Commentary: Respirators can be reused, but decontamination not well studied

A person holds an N95 respirator mask in their hand.

Note:  this commentary was originally published by the Center for Infectious Disease Research and Policy.

Perhaps lost in the recent debate and federal recommendations on cloth masks for the general public has been the key issue of protecting those putting their lives on the line daily: hard-working healthcare workers as hospitals face limited supplies of N95 filtering facepiece respirators (FFRs) during this pandemic.

Key to such protection are issues surrounding respirator reuse and decontamination to optimize use of FFRs, which are personal protective devices constructed largely from filter material worn on the face that prevent inhalation of viral aerosols by the wearer.

This commentary assumes, however, that healthcare institutions are already implementing all other engineering controls—such as isolation rooms, physical barriers, and ventilation systems—and administrative controls—steps like limiting patients being admitted, practicing telemedicine, and cohorting healthcare workers—as recommended by the Centers for Disease Control and Prevention (CDC) in its conventional and contingency strategies for optimizing respirator supplies.

In China, the response to COVID-19 included cohorting patients in separate locations. Those with mild symptoms were housed in temporary hospitals fashioned from large open buildings such as sports stadiums. Those with more serious or life-threatening symptoms were cohorted in wards or hospitals with more medical resources and higher levels of engineering and administrative controls.1This allowed healthcare workers to wear a single respirator throughout a shift.

These same practices, recommended by the CDC in its conventional capacity strategies, along with respirator reuse, would greatly limit the number of respirators required to care for patients with mild symptoms, thus saving them for healthcare workers in higher-risk settings.

Extending respirator use in a single shift

When caring for multiple COVID-19 patients, there should be no concern for cross-contamination and no need to remove the respirator between patients. Healthcare workers had no detrimental physiologic health effects when wearing an N95 FFR throughout a 12-hour shift but did report discomfort and feeling out of breath.2

Discomfort is most likely caused by buildup of temperature and humidity inside the facepiece. A respirator with an exhalation valve would do much to alleviate discomfort.

If multiple donnings are required throughout the shift, then hand contamination may be a problem. Careful doffing and storage followed by immediate hand washing may be a solution. Other solutions include wearing a face shield or a surgical mask over the respirator. The former is preferred because it will have no impact on respirator fit. The latter has been shown to have no adverse physiologic effects,3 but the impact on respirator fit has not been studied.

Wearing a single respirator throughout a shift would somewhat alleviate the number of respirators needed overall during a pandemic. We recommend adopting this approach, now, for extending the limited supply of respirators and helping healthcare workers feel protected.

Extending respirator use over multiple shifts

Fit can be compromised after multiple donnings, as straps and other components begin to fail. A study of six N95 respirators found increasing rates of failure to achieve passing fit factors as the number of donnings increased, although failure rates differed with the respirator model.  Fit was most consistent on the first five donnings, and failure rates increased to 53% to 76% after 15 donnings.4

Our recommendation, in pandemic situations, would be to use the same respirator over a 5-day period, as long as it is not being removed multiple times during a shift.

Respirator decontamination

Another possible approach to stretching respirator supply is decontamination of the FFR, but there are many caveats.

While extended use of FFRs is addressed in CDC COVID-19 pandemic guidelines as a possible contingency capacity strategy, FFR decontamination is not. Nor is FFR decontamination addressed in National Institute for Occupational Safety and Health (NIOSH) pandemic planning recommendations for FFR extended and limited reuse in healthcare settings.

Decontamination of FFRs is not an approved practice by the CDC, NIOSH, the Occupational Safety and Health Administration (OSHA), or the Food and Drug Administration (FDA). Modifying a respirator invalidates its NIOSH certification, and OSHA requires that only NIOSH-certified respirators may be used in workplaces for worker protection. Respirator manufacturers have stated that FFR decontamination is not acceptable for their respirators.

Research on decontamination has been conducted or supported by NIOSH and the FDA, but only for the purpose of exploring its use in pandemic situations, as the result of recommendations from the National Academies of Science Institute of Medicine.5

In the case of a pandemic, an acceptable decontamination method must render the organism (or a closely related surrogate) non-viable and not diminish filter and fit performance, respirator integrity and structure, or comfort, odor, and wear.

These are minimal criteria. A full health and safety assessment should be conducted before bringing any decontamination method on-line. This should include consideration of the impacts on health and safety of workers involved in the decontamination process as well as those wearing the decontaminated respirator. Fisher and Shaffer6 detail the many factors that should be considered before selecting decontamination as a means of extending respirator supplies.

