Annals of Occupational Hygiene Advance Access originally published online on March 2, 2007
Annals of Occupational Hygiene 2007 51(4):415-421; doi:10.1093/annhyg/mem005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Racial Differences in Respirator Fit Testing: A Pilot Study of Whether American Fit Panels are Representative of Chinese Faces
1 Department of Energy & Environmental Engineering, Zhongyuan University of Technology, No.41 Middle Zhongyuan Road, Zhengzhou, 450007, China
2 School of Environmental Science & Engineering, Donghua University, No.1882 West Yan'an Road, Shanghai, 200051, China
* Author to whom correspondence should be addressed. Tel: +86-21-6237-4807; fax: +86-21-6779-2522; e-mail: lyang.pa{at}gmail.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The current respirator fit test panels (RFTPs) developed by Los Alamos National Laboratory or improved by the National Institute for Occupational Safety and Health were based on facial anthropometric survey of American groups including few Chinese examples. Facial anthropometric dimensions for other race/ethnic groups are different. Despite the fact that China has been the largest national source of industrial labor supply, and Chinese is one of top 10 races for immigration to America and Britain, there have been few related facial anthropometric studies on RFTPs for Chinese. This pilot study was conducted to investigate the difference of facial anthropometric dimensions between Chinese and Americans, and whether American RFTPs are applicable to Chinese. The results show that facial anthropometric measurements of 461 Chinese subjects are different from those of American groups described in the literature. About 12–35% of the Chinese subjects fall outside the ranges derived from American panels. Chinese may have shorter and wider facial character than American groups. Current RFTPs, which are based on American facial anthropometric surveys, may not fairly represent the facial anthropometric characteristics of Chinese groups. The results can also help to construct new RFTPs for Chinese in the future. Further studies are needed to verify whether the results in this study would be in agreement with those of facial anthropometric survey in wider samples.
Keywords: ergonomics • fit testing • racial differences • respirator testing
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
There are many types of respirators used in industries for individual respiratory protection. The design and fit testing of respirators are based on respirator fit test panels (RFTPs), but facial dimensions depend on racial type, and this affects goodness of fit (Han and Choi, 2003; Kim et al., 2003; Yokota, 2005). The RFTP with the proper facial anthropometric dimensions should specify tightness of fit satisfactorily for >95% of the targeted race group (Hyatt et al., 1972; NIOSH, 1972; Hack and McConville, 1978; Oestenstad et al., 1990; Zhuang, 2004). Current RFTP used in China and other Countries were developed by Los Alamos National Laboratory (LANL) in 1978 (Hack and McConville, 1978), based on American groups including few Chinese examples. These RFTP have been widely utilized to determine how respirators fit the general working population. This raises the question of how well the American specification applies to Chinese faces.
Recently, China has been the largest supplier of national and international labor, and needs many respirators with good faceseal. In addition, Chinese is one of top 10 races for immigration to America (U.S. Census Bureau, 2006) (http://www.census.gov/statab/www/) and Britain (U.K. ONS, 2006) (www.statistics.gov.uk). To a certain extent, immigration from China may have changed demography of those countries. So as migration and globalization end the possibility of focusing on just one race, the need to obtain anthropometric data effortlessly on different race groups has been growing rapidly, and new RFTP based on facial characters of Chinese are needed within China and in other places where Chinese work. In 1988, head-face dimensions of 393 adults have been investigated, but no RFTP was constructed (CNIS, 1998). This study investigated differences of facial anthropometric dimensions for establishing RFTP between Chinese groups and American groups. Furthermore, this study also evaluated the application of the current RFTP and NIOSH panel for Chinese, and tried to demonstrate whether the desirable result of covering >95% of Chinese surveys can be realized by those panels.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
A total of 461 university students and teachers from Zhongyuan University of Technology and Donghua University volunteered for this facial anthropometric survey. The volunteer subjects consisted of 270 males and 191 females. The range of ages of the test subjects was 23–43 year old, covering most of the age range of Chinese workers, which is
20–60 years old (Adamchak, 2001).
