A measure of last resort: Homemade respirator masks in a pandemic

Author(s): Roschelle Fritz, PhD, RN; Marian Wilson, PhD, MPH, RN; and Shawn Brow, MAE, CRNA

Nurses design and test masks to advance innovation during a time of need.

Editor’s note: Commercially manufactured single-use respirator masks are standard of care. Homemade masks should be a last resort option when commercially manufactured masks aren’t available. This article is an early release article and will appear in the upcoming November 2020 issue of the American Nurse Journal.

As the COVID-19 pandemic escalated and a shortage of N95 masks ensued, a big question for many frontline nurses was, “Will I be asked to go into a patient’s room without adequate protection?” At the same time, nurse managers and leaders at all levels were worried about their ability to secure personal protective equipment (PPE) at anticipated levels of consumption.

One Washington state critical access hospital outside of Seattle experienced the stress of not receiving PPE shipments while anticipating possible spread of the virus to their community. In preparation, community members began making homemade masks, face shields, and washable protective gowns and booties for the hospital staff. People dropped off hundreds of masks of various materials and designs. The problem? Nurse leaders and infection prevention staff knew the masks wouldn’t provide respirator-level protection, which limited their use with a virus that could spread via airborne transmission.

As nurse leaders associated with hospitals across the Pacific Northwest, we wondered about the role of homemade masks. We delved into the literature, looking for any evidence about material and design efficacy. Some evidence is now available, but at the time we didn’t find any data supporting the practice of using homemade masks. In response to our concern for frontline staff, we began considering homemade possibilities with respirator-level protection.

Respirator masks are regulated and must block 95% of airborne particles, but we didn’t have equipment to test the number and size of particles passing through the mask. So, our goal was to create a mask, using materials found in the home or purchased at big box stores, that would pass the Occupational Health and Safety Administration’s standardized fit-testing. We thought it might be possible to use such a mask as an emergency stopgap—and only as a stopgap—in the event that clean, intact medical-grade respirators (air-purifying respirator or N95) weren’t available.

Making masks

With the assistance of two seamsters, we developed two masks for fit-testing.

Mask 1

The first mask was made of a tightly woven outer and inner layer, with a removable double layer of 3M Filtrete™ 1500 furnace filter in between. Although 3M doesn’t certify their filter for facemask use, a 3M representative confirmed that the material doesn’t contain fiberglass, a known respiratory irritant. The filter is made of a nonwoven, melt-blown polypropylene material, similar to what’s used as interfacing in typical medical masks. Melt-blown material is made by pushing polypropylene micro- and nanofibers through small nozzles surrounded by high-speed blowing air, causing it to land on a flat surface. When cooled, the fibers stick together. Think cotton candy, but tougher, like a medical-grade mask. We considered using a medical-grade interfacing material but because of the pandemic, a company representative wasn’t available. Our decision to use the 3M product was similar in logic to off-label use in the pharmaceutical industry—a step toward advancing scientific data in mask design. (See Figure 1.)

We cut the filtering material to size and pressed the folds with an iron on low heat. We placed two layers inside a pocket in the mask. A metal wire was inserted so the mask could be molded over the nose bridge, and we used a single toggled shoelace to secure the mask to the face. We wanted to create a comfortable, well-fitting mask that would pass fit-testing, require minimal work of breathing, and be easy to don and doff. To make the masks environmentally friendly, we wanted the insert to be removable and the mask to withstand laundering.

We experimented with several freely available patterns before creating our final design. The Olson mask (made by Clayton Olson and Rose Hedges and distributed by Unity Health, Cedar Rapids, IA and available here) inspired us to consider the filtering pocket. To create a superior overall seal, instead of using ear bands and adhesive tape on the nose bridge, we incorporated a cinching tie and toggle design. We broadened the lower dimension at the chin to create an anchoring action across the cheeks and along the mandible and widened the filter and extended the pocket accordingly. In lieu of the Olson mask’s 0.3 micron-rated vacuum bag as the filter, we used the 3M furnace filter.

