2

Perspectives from Users, Manufacturers, and Distributors

In the workshop’s first panel session, those who use, manufacture, and distribute N95 respirators were asked to address the following in their presentations and discussions:

• What N95 respirator attributes need to be tested to ensure worker safety and health in health care settings (e.g., filtration, flammability, fluid resistance, biocompatibility, others)?
• What, if any, are the current issues being faced with having two types of N95 respirators (surgical N95s and standard N95s)?
• In your opinion, what are the priorities for research, testing, and post-market surveillance to improve N95s for health care workers’ safety and health? What are the priorities to be considered in the integration of FDA and NIOSH evaluation processes for N95s?

USER PERSPECTIVE: MAYO CLINIC

Jeffrey Nesbitt, Mayo Clinic, Minnesota

The health care respiratory protection program at the Mayo Clinic is responsible for conducting annual fit testing¹ of respiratory protective devices for nurses, nurse practitioners, residents, physicians, and other personnel who care for patients in both inpatient and outpatient settings, as well as for those who work in clinical laboratories and who conduct

___________________

¹Fit testing is the process by which the appropriate respirator model and size is identified for the respirator user. Fit testing protocols using qualitative or quantitative tests are specified by the Occupational Safety and Health Administration in 29 CFR 1910.134.

autopsies. The staff members are from 62 units at the institution that support or are part of 13 patient care units. In total, some 1,700 staff members are tested annually. Currently, explained Nesbitt, the Mayo Clinic uses eight models of N95 respirators from three manufacturers to try to achieve fit for all employees who need respirators. All of the N95s that they use are NIOSH approved. The respirators from one manufacturer are FDA-cleared surgical N95s, another are not FDA-cleared surgical N95s, and the third has a product that is not FDA-cleared but has passed the FDA-specified fluid resistance performance test (ASTM F1862—Standard Test Method for Resistance of Surgical Mask to Penetration by Synthetic Blood).

From this selection, Nesbitt noted that 92 percent of the staff were able to find a respirator that passed the fit test on their faces but that the fit test process can be quite time intensive with health care workers often having to try on several makes, models, and sizes. Additionally, there are significant monetary and staff costs associated with warehousing the requisite sizes and brands for meeting operational and emergency response needs and for keeping track of the expiration dates on the respirators and restocking as needed. Record keeping is also extensive and aims to inform managers about whether their employees are medically cleared for respirator use, have been fit tested, and have received the required training in their use.

In closing, Nesbitt said that performance and comfort are the most important attributes that need to be tested to ensure worker safety and health. The main issues he sees with having two types of N95 respirators for health care workers are that it creates confusion and there is a default assumption that fluid resistance is adequate in any approved N95 product. With regard to research priorities, Nesbitt said that developing better-fitting N95 respirators would be a significant improvement for organizations with a large number of employees that use these respirators and that more research is needed to establish guidelines on the appropriate reuse of N95 respirators.

USER PERSPECTIVE: JOHNS HOPKINS UNIVERSTIY

Geeta Sood, Johns Hopkins University

As an infectious disease specialist, Sood’s experience with N95 respirators comes from personal use but also from chairing her facility’s infection control committee, which receives reports of medical center staff who do not use their N95s when appropriate. She explained infec-

tious disease transmission can be through direct contact, via large droplets, and also by airborne transmission of droplets less than 0.5 microns in diameter, or by some combination of those routes. Respirators are particularly useful when concerned about airborne transmission of varicella, tuberculosis, influenza, or severe acute respiratory syndrome (SARS) and are particularly useful when intubating a patient or performing a bronchoscopy, so the filtration performance of an N95 respirator is important in those settings.

Sood’s main concern is that health care workers do not use N95 respirators as often as they should, and even when they do, they often do not use them appropriately. Discomfort is the main reason people do not use these respirators—she said every time she puts one on she counts the seconds before she can remove it. She noted that not all of the relevant staff at her institution get fit tested annually as required, and this is a particular challenge when someone has facial hair or experiences weight changes that make a previous fit unreliable. The availability of different sizes from different manufacturers can be a challenge when someone needs an N95 at a patient’s bedside and the model and size for which that person has been fitted is not available. Even when the right respirator is available, many people do not put them on properly and can even self-contaminate the respirator. The latter is such a significant problem, said Sood, that there is a great deal of research on self-contamination in the infection control field.

In her institution, Sood noted, they use N95 respirators from two manufacturers and each comes in multiple sizes. The many options create challenges for fit testing, storage, and record keeping as noted by the prior speaker.

