Document gQQrNrJZJrMQKGqrjDEbqvzL

PLAINTIFF'S EXHIBIT DUP-2375 RESPIRATOR WORKPLACE PROTECTION FACTORS FOR ASBESTOS SUMMARY AND BACKGROUND Workplace protection factors for elastomeric and single use half-mask respirators were measured during asbestos removal operations. The operations consisted of asbestos removal from a ceiling (fire proofing) and removal of pipe insulation. The U6e of water/surfactant solutions to spray fire proofing and insulation resulted in conditions of high humidity for in-doors operations. Out-of-doors operations were also under high humidity conditions since light rain occurred on two of four days of out-of-doors sampling. Temperatures for both operations were in the 65-85*F range. The respirators were used in the context of a respirator program that followed the guidance in 29 CFR 1910.134 and ANSI 288.2 (1980) including proper respirator selection and fit testing. A total of seventeen volunteers participated in the study. One or two workplace protection factors were determined for each respirator for which the fit test had been passed. The respirators tested were: One successfully fitted elastomeric facepiece respirator chosen for comfort from six respirators (MSA Comfo II and Survivair 2000 brands - small, medium or large sizes). These respirators were tested using a dust, fume and mist type filter and a high efficiency filter. M I U400 2 Each of three disposable respirators successfully fitted (3M 9910, 3M 8710 and American Optical R1050). ' A successfully fitted North 7700 elastomeric facepiece respirator, chosen from the small, medium or large size facepieces. In addition two workplace protection factor measurements were made for a self-contained breathing apparatus operated in the pressure-demand mode. Results showed that, with the exception of the American Optical (AO) R1050, all respirators tested reliably provided workplace protection factors of ten or greater based on estimates of the lower 5th percentile. The AO R1050 reliably provided a protection factor of five or greater. These results are consistent with information supplied to the record by Los Alamos National Laboratory which reported on the penetration of asbestos through respirator filter media. Results for the self-contained breathing apparatus suggest that it may not provide appreciably better protection than the two better disposable respirators and the elastomeric facepiece respirators, though it creates significantly greater hazard of falling because of the weight and awkwardness of the tanks. Myhre et al.^2^ and Raven^3^ have 6hown that pressure-demand type respirators do not maintain positive pressure inside the facepiece when workloads are increased from n I !! 4 a 1 3 reet conditions. This may explain in part why no significant difference exists between half facepiece negative pressure and pressure-demand respirators. PROCEDURE Prior to testing, each person was trained in the selection and fit testing of respirators using the National Paint and Coatings Association Training Program^4^. Each was then fit tested using the saccharin fit test^4^. Respirators selected for use by participants are listed above and in Table I. Two individuals were also trained in use of a pressure-demand self-contained breathing apparatus. Fit testing was not done, as it is not required for pressure-demand equipment in the ANSI 286.2 (1980) standard. During sample collection, each participant was observed continually to ensure that the sample train remained intact and did not interfere with the respirator seal on the face, and to note occurances which might have affected respirator performance, such as movement of the respirator on the face. The participants were very busy with their work activities and 6oon ignored the presence of the researchers. AIR SAMPLING AND ANALYSIS To determine the workplace protection factors, concurrent samples were taken from inside