Mask 2

The second mask was made of a tightly woven cotton outer and inner layer with a removable insert made of eight layers of t-shirt cotton. A small wire allowed for molding the mask over the nose bridge, and tie straps were sewn to the corners of the mask. We used a disassembled surgical-grade mask as a pattern. The material and number of layers were chosen based on scant literature pointing to eight layers as the sweet spot between protection and breathability. (See Figure 2.)

Fit-testing

We fit-tested both masks on 28 frontline staff at Arbor Health Morton Hospital (a critical access hospital in the Cascades outside of Seattle). We included nurses, physicians, housekeepers, respiratory therapists, and others. Twenty-four women and four men participated. The employee health department assisted with fit-testing, and the hospital pharmacy provided a sweet (saccharine) solution for testing how well the mask filtered out smell or taste. The bitter solution that’s normally used in the hospital’s fit-testing protocol wasn’t available from distributors due to the pandemic.

To fit-test staff, we placed a hood over the participant’s head and sprayed aerosolized saccharine inside the hood through a port near the face. To get each staff member’s smell or taste sensitivity baseline, we sprayed the solution in sets of 10 sprays, until the participant tasted or smelled it. We used the baseline number of sprays for each of the 12 steps in the fit-testing procedure. (See Figure 3.)

We first fit-tested Mask 1 without taping the mask borders. If the mask failed (the participant could taste or smell the sweet solution), we had the staff member tape the entire perimeter of the mask with 1-inch wide pieces of silk cloth tape to create a seal and force air to enter only through the filtered area. We then repeated fit-testing.

After we completed Mask 1 testing, we asked the staff member to don Mask 2. We anticipated Mask 2 would fail fit-testing so we began by first having the staff member tape all perimeters using the cloth tape. Mask 2, with all perimeters sealed by tape, had a 100% fail rate. Twenty-five staff members smelled or tasted the sweet solution in the first three fit-testing steps. We didn’t test Mask 2 without the perimeters taped.

Mask 1 showed promise. It passed fit-testing in 12 of 28 participants. Four participants completed the entire fit-testing protocol without tasting the sweet solution and without having the perimeters of their mask sealed with tape. Eight additional participants passed fit-testing with the perimeters of their masks taped. We believe that more participants may have passed fit-testing if we had taped the mask perimeters ourselves on all participants. Initially, we asked participants to tape themselves because we recognized that in a pandemic frontline staff may have to tape their own masks. However, given that our main focus was to test the efficacy of this homemade mask, after the first 10 participants, we did the taping ourselves. Thereafter, more participants passed fit-testing. This tells us that one of the most important considerations in using this homemade mask is correctly fitting and sizing the mask to the face.

Staff said they liked the feel of the mask’s cotton material against their cheeks. They reported the double layer filter was as easy to breathe through as an N95, and they liked the ability to wash and reuse the mask. An area requiring further design attention was the shoestring tie and toggle. It worked better for people with curly, thick, or coarse hair. The tie and toggle, intended to be placed at the crown of the head, tended to slip in participants with thin, slippery hair. Some staff indicated that N95 masks have more room between the mask and mouth, making talking easier. All those who underwent fit-testing (except for one) wanted to keep the mask after testing, indicating they liked it and would use it again.

Mask-making instructions

We’re sharing the Mask 1 pattern and sewing instructions and encouraging nurses to continue to improve the design and further test its efficacy. We hope a homemade respirator mask will never be needed in clinical practice. At the same time, we never thought that mask conservation policies, such as reusing N95s, would be implemented in U.S. hospitals. Reuse always has been considered substandard infection prevention practice. Also, nurses and other frontline workers aren’t the only ones who may need protection from airborne transmission. Many family and community members working in service industries and home-based caregivers will want to optimize their protection when access to an N95 is limited. In such cases, a homemade mask similar to Mask 1, when properly fitted, might provide better protection than a cotton mask. (See Mask instructions.)

Mask instructions

A full set of instructions and a pattern for making this mask are available free of charge here. These instructions were developed by seamster Georgette Thornton.

