Document jyomK7OGpVyBvNL6EbJkz7jqk

International Journal of Environmental Health Research 10, 5-19 (2000) An evaluation of personal airborne asbestos exposure measurements during abatement of dry wall and floor tile/mastic JOHN H. LANGE1 and KENNETH W. THOMULKA2 1Envirosafe Training and Consultants P.O. Box 114022 Pittsburgh, PA 15239, 2University of the Sciences of Philadelphia*, 600 South Forty-third Street Philadelphia, PA 19104 USA This investigation provides information on exposure to workers during asbestos abatement of dry wall and floor tile/mastic. Personal airborne exposure concentrations were collected during an asbestos abatement project involving dry wall material and floor tile/mastic. Twenty-five dry wall and twenty-three floor tile/ mastic personal air samples were collected during abatement. Exposure concentrations for dry wall and floor tile/mastic abatement were 0.85 fcm"3-TWA (time-weighted average) and 0.04fcm"3-TWA for arithmetic means and 0.72fcirf 3-TWA and 0.03 fcm''-TWA for geometric means, respectively. One outlier was determined for dry wall and none for floor tile/mastic. Sample distribution exhibited variability and was non-normal (logarithmic) for both types of materials abated. Probability of exceeding the Occupational Safety and Health Administration Permissible Exposure Limit for floor tiie/mastic was low. Employment of respirators during abatement of floor tile/mastic is not required by regulatory standards, but is necessary for dry wall abatement. Comparison of exposure for within- andbetween-workers suggests that the process of abatement is the important factor for source of exposure. These exposure data suggest that dry wall abatement workers are not a homogeneous group, but floor lile/mastic workers are a homogeneous group. Keywords', respiratory disease, occupational protection, asbestos fiber distribution, occupational groups. Introduction Exposure to airborne asbestos has been reported to be the causative agent for a number of diseases, especially associated with the respiratory system (Selikoff ex al. 1979, Dement el al. 1982, Nicholson 1982, Weill and Hughes 1986, Mossman and Gee 1989, Mossman et al. 1990, Health Effects Institute - Asbestos Research - HEI-AR 1991, Oliver et al. 1991, Bresnitz et al. 1993, Waage et al. 1994, Garcia-Closas and Christiani 1995, Koskinen et al. 1998). As a result of these potential diseases many countries have established regulations for asbestos-containing materials (ACM) and its control (Occupational Safety and Health Administration - OSHA 1986, 1998a; HEI-AR 1991; Lange et al. 1994, 1996a, Mlynarek et al. 1996). These regulations are designed to control exposure from ACM (Mlynarek et al. 1996, Lange et al. 1996b), with numerous reports published on both non-occupational and occupational airborne levels of Correspondence: John H. Lange. ISSN 0960-3123 printed/ISSN 1369-1619 online/00/010005-15 2000 Taylor & Francis Ltd HiiniiiiiTT' .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAINTIFF EXHIBIT SOA 5655 6 Lange and Thomulka asbestos in buildings (Lumley et al. 1971, Sawyer 1977, 1985, Bozzelli and Russell 1982; Paik et al 1983, Sawyer et al. 1985, Massey and Fournier-Massey 1987, Jaffery et al. 1988, Chessen et al. 1990, Com et al. 1991, HEI-AR 1991, Keyes et al 1991, Ganor et al. 1992, Lange et al. 1995a, 1995b, 1996a, Lange 1999, Mlynarek et al. 1996). From an occupational prospective, the purpose of air sampling is to provide information for protection to workers. Exposure concentrations are used to determine requirements for engineering controls, work practices, personal protective equipment (PPE) and hygiene (US Environmental Protection Agency - EPA 1987, OSHA 1998a). Implementation of specific practices, such as use of PPE, is determined by whether exposure exceeds the regulatory defined safe exposure level. This level in the US for occupational exposure is called the Permissible Exposure Limit (PEL) and is 0.1 fcnT3-time-weighted average (TWA; OSHA 1998a). Relatively few studies have been reported on exposure to asbestos during abatement of specific types of ACM (HEI-AR 1991, Lange et al. 1996a). Exposure data associated with abatement of a specific type of building material can provide historical information for future projects and establish an exposure assessment for epidemiological and occupational hygiene analysis (Environ Corp. 1992, OSHA 1994, Lange et al 1996a, Phillips and Esmen 1999). Identification of the source of exposure is important for determining applicable controls (Gardiner 1995). Comparative evaluation of within- and between-worker exposure variability measurements can be used to provide evidence as to whether individual practices or process is the most `important' contributor of the `source of exposure' (Kromhout et al. 1993, Gardiner 1995, Peretz et al. 1997). Variability of exposure for within- and between-workers is represented as a geometric standard deviation (GSD; Peretz et al. 1997). When within-worker variability measurements are larger than between-worker exposure variability individual practices are suggested to be the `source' of exposure, while a numerically lower between-worker exposure variability value as compared to within-worker variability measurement(s) is