Millions of Americans use respirators in their daily work to protect themselves from potential exposures to toxic vapors and gasses; harmful particulate matter; airborne bacteria, fungi, and viruses; or other respiratory hazards. To provide an appropriate level of protection, it is essential for respirator facepieces to mold to the user’s face in such a way that no air from the outside may leak in, even when the user is moving or speaking. The respirator-wearing workforce of the United States is quite diverse, thus manufacturers are required to provide respirators that will fit the faces of a diverse range of face size and shape.
It is the responsibility of the National Institute for Occupational Safety and Health (NIOSH) to certify that respirators from manufacturers meet certain minimum performance levels. Respirator “fit-test” panels derived from human facial size and shape data are used in NIOSH certification. The dimensions of the face panel depend on whether it is to be used to certify a full- or half-face respirator. Each face panel consists of about 25 persons who are chosen because their facial dimensions represent various categories of size and shape. NIOSH considers a mask (or set of masks) that passes a qualitative fit test to adequately fit the panel and be suitable for the marketplace (Chapters 4 and 5).
The data for, and design of, the face panels have not been updated since the establishment of the 1972 Los Alamos National Laboratory (LANL) anthropometric face panel. In 2001 NIOSH contracted with Anthrotech, Inc., to perform a study, “A Head-and-Face Anthropometric Survey of U.S. Respirator Users,” that would gather facial size data for use in the development of new fit-test face panels (Appendix C). These proposed face panels in turn are to be used in testing of respirators to
ensure that respirators certified by NIOSH fit as many respirator-wearing workers in today’s workforce as possible (Anthrotech, 2004).
This report is a review of that 2001 NIOSH-sponsored Anthrotech study.
Because it is not feasible to test the fit of a respirator on every worker who might use it prior to release, the historical practice has been to test each respirator’s fit against a representative collection of subjects who have faces of various sizes and shapes. In 1934, for instance, the U.S. Bureau of Mines issued its first test schedule for dust respirators (Schedule 21). Later, revised regulations required three men of differing facial features (lean, average, and full) to wear respirators in an atmosphere of coal dust (Bureau of Mines, August 20, 1934, and March 23, 1965). In the 1960s, tests for self-contained breathing apparatus and certain other types of respirators used 15 to 20 men with widely varying facial shapes and sizes to test the suitability of fit (Bureau of Mines, March 23, 1965). Other fit tests analyzed the subjects’ nasal secretions, and those that were free of coal dust were considered to have an appropriate fit (Drinker and Hatch, 1954). In 1972 new regulations (Code of Federal Regulations [CFR], 30 CFR 11) required facepieces to be designed and constructed in a range of sizes in order to fit persons with various facial shapes and sizes. Fitting requirements varied slightly from one type of device to another. For certain devices (such as gas masks and pesticide respirators), the manufacturer had the option to specify which facial sizes each mask was designed to fit, and the mask was then tested on faces in those size ranges (Code of Federal Regulations, 1972).
NIOSH and the Occupational Safety and Health Administration (OSHA) share responsibility for overseeing respiratory protection in the workplace and have established regulations for this purpose. Specifically, NIOSH has issued regulations which define respirator testing and certification in 42 CFR 84 (National Institute for Occupational Safety and Health, June 2, 1995). OSHA has issued regulations which define conditions under which employers are required to maintain respiratory protection programs in general industry (29 CFR 1910.134); shipyards (29 CFR 1915.154); marine terminals (29 CFR 1917.92); and construction (29 CFR 1926.103) (Occupational Safety and Health Administration, 1998). This report focuses on the establishment of the scientific base re-
quired for certification standards of respirators, not their use in the workplace. However, the committee does recognize that effective certification is not enough to guarantee that respirators do their job properly. Even the best respirators may not be protective if they are not used in the context of a respiratory protection program as described in the OSHA regulations.
CERTIFICATION OF FACE MASK RESPIRATORS
To ensure proper fit of respirators, manufacturers must demonstrate that the respirators meet NIOSH’s certification requirements (42 CFR 84.135), which read:
Half-mask facepieces and full facepieces shall be designed and constructed to fit persons with various facial shapes and sizes either by providing:
1. more than one facepiece size; or
2. one facepiece size that will fit varying facial shapes and sizes (NIOSH, 1995).
This regulation does not specify how the face masks are to be designed, selected, and fitted to the various facial shapes and sizes, but historically NIOSH has certified respirators by testing them on panels of human subjects (Code of Federal Regulations, 2004). Because of the cost and time required by testing masks on human subjects, anthropometric face panels typically contain no more than about two dozen people. Therefore, each panel member must represent a large subset of the total population of respirator users and the entire anthropometric face panel must represent an overwhelming majority (e.g., 95 percent) of the same population (Campbell et al., 2001). This in turn means that the reliability of the certification process depends heavily on how well the anthropometric face panel represents the target population.