We reviewed a few of the early NIOSH studies of respirator decontamination,7-10 The most promising methods appeared to be ultraviolet germicidal irradiation (UVGI) with UV-C light and hydrogen peroxide vapor (HPV). We explore these two methods in more detail here but do not claim to have reviewed all of the data available for either method, and there may be other methods to consider, as well.

Ultraviolet germicidal irradiation

UVGI disinfection involves placing respirators in direct line with UV-C lamps. Any exposed surface of the mask will be disinfected. This process generally requires the masks to be flipped after some period to ensure both sides of a mask are disinfected or requires a setup involving two UV-C lamps on either side of the respirator.

For samples of FFR filters treated with SARS (severe acute respiratory syndrome) and MERS (Middle East respiratory syndrome) viruses, no viable virus was found following a dose of 1 joule per square centimeter (J/cm2) of short wavelength (254 nanometers) UV-C light, even in the presence of artificial skin oil and saliva.11 Researchers found no impact of doses ranging from 3 to 20 J/cm2 on filter performance.9-12

A similar dose range did not compromise fit on three N95 FFRs worn by 10 volunteers.12 Doses ranging from 3 to 6.5 J/cm2 did not alter fit performance on human subjects; doses up to 20 J/cm2did not significantly change fit performance on respirators mounted on a static head form connected to a breathing machine.11

Lindsley et al13 tested the impact of much higher doses, ranging from 120 to 950 J/cm2, on respirator filter samples and 590 to 2,360 J/cm2 on respirator straps. For most samples, filter efficiency decreased after UVGI treatment but remained above 95% at all dose levels for the four tested respirator models. Strength of the different layers decreased at all doses, with more layers showing decreased strength with increasing dose. The breaking strength of straps also decreased at all doses, ranging from a loss of 10% to 21% at the lowest dose of 590 J/cm2 and 20% to 51% at the highest dose of 2,360 J/cm2. Respirators treated with a single decontamination of UVGI showed no significant difference from untreated respirators in odor, donning ease, comfort, and respirator fit.9

Altogether, these data suggest that respirators could be decontaminated by UVGI for up to 20 cycles at a dose of 1 J/cm2 per cycle, if the fit performance data from Heimbuch and Harnish11 using a static head form are representative of fit on humans. Limited human data on a small number of N95 FFR models suggest that 3 cycles at this dose will not degrade fit.

While respirator filter performance (and perhaps fit) were not adversely affected at the higher doses studied by Lindsley et al,13 because the lowest dose level of 120 J/cm2 caused a loss of strength of at least one filter layer, extending UVGI treatment beyond 20 cycles does not seem appropriate at this time.

More research is needed on off-gassing, process implementation, and worker health and safety (for process workers and respirator wearers). Without published studies describing implementation of this method in real-world settings, we are hesitant to recommend UVGI during the COVID-19 pandemic. If absolutely necessary for extending limited supplies, it would be precautionary to limit its use to no more than 5 cycles at 1 J/cm2 per cycle.

Hydrogen peroxide vapor

HPV involves generating a hydrogen peroxide vapor at a concentration able to deactivate organisms. The generated vapor needs to be able to reach all sides of the respirator.

Two studies have demonstrated six-log spore reduction with biological indicators (Geobacillus stearothermophilus) using:

  1. In a 64-cubic-meter room, exposure of 8 grams per cubic meter (g/m3) HPV for 3 cycles (total dose of 24 g/m3, total single cycle time of 125 minutes with 15-minute dwell) using a Bioquell Room Bio-Decontamination Service HPV generator.10
  2. A 310-liter test chamber exposure of 323 g/m3 (10 minutes conditioning, 20 minutes at 2 g/min gassing, 120 minutes at 0.5 g/min dwell) using a mobile HPV generator (Bioquell Clarus C).14

It is assumed that a dose sufficient to deactivate bacterial spores will be sufficient for deactivation of less hardy viral organisms. No data are available, however, for the dose required to deactivate SARS-CoV-2, the virus that causes COVID-19.

After 3 or 50 rounds of disinfection in these two studies, respectively, neither group found any significant reduction in filter performance. Respirator fit was measured only in the second study, using a static head form and breathing machine. After 20 cycles, the researchers found no difference in fit between treated and control respirators, but there was noticeable and significant damage to the straps.14 (Richter 2016) No off-gassing of HPV from the respirator was measured after 5 hours of aeration in the second study14;off-gassing was not explored in the first.

Thus, HPV appears to meet the minimum threshold of our four criteria, although there are some important limitations in the data. There are no data for the dose required to inactivate SARS-CoV-2 or other similar coronaviruses.

It is not clear why the doses required to achieve a 6-log reduction differ by an order of magnitude between the two methods. Richter et al14 conducted the most extensive set of experiments but evaluated only a single respirator model; the data are available only in a non-peer-reviewed final contract report. Bergman et al10 conducted fewer assessments on six respirator models and did not evaluate fit.

