Covid-19: The Mask Dilemma

What are the types of mask?

First of all, lets recognize that when people talk about masks to prevent the transmission of Covid-19 they are not all talking about the same thing. There are two main types of mask:

1. N95 masks. These were originally designed for industrial use, but were later adopted for medical use. These are tight-fitting masks designed to stop the inhalation of most particles. (Depending upon the national standard used, these may also be referred to as KN95 or FFP2 masks). There is a saying about wearing them: if it's comfortable, you are wearing it wrong.
2. Surgical masks. These are the loose-fitting type you typically see in TV hospital dramas when the surgeon is performing a life-saving operation. These have two functions: (a) to reduce the chance of the surgeon infecting the patient by breathing out infectious particles, (b) to protect the surgeon from blood or other bodily fluids spurting into their nose or mouth.

There are also Surgical N95 masks (which pass both the approval tests for N95 and for surgical masks), and industrial N95 masks with relief valves where the user breathes out through an unfiltered valve for easier breathing — not something that's a good idea for preventing infection. (3M has a useful information sheet on their range of masks and their intended use.)

So if you see someone wearing a surgical mask you should thank them for doing something which primarily protects you from infection, if they are wearing an N95 mask it should help you both, but if they are wearing the industrial type N95-type mask with a relief valve on the side they are only thinking of themselves.

Concerns with masks

Masks obviously do something, but there are some concerns:

• Masks are only designed to stop larger droplets. The virus responsible for Covid-`9 is small enough that it could penetrate a mask, although most transmission is believed to be by respiratory droplets >5µm. (The 95 refers to the percentage of particles of size 0.3 µm that will be filtered.)
• Medical tests on mask efficacy have typically been performed on influenza viruses, a different pathogen with a different size.
• It's easy to wear masks wrongly, or take them off in a way that will infect the wearer. (Yes, medical staff actually get trained how to put then on and how to take them off correctly).
• When used for protection against a biological hazard, the mask must be discarded or sterilized after use.
• Covid-19 can stay on face masks for up to a week.
• Initially it was thought likely that if someone was asymptomatic, they were not infectious, although there was no evidence for or against this.

So it's entirely legitimate to point out that masks aren't perfect, are of unknown efficacy, and may give mask users a false sense of security. In addition at the time advice was being given they weren't definitely known to be necessary unless you had symptoms of Covid-19, or were likely to come into contact with people who have symptoms of Covid-19.

But if the objective is to just reduce the Effective Reproduction Rate (R) of the virus — to flatten the curve, not to reduce transmission to zero, surely if everybody wears a mask it will help? Even if some people will wear their masks incorrectly and get infected as a result, that doesn't mean all people will or that those who wear a mask incorrectly will be worse off than if they hadn't worn a mask at all.

Where's the dilemma?

So why not just apply the precautionary principal and advise everybody to wear masks?

To understand the ambiguous and changeable advice governments have been giving, I think you have to look not just at whether wearing masks reduces virus transmission (they clearly do), but also look at the possible effects of the advice itself.

If the government advises everybody to wear a mask, what happens?

• First, everybody rushes out to buy masks.
• Normally only health care professionals and people in hazardous workplaces buy masks. The mask industry is sized for this level of demand.
• Initially there won't be enough masks around, so prices will go up.
• In a market economy, masks go to those willing to pay the most for them. This isn't (unfortunately) health care providers, workers in hazardous workplaces, workers in care homes, and other people who clearly have the greatest need. It is the countries and people with the most money, and those trying to speculate on the rise in the price of masks.
• Eventually, production will increase. New companies will start making masks, and, in a competitive market, prices will fall to a level representing the cost of production plus a reasonable profit.

The key word here is eventually. How long will this be?

In 2016, the South China Morning Post estimated world production of masks (type not specified) as 25 million masks per day.

If we assume 20% off the people in the work need on average just one mask a day, then production needs to increase to about 1.4 billion masks / day. That's a fifty fold increase in world production. That's a big increase in production.

If you are not familiar with production and logistics, imagine the difference between inviting six friends to dinner and inviting 300 friends to dinner: you can't just increase the amount of groceries you buy and spend a bit longer in the kitchen. You will need to build a temporary kitchen, bulk order foods from butchers and bakers, arrange transport, hire extra chefs, install extra toilets, and so on.

It's the same with scaling up manufacturing.

You can't just ask a few people to work longer. It's setting up and training people to work on new production lines, sourcing new equipment, sourcing new suppliers, and so on. One of the key materials in masks is a melt-blown non-woven fabric. There's a limited number of suppliers, with limited capacity. It needs custom equipment for production. So once all the existing production lines are running full out, and it will take some time before capacity can be increased.

By February 2020, China had managed to increase production by a factor of twelve to 116 million masks per day, but that was still insufficient to meet world demand. In March, they exported almost 4 billion masks. At the same time the price on the open market for melt-blown non-woven fabric went from US$2,825 / tonne to US$70,000 / tonne.

So issuing advice for the general population to wear masks::

• reduces the effective reproduction rate by an unknown amount (assuming a significant number of masks are available),
• creates shortages of masks where are they are most needed,
• increases the price of masks significantly, making them less affordable for those who need them to look after patients,
• creates panic among those who cannot afford or cannot get masks,
• and may raise questions about the recommendation due to a lack of scientific evidence.

So what is the solution?

It's a tricky dilemma. Advising general mask wearing makes things better for some, but much worse for others.

I think concern about the consequences described above explains the official recommendations that masks were only needed by people in high risk of coming into contact with Covid-19 patients and the patients themselves. As production of masks increased, and as more people wore masks tanyway, the downside of advising against their general use began to disappear.

• The WHO changed it's recommendation on 3rd April
• The government of Singapore also changed it's recommendation on 3rd April
• The CDC recommended wearing non-medical cloth masks on April 3rd.
• Various US states and cities introduced laws requiring face masks in the following weeks.
• On 27 April the German government brought in national rules requiring masks to be worn in shops and on public transport.

I cannot say if concern about the unintended consequences of a sudden demand for masks was the reason behind the suggestions that masks were unnecessary and would not help. However, I can say that this is the only rational reason I can think of not to recommendation that everybody should wear masks, and that the reasons offered for not recommending mask wearing did little for anyone's future credibility.