As far as the attributes of N95 respirators that need to be tested, Sood listed filtration ability, comfort, ease of proper use, fluid resistance, and self-contamination risks. In Sood’s opinion, post-market surveillance should be prioritized, and research is needed on self-contamination and to develop respirators that are easier to implement in terms of supply, fit testing, and comfortable use.

USER PERSPECTIVE: UNIVERSITY OF MARYLAND MEDICAL CENTER

James Chang, University of Maryland Medical Center

The University of Maryland Medical Center (UMMC) is an academic medical center in Baltimore, Maryland, with approximately 9,000 staff

and faculty members. The medical center uses respiratory protection in two major scenarios, explained Chang. The first is to protect staff from airborne infectious diseases, including tuberculosis, varicella, measles, and more recently, novel pathogens, including H1N1 influenza and Middle East respiratory syndrome; and the second is for protection from airborne hazardous medications. UMMC currently uses elastomeric respirators with P100 cartridges due to supply issues that began during the H1N1 influenza pandemic. UMMC is in the process of replacing these elastomeric respirators with disposable N95s in response to user reports that it is inconvenient for mobile staff to access respiratory protection when moving between locations or patients. Also, it has been observed that staff are not always proficient at cleaning their elastomeric respirators after use. In their place, will be three different disposable N95 models and PAPRs. Chang noted that they have received somewhat surprising user feedback indicating that approximately 25 percent of those who use elastomeric respirators want to stay with their reusable respirator and not transition to the disposable N95s, citing comfort and a greater feeling of safety as the two main reasons for not switching.

Fit testing with N95 respirators is of concern to Chang. It can take him 15 minutes or longer to get an N95 respirator to fit someone properly. Proper fitting of disposable N95s can involve making several adjustments to the straps or to the respirator’s position on the face and he worries that it may be challenging for that person to replicate the steps it took to achieve a reliable fit/seal when it comes time to use the respirator months later. There is no easy way for the user to know that he or she has achieved a reliable fit as the recommended fit check is highly subjective and difficult to execute. The medical center will continue to provide PAPRs for those individuals who require respiratory protection sporadically as well as those who are not fit tested successfully.

Addressing the questions specific to surgical N95s, Chang said that the three brands of N95s used at UMMC are surgical N95s. The only time that he is aware that the issues regarding surgical versus standard N95s have come up at his institution was in planning for an influenza pandemic, during which the medical center was looking at home improvement centers and industrial supply houses as potential sources for N95 respirators.

With regard to performance characteristics needed for N95 certification, UMMC does not consider flammability to be an issue. Citing the comparison between a billowing surgical or isolation gown that might come in contact with an ignition source, Chang noted that respirators worn on the face in health care settings are unlikely to be brought near an

ignition source or flame. However, fluid resistance and filtration performance are of concern particularly regarding liquid splashes. He noted that discussions regarding the U.S. Pharmacopeial Convention 800 standard for handling hazardous medications note that disposable respirators offer little protection against direct liquid splashes (USP, 2016). Chang also reiterated Sood’s concern about cross contamination and wonders about the validity of antimicrobial claims from some manufacturers.

Chang concluded his comments with three suggestions to improve the effective use of disposable N95s in health care settings: develop respirators with reliable and consistent fit that have a realistic means for users to check the fit, create guidance on the resistance to hazardous medication splashes, and prepare uniform guidance on the new antimicrobial-treated N95s to determine how effective they are at preventing cross contamination with and transmission of infectious agents.

A MANUFACTURER’S PERSPECTIVE

Craig Colton, 3M

From a manufacturer’s perspective, Colton noted, the N95 respirator attributes that need to be tested depend on where it will be used and the airborne hazards found in those settings. When it comes to respirator filtration performance, the scientific literature shows that bioaerosols behave the same as other aerosols in workplace settings as well as aerosols used for filter testing. However, the exposure of health care workers to bloodborne pathogens suggests that different attributes other than filtration, such as fluid resistance, are of concern. Colton said his company has also received requests for respirators to be used in health care settings that have other attributes not typically considered, such as blocking nuisance odors, including fecal odors. Flammability hazards in health care settings appear to be low and as a result, testing for flammability resistance may not be necessary for N95 respirators intended for use in health care settings. It goes back to assessing risks and location of risks, he noted. Other industries have similar issues regarding respirator attributes and the need to ensure that workers have access to the types of personal protective equipment that have the protective attributes appropriate to the tasks. Colton stated that in his opinion the current test for fluid resistance is subjective and may be a cause for concern. “There is a lot of flexibility in that particular test and we believe the test procedure can be improved by having more standardized tests,” said Colton.

