Which masks best stop the spread of COVID-19? Duke researchers tested them to find out
Dr. Eric Westman was helping a Durham nonprofit organization provide free masks to people who need them when he decided he should find out whether the masks would really prevent the spread of coronavirus.
But it wasn’t simply a matter of searching online to see which masks worked best. This was April, and at that time researchers hadn’t tested the variety of home-made masks and improvised face coverings that most people are wearing now, Westman said.
“No one had studied that before,” he said in an interview. “The situation was new.”
So Westman, an associate professor at Duke University’s medical school, asked the Duke physics department if it could test various types of masks and face coverings for him.
The results, published Friday in the journal Science Advances, not only helped the nonprofit Cover Durham choose the right masks, Westman said, they also show that testing the effectiveness of face coverings is not all that difficult.
And to the surprise of the Duke researchers, the study also found that some types of coverings are worse than wearing no mask at all.
Martin Fischer took on the task of devising the test. Fischer, who directs the Advance Light Imaging and Spectroscopy facility at Duke, used a method developed by researchers at the National Institutes of Health to measure the small droplets of moisture that get expelled from our mouths when we speak. Those droplets can carry coronavirus from a person who is infected.
Fischer and his team had someone speak into the side of a box, saying “Stay healthy, people,” over and over again, first without a mask and then wearing 15 different types of coverings. A laser light passing through the box lit up the droplets in the air, which were then recorded by cellphone video.
Fischer, who usually works with solid materials, said he was amazed at what he saw.
“It was just an eye-opener to see all those particles light up in our laser beam,” he said.
With the help of a computer, the team then counted the droplets visible in the videos to determine which coverings let the most through.
Not surprisingly, the fewest droplets (almost zero) were detected with a fitted N95 mask, the ones used by doctors, nurses and other health care workers in the highest-risk settings. A surgical mask was a close second.
Masks made of polypropylene also did very well, but so did various kinds of double-layer cotton masks. Several of the cotton masks produced less than 20% as many particles as someone wearing no mask at all.
Less effective was the bandana, which let through about 50% as many particles as someone not wearing a mask at all.
But the worst performer was the gaiter, which people wear around their necks and pull up over their faces like a mask. The one the Duke team tested actually made things worse. Rather than preventing the droplets from escaping, it turned the larger droplets into a cloud of smaller ones that hung in the air longer.
“The use of such a mask might be counterproductive,” the researchers wrote in their paper. Or, as Westman put it Monday, “The idea that something’s better than nothing may not be true.”
The paper refers to the gaiter as a “neck fleece,” but Fischer says it was made out of polyester mixed with a little spandex, not cotton fleece. He said it appears that the act of pulling the gaiter up over your face allows particles to get through.
“You can see the material was pretty thin,” he said. “So if you stretch it out over your mouth, it gets even thinner.”
Fischer points out that the Duke researchers tested only one type of neck gaiter. In fact, he said, the study was never meant to be a comprehensive test of all masks and mask materials.
“The focus of the study was to develop a simple technique for mask testing that people could duplicate and set up themselves,” he said. “This is, as far as we know, by far the simplest and easiest way to do this. We set this up on a weekend with stuff that we had lying around in the lab.”
With the COVID-19 pandemic showing no signs of going away, interest in the Duke study has been high. Fischer said he had 130 emails about it by Monday morning, including requests for interviews from newspapers, wire services, CNN, CBS and the BBC.
One email came from a doctor who hopes to replicate the tests in Central Africa. Several came from people who simply want the Duke team to test their masks (Duke doesn’t have the manpower to do that, Fischer said).
Westman said all the attention on the study may also highlight that a person doesn’t need to cough or sneeze to spread coronavirus.
“If you didn’t know that speaking can spread particles that can spread virus, you need to know that,” he said. “That was confirmed over and over and over in the hundreds of trials that we did.”
This story was originally published August 10, 2020 at 12:07 PM with the headline "Which masks best stop the spread of COVID-19? Duke researchers tested them to find out."