suggestive of the process being the `source' of exposure (Gardiner 1995). This paper presents personal air sample results that were collected during asbestos abatement of floor tile/mastic and dry wall material. These data along with other published exposure information can be used for establishing historical air concentrations of asbestos during abatement of asbestos-containing dry wall and floor tile/mastic (Lange et al. 1996a). Importance of air sample data for occupational protection is discussed. Materials and methods Personal air samples were collected from workers during asbestos abatement of dry wall and floor tile/mastic. This abatement was performed in a multistory building that is located within the Commonwealth of Pennsylvania, USA. All work was conducted during the summer of 1998. The air sampling technician selected the individuals to be monitored at random. Thus, sampling was conducted as would be performed for any abatement project. Each type of abatement, dry wall and floor tile/mastic, was performed separately. During abatement activities of dry wall and floor tile/mastic there were 25 and 23 measurements collected, respectively. Both building components were ACM (greater than 1.0% asbestos by volume) as defined by regulatory criteria established by EPA and Pennsylvania Department of Labor and Industry (DLI; EPA 1987, DLI 1991). Concentration of ACM was determined by polarized light microscopy (PLM) before abatement was started (EPA 1987). Asbestos abatement procedures were conducted as described and required by the OSHA and EPA standards for asbestos in the construction industry (EPA 1987, OSHA 1998a). All workers Airborne asbestos exposure 7 were trained as required by OSHA and DLI. Samples reported in this study are from abatement activities and do not include non-abatement activities (e.g. setup). Floor tile was abated by a lift and scrape technique (Lange et al. 1996a). Little if any wetting was used for removal of the floor tile. Mastic was abated by shot blasting (US Filter Blastrac Track, Oklahoma, OK) and chemical methods (Lange et al. 1996a, US Filter Blastrac Track 1998). All areas were abated by shot blasting except edges and locations where the machine could not be employed. Dry wall was abated by wet techniques and physical destruction methods. Due to the thickness, consistency and previous painting of dry wall, wet methods were difficult to perform effectively. This resulted is some material not being adequately wet during removal. Cleanup was part of this abatement activity and was conducted during removal and at the end of each shift. The number of workers per shift was approximately four. Each shift was 8h in length with one shift per day. Personal samples were collected on 25 mm diameter electrically conductive extension cowl cassettes containing a mixed cellulose ester membrane filter using a personal sample pump as described by OSHA (1998a). All measurements were from the breathing zone and had a sample flow rate of 2.01 min-1 (nominal) as determined by a calibrated rotameter (ASTM, 1990a, 1990b, 1993, Lange et al. 1996a, OSHA 1998a). Samples were collected in a downward and open face position (OSHA 1998a). Filters were analyzed using the NIOSH 7400 method that employs Phase Contrast Microscopy (PCM; OSHA, 1998a). The detection limit for this study was set at 0.01 fcm"3. All samples are reported as a time-weighted average (TWA). Exposure data, separately for dry wail and floor tile/mastic, were summarized by measures of central tendency (Lange et al. 1996a, 1998) using a computer program (Timko and Downie 3992). Distribution and existence of outliers for both non-transformed and transformed (natural logarithm) data were determined using the Shapiro-Wilk W test (Shapiro and Wilk 1965, EPA 1992) and Grubbs tests (Grubbs 1950, Grubbs and Beck 1972, Taylor 1990), respectively. Confidence interval (Cl), at 95%, for arithmetic means, with and without the outlier, were determined using non-transformed data for both dry wall and floor tile/mastic samples using a technique for non-normal populations (Daniel 3991). Skewness was determined using a computer program for non-transformed data (Timko and Downie 1992). All statistical calculations were performed at a 5% level unless otherwise noted. Sample data were evaluated for both floor tile/mastic and dry wall separately. These data were evaluated for total-, within- and semi-between worker exposure distributions (Kumagai et al. 1997). Total-worker exposure distribution included both categories of sample measurements, dry wall and floor tile/mastic. Distribution for total-worker exposure is actually a subtotal value since there were varying and non-uniform sample numbers among workers (Kumagai et al. 1997). Between-worker exposure is actually a semi-between determination since not all workers were included, some workers only had one or two measurements and the sample size is small (Kumagai et al. 1997). Calculation for between-workers was only determined for dry wall abatement. Due to the small number of within-worker measurements for floor tile/mastic, calculation of between-worker variation was not feasible. Within-worker calculations were conducted for both dry wall and floor tile/mastic abatement. Outliers were not determined for within and between sample measurements. Comparative evaluation of within- and between-workers was performed for determination of the source of exposure (Gardiner 1995). A low between-worker GSD as compared to within-worker GSD suggests the source of exposure is the process. However, a low within-worker GSD as compared to between-worker GSD is suggestive of individual practices being the source of exposure. 