In regard to priorities for research, testing, and post-market surveil-

lance, he would like to see fluid resistance testing improved and validated. Research is also needed, said Colton, to understand what char- acteristics of a respirator would make it suitable for use in hot and humid environments such as one might encounter in the field as opposed to the climate-controlled conditions inside a hospital. A better understanding of the attributes that create a comfortable and tolerable respirator and how to measure those attributes is also needed. He noted that from a manufacturer’s perspective, NIOSH’s post-market surveillance system for respirators is more robust than the FDA system.

A DISTRIBUTOR’S PERSPECTIVE

Akhil Agrawal, American Medical Depot

Addressing supply chain issues, Agrawal noted that his company distributes products from more than 2,000 manufacturers to some 6,000 health care customers, including Department of Veterans Affairs and Department of Defense facilities. Globalization, he explained, has profoundly affected how supply chains are managed both strategically and operationally. For respirators, the main supply chain issue stated by Agrawal is that more than 90 percent of the products are not manufactured in the United States. This creates significant risks because during a true global pandemic those supplies that come from outside of the United States would likely be immediately unavailable, resulting in the U.S. supply chain being unable to meet the demand for respirators. The shortage would be further complicated by the lean inventory process employed by most health care facilities that depends on just-in-time fulfillment to meet normal operational demands for respirators and other protective equipment.

While there have been no true disruptions in the supply chain, there have been disturbances from which there are lessons to learn about the fairly “frail” supply chain for respirators noted Agrawal. These disturbances include the 2003 SARS outbreak, the 2009 H1N1 outbreak, and the 2014 Ebola outbreak. What happened during these disturbances is that health care officials placed multiple orders for respirators with multiple vendors, creating what Agrawal called an enormous amount of phantom demand accompanied by hoarding. In addition, there was a large increase in what he termed nontraditional demand from the public, clinicians, first responders, and others. As a result, the respirator manufacturers put each of the 10 major distributors (that account for 98 percent of all products moved in the health care sector) on an allocation

within 3 to 4 days. In turn, that meant that within 3 to 4 days, distributors were providing customers with 90 percent of what had been their usage over the preceding 6 months. Within a week or two, that dropped to 60 percent of a customer’s historic usage, and items went on backorder after approximately 3 weeks of elevated demand. Agrawal noted that the Department of Defense’s supplies were not reduced. Depending on the nature of future situations, health care facilities may have to give up part of their supplies to meet urgent needs of other facilities.

The traditional strategies for managing supply chain risk are to stockpile inventory, diversify the supply of product, identify backup suppliers, manage demand, strengthen the supply chain by working with manufacturers, and more effectively use the existing supply of product (see Table 2-1). None of these strategies would be sufficient, however, if a true pandemic situation existed, said Agrawal. “The order-of-magnitude difference for respirators and masks in terms of utilization in case of airborne pathogen transmission is so great that none of the traditional supply chain strategies that are available would ever be sufficient to meet the demand and the expectations that our 6,000 or so health care customers would have,” he explained.

What is needed, he said, is innovation, and one possible solution for a pandemic situation could be one of the newly developed reusable antimicrobial respirators. While this may not be the perfect solution from a science perspective, in the event of a high-risk pandemic it may represent an acceptable solution. Agrawal wondered if there is a role for the Biomedical Advanced Research and Development Authority to play in addressing the regulatory issues, but regardless, he said there needs to be a clear path between NIOSH and FDA for approval of N95 respirators specifically and guidance regarding reusability of respirators in general.

In Agrawal’s opinion, providing regulatory clarity would open up funding to innovators who would work to solve this part of the supply chain problem; and even with that, however, there would still be a need to enhance domestic production capability for respirators. He noted that there are companies that produce these materials and do the molding domestically, but they account for a small percentage of the respirators used on a normal basis in the United States.

TABLE 2-1 Traditional Strategies for Managing Supply Chain Risk

SOURCE: Presented by Akhil Agrawal, August 1, 2016.

DISCUSSION

In the discussion session with workshop participants, a number of issues regarding the use of respirators in health care settings were discussed, including issues regarding streamlining the regulatory processes. Several participants noted the potential confusion in the health care workplace regarding the use of surgical and nonsurgical N95 respirators and in addition, the complication of another type of product—surgical masks—which are not respiratory protective devices. Issues relevant to specific types of respirator testing are discussed in the following chapter.