Measurement of facial dimensions
Ten critical facial dimensions illustrated in Figure 1 were measured in this study because of their close relationship to the performance of respirators (Oestenstad et al., 1990; Han and Choi, 2003). To ensure that each dimension was measured accurately and consistently from one subject to the next, the dimensions were defined in term of body landmarks in accordance with 1988 Anthropometric Survey of the U.S. Army Personnel Project (Cherverud et al., 1990) and Anthrotech (Anthropology Research Project, Inc., Yellow Springs, OH). The abbreviations and description of the dimensions are shown in Table 1. The measurements of all facial anthropometric dimensions were made to 0.1 mm using sliding and spreading anthropometric calipers and plastic tape (SiberHegner & Co. Ltd, Zurich, Switzerland). In addition, Student's t-test was used to test gender differences in facial dimensions.
|
|
RFTPs
Anthropometric panels of facial dimensions are relied upon to provide size reference for respirators in many applications, such as respirator design and development. Current RFTP were developed in America by LANL based on 1967–1968 U.S. Air Force surveys. No special RFTP had been utilized in Chinese occupations and the design of respirators for Chinese occupational protection was commonly based on LANL RFTP.
LANL suggested that lip length (LIPLGTHH) and face length (MENSELLH) were key facial dimensions for half-mask RFTP, and the distribution of panel members by these dimensions is illustrated in Figure 2. Face width (BIZBDTH) and face length (MENSELLH) were the dimensions for defining full-facepiece RFTP, shown in Figure 3. Both LANL panels included 10 cells and 10 size categories. These dimensional criteria can be used to design respirators to be certified. For convenience, each size category contained a 10 mm range of face length and a 9 mm range of lip length (half-mask RFTP) or face width (full-facepiece RFTP). The facial anthropometry of LANL panels for respirators was assumed to be representative of U.S. adults with the expectation of accommodating 90–95% of the U.S. population (McConville et al., 1972). The number of subjects to be used in the panel was set at 25. This was a compromise between the need for a sufficient number of tests to develop good statistics and the requirement to test all devices submitted for approval in a reasonable length of time. Members in each cell referred to the number of test subjects, male or female, to be selected for the test panel. A total of 25 subjects were used, with the exact number in each cell being the best approximation to the population distribution from 1967 to 1968 U.S. Air Force surveys (Hack and McConville, 1978). In this study, the 10 cells are numbered as shown in Figures 2 and 3.
|
|
Recently, many analyses of the correlation between facial dimensions and respirator fit have demonstrated that face width (BIZBDTH) and face length (MENSELLH) should also be considered for defining half-mask RFTPs, not just full-facepiece panels (Liau et al., 1982; Gross and Horstman, 1990; Oestenstad and Perkins, 1992; Zhuang et al., 2005). Secondly, the demography of the U.S. population has changed over the last 30 years (Gordon et al., 1989). The old RFTP may not fairly represent the diversity of face sizes because
15% of the Civilian American and European Surface Anthropometry Resource (CAESAR) surveys and 19.9% of the U.S. Army personnel surveys were beyond the limits of LANL PFTP (Guan et al., 2002; Zhuang, 2004). So a new RFTP with 25-member for American groups has been suggested by NIOSH researchers to accommodate >95% of new facial anthropometric surveys on American groups (Zhuang, 2004). The NIOSH panel was defined with the abscissa of face width (BIZBDTH) and the ordinate of face length (MENSELLH). The NIOSH panel was divided into 10 cells as illustrated in Figure 4, which were recommended for selecting test subjects for the NIOSH total inward leakage testing.
|
We used the LANL RFTP and the NIOSH panel for analyzing whether they can fairly represent the facial anthropometric characteristics of Chinese groups.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The descriptive statistics of facial anthropometric measurements for all 461 Chinese subjects are summarized by gender in Table 2. The results show that all measurements were statistically different between genders (P < 0.001). Hence, the following results are comparative analyses between Chinese and American by gender. Table 3 shows the comparison between Chinese and American groups. From Table 3, lip length (LIPLGTHH) of Chinese males in this study is relatively larger than results of Amerian males from most of other studies except Liau et al. (1982). For lip length (LIPLGTHH) of females, the mean value in this study is also relatively large and only smaller than the result from Gross and Horstman (1990). Being similar to lip length, the mean value of face width (BIZBDTH) of all Chinese genders in this study is larger than those of all American surveys. However, the reverse is true of facial length (MENSELLH), Chinese faces being shorter than American.