In addition, neither study evaluated fit on human subjects. The application in a real-world setting for either approach has not been described in a peer-reviewed publication. Given these many limitations, we hesitate to recommend the use of HPV unless absolutely necessary in the most dire of pandemic situations. It would be precautionary to limit the number of cycles to no more than 5 for either of the two protocols.

More decontamination caveats

None of the methods described here have received a thorough health and safety process review, and none have been adequately assessed in clinical settings. More information is needed on the health and safety of workers involved in the decontamination process; short- and long-term costs of equipment purchase and installation, training, maintenance, program administration, quality control, process time, etc.; and whether the method introduces new hazards for respirator wearers.

Respirator users should be involved in the assessments. We should expect to see well-designed pilot studies in real-world settings prior to a pandemic. The literature and research do not yet appear to be at that point.

Recent guidelines from NIOSH confirm UVGI and HPV as possible methods for decontaminating FFRs but also suggest that moist heat may acceptable. Our criteria should be considered for any additional decontamination methods that were not reviewed in detail here.

Exhausting all means to extend protection

Organizations should be implementing as many of the controls described in the CDC conventional and contingency strategies for extending respirator supplies before resorting to FFR decontamination, which, as noted above, is not an approved or even suggested strategy by any federal government agency.

We ask the public to support their healthcare workers and insist that healthcare institutions implement all possible administrative and engineering controls that would ensure ongoing supplies of respirators throughout the pandemic.

About the authors

Lisa Brosseau, ScD, retired as a professor of environmental and occupational health sciences from the University of Illinois at Chicago (UIC) School of Public Health in 2019.  She is a national expert on respiratory protection and infectious diseases.

Margaret Sietsema, PhD ’15, MS ’11, is an assistant professor of environmental and occupational health sciences at the UIC School of Public Health whose expertise and research focuses on respiratory protection.


  1. WHO. Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19). Feb 28, 2020
  2. Rebmann T, Carrico R, Wang J. Physiologic and other effects and compliance with long-term respirator use among medical intensive care unit nurses. Am J Infect Control 2013 Dec;41(12):1218-23
  3. Roberge RJ, Coca A, Williams WJ, et al. Surgical mask placement over N95 filtering facepiece respirators: physiological effects on healthcare workers. Respirology 2010 Apr;15(3):516-21
  4. Bergman MS, Viscusi DJ, Zhuang Z, et al. Impact of multiple consecutive donnings on filtering facepiece respirator fit. Am J Infect Control 2012 May;40(4):375-80
  5. Bailar JC III, Burke DS, Brosseau LM, et al. Reusability of facemasks during an influenza pandemic. Institute of Medicine, National Academies Press, 2006
  6. Fisher EM, Shaffer RE. Considerations for recommending extended use and limited reuse of filtering facepiece respirators in health care settings. J Occup Environ Hyg 2014 Jun 27;11(8):D115-28
  7. Viscusi DJ, King WP, Shaffer RE. Effect of decontamination on the filtration efficiency of two filtering facepiece respirator models. J Int Soc Respir Prot 2007 fall/winter;24(III-IV):93-107
  8. Viscusi DJ, Bergman MS, Eimer BC, et al. Evaluation of five decontamination methods for filtering facepiece respirators. Ann Work Expo Health 2009 Nov;53(8):815-27
  9. Viscusi DJ, Bergman MS, Novak DA, et al. Impact of three biological decontamination methods on filtering facepiece respirator fit, odor, comfort, and donning ease. J Occup Environ Hyg 2011 Jun 11;8(7):426-36
  10. Bergman MS, Viscusi DJ, Heimbuch BK, et al. Evaluation of multiple (3-cycle) decontamination processing for filtering facepiece respirators. J Eng Fiber Fabr 2010 Dec;5(4):33-41
  11. Heimbuch B, Harnish D. Research to mitigate a shortage of respiratory protection devices during public health emergencies. Applied Research Associates. Mar 22, 2020
  12. Bergman, MS, Viscusi DJ, Palmiero AJ, et al. Impact of three cycles of decontamination treatments on filtering facepiece respirator fit. J Int Soc Respir Prot 2011 Jan;28(1):48-59
  13. Lindsley WG, Martin SB Jr, Thewlis RE, et al. Effects of ultraviolet germicidal irradiation (UVGI) on N95 respirator filtration performance and structural integrity. J Occup Environ Hyg 2015 Jul 1;12(8):509-17
  14. Richter W, Hofacre K, Willenberg Z. Final report for the Bioquell hydrogen peroxide vapor (HHPV) decontamination for reuse of N95 respirators. Battelle. Jul 22, 2016