the respirator (in-mask) at nose level and outside the respirator at the A I !. J 4 8 2 4 lapel (lapel). Closed-face Glasrock (;505) cassette filter holders fitted with half inch extenders and probe6 developed by KIOSH^5^ were used for both the in-mask and lapel samples. The sample probes are specially designed to minimize the los6 of particulates. The cassettes contained O.S urn, 25mm cellulose ester filters. The sample trains were calibrated before and after taking each sample U6ing a mass flow meter. The mass flow meter was calibrated against a bubble flow meter at the beginning and end of the study. A flow rate of 2 Lpm was used for in-mask samples. Lapel samples were taken at 0.5-1.0 Lpm to avoid overloading. Samples were collected for a one to two hour period. Each sample represented a single wearing of the respirator. Fiber counts were done per NIOSH procedure P&CAM 239^6* (phase contrast microscopy), except that the triacetin/acetone mounting method was used as described in NIOSH method 7400v . Counting was don according to the MA" counting rules in method 7400. Five hundred fields were counted for in-mask samples to increase analytical sensitivity. All analyses were done by one counter who participates successfully in the American Industrial Hygiene Association's Proficiency Analytical Testing quality assurance program. Concentrations were calculated from the actual number of fibers counted (even if fewer than 50 fibers were counted) per the formulas in the NIOSH method. a i 0483 5 NIOSH method P&CAM 239 reports a coefficient of variation of 0.12 which applies to lapel samples in this study. We estimate that the coefficient of variation for the in-mask samples is 0.4 at the median in-facepiece concentration and fiber count (Table III). SAMPLING AND ANALYSIS METHOD VERIFICATION In-mask sampling required use of closed-face, filter cassettes. The Glassrock #1505 25mm filter cassettes used have a unique tapered design for the outlet and a half inch extender which improves the deposition patterns^8^. To minimize sample loss, the NIOSH^^ probe was used to withdraw in-ma6k samples. To eliminate possible bias, the 6ame closed-face/probed cassette was used for both in-mask and lapel samples. To document whether concentrations determined from closed-face/probe cassette sampling differed from open-face sampling, twelve area samples were taken using both methods. Besults are compared in Table II with Student'6 and Variance Ratio tests. The two sampling methods were found to give similar results with no significant bias or difference in precision. RESULTS AND DISCUSSION Transmission electron microscopic analyses were made on four lapel samples to document the distribution of fiber diameters and lengths. Distributions are shown in Figures 1 and 2. Fiber diameters and lengths covered a wide range. ^ Q i? ["it -6The overall Median concentrations of fibers for in-nask and lapel samples are given in Table III. The Median number of fibers counted for in-mask (500 fields counted) and lapel (<.100 field counted) samples are also given in Table III, Cumulative distributions, geometric means, geometric standard deviations, be6t estimates of 5th percentiles and lists of workplace protection factors are given in Figures 3 through 8 for each respirator. The distributions of workplace protection factors were found to be approximately lognormal for all respirators studied. Therefore, logs were UBed for the plots, geometric means are reported as the proper measure of central tendancy and geometric standard deviations are reported as the proper measure of variability. Best estimates of 5th percentiles represent the workplace protection factors that most (95\) respirator users are expected to obtain from each respirator. Distribution plots and geometric standard deviations include variability from the sampling and analytical method as well as from the workplace protection factors. The unavoidable inclusion of the sampling and analytical variability in the plots re6ult6 in be6t estimates of 5th percentile protection factors which are conservative (lower than actual). This should be considered in evaluating results. M / !. 1 4 5 -7 Workplace protection factors were determined for a self-contained breathing apparatus for two individuals. Results are reported in Table IV. Table V gives geometric means and standard deviations for all respirators studied. A Bonferroni*9) test for differences (p-0.05) was done and results are given in Table V. Workplace protection factors for the 3M 9910 were found to be significantly higher than those for the AO R1050. No other significant differences were found. The relatively low results for the AO R1050 are likely due to poorer filter efficiency for asbestos, reported by Dr. Ortiz of Los Alamos National Laboratory.^ The unexpected comparability of the results from the 3M disposable respirators, and elastomeric facepiece respirators with both du6t. mist and fume filters and high efficiency filters may be explained by several factors. First, the limiting factor in performance with the respirators is likely face fit, not filter efficiency (except for the AO R1050 as noted above). Since the same fit test was used to select good fitting respirators for the study, comparable results would be expected. In addition, the researchers noted a tendency for elastomeric respirators to slip around when some wearers' faces became wet with the water/surfactant 6pray. This did not occur with the disposable respirators because their fibrous material of construction clung to the 6kin. /a i114 8 6 -B- It i.6 particularly noteworthy that the workplace protection factors for the self-contained breathing apparatus were not superior to the better disposable and elastomeric respirators. Myhre*2* and Raven*3* have Bhown that at workloads exceeding approximately 35t of maximal aerobic capacity pressure inside the facepiece of pressure-demand type respirators does not remain positive with respect to the surrounding atmosphere. Thi6 would provide an opportunity for asbestos to be drawn into the facepiece. The volunteers who wore the self-contained breathing apparatus complained of stress and discomfort due to its weight and bulk. Movement and balance were difficult and one volunteer almost fell from a step ladder as a result. Based on their observations, the researchers believe that routine use of self-contained breathing apparatus for aBbestos removal in the chemical industry would likely result in injuries, in view of the large amount of climbing involved. Certainly, acceptance will be much poorer than for the other respirators studied, due to discomfort which would discourage respirator U6e. Transmission electron microscopic analyses were done on in-mask and lapel samples. Resultant workplace protection factors are reported in Table VI. Results for small diameter fibers (<0.2um - the diameter below which fibers cannot be resolved with the standard optical microscopic method) do not appear to differ from those for large diameter (>0.2um) fibers, for 5 um or longer lengths. Thus, results for fibers visible t *>48/ with the phase contrast aethod (>0.2um diameter) appear to be representative of those for smaller diameter fibers for the 3M 8710 and 9910. and for elastomeric respirators with a dust, fume and mist filters and high efficiency filters. - 10 CONCLUSIONS 1. All respirators tested provided protection against asbestos. The American Optical R1050 disposable respirator reliably provided workplace protection factors of 5 or greater. The 3M 9910. 3M 8710. and elastomeric half-face respirators with both dust, fume and mi6t and high efficiency filters reliable provided workplace protection factors of 10 or greater. 2. Use of higher efficiency filters did not appear to improve the workplace protector factors of the elastomeric respirators. 3. Workplace protection factor measurements for a pressure demand self-contained breathing apparatus were not significantly higher than those for the better disposable and elastomeric respirators. Comments from the test volunteers and observations indicate that acceptance of the self-contained breathing apparatus will be poorer than that of the other respirators tested. In addition, they present a significant risk of tripping or falling where removal work requires climbing ladders and working from scaffolding. 4. Transmission electron microscopic analyses indicate that overall workplace penetration of small diameter (<2um) fibers (longer than 5 urn) is not discernibly higher than that of larger diameter fibers for the 3H 8710 and 3M 9910 disposable respirators nor for elastomeric respirators with dust, fume and mist or high efficiency filters. i (14 8 9 iiiiliinil--iBHiMBMMi - 11 REFERENCES 1. Ortiz. L. W. et *1: "Interim Report: Penetration of Respirator Filters by an A6besto6 Aerosol." May 2, I960 .(Los Alamos National Laboratory). 2. Myhre. L. G.. R. D. Holden. F. W. Baumgardner and D. Tucker: Physiological Limits of Firefighters. Unclassified Report #ESL-TR-79-06. AFESC, Tyndall AFB. Florida (June 1979). 3. Raven. P. B. et al.: "Physiological Response to 'Pressure-Demand' Respirator Hear". Am. Ind. Hyg. Assoc. J. 43 (10): 773-781 (1982). 4. "Guide to Respirator Fit Testing" (1981). National Paint and Coating Association. Washington. D.C. 5. Liu. B. Y. U. et al: "In-Ma6k Aerosol Sampling for Powered Air Purifying Respirators," Am. Ind. Hyg. Assoc. J. 45(4):278-283 (1984). 6. U.S. Department of Health. Education and Welfare; NIOSH Manual of Analytical Methods. Second Edition, Volume I (April 1977). 7. National Institute for Occupational Safety and Health, Method 7400. February 15, 1984. 8. Adams. M. J. et al, "Cassette Extenders: Effects on Chrysotile Fiber Deposition and Concentration Measurements". Presented at the American Industrial Conference. May 24. 1983. 9. Snedecor. G. W. and W. G. Cochran: Statistical Methods, 7th Ed.. Iowa State University Press. Ames. Iowa (1980). a i U490 Respirator Name Survivair 2000 Comfo II 3M 8710 3M 9910 American Optical R1050 North 7700 Scott Air-Pak TABLE I Respirators Studied Tvpe Elastomeric (silicone rubber) with dust, fume and mist and high efficiency filters Elastomer (neoprene) vith du6t, fume and mi6t and high efficiency filters Disposable vith nonadjustable straps Disposable vith adjustable straps Disposable vith nonadju6table straps Elastomeric (6ilicone rubber) vith high efficiency filters Self-contained breathing apparatus, pressure-demand Vendor U.S.D. Corp. Mine Safety Appliance Co. 3M Company 3M Company American Optical Company North Company Scott Aviation ! !>4y 1 TABLE II COMPARISON OF ASBESTOS SAMPLING METHODS (OPEN-FACE VERSUS CLOSED-FACE WITH PROBE) Mean Parameter Standard Deviation T Statistic* Variance Ratio** Data Open-face cassette: Closed-face cassette: Value 2.2 fibers/mL - open face filter cassette 2.3 fiber6/mL - closed face cassette with probe 0.264 fibers/mL - open face cassette 0.286 fibers/mL - closed face cassette with probe 1.28 1.17 1.9. 1.9. 2.0. 2.1. 2.1, 2.1, 2.2. 2.2, 2.2. 2.2. 2.3. 2.9 2.0. 2.0. 2.0. 2.2. 2.3. 2.3. 2.3. 2.4. 2.4. 2.4. 2.5. 3.0 * 1.28 is less than a comparison t statistic of 1.78 (p~0.05). Therefore, the means do not differ significantly. ** 1.17 is less than a comparison F statistic of 2.69 (p=0.05). Therefore, the variances (precision) of the methods do not differ significantly. '4 i i) 4 y 2 m iitg TABLE III MEDIAN ASBESTOS CONCENTRATIONS AND COUNTS FOR WORKPLACE PROTECTION FACTOR STUDY FOR ALL RESPIRATORS COMBINED Concentrations Inside Respirator Outside Respirator 0.006 fibers/mL (8 HR. TWA ~ 0.003)* 2.6 fiber6/mL (8 HR. TWA in 1.3)* Count6 Inside Filter Outside Filter 9 fibers 130 fibers No more