August 28, 2020

Gaitergate, explained

NCSE Executive Director Ann Reid goes past the headlines to tackle the question: is wearing a neck gaiter as a mask worse than wearing no mask at all?

Runners

Check out our entire series explaining the science involved in the coronavirus pandemic. Sign up to receive our coronavirus update each week.

No more than a couple of hours after the Washington Post published this article about the efficacy of various masks, an NCSE staffer posted on Slack (yes, of course we have a Slack channel at NCSE; how did you think we communicate, semaphores?):

“Turns out, the gaiters [my spouse] wears on her morning walks and during our hikes break up her large particles into small ones, that linger longer in the air. Thank goodness she hasn't been sick, because she would have been spreading the virus. All those runners/hikers/bikers need to switch to the thicker cotton fabric masks, which is tough because it's really hard to breathe through them during exertion.”

If a savvy NCSE staff member was alarmed by the Post’s story, we can be sure that lots of other people were too. Arguably, the whole point of the article (or at least the headline) was to alarm — even though, as it turns out, the researchers who did the study never claimed that their work should be considered definitive about how well different masks stop the spread of the coronavirus! The researchers did, however, indulge in some speculation that got out pretty far ahead of what they’d actually shown.

Reading the introduction to the research study with your class, then following up by examining some of the articles reporting on the study in the popular press, makes for an excellent exercise in thinking critically about media coverage of science. In addition to the Washington Post article cited in the first paragraph, here is a follow-up article written by same reporter. And here is a more deeply reported discussion of what the original research study actually showed.

I do want to pause for just a second to insist that my intention is not to bash science journalists. Science journalists do our world an immense service. However, sometimes the desire to make research results sound more exciting, or more definitive, or more controversial than the results justify lead both reporters and scientists to focus on the “newsiest” finding, even if it’s not the strongest finding. That’s why it’s so important that students learn to read science coverage critically. And boy-howdy does this story ever illustrate the problem.

First, the original Washington Post headline:

Washington Post headline

Your first question might be, “What the heck is a gaiter? They’re those stretchy tube-style thingies that you can sling around your neck and pull up over your mouth and nose when you get too close to someone. Lots of runners and hikers use them because they’re pretty easy to breathe through.

Pop quiz: Find the gaiter!

A neck gaiter and an alligator

Now that you know what a gaiter is, let’s take a look at that original study. Right in the abstract — the summary of a research study that appears at the beginning of all scholarly articles — it is clear that the goal of this study was NOT to assess the efficacy of different kinds of masks. Here’s what it says:

We have demonstrated a simple optical measurement method to evaluate the efficacy of masks to reduce the transmission of respiratory droplets during regular speech. In proof-of-principle studies, we compared a variety of commonly available mask types and observed that some mask types approach the performance of standard surgical masks, while some mask alternatives, such as neck fleece or bandanas, offer very little protection.

The goal of the research was to demonstrate “proof of principle” that a simple and inexpensive measurement technique that the authors rigged up in their lab could detect a difference in the amount of respiratory droplets that passed through various kinds of masks during regular speech. A “proof-of-principle” study only claims to show that its basic premise is sound. In this case, the goal was to show that the technique detected a difference among the masks. And the technique appeared to be successful — clinical grade N95 masks blocked the most droplets, followed by surgical masks, then cotton masks, with bandanas and neck fleeces blocking the fewest droplets.

Can we agree that these results are less than mind-blowing? They are exactly what you would have expected. And, indeed, the fact that the measurement set-up detected what you would expect is exactly what a proof-of-principle study should find! If the authors had found that their “simple optical measurement method” couldn’t detect any difference among the masks, they would have had to go back to the drawing board, right?

The authors did exactly as much as they needed to do to demonstrate that their measurement technique measured something. Specifically, they asked one speaker to say the words “Stay healthy, people” five times. They repeated that test 10 times. Got that? One speaker, repeating the test 10 times. For three of the mask types, they had four different speakers repeat the test ten times. Let’s look at the results. The different kinds of masks are along the x axis and the relative number of droplets on the y axis:

Graph

What do you see? The fitted N95 masks basically blocked all droplets. Surgical masks and two different cloth masks performed only slightly less well. As you move along the x axis, the masks are allowing more and more droplets through. When you get to the far right, you’ll see that the bandana was not that different from no mask at all. And the fleece (aka gaiter) appears to allow even more droplets through than no mask at all.

One more thing to notice. The results with the open circles (surgical, cotton5, and bandana) represent the masks that were tested on four different subjects. What do you notice about them? You should notice that their error bars (which measure the variability of the results) are bigger than the error bars on the solid-circle result for the same mask. This makes sense — when you test the mask on four different people, you get more variation in your results. You can see that the surgical mask performed consistently well on all four subjects. The bandana, however, clearly gave wildly different results across the four subjects. Again, that makes sense. There’s really only one way to put on a surgical mask, whereas everyone ties bandanas differently.

And the gaiter? It was tested on just one person, and the 10 replications gave very variable results. In fact, the level of variability is so large that I would suspect the gaiter result is not statistically different from the no-mask negative control.

And yet, the darn gaiter got virtually all the attention in media coverage of this study. Why? Because the authors and the press couldn’t resist speculating about why it appeared to perform very slightly — or maybe not really at all — worse than no mask at all. In the research paper itself, the authors write:

“We noticed that speaking through some masks (particularly the neck fleece) seemed to disperse the largest droplets into a multitude of smaller droplets.…which explains the apparent increase in droplet count relative to no mask in that case. Considering that smaller particles are airborne longer than large droplets (larger droplets sink faster), the use of such a mask might be counterproductive.” [emphasis mine]

The press picked up on “might be counterproductive” and pursued it in all the follow-up interviews. That’s what showed up in headlines, and in the lead-in to radio and TV stories. And that’s really too bad, because the authors did not actually demonstrate that droplets were being broken up into smaller droplets by the gaiter — that is just one possibility. So to then jump to the conclusion that wearing the gaiter might be “worse than nothing”, as most of the press coverage suggested, is unjustified. Maybe gaiters are worse than nothing, maybe they’re not; this study really can’t tell you that.

It’s too bad that unsupported conclusion stole all the thunder, because the rest of the results, while not exactly headline grabbing, show that the authors’ original idea — to develop a cheap test to compare the relative performance of different mask designs — looks pretty sound. If I were looking into producing or buying masks for my workforce or family, the Poly/Cotton and the PolyProp masks perform nearly as well as a surgical mask. Those are the ones I’d want my volunteers sewing up for me, not, for example, the Cotton 3 design. And, although it saddens me as a devoted knitter, it’s pretty clear that knit masks might not be the best way to go.

OK, that’s about it for gaitergate. If a runner goes by you wearing a gaiter, this study provides no evidence to suggest that you’re in more danger than if she’d not been wearing a mask at all. Should kids be wearing gaiters (or bandanas, for that matter) to school where they’ll be in close proximity to lots of people indoors for hours? No. No they shouldn’t. Get a multi-layer fabric mask that fits. And follow our tips for helping your child stay in the mask all day.

NCSE Executive Director Ann Reid

Ann Reid is a former Executive Director of NCSE.

reid@ncse.ngo