  1. Cut mask Face, Cheek, and Pocket out of cotton fabric.

  1. Sew Face fronts and Pocket fronts together with a ¼-inch seam. Cut pie notches in both. Press notched seams of Face to one side and press Pocket notched seams to opposite side. (Seam allowances will go in opposite directions when you pin right sides together.)
  2. Fold wider Cheek ends in ¼ inch and stitch a ⅛-inch hem.

  1. Pin Pocket to Face section, matching seams. (Note: Pocket curve is slightly shorter than Face curve to eliminate extra fabric when turned.) Line up the bottom of each piece. (Note: Notches don’t line up on the Pocket. These were added to designate the top of the Pocket piece.)

  1. Place Cheek sections on each side, lining up all edges. Re-pin where necessary.

  1. Stitch ¼-inch seam around the whole mask. Clip the corners.

  1. Turn the Cheek ends right-side-out and push out the corners.

  1. Push the Cheek side through the Pocket and turn all right-side-out. Press.

  1. Sew ¼-inch nose bridge channel from end to end on the Pocket.

  1. Fold each Cheek end in a ½ inch and stitch, reinforcing each end.

NOTE: The finished mask should include a moldable wire nose/cheek bridge and white ties with a toggle (not shown here).

What nurses should know

As the onslaught of information about homemade masks grows, nurses should keep in mind that most of the information is about droplet spread protection, not airborne level protection. Little robust scientific evidence exists, including our own testing, to support using homemade respirator masks. Our results only show that the potential exists to make a mask that can pass fit-testing from nonmedical sourced materials.

When picking a homemade mask, consider the desired level of protection. If a high filtration rate is needed (≥95%), then a mask using only cotton or polyblend cotton material won’t be effective—no matter the number of layers. High filtration levels also require that the homemade mask fit the individual’s face. Different people require different types of N95s for best fit, and the same is true of homemade masks. (See Homemade mask considerations.)

A well-fitting mask should not allow for gaps where air can seep through and should ensure adequate airflow and speech. Before beginning to construct a homemade mask, consider the design’s overall fit. The mask’s ability to seal closely to the face is a top priority for maximizing filtering capability. Does the mask stay in place when turning the head, flexing the neck, or speaking? Typical tri-fold medical mask styles don’t form-fit to the skin, especially around the outer cheek perimeters and chin. Measure from the highest point of the nose bridge to just under the tip of the chin to determine the correct size for optimal fit. People with specific mask fit challenges—such as smaller chins, broader nose bridges, or facial hair—may wish to connect with a professional seamster to customize and revise our mask pattern for optimal fit.

The fabric used should be a tightly woven cotton or cotton blend with a high thread count. We used an organic 300-thread-count cotton, which supports filtering and laundering. If choosing 100% cotton, pre-shrink the fabric before construction. Although our study used a widely available furnace filter material for the removable filter in Mask 1, other materials may warrant consideration as research continues. If you don’t have access to fit-testing supplies, you can conduct a seal check by first placing your hands slightly above the upper and lower mask perimeters while exhaling and feeling for air flowing out around the sides of the mask. With a good fit, the air should exit only through the front of the mask, not the sides. Then inhale slowly and steadily and see if the mask collapses in, indicating air is not entering the mask through the sides. Masks are only effective if entering and exiting air goes through the filtered area. Taping the perimeter of any mask, as we did in this study, can reduce airflow at the sides. 

Homemade mask considerations

If circumstances make it necessary to use a homemade mask, keep these factors in mind.

  • Fabric—The fabric should be a tightly woven cotton or cotton blend with a high thread count to support filtering and washing.
  • Filtering material—The filtering material should be made of a nonwoven, melt-blown polypropylene material, similar to interfacing in typical medical masks.
  • Overall fit to face—No gaps should exist where air can seep through and the fit should allow for adequate airflow and speech.
  • Face measurements—Measure from the highest point of the nose bridge to just under the tip of the chin for optimal fit.
  • Seal to the face—Perform seal check to ensure air exits only through the front of the mask, not the sides.
  • Ability to stay in place—A wire can help mold the mask over the nose and straps should secure the mask in place.

