Why N95 masks may not be as protective as you think

During the pandemic, the prevailing thought has been that a high-filtration mask—such as an N95, KN95, or FFP3—should offer better protection than a surgical or fabric mask. However, recent research I conducted with colleagues at the University of Cambridge suggests this isn’t necessarily the case. For these masks to be highly protective, they must properly fit the wearer, but often they don’t.

This is itself an issue, but our research also discovered another worrying thing: People aren’t very good at assessing whether their mask fits. Frequently, they use a self-assessment method to check fit, which we found to be unreliable. This suggests people may be acting as though they’re wearing a well-fitting mask when they aren’t, which increases the risk of them catching and spreading the coronavirus.

Given this, we decided to conduct a second piece of research, to see if we could improve the fit-checking method. This resulted in us creating a potential new way of testing whether a mask fits that can be performed at home with materials costing approximately $28.

Good fit isn’t universal

Our first piece of research measured how well different masks blocked ambient air particles from entering them when being worn. This involved taking continuous measurements of the concentration of air particles in the ambient air outside the mask as well as in the air inside the mask (via a valve connected to it) and then looking at differences between the two to see if air was seeping in.

This is what’s known as quantitative fit testing. It’s a commonly used and highly reliable way to test mask fit. The U.S. Occupational Safety and Health Administration (OSHA) mandates that for an N95 or similar mask to be deemed suitable for use, it must score above 100 in such a test.

We ran quantitative fit testing with seven participants, who each tested five different types of N95 mask, as well as a KN95 mask and a surgical mask. Three participants also tested some fabric masks. This meant N95 masks were tested a total of 35 times; only in six instances did their scores exceed 100. When poorly fitting, we found the N95 masks provided not much more protection than the surgical or cloth masks. The KN95 masks also performed very poorly when they didn’t fit, offering no more protection than a cloth or surgical mask.

It therefore can’t be assumed that high-performance masks are universally highly protective. Proper fit is necessary for protection, and even the best performing mask, the 8511 N95, fit only three out of seven participants. Some N95 masks, such as the Aero Pro and Xiantao Zong, didn’t fit any of the participants adequately.

Self-assessing fit often inaccurate

Before measuring quantitative fit, we had also had the participants themselves assess whether they thought each mask fit (what’s known as qualitative fit testing). They did this by following the U.K.’s National Health Service (NHS) guidelines: visually and manually inspecting the mask each time, carefully feeling around the edge for air leaks, and adjusting the fit as necessary. We then had participants rate each mask on two criteria—whether they believed the mask fit and how confident they were in their decision.

We then compared these self-assessments with the highly reliable, quantitative fit data that we gathered. In the linked graph, you can see how the actual and estimated fit of the five N95 masks compared across the seven participants. (There were four women and three men, aged between 18 and 74. They are listed on the X axis.)

The height of the bars shows how well the masks fit—as mentioned previously, only six times did an N95 mask fit well enough to pass the OHSA’s 100-point benchmark. The color of the bars shows what was predicted in each self-assessment: green means the participant thought it did fit, red that it didn’t, and the darker the color the greater the confidence in their prediction.

If predicting fit accurately, bars above the 100-point mark are green and those below are red. The participants weren’t particularly accurate in assessing the fit of the various N95 masks. They correctly identified masks that did fit, but regularly misidentified poorly fitting masks as being adequately able to protect them. There’s also no correlation between their confidence in their predictions and the predictions’ accuracy.

These findings suggested that the self-administered method of checking fit used in this study isn’t reliable and that an alternative is needed.

How to improve self-assessment

We then had a go at developing a better method of qualitative fit testing that could be performed at home for a reasonable price.

In healthcare settings, a different method is used to test high-filtration masks. Rather than assess fit visually and by touch, a substance with a certain taste is released into the air around the mask wearer. If they can taste the substance, it indicates the mask doesn’t fit properly.

This technique requires a suitable enclosure, as well as a flavored test solution and a nebulizer to spray it into the air. These supplies can be expensive and, during COVID-19, difficult to obtain. So, we showed how to replicate this method at home using inexpensive alternatives purchased on Amazon for around $28.

An aroma diffuser can replace the nebulizer, and can be used to aerolize saccharin dissolved in water into an enclosed space. The participant then tests to make sure they can taste this without a mask on, and then sees if they can taste it with a mask donned. To ensure a thorough test, the user can move the diffuser around so that its vapor touches all edges of the mask, or they could cover their head with something—such as a densely woven towel—to help concentrate the mist around their face.

It’s important to note that while this method was shown to be effective in our study, it is not regulator-approved, and so should be used at one’s own risk. We do not certify its safety.

Eugenia O’Kelly is a PhD Candidate in the Department of Engineering at the University of Cambridge. This article is republished from The Conversation under a Creative Commons license. Read the original article.