|
|
In this study, the Chinese facial anthropometric dimensions were compared with the LANL panels and the new NIOSH panel. Figure 5 shows that most measurements of lip length and face length for Chinese subjects are within the LANL half-mask RFTP range. Table 4 gives the details of percentage distribution of Chinese data against LANL half-mask RFTP. The range of the LANL half-mask RFTP includes about 85.5% males and 91.8% females, or together >88% of the total Chinese facial measurements.
|
|
Anthropometric values of face width and face length from Chinese subjects were compared with full-facepiece RFTP by gender in Figure 6. When the Chinese bivariate distributions of face width (BIZBDTH) and face length (MENSELLH) are compared with LANL full-facepiece RFTP, the Chinese subjects are mainly scattered at the upper right side of the panels for males and at the lower right of the panels for females (see Figure 6). From Table 5, it can be calculated that
38.2% of Chinese males and 30.3% of Chinese females exceed the panel limits. Unexpectedly, the LANL full-facepiece RFTP excludes 35% of the total Chinese subjects.
|
|
Lastly, bivariate distributions of the face width and face length of Chinese subjects are shown against the NIOSH panel in Figure 7, and this shows that, the distributions of face width (BIZBDTH) and face length (MENSELLH) shift beyond the NIOSH panel limits. Although many scatter plots in Figure 7 are distributed within the cells of the right part of NIOSH panels, 11.5% male subjects appear outside the panel limits (see Table 6). For females, Figure 7 illustrates that many plots are located at the lower cells of NIOSH panels, but 18.8% measurements are beyond the panel limit (see Table 6). The number of total measurements included in the NIOSH panel is summed as 394, which is 85.5% of all subjects.
|
|
For further analysis, results of fitting Chinese data into the LANL half-mask RFTP, LANL full-facepiece RFTP, and NIOSH panel are detailed in Tables 4–6, respectively. The number in parentheses in the tables is the percentage of the bivariate distribution of key facial dimensions against the corresponding cell of the panels. Table 4 shows that each of cell 2, cell 4, cell 5 and cell 8 accommodates >10% of Chinese population, with the summed percent of 59.4%. Both cell 2 and cell 4 in the LANL half-mask RFTP are occupied by >26% of Chinese female subjects. This means that the facial anthropometric characteristics of Chinese females are predominantly in those cells—lip length from 43.5 to 52.5 mm and face length from 93.5 to 113.5 mm. Moreover, cell 8 accommodates 26.7% of Chinese male subjects, which means a large part of Chinese male subjects have lip length (52.5–61.5 mm) and face length (113.5–123.5 mm). From Table 5, only cell 5 accommodates >25% of Chinese male subjects. About half of all Chinese subjects have face width larger than the limit of LANL full-facepiece RFTP. Furthermore, a large part of Chinese male subjects has face length lower than limit of LANL full-facepiece RFTP. Although the NIOSH panel accommodates a similar total Chinese subjects as LANL half-mask RFTP, Table 6 shows that the most populous cells are cell 2 and cell 5 located at the right-bottom corner of NIOSH panel. Giving an overview of Tables 4–6, LANL RFTP can accommodate more Chinese female subjects than the NIOSH panel, but the result is reverse for Chinese male subjects.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
There are significant differences between the facial dimensions of Chinese and those of Americans. Compared with most Americans, the facial character of Chinese is short and wide. This indicates that it may be necessary to construct new RFTPs for Chinese. Standard deviation (SD) of face width (BIZBDTH) measurements from Chinese males in this study is relatively high, but SD for other measurements from Chinese is lower than those of other studies. The high SD of face width (BIZBDTH) illustrated in Table 3, 8.9 mm, indicates the large dispersion of anthropometric measurements of Chinese males. It can be attributed to the diverse region background of volunteer subjects in Zhongyuan University of Technology and Donghua University. Subjects from the north of China have relatively large face width, but subjects from the south of China have smaller face widths (Zheng and Lu, 1997; Zhou and Wu, 2001). Similar results can be found from Table 2, in which SDs of BIGBRH, MENSELLH (face length), TRGSUBNA and TRGMENA are also higher than 7.0 mm.
Most of facial anthropometric surveys (88% of total surveys) for Chinese subjects can be fitted into LANL half-mask RFTP. Without new RFTP constructed for Chinese, the LANL half-mask RFTP may be temporarily applicable to Chinese groups, especially for Chinese females. Results show that the LANL full-facepiece RFTP cannot be used in respirator fit testing for Chinese groups. As a temporary full-facepiece RFTP before new panel being constructed for Chinese, the NIOSH panel is recommended. Nevertheless, a new RFTP should be developed in order to cover >95% of Chinese populations as its original purpose.
For the three dimensions (lip length, face width and face length) currently used to define respirator test panels, all mean values of Chinese subjects are different from those of American. Previous studies demonstrated that face width differed more between race and ethnic groups than other dimensions, so caution must be exercised in selecting face width for constructing RFTP. A case in point is that NIOSH panel is better than LANL full-facepiece RFTP for Chinese, because the mean value of face width surveyed by NIOSH (Zhuang and Bradtmiller, 2005) is closer to new Chinese facial anthropometric survey than that surveyed by LANL (Hack and McConville, 1978).
Although Chinese can temporarily use respirators designed and fit tested by LANL panels and the NIOSH panel, there may be high probabilities of hazard when they wear those respirator products to prevent dangerous contaminants (Yang and Shen, 2005). Consequently, it is critical to establish new RFTPs basing on facial anthropometric surveys of Chinese.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The recent anthropometric survey in this study may be more representative of facial characteristics of the current Chinese civilian population. The study can provide respirator industries the anthropometric data on different race groups, especially Chinese, to meet the needs of their customers. The difference between the facial dimensions of Chinese and those of American indicate that it may be necessary to construct new RFTPs for Chinese individuals.
This study demonstrates that Chinese have shorter and wider facial characteristics and larger lip width than American groups. Current panels based on American facial anthropometric surveys exclude about 12–35% of the total Chinese subjects. Full-facepiece RFTPs in the U.S. cannot be applicable to Chinese individuals. However, LANL half-mask RFTP and NIOSH panel accommodated >85% of the all Chinese subjects and may be used as interim measure. But Chinese groups need the optimal RFTP based on their facial anthropometry.
Because China has more than 1.3 billion populations, relatively small number of samples in this study may make facial anthropometric measurements biased. Further research should be focused on large survey of different ethnic groups in diverse regions of China.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
The authors wish to acknowledge Dr. Ziqing Zhuang, CDC/NIOSH National Personal Protective Technology Laboratory, Pittsburgh PA, for his critical review of this research.
Received September 29, 2006; in final form January 6, 2007
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCES |
|---|
Adamchak DJ. The effects of age structure on the labor force, retirement in China. Soc Sci J (2001) 38:1–11.[CrossRef][Web of Science]
Cherverud J, Gordon CC, Walker RA, et al. 1988 Anthropometric Survey of U.S. Army Personnel: Correlation Coefficients and Regression Equations, Part 1 Statistical Techniques, Landmark and Measurement definition (TANICK/TR-90/032). (1990) Natick, MA: U.S. Army Natick Research, Development and Engineering Center Evanston. 48–51.
China National Institute of Standardization. Head-face dimensions of adults by. Xiao H, Hua DH, Yang TX, Zhang ZB, Bi GX, Liu W, Liu JM, eds. (1998) Beijing: General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. CNIS GB/T 2428:1998.
Gordon CC, Churchill T, Clauser CE, et al. 1988 Anthropometric survey of U.S. Army personnel: Summary statistics interim report (NATICK/TR-89/027). (1989) Natick, MA: U.S. Army Natick Research, Development, and Engineering Center. 264–312.
