requiring people to wear masks only if there is actual risk of infection as described above
  • Rigorous enforcement especially in places where there are lots of people around (public transit, dense work places, schools and so on)
not realizing that a well fitting!!! reusable silicone mask is actually much less unleasant to wear than FFP masks.

Do you have more information about that? How do I go about getting such a well-fitting reusable silicone mask for myself?


Well, difficult at the moment. I would try the M3 6000 or 6500 series first since they are available in 3 sizes. But you basically have to shop around until you find something, which is difficult when most shops are sold out.

The outlook of kids with a respiratory is kind of scary, as shown in HK. I did not see any other issue with this strategy.

Sufficient flattening would be extremely difficult to achieve

The range for effective R0 in which the virus spreads to the point of herd immunity in a reasonable time, without blowing out the hospital system and without infection rates collapsing, is extremely narrow.If you want to be very sure to avoid a blowout you have to basically be quite likely to produce a collapse in # cases. Maybe you could turn the isolation on and off but that is also very tricky.

Even if achieved, flattening would be a protracted process, lasting 1-2 years, with all that implies in terms of the economy.

As OP mentioned, the cost in lives lost would also be high for a flattening where 60+% of the community is infected. People argue that "only" old people/boomers are killed but... 1) the average years of life lost seems to be substantial (~11 years). This is not people dying a couple of months early 2)

See the modeling done for the Australian government here - if you read between the lines you can see this confirms the above points, which originally came from a simple spreadsheet I did.

Finally, daily reminder that Taiwan has very few cases, one death so far in April, without shutting down their economy. This idea that you have a dilemma of health versus economy is quite false. The dilemma is competence versus stupidity.

On the masks issue, OP's proposal seems plausible, reducing effective R0. I am working on a program to model this and I will add this option and see what happens. It is a pity the authorities lied to us about masks, in order to preserve them for those with greater need when they were in very short supply.

In my program I assume

Fraction of people compliant with the masks policy(asymmetric-distancing-fraction-compliant-fv) 0.7
Fraction of infection that still gets through from mask wearer to other person (asymmetric-distancing-outbound-ineffectiveness) 0.3
Fraction of infection that still gets through from non-mask-wearer to mask-wearing person (asymmetric-distancing-inbound-ineffectiveness) 0.8

Given this, and numerous other assumptions including no other measures taken, the death rate falls from 0.65% to 0.48% of the population. This is a good benefit but not a total solution.

If you have better numbers for mask effectiveness than the ones I guesstimated above please let me know.

The other main dubious assumption in my model (other than no other measures taken) is uniformity of people. I am adding some options on that tomorrow.

Note: the CDC now recommends public mask-wearing, even DIY cloth masks, (reversing their previous advice):

Please promote #Masks4All wherever you can: