Future Directions: Additional Analyses and Research to Practice
The National Institute for Occupational Safety and Health (NIOSH) requested that the committee make recommendations for areas that require additional research and analysis for moving from research to practice. This chapter discusses those recommendations and the rationales behind them.
FACE PANELS REQUIRE PERIODIC UPDATES
As described in detail in the previous chapters, there are several potential weaknesses in the NIOSH-sponsored Anthrotech study. However, the proposed face panel represents a clear improvement over the existing LANL face panel, which only accommodates 84 percent of the civilian population (Zhuang, 2001). The application of the proposed face panels will improve the availability of respirators that fit a broader segment of the workforce.
New technologies, particularly in the area of digitized imaging, may allow NIOSH to create a continuously updated virtual face panel, which could be used to create digital or actual patterns. This can be done by considering the proposed face panelwith appropriate revisions as recommended throughout this reportas the baseline.
The demographics of the workforce, especially racial and ethnic, are shifting at a rapid pace. For example, one-third of those employed in the construction trade are Hispanic, which is a 50 percent increase from 5 years ago (U.S. Census Bureau, 2000; U.S. Department of Labor, 2006). This rate of change is expected to continue and most likely intensify as
increasing free trade leads to easier movement of workers from one country to another.
The committee recognizes the significant investment NIOSH has made in developing the proposed Anthrotech face panel and NIOSH’s limited resources. It is the committee’s opinion that, while there is room for improvement in future NIOSH studies, it would be a disservice to the community of respirator users if the “good” features of the revised panel are abandoned in pursuit of longer-term “better.” The LANL face panel was established in 1972, 34 years ago (Hack et al., 1974). The changing demographics of the respirator workforce requires that NIOSH face panels be more frequently updated. The development of these future panels would be greatly improved if NIOSH assured that the samples included in the panels were representative of workers who should be using respirators, and if NIOSH incorporated and validated three-dimensional measures, and measured quantitative fit.
The proposed NIOSH-sponsored Anthrotech face panel represents an improvement over the LANL face panel, and its application is likely to improve the availability of respirators that fit a broader segment of the workforce. However, the Committee also found that this study could have been greatly improved. In addition, the NIOSH face panels require periodic updates.
Recommendation 5-1: Update the Panel More Often, Using a Scientifically Valid Design.
NIOSH should plan to update the face panel more often to reflect the rapidly changing demographic structure of the U.S. workforce. To do so, it should (1) establish a valid sample of the target population for its respirator certification, (2) assure that the sample that is selected to establish a valid panel is representative of the target population, (3) apply three-dimensional measures to describe the essential fit characteristics of the panel, and (4) rely on
quantitative fit testing to determine the extent to which a respirator covers the fit characteristics of the face panel.
FACE PANEL VALIDATION
As discussed in Chapter 4, there are several concerns regarding the validity of the panel proposed by the NIOSH-sponsored Anthrotech study as currently designed. In addition to the analyses described in the preceding chapters, the committee recommends that NIOSH use its existing fit and facial dimensions data, and new quantitative fit tests, to evaluate the current LANL face panel, the proposed bivariate face panel, and the principal component analysis-derived face panel. Specifically, NIOSH should examine the following questions:
Using the same brand of respirators, do the proposed face panels have an improved pass rate compared to the LANL face panels?
Is there a specific subpopulation of respirator wearers selected by either of the proposed face panels who fail to obtain a satisfactory fit and who would have not been included in the proposed face panel?
Do the proposed face panels exclude individuals who fail to obtain an adequate fit and who would have been included by the LANL face panel?
If there are no substantial differences in identifying wearers who fail to obtain a satisfactory fit among the different face panels, the identification of a new anthropometric face panel is more of an intellectual exercise than a practical improvement. Conversely, if one or both of the new panels do identify such test subjects, respirator manufacturers will have to develop new or redesign facepieces to meet certification requirements.
QUALITATIVE FIT TESTING
NIOSH currently uses a qualitative fit test as a part of its certification process. This involves the use of isoamyl acetate (IAA), also known as banana oil. Test subjects are asked to wear different respirators, while
IAA is introduced to the outside air. If a test subject indicates that they can sense the presence of IAA, the respirator fails the fit test. There are many potential difficulties that arise by using a qualitative test design, including the following:
Individual variability in IAA sensitivity thresholds. Because of individual variations in sensitivity, it is not possible to correlate a failed fit test with how much of the test substance leaked into the facepiece. The committee heard from NIOSH personnel that some respirator fit failures may be a result of above-average sensitivity of some subjects to the odor of IAA rather than from a poorly fitting respirator.