In response to a question about how big of a problem occupational illness is in the health care setting, Sood said the numbers in normal endemic hospital functions are small and that in 10 years she has seen at most three cases of occupational transmission. Tuberculosis is very non-transmissible and while varicella has a bigger potential to pass from a patient to a staff member, most health care workers have been vaccinated and so it is seen infrequently in health care workers. In a pandemic situation, she expects these numbers to be higher. She also remarked that in her experience the biggest cause of occupational health failures is people forgetting to use respirators and face shields, which she attributed largely

to discomfort. Session moderator Barbara DeBaun from Cynosure Health noted that the health care field experienced the same challenge in the 1980s when gloves were first gaining traction in health care. “They were uncomfortable and people would not wear them because they were uncomfortable,” she said.

Another issue raised by David Prezant from the New York City Fire Department was the extent of the evidence regarding expiration dates on respirators, particularly in the context of supply chain problems. He asked whether incorporating fluid resistance properties would affect the expiration date. Colton replied that there can be a number of attributes contributing to a particular respirator’s expiration date, including fluid resistance, flammability, and the material from which the straps are made. Expiration dates, Colton explained, have only been added fairly recently. Although there is some scientific evidence to determine those dates—collected through a combination of accelerated aging testing and experience with specific materials—much still remains to be learned. Thus, expiration dates tend to be set toward the conservative side.

There was discussion about the use of PAPRs that have the advantage of providing both respiratory and splash protection without requiring annual fit tests; but as noted by Sood, the tradeoffs are that PAPRs are expensive and difficult to clean. She also noted that some of the hospitals in the Johns Hopkins health system tried to use PAPRs universally with little success.

The issues of fit testing came up in the discussion session with Kate Bradford from the Department of State asking if fit testing at the panelists’ institutions was qualitative or quantitative. From the panelists’ responses, qualitative testing is the norm for group fit testing at their institutions, and quantitative testing is used for new employees and when someone fails qualitative testing. A suggestion was made by Jonathan Rosen from AJ Rosen & Associates, LLC, to study the varying types of N95 respirators on the market today and compare their abilities to produce a good fit. Nesbitt noted that if a small number of models that would fit 90-plus percent of his staff could be identified, it would be a big boon in significantly reducing costs and logistical issues.

Rosen brought up the two Respiratory Evaluation for Acute Care Hospitals studies that NIOSH has funded. These studies demonstrated there were large gaps in complying with Occupational Safety and Health Administration (OSHA) requirements regarding respirator use in health care settings, either because staff members were not using the respirators or not wearing them correctly. He also recounted an anecdote in which an industrial hygienist was training and fitting 15 employees at a time in

short (15-minute) increments, suggesting that training and education about the use of respirators are lacking.

Mark Catlin from the Service Employees International Union said it is important to separate certification approval issues from those resulting from managers not understanding how to run a respiratory protection program. One issue he has seen is that when health care workers ask for respirators, they are told surgical masks are good enough. He also noted that NIOSH and other researchers have identified the need for health care administrators to figure out how to better run these programs. With regard to approval certification, an issue he would like to see addressed concerns the maximum length of time disposable N95s should be worn. “They are often thought of as disposable, but in the health care industry people often wear them all shift long, they take them on and off, they crease them, and they put them in pockets,” noted Catlin. He said that the union’s members have reported they are told that their respirators have to last for periods ranging from an entire shift to a week, a month, or an entire year because they are expensive. The panelists agreed that reusability is an issue given that many of the microorganisms of concern can be transmitted through contact with a contaminated surface such as the outside of a respirator.

James Johnson from JSJ and Associates asked the panelists to comment on the use of reusable versus disposable N95 respirators noting that in other industries, such as the radiation research laboratories where he has worked, the lab employees did not have to clean their respirator; a separate staff conducted the respirator cleaning as part of their responsibilities. Both Chang and Sood replied that infection control is a major reason for using disposable N95s (particularly as health care providers move between patients of varying risks), as well as the convenience of not having to carry a reusable respirator for an entire shift. Agrawal highlighted the issue of reusability and the potential for a positive impact on the supply chain. Rosen noted that the Joint Commission’s guidance on implementing hospital respiratory protection programs highlights the Texas Center for Infectious Disease’s protocols for successfully reusing N95s in the tuberculosis setting (Joint Commission, 2014).