8 Lange and Thomulka Homogeneous or monomorphic group determinations were evaluated for both dry wall and floor tile/mastic workers (Rappaport 1991, Gardiner 1995). If the mean (arithmetic mean, AM) exposure concentration value for a group (either dry wall or floor tile/mastic) is within two-fold of 95% or two standard deviations of the sample population, the group is considered to be monomorphic or homogeneous (Rappaport 1991, Gardiner 1995). Probability (confidence coefficient) for occupational exposure (occupational exposure limit OEL or OSHA PEL), of at least 5% of the employees (workers, samples) exceeding the OSHA PEL or protection factor for a respirator category, was determined by `employee over exposure risk' calculations (Rappaport 1991) and graphic methods (Leidel etal. 1977). These calculations were conducted for dry wall and floor tile/mastic exposure using GSD and average (AM) exposure concentrations (Leidel et al. 1977, Rappaport 1991). For graphic methods a ratio of the exposure to the standard is required. The standard used in calculating the ratio was the OSHA PEL. Protection factors for a half and full face mask are 10- and 50-fold of the OSHA PEL or represented by an exposure value for a standard are 1.0 and 5.0 f cm-3-TWA (outside the mask), respectively. Pre-abatement (background) and post-abatement (final clearance) samples were collected using high volume sampling procedures. Results The asbestos content for floor tile, mastic and dry wall was 2-7%, 1-3% and 25-35%, respectively. These results suggest that each of the materials abated meet the definition as being ACM (DLI 1991). Abatement of dry wall material and floor tile/mastic was performed at different times. Dry wall was abated first then floor tile/mastic, both under the same containments. All work was performed within one containment. Pre-abatement (work) and final clearance samples were all below 0.01 fcrrf3. Quantity of material abated for both floor tile/mastic and dry wall exceeded the National Emissions Standard for Hazardous Air Pollutants (NESHAP) threshold values (EPA 1999). Table 1. Descriptive data and summary statistics for results of fiber counts (fcrrf3 TWA) Type N Mean SD GM GSD Range Sample disiribution Dry Wall 25 Dry* Wall 24 FT/Mastic 23 0.85 0.76 0.04 0.72 0.63 2.08 0.57 0.59 1.94 0.02 0.03 1.71 0.12-3.16 0.12-2.03 0.01-0.08 lSD-80.0% 2SD-96.0% 3SD-96.0% lSD-75.0% 2SD-91.7% 3SD-S00% lSD-78.4% 2SD-100% 3SD-100% N, number of samples; Mean is arithmetic mean; SO, standard deviation; GM geometric mean; GSD. geometric standard deviation; Range, arithmetic range; FT, floor tile; ^without outlier. Airborne asbestos exposure Table 2. Summary data for within and between measurements for individuals (within- and between-workers) from abatement of dry wall material (fcirf 3-TWA). 9 Type N Mean SD GM GSD Range Distribution Within Worker A Worker B Worker C Worker D Between Workers A-D 3 4 5 7 4 1.51 0.93 0.54 0.69 0.92 1.10 0.63 0.50 0.57 0.43 1.17 0.69 0.37 0.55 0.85 1.92 1.84 2.20 2.05 1.56 0.63-3.06 0.30-1.72 012-1.50 0.27-2.03 0.54-1.5! Non-normal* Non-normal* Non-nonnalf Non-normal? Non-normal? N, number of samples; Mean is arithmetic mean; SD, standard deviation; GM, geometric mean; GSD, geometric standard deviation; Range, arithmetic range. All measurements are persona! samples, fcirf^-TWA. Transformed values were not normal for distribution; ("transformed values were normal for distribution; ^Non-normality existed for distribution at 5% and was normal for distribution at 1%. Summary statistics for total-, within- and between-workers are shown in Tables 1, 2 and 3. One outlier was detected for non-transformed air samples measurements during abatement of dry wall. This outlier was the highest exposure value, 3.06fcm~3-TWA. There were no outliers when these data were transformed. Median values, in ferrf 3-TWA, for exposure concentrations from dry wall, with and without the outlier, were 0.60 and 0.58, respectively. CIs for dry wall exposure, with and without the outlier, were + /-0.28 and +/-0.23, respectively. When comparing Cl ranges for exposure from dry wall, with and without the outlier (0.57 to 1.13 fcnrf3-TWA and 0.53 to 0.99fcm~3-TWA, respectively), to the OSHA PEL (OSHA, 1998a), exposures were above this criterion and would require at a minimum employment of a half mask respirator. Eight out of 25 samples collected during abatement of dry wall were above the exposure level established for a half mask respirator and by these individual measurements would require at a minimum a full face mask respirator (OSHA 1998a). Mean values suggest that the work area would be defined as a regulated area for dry wall abatement based on OSHA regulations (OSHA 1998a). Even when the single outlier is removed summary data and individual values suggest that the average airborne concentration is above OSHA PEL. Table 3. Summary data for within measurements for individuals (within-workers) from abatement of floor tile/mastic material (fem 3-TWA). Type N Mean SD GM GSD Range Distribution Within Worker A Worker B 4 5 0.06 0.04 0.03 0.05 0.03 0.03 1.51 0.02-0.08 Normal* 1.81 0.01-0.08 Non-normalt N, number of samples; Mean is arithmetic mean; SD, standard deviation; GM, geometric mean; GSD, geometric standard deviation; Range, arithmetic range. All measurements are personal samples. Transformed values were not normal for distribution; ('transformed values were normal for distribution. 