Uncertainty of individuals as to whether they detect IAA. NIOSH personnel informed the committee that during some fit-test exercises, test subjects are uncertain as to whether they briefly detected the odor of IAA. This raises doubts among the committee about these test results.
Inability to determine quantitative numbers for respirator fit. The IAA test is designed to produce a single pass/fail criterion. It does not assess how much better (or worse) the respirator may fit the subject than the pass criteria.
Qualitative fit testing is a subjective process and does not provide NIOSH certification personnel with a specific value to analyze leakage around the facepiece.
Recommendation 5-2: Replace Isoamyl Acetate with Quantitative Measures.
NIOSH should use quantitative measures for respirator fit-test certification. The current use of qualitative measures as a fit-test agent for certification, for example isoamyl acetate, should be discontinued.
CERTIFICATION OF FACE MASKS
Although certifying respirators against an anthropometric face panel does not eliminate the need for individual fit testing, it increases the likelihood that certified respirator facepieces will fit the workers they are designed to protect. As discussed in Chapter 4, the committee heard anecdotal evidence from a fit-test expert that in the case of half- or full-face respiratorswhich are certified against fit-test panelsit is possible to fit most respirator wearers if proper training is provided and different respirator brands and sizes are available (McKay, July 10, 2006). For example, in a recent study performed in the auto body repair and refinishing industry, it was shown that for half-face cartridge respirators, 92 percent of workers (both new and previously fit-tested workers) can be fitted with a respirator after two tries. Those without previous testing experience had a 72 percent pass rate (Liu et al., 2006). However, the same is not true for filtering facepiece respirators, which are not currently certified against a fit-test panel. Anecdotal evidence from experts who met with the committee and NIOSH suggests that there are brands that fit only a small proportion of faces in the workforce (McKay, July 10, 2006).
The failure to use anthropometric face panels for certification of filtering facepiece respirators may result in families of respirators that do not adequately fit some of the population of workers who should be using respirators.
Recommendation 5-3: Utilize the Revised Anthropometric Face Panel for Filtering Facepiece Respirators.
NIOSH should include filtering facepiece respirators in the revised anthropometric face panel used for certification of half-facepiece respirators. Plans for any additional data gathering should be developed in consultation with experts in statistical sampling and measurement.
Challenges for Manufacturers
As described in the previous chapter, the LANL face panel has specific certification requirements for single-size facepieces (e.g., one size fits all) that are different from those facepieces that are available in multiple sizes (e.g., small, medium, large). NIOSH representatives told the committee that, because of these different certification requirements, it can be more difficult for manufacturers that produce multiple-size facepieces to meet the certification requirements than for those manufacturers that produce single-sized facepieces. NIOSH certification rules permit masks that are designed as “one size fits all” to fail on up to two panel members; whereas, masks that are available in multiple sizes will need to fit all panel members within specific panel categories, including the segments that border the next sizes. NIOSH should examine their certification requirements to ensure that they do not introduce an undue burden on manufacturers that make multiple-sized facepieces. The committee is concerned that certification requirements appear to make it more difficult for a set of multiple-sized respirators to pass than a single-sized respirator.
Given the indicated difficulty in certification that some manufacturers of multiple-sized facepieces have, and the ultimate goal of ensuring that everyone who needs a respirator is protected, the committee recommends that NIOSH certify a family of respirators (e.g., small, medium, and large) against a fit-test panel and not specify what portion of the panel the respirators must fit, provided that the family adequately covers the entire panel. This will eliminate current inequities among those manufacturers submitting a single size versus those submitting multiple sizes.
Future NIOSH certification regulations may wish to also take into account the importance of market forces in the development of respirator facial sizes, and the agency should do nothing to discourage those manufacturers who may wish to target a specific underrepresented range of wearers or facial shapes and sizes.
Manufacturers of multiple-sized facepieces often have difficulty obtaining certification for each individually sized facepiece.
Recommendation 5-4: Modify Certification Requirements.
NIOSH should modify its certification requirements to encourage manufacturers to develop specific sizes designed to fit underrepresented anthropometric categories. Certification requirements should be modified to allow families of respirators (e.g., small, medium, and large) to be certified against a fit-test panel and not specify what portion of the panel each individual size respirator must fit, provided that the family adequately covers the entire panel.
Clarifying Face Mask Packaging Requirements
The committee heard from experts and NIOSH staff that it is current industry practice to limit designation of fit attributes of half-face respirators to size—that is, small, medium, and large (McKay, July 10, 2006). This is largely due to inadequate details on the potential sizing of the facepieces on their packaging. As a result, fit testers and respiratorwearing workforces have difficulty determining which size and brand of facepiece is most appropriate for specific segments of the population. Respirator users would be helped by having more information on leakage characteristics of each respirator, in addition to the size information. This could be obtained by both manufacturers and certifiers producing data on fit characteristics and providing additional fit attributes, such as type of face (round, long, etc.) that would capture more facial features.
The current size designations of small, medium, and large for respirators in product literature are not adequately informative for wearers and fit testers.
Recommendation 5-5: Develop Improved Descriptions of Face Mask Sizes.
NIOSH should encourage manufacturers to develop improved methods of describing facial sizes and shapes in product literature. For example, NIOSH and manufacturing companies should further explore the use of face panel images, and respirator containers should include corresponding pictures of small, medium, and large sizes, and long/narrow and small/wide face pictures.
PROVIDE ACCESS TO PRIMARY DATA
The committee believes that a number of stakeholders may benefit from access to NIOSH’s revised face panel and the primary data that were used to develop the face panels. Special care is recommended to ensure that only good quality and reliable data is made accessible to stakeholders. In addition, guidelines to approve requests for use of the data need to be established to ensure good use of the data. They include the following:
Regulators. NIOSH for its certification program and OSHA in future revisions of respiratory protection standards
Respirator manufacturers. By introducing the design of newer face piece models that will better fit the needs of their customers
Respirator program administrators. By improving the selection and fit of respirators, and by fitting more of their employees
Respirator users (workers). By having a selection of facepieces that better fit the variety and sizes of their faces (especially those in the outlying areas of design)
Researchers in the field of respiratory protection. By using data from the panel to validate fit-testing methods and protection factors for different types of respirators
The international community of respirator manufacturers, regulators, and researchers. By identifying basic approaches to the collection and analyses of anthropometric data necessary for identifying the sizes and shapes of respirators to fit the ethnic diversity of workers. It may also be possible to establish one or more universally accepted face panels for this purpose.
IMPROVING THE PRACTICE OF RESPIRATORY PROTECTION
NIOSH can enhance the practice of respiratory protection in two broad areas: through periodic recertification of respirators against a regularly updated panel (see Recommendation 5-1) and through improvements in the practice of using respirators in the workplace. It is difficult for the committee to define the exact frequency of the periodic recertification as it would be based on many changing variables, such as changes in the workforce demographics.
The committee notes that, although certification of respirators is important and necessary, certification alone does not ensure that workers are properly protected. Limited research studies have confirmed that the selection of a respirator based solely on its size, without the benefit of individual fit testing, will produce a poorly-fitting respirator 30 to 40 percent of the time (Coffey et al., 1999). The effectiveness of the protection offered depends on the appropriate fit to a worker’s face, which in turn is affected by the user’s facial size and contours, the training in selecting the right respirator, and wearing it properly. Therefore, in addition to certification, it is critical that individual fit testing be conducted by employers, as currently required by OSHA. This step will ensure, not only that users are provided with respirators that fit them appropriately, but it also ensures that they have the competence necessary to don and use the devices properly. As such, in circumstances where fit testing can not be realistically conducted, it may behoove the employer to, instead, offer hoods, or other loose-fitting respirators that offer protection without needing to be fitted to each wearer.
Although the characteristics of the workforce are dynamic, NIOSH’s certification of respirators is static: once certified, a respirator is certified
forever. NIOSH may wish to examine the possibility of updating certification regulations to keep up with changes in the workforce. That is, NIOSH might wish to periodically retest certified respirators by comparing them to a fit-test panel that is updated regularly.
The Occupational Safety and Health Administration (OSHA) has issued regulations that require employers to maintain respiratory protection programs in general industry, shipyards and marine terminals, and construction. The implementation of a respiratory protection program requires significant technical expertise. However, most employers are very small and do not have the necessary resources readily available. In addition, there is a growing trend for workers to be self-employed as independent contractors. For instance, in the construction industry, the self-employed portion of total employment now approaches one-third, up from about 7 percent 30 years ago (U.S. Department of Labor, 2006). Thus, in the real world, many workers who should be covered by respiratory protection programs work under conditions of minimal safety supervision and often with no respiratory protection.