Looking to the future

An important lesson from the COVID-19 pandemic is that nurses are needed in healthcare innovation. We found research literature to guide our mask development, but it’s clear that more studies are needed about mask materials, decontamination or laundering, wear time, and more. Designing and testing reusable protective gear that can be made from readily available materials should be a high priority for all healthcare organizations. Potentially, innovations can be more environmentally friendly and will allow nurses to remain autonomous in their ability to protect themselves and others. When normal channels of sourcing PPE fail, nurses must rely on their scientific knowledge and use infection-control principles and creativity to problem-solve and develop solutions.

Roschelle Fritz is an assistant professor (Vancouver), and Marian Wilson is an associate professor (Spokane) at Washington State University College of Nursing. Shawn Brow is an adjunct professor at the Elson S. Floyd College of Medicine in Spokane, Washington.

Acknowledgments

Mask development and testing were made possible with financial support from the Washington State University – Vancouver Office of Research. We wish to the thank Arbor Health and staff in Lewis County, Washington, for hosting and participating in fit-testing and seamsters Georgette Thornton and Louise Fisher for sewing the masks for this study, and Jo Phillips for assisting with Mask 2 design.

Figure 1. Mask 1 with 3M Filtrete™ insert. Photo by Roschelle Fritz.

Figure 2. Mask 2 with eight-layer cotton insert. Photo by Roschelle Fritz.

Figure 3. Fit-testing hood. Note front port inside black box. Photo by Roschelle Fritz.

References

3M. Material Safety and Data Sheet. Filtrete™ commercial high performance HVAC utility filters and Filtrete commercial mid-performance HVAC utility filter. May 29, 2007. www1.mscdirect.com/MSDS/MSDS00023/53337903-20111224.PDF

Center for Disease Control and Prevention. NIOSH science blog: Proper N95 respirator use for respiratory protection preparedness. August 14, 2020. blogs.cdc.gov/niosh-science-blog/2020/03/16/n95-preparedness

Davies A, Thompson KA, Giri K, Kafatos G, Walker J, Bennet A. Testing the efficacy of homemade masks: Would they protect in an influenza pandemic? Disaster Med Public Health Prep. 2013;7(4):413-8.

Feng S, Shen C, Xia N, Song W, Fan M, Cowling BJ. Rational use of face masks in the COVID-19 pandemic. Lancet Respir Med. 2020;8(5):434-6.

John Hopkins Bloomberg School of Public Health. 2019 Novel coronavirus research compendium (NCRC). 2020. ncrc.jhsph.edu

MacIntyre CR, Seale H, Dung TC, et al. A cluster randomised trial of cloth masks compared with medical masks in healthcare workers. BMJ Open. 2015;5(4):e006577.

The National Academies of Sciences, Engineering, Medicine. Rapid expert consultation on the effectiveness of fabric masks for the COVID-19 pandemic. April 8, 2020. nap.edu/catalog/25776/rapid-expert-consultation-on-the-effectiveness-of-fabric-masks-for-the-covid-19-pandemic-april-8-2020

The Ohio State University College of Nursing. Evidence-based COVID-19 resources. 2020. fuld.nursing.osu.edu/covid19resources

Rengasamy S, Eimer B, Shaffer RE. Simple respiratory protection—Evaluation of the filtration performance of cloth masks and common fabric materials against 20–1000 nm size particles. Ann Occup Hyg. 2010;54(7):789-98.

United States Department of Labor Occupational Health and Safety Administration. Standard 1910.134 App B-1: User seal check procedures (Mandatory). January 8, 1998. osha.gov/laws-regs/regulations/standardnumber/1910/1910.134AppB1

United States Department of Labor Occupational Health and Safety Administration. 1910.134 App A – Fit testing procedures (Mandatory). August 4, 2004. osha.gov/laws-regs/regulations/standardnumber/1910/1910.134AppA

LEAVE A REPLY

Please enter your comment!
Please enter your name here