Gross SF, Horstman SW. Half-mask respirator selection for a mixed worker group. App Occup Environ Hyg (1990) 5:229–35.
Guan Jh, Hsiao HW, Zhuang Z. Recapturing the sizing issues of respirator fit-test panels for emergency response. (2002) Presented at Internaltioanl Society for Respiratory Conference: Edinburgh, U.K. 2002.
Hack AL, McConville JT. Respirator protection factors: Part I—Development of an anthropometric test panel. Am Ind Hyg Assoc J (1978) 39:970–5.[Web of Science][Medline]
Han DH, Choi KL. Facial dimensions, predictors of fit for half-mask respirators in Koreans. Am Ind Hyg Assoc J (2003) 64:815–22.
Hyatt EC, Pritchard JA, Richards CP. Respirator efficiency measurements using quantitative DOP man tests. Am Ind Hyg Assoc J (1972) 33:635–43.[Web of Science][Medline]
Kim H, Han DH, Roh YM, et al. Facial anthropometric dimensions of Koreans, their associations with fit of quarter-mask respirators. Ind Health (2003) 41:8–18.[Web of Science][Medline]
Liau YH, Bhattacharya A, Ayer H, et al. Determination of critical anthropometric parameters for design of respirators. Am Ind Hyg Assoc J (1982) 43:897–9.[Web of Science][Medline]
McConville JT, Churchill E, Laubach LL, et al. Anthropometry for Respirator Sizing. (1972) 14. Final Report. Prepared under Contract No. HEW Contract No. HSM 099-71-11, Cincinnati, OH: U.S. Department of Health, Education, and Welfare, Health Services and Mental Health Administration, National Institute for Occupational Safety and Health.
National Institute for Occupational Safety Health NIOSH, DHEW/NIOSH TR-004-73. McConville JT, Churchill E, Laubach LL, eds. (1972) Cincinnati, OH: National Institute for Occupational Safety and Health. 1–44. Anthropometry for respirator sizing by.
Oestenstad RK, Perkins JL, Rose VE. Distribution of faceseal leak sites on a half-mask respirator, their association with facial dimensions. Am Ind Hyg Assoc J (1990) 51:285–90.[Web of Science][Medline]
Oestenstad RK, Perkins LL. An assessment of critical anthropometric dimensions for predicting the fit of a half-mask respirator. Am Ind Hyg Assoc J (1992) 53:639–44.[Web of Science][Medline]
U.K. The Office for National Statistics. (2006) ONS, Series MN no.31, ISSN 0140-900X, International Migration—Migrants entering or leaving the United Kingdom and England and Wales (2004), November 29 2006.
U.S. Census Bureau. (2006) U.S. Census Bureau, Statistical Abstract of the United States: 2006, November 29 2006.
Yang L, Shen H. The assessment on the leakage of simple respiratory protection in ambient air. (2005) Proceedings of 10th International Conference on Indoor Air Quality and Climate, September 2005: Beijing, China. 1011–1015.
Yokota M. Head, facial anthropometry of mixed-race U.S. army male soldiers for military design and sizing: A pilot study. App Ergonom (2005) 36:379–83.[CrossRef]
Zheng L, Lu Sh. Cluster analysis, principal component analysis on the physical characters of 23 populations in China. ACTA Athropo Sinica (1997) 16:151–8.
Zhou W, Wu X. Quantitative study of some facial features of the modern human skulls with the phase-approaching method of project grating. ACTA Athropo Sinica (2001) 20:81–92.
Zhuang Z. (2004) Anthropometric research to support RFTPs. Presented at the CDC workshop on respiratory protection for airborne infectious agents, Atlanta, GA, November 2004.
Zhuang Z, Berry Ann R, Coffey CC. The effect of subject characteristics and respirator features on respirator fit. J Occup Environ Hyg (2005) 2:641–9.[CrossRef][Web of Science][Medline]
Zhuang Z, Bradtmiller B. Head-and-face anthropometric survey of U.S. respirator users. J Occup Environ Hyg (2005) 2:567–76.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||