Simplification of respirator use is therefore critical if the goal is to increase use of appropriate respiratory protection. Additional interdisciplinary research is needed to develop respirators that do not require fit testing, are more user friendly, and require minimal maintenance. To this end, using virtual face panels to model future respirators could enhance product development. Included in these interdisciplinary teams should be researchers who focus on the impact of human factors, biomechanics, biomaterials, respiratory protection, and others. It could also make possible the progressive development of low-cost disposable respirators that provide significantly greater protection against a much broader range of hazards.
Increased Confidence in the Protection Factors
The proper use of the face panel, along with quantitative fit testing, can define protection factors used by employers. NIOSH can determine minimally acceptable protection factors for each style of respirator (e.g., half-face negative pressure, full-face negative pressure), and use those standards as part of the certification process. If most half-face respirators can provide a fit factor of 500 (the current minimum acceptable level is 100), this could become the new standard (Code of Federal Regulations,
1998). However, these efforts are dependent on NIOSH beginning to use quantitative fit testing in their certification process.
DEVELOPING FACE PANELS APPLICABLE TO THE WORLD’S WORKFORCE
Representatives from large respirator manufacturers described to the committee that their companies conduct fit testing using the current LANL face panel. But because they also sell respirators outside the United States, these companies must also examine facial sizes from other countries (e.g., the Asian and Pacific rim nations) to determine what modifications in their facepieces will improve the fit to workers from other countries and to produce respirators that may target a specific range of facial sizes or unique facial shapes.
Although the proposed NIOSH-sponsored Anthrotech face panel was designed to represent only the U.S. workforce, NIOSH personnel indicated an interest in using it as the foundation for the development of international standards. Maintaining an up-to-date head and face panel can be both financially and scientifically challenging. It is also an area where international scientific collaboration could be beneficial. Because the U.S. population is among the world’s most diverse, with representation from most racial groups and ethnicities, it makes sense to start with a face panel that is representative of the U.S. population and build on that to accommodate ethnicities that are less well-represented in the United States than they are elsewhere. The following actions would support such development:
Validate a digitized scanning protocol that investigators can use to develop head and facial images.
Develop a detailed ethnic, geographic, and demographic classification system to be applied to head and facial images derived from digitized scanning.
Establish a database to store an ever-expanding number of images.
Enlist and approve or register investigators from around the world to perform digital scanning of images with certification to assure that scanning is done according to protocol.
Develop an electronic system for submission of images from investigators who have been approved.
Make the database available to approved investigators so that they can create virtual head-and-face panels suitable to their research interests or needs.
Code of Federal Regulations. 1998. Respiratory protective devices: 29 C.F.R. § 1910, 1915, 1917, 1918, 1926.
Coffey, C. C., D. L. Campbell, and Z. Zhuang. 1999. Simulated workplace performance of n95 respirators. Am Ind Hyg Assoc J 60(5):618-624.
Hack, A. L., E. C. Hyatt, B. J. Held, T. D. Moore, C. P. Richards, and J. T. McConville. 1974. Selection of respirator test panels representative of U.S. adult facial sizes. Los Alamos, NM: Los Alamos Scientific Laboratory.
Liu, Y., M. H. Stowe, D. Bello, S. R. Woskie, J. Sparer, R. Gore, F. Youngs, M. R. Cullen, and C. A. Redlich. 2006. Respiratory protection from isocyanate exposure in the autobody repair and refinishing industry. J Occup Environ Hyg 3(5):234-249.
McKay, R. T. July 10, 2006. Presentation to Committee for the Assessment of the NIOSH Head-and-Face Anthropometric Survey of U.S. Respirator Users: Respirator fitting issues. Pittsburgh, PA: National Personal Protective Technology Laboratory.
U.S. Census Bureau. 2006. Occupations: 2000. http://www.census.gov/prod/ 2003pubs/c2kbr-25.pdf (accessed December 15, 2006).
U.S. Department of Labor. 2006. Career guide to industries: Construction. http://www.bls.gov/oco/cg/cgs003.htm (accessed December 15, 2006).
———. 2006. Employed persons by detailed occupation, sex, race, and Hispanic or Latino ethnicity. http://www.bls.gov/cps/cpsaat11.pdf (accessed December 15, 2006).
Zhuang, Z. 2001. Anthropometric survey of respirator users: Study protocol. National Institute for Occupational Safety and Health.