LOS ALAMOS NATIONAL LABORATORY PANEL
The test panels in use today are based on military anthropometric data collected from 1967–1968 (Hack et al., 1974; Hack and McConville, 1978). Researchers at LANL utilized facial and other measurement data available from a large number of young men and women in the U.S. Air Force to design respirator face panels (Hack et al., 1974). The original panel consisted of 16 men and was later expanded to include 9 women for a total of 25 persons. For more than 30 years, new respirators have been tested and certified by NIOSH against groups of 25 people, 16 male and 9 female, chosen to have facial shapes and sizes representative of that 25-person LANL panel (Campbell et al., 2001).
Military personnel, however, tend to have physical characteristics that differ from those of the general civilian workforce. Specifically, they tend to be younger, they must adhere to strict height and weight requirements, and they generally have a lower percentage of body fat. Thus it is reasonable to ask whether personal protective equipment designed and sized according to data from U.S. Air Force personnel would provide appropriate protection to the full range of civilian workers.
Respirators sized for the general working population must take into account the differences among workers in age, health status, and ethnicity—which the LANL anthropometric face panels do not do. Furthermore, the demographics of the U.S. workforce have changed significantly since 1972. The result is that there has been a growing concern among those who regulate respirators that the LANL face panels do not adequately represent the modern U.S. workforce.
Several studies add weight to this concern. In 1975, Leigh studied 1,467 Dow Chemical employees in the Atomic Energy Commission’s Rocky Flats Plant in Colorado and found that 10.3 percent of half-face respirator users and 12.6 percent of full-face respirator users had measurements that fell outside the LANL face panel specifications (Leigh, 1975). A U.S. Bureau of Mines survey of 48 male mine-rescue workers also indicated a need for a more extensive survey of the industrial population (Stein, 1978). And in 2001, when NIOSH compared the face length and width measurements of the 2,391 civilian subjects in the Civilian American and European Surface Anthropometry Resource (CAESAR) survey with the same measurements from the subjects of the 1972 Air Force study, it found that approximately 16 percent of today’s workforce have facial measurements that are outside the range of measurements in the Air Force’s LANL face panel dataset (Zhuang, 2001).
Researchers have also found that face length and lip length—which were used to define the LANL half-face fit-test panel—are not as predictive of fit as other facial dimensions, such as interpupillary distance, nose breadth, and nose protrusion (Zhuang, 2001). This raises the question of how well the LANL half-mask face panel may determine whether a particular respirator design will fit most of the workers who will need to use it.
THE 2001 NIOSH-SPONSORED ANTHROTECH STUDY
As previously mentioned, in 2001 NIOSH contracted with Anthrotech, Inc., to perform a comprehensive anthropometric study of respirator users in the U.S. workforce and to propose new fit-test face panels based on their findings. To accomplish these aims, Anthrotech recruited 4,026 subjects from 41 different sites in eight states. The subjects were employed in various industries and organizations that required either the regular or the occasional use of respirators. Each subject was measured for 21 dimensions, such as Menton-Sellion length, bizygomatic breadth, and lip length, using traditional anthropometric tools (see Chapter 2). Three-dimensional (3D) surface scans were also acquired on approximately a quarter of these individuals, allowing investigators to collect linear measurements that were used in the development of the LANL face panel, while also recording the various curves and contours of the subjects’ faces (see Chapter 2) (Anthrotech, 2004).
Based on the data collected Anthrotech proposed two new panels for half- and full-face respirator fit testing. These panels were directly comparable with the panels based on the earlier LANL data, as each was based on the same pair of manually collected linear measurements used in the corresponding LANL face panel (e.g., Menton-Sellion length and lip length for the half-face panel). In other words, though new data were used, the data were analyzed using many of the same methods employed for the earlier panels. NIOSH also performed a principal component analysis (PCA) to identify the combination of facial dimensions that best accounted for the variation among faces in the data set and designed a panel based on those facial dimensions as well. NIOSH suggested that perhaps this new PCA-based face panel might be superior to the traditional two-dimensional panels for the purpose of testing respirators.
The final report provided by Anthrotech describes the sampling and measurement procedures and discusses the fundamentals of the analysis
of the data collected. It also includes its proposed anthropometric face panels, summary tables of anthropometric statistics, and a brief examination of the 3D data collection and interpretation.
The results of this survey and the analysis that followed were detailed in the 2004 Anthrotech report to NIOSH, “A Head-and-Face Anthropometric Survey of U.S. Respirator Users” (Anthrotech, 2004) (Appendix C) and supporting documents, (Zhuang et al., 2004; Zhuang and Bradtmiller, 2005; Zhuang et al., 2005).