than four of eight work hours were spent doing asbestos removal. Therefore, a factor of 0.5 (4 HRS/8 HRS) was applied to concentrations to estimate 8 HR TWA concentrations. U If 14 y 3 TABLE V COMPARISON OF WORKPLACE PROTECTION FACTORS FOR SEVEN RESPIRATORS Respirator Number of Measurements Workplace Protection Factor Geometric Mean fGeom. S.D.) Significant Differences 3M 6710 16 310 (5.3) 3M 9910 14 560 (4.2) Higher than R 1050 American Optical R1050 7 52 (4.2) Lover than 3M 9910 Elastomeric/Dust. Fume and Mi6t Filter 17 240 (6.3) Elastomeric/High Efficiency Filter 14 94 (3.0) North 7700 High Efficiency Filter 14 250 (6.9) Self-Contained Breathing Apparatus Pressure-Demand 2 620 Lover 5th Percentile 20 55 5 12 16 11 Based on a Bonferroni te6t(5) at a 0.05 level of significance, only the 3M 9910 and AO R1050 differed. !'J 4 y 5 TABLE VI WORKPLACE PROTECTION PACTORS FROM TRANSMISSION ELECTRON MICROSCOPIC ANALYSES : _________________ Workplace Protection Factor______________ Respirator Fibers Exceeding 0.2 um Diameter and S urn Length* All Diameter Fibers Exceeding 5 un Length 3M 8710 330 120 3M 9910 47 41 Survivair 2000 with 28 Dust. Mist and Fume Filter - 29 Survivair 2000 with High Efficiency Filter 21 19 * Those which would be counted by the standard phase contrast microscopy method. *4y6 Figure 3 - Workplace Protection Factors for 3M 8710 Respirator Cumulative Distribution: (Lognormal) Geometric Mean: 310 _ Geometric Standard Deviation: 5.3 A I U4! Best Estimate of 5th Percentile: 20 Data: 7.4, 15, 61, 110, 110, 180, 200, 310, 400, 420, 430, 1000, 1000, 1100, 1500, 1600, 1800, 3200 Figure 4 - Workplace Protection Factors for 3M 9910 Respirator Cumulative Distribution: (Lognormal) WPF Geometric Mean: 580 ,, .,, 4 y Q Geometric Standard Deviation: 4.2 Best Estimate of 5th Percentile: 55 Data: 94, 110, 150, 150, 170, 280, 550, 630, 710, 1300, 2400, 3000, 3700. 5800 Figure 5 - Workplace Protection Factors for AO R1050 Respirator Cumulative Distribution: (Lognormal) Geometric Mean: 52 Geometric Standard Deviation: 4.2 Best Estimate of 5th Percentile: 5 Data: 9.7, 26, 28, 38, 52, 75, 970 Figure 6 - Workplace Protection Factors for Survivair 2000 and MSA Comfo Halfmask Respirators with Dust, Fume and Mist Filters Cumulative Distribution: (Lognormal) WPF Geometric Mean: 240 Geometric Standard Deviation: 6.3 Best Estimate of 5th Percentile: 12 Data: 15, 24, 45, 46, 47, 74, 88, 140, 190, 370, 480, 1000, 1100, 1100, 1800,4100, 4200 Figure 7 - Workplace Protection Factors for Survlvair 2000 and MSA Comfo Halfmask Respirators with High Efficiency Filters Cumulative Distribution: (Lognormal) Best Estimate of 5th Percentile: 16 A 1 050 1 Data: 12, 28, 32, 42, 65, 85, 110, 120, 140, 160,160, 220, 220, 780, 7900* Removed from plot analysis, as an outlier. Figure 8 - Workplace Protection Factors for North 7700 Halfmask Respirators with High Efficiency Filters Cumulative Distribution: (Lognormal) WPF Geometric Mean: 250 ^ 10 5 0 2 Geometric Standard Deviation: 6.9 Best Estimate of 5th Percentile: 11 Data: 12, 20, 36, 60, 74, 110, 260, 350, 400, 1000,1900, 2000, 2400, 3100 DISTRIBUTION LIST GENERAL OCCUPATIONAL HEALTH COMMITTEE B. L. BARTON R. L. BUJALSKI J. M. CLEVERDON T. L. COLLINS V. D. DUTCHER T. A. GAGNER W. F. GORHAM D. A. GOSSELIN A. J. HART S. E. HINE E. K. JENSEN V. H. JOHNKOSKI R. M. LAWTON J. A. LEONARD W. H. LONG D. E. MACNAB D. PEARSON R. E. PLEVAN T. P. RABY H. D. RHODES J. F. ROONEY H. D. SMILIE CHEMICAL PLANT INDUSTRIAL HYGIENISTS J. N. DERMIT F. GARCIA-SHARP D. GEARY W. D. NEAL J. E. NEFF R. E. PEELE S. E. ROBINSON K. WILLIAMS A. P. YALCINKAYA W. C. YOUNG L. W. YOUNGBLOOD CORPORATE INDUSTRIAL HYGIENE STAFF E. G. BROWN R. W. COPE W. S. EVERETT 0. P. MUKHEJA R. D. ONDOCSIN CORPORATE HS&EA B. BALLANTYNE L. A. CRISORIO N. W. GAINES D. L. HEYWOOD T. G. FORTNEY A. A. LANG H. C. LEWINSOHN T. A. LINCOLN R. R. RANKIN T. R. TYLER R. VAN MYNEN k 1 050 3