THE INSTITUTE OF MEDICINE REVIEW OF THE NIOSH-SPONSORED ANTHROTECH STUDY
In 2005, NIOSH contracted with the Institute of Medicine (IOM) to establish an ad hoc committee to review the NIOSH-sponsored Anthrotech study (Box 1-1). The committee’s charge was to study the NIOSH-sponsored Anthrotech study along with its supporting information and reports, and to examine and report on the adequacy and relevance of the study protocol, the analyses conducted, the resulting anthropometric dataset, and the appropriateness of the respirator fit-test panels derived from the new dataset. In particular, the committee was requested to assess how well the revised panel of facial characteristics represents the diverse U.S. workforce; whether the sample population, data collection methods, and data analyses were appropriate; whether the selected anthropometric features and parameters were adequately considered in the new respirator face panels; and whether the conclusions were well supported by the study results.
The committee was also asked to make recommendations concerning additional analyses that NIOSH may wish to undertake and additional information NIOSH might derive from current and future research. Finally, the committee discussed a model for translating the suggested respirator criteria derived from research into practice, with consideration given to the potential economic and social impact these changes would have on both manufacturers and the workforce.
STATEMENT OF TASK
The committee will review the NIOSH-sponsored Anthrotech report, “As-sessment of the NIOSH Head-and-Face Anthropometric Survey of U.S. Respi-rator Users.” The Committee will examine the adequacy and validity of the NIOSH study, the data collected, and the recommended revisions to the set (panel) of facial characteristics (anthropometric features) that are to be used in testing the fit of respirators.
Finally, the committee will suggest methods for translating the suggested respirator criteria derived from research into practice.
In assessing the NIOSH “Head-and-Face Anthropometric Survey,” the IOM committee received presentations from NIOSH’s National Personal Protective Technology Laboratory as well as from experts from the military and industry. It reviewed an extensive body of journal articles as well as manuscripts and other supporting materials provided by NIOSH. A full discussion of the committee’s process can be found in Appendix A.
The product of the committee’s work is this report. Chapter 2 describes and analyzes the anthropometric measurements performed by Anthrotech for its NIOSH-sponsored study. Chapter 3 looks at the survey methods used by Anthrotech in the study. Chapter 4 examines how Anthrotech analyzed its data to derive fit-test panels and suggests some ways that analysis could be improved. And Chapter 5 discusses future
directions, pointing toward further analyses of the data and offers suggestions for moving from research to practice. Although the revised panel is likely to have some impact on respirator manufacturers and manufacturing practices, discussion of this impact was outside the committee’s statement of task and will not be included in this report.
Anthrotech. 2004. A head-and-face anthropometric survey of U.S. respirator users: Final report. Prepared by B. Bradtmiller and M. Friess for NIOSH/NPPTL.
Bureau of Mines. August 20, 1934. Procedure for testing filter-type dust, fume, and mist respirators for permissibility: Schedule 21b. Washington, DC: Bureau of Mines.
———. March 23, 1965. Filter-type dust, fume, and mist respirators, requirements investigation, testing, and certification: Schedule 21b. Washington, DC: Bureau of Mines.
Campbell, D. L., C. C. Coffey, and S. W. Lenhart. 2001. Respiratory protection as a function of respirator fitting characteristics and fit-test accuracy. American Industrial Hygiene Association Journal 62:36-44.
Code of Federal Regulations. 1972. Respiratory protective devices: 30 CFR 11.
———. 1995. Respiratory protective devices: 42 CFR § 84.
———. 1998. Respiratory protective devices: 29 CFR § 1910, 1915, 1917, 1918, 1926.
———. 2004. Respiratory protective devices: 42 C.F.R. § 84.135.
Drinker, P., and T. Hatch. 1954. Dust respirators and air masks. In Industrial dust: Hygienic significance, measurement, and control. 2nd ed, edited by P. Drinker, and T. Hatch. New York: McGraw-Hill. Pp. 333-346.
Hack, A. L., and J. T. McConville. 1978. Respirator protection factors: Part I — development of an anthropometric test panel. Am Ind Hyg Assoc J 39(12):970-975.
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.
Leigh, J. 1975. Dow chemical corporation: Quantitative respirator man-testing and anthropometric survey. Dow chemical corporation report to U.S. energy research and development administration. Golden, CO.
Stein, R. L. 1978. Selected head and facial dimensions of mine rescue team personnel. American Industrial Hygiene Association Journal 39(7):576-578.
Zhuang, Z. 2001. Anthropometric survey of respirator users: Study protocol. National Institute for Occupational Safety and Health.
Zhuang, Z., and B. Bradtmiller. 2005. Head-and-face anthropometric survey of U.S. respirator users. J Occup Environ Hyg 2(11):567-576.
Zhuang, Z., J. Guan, H. Hsiao, and B. Bradtmiller. 2004. Evaluating the representativeness of the LANL respirator fit test panels for the current U.S. civilian workers. Journal of the International Society for Respiratory Protection 21:83-93.
Zhuang, Z., Coffey, C.C., and R.B. Ann. 2005. The effect of subject characteristics and respirator features on respirator fit. J Occup Environ Hyg 2(12):641-649.