10 Lange and Thomulka Airborne samples collected during abatement of floor tile/mastic did not exhibit any outliers. Exposure concentrations for floor tile/mastic were all below the OSHA PEL (OSHA 1998a) with the highest sample concentration being 0.08fcm~3-TWA. The Cl for these data was + /-0.008, suggesting a range, at 95% of approximately 0.03 to approximately 0.05 fern'3-TWA. Evaluation of the Cl also suggests that the exposure level for this abatement is below the OSHA PEL and the work area would not be considered a regulated area based on OSHA requirements (OSHA, 1998a). The median concentration for these data was 0.03 f crrf3-TWA. Evaluation of the over exposure concentration and variability for determining the probability of at least 5% of the employee exposure will be over the OSHA PEL for floor tile/mastic is about 0.1 or 10% for both graphic and calculation methods (Leidel et al. 1977; Rappaport 1991). The probability of at least 5% of exposures exceeding the protection factor for a half and full face mask for dry wall material are approximately 0.3 (30%) and 0.03 (3%), respectively, using the calculation method. Both airborne concentrations for abatement of dry wall and floor tile/mastic were nonnormally distributed when non-transformed and normally distributed when transformed. Since these data are normal when transformed using a logarithm, this distribution would be suggested to be log-normal (Kumagai et al 1997). Standard deviations (SD) suggest these data are not normally distributed (Lange et al. 1996a). A non-normal distribution is also suggested by the GSD values and that these data have a right sided tail (Leidel et al. 1977). For airborne concentrations associated with dry wall, both with and without the outlier, the difference between the means, and similarity of the geometric mean (GM) and median support a non normal distribution (Lange et al. 1996a). The differences between means, and similarity for GM and median are not as evident for abatement of floor tile/mastic. Skewness values for abatement of dry wall, with and without the outlier, were 1.05 and 0.99, respectively. Both of these values are within a range suggested for normality at 1 %, but are both non-normal at 5% (Taylor 1990). The skewness value for abatement of floor tile/mastic was 1.27 which is suggested to be non-normal at both 3 % and 5% (Taylor 1990). Since both dry wall and floor tile/mastic airborne sample data have a positive skewness, this is suggestive of a tail to the right (Taylor 1990). Airborne concentration data for within-workers are shown in Table 2. Four workers in this study had three or more measurements. These data suggest that a wide range of average exposures exist for within each worker evaluated. The AM for these individual workers ranged from 0.54 to 3.51 fcm~3-TWA. GM showed a similar wide range for concentration values. Variability of these data are high as evident from both the GSD values and ranges for individuals. Three workers had two measurements and are not presented in the table. When these workers were evaluated a GSD of around 1.3 was determined for each individual. These data suggest low exposure variability of workers with two measurements and numerically, by GSD, an approximated normal distribution for exposure to these individual workers. AM values for these individuals were 1.52, 0.42 and 0.85 f cm~3-TWA. GM were similar to the AM and were i.45, 0.39 and 0.83fcm'3-TWA. Comparison of GSD values for within- and between-workers suggest that process of asbestos abatement associated with dry-wall is the `primary' source of exposure (Gardiner, 1995). No comparative evaluation can be formulated for floor tile/mastic due to the small number of individual comparison measurements. Airborne concentrations for individual workers that performed floor tile/mastic abatement are shown in Table 3. Two workers had three or more measurements with three workers having two measurements. Average exposure levels are similar for all workers with multiple measurements. Airborne asbestos exposure 11 No sample for either dry wall or floor tile/mastic abatement was below the limit of detection. The lowest value reported was 0.01 fcm_3-TWA. Thus, all calculations were performed using the actual numerical value (Burstyn and Teschke 1999). Distribution for each individual worker that performed abatement of dry wall is shown in Table 2. When evaluated as non-transformed values the distributions were non-normal. After transformation both normal and non-normal distributions were found. A similar finding for distribution was observed for abatement of floor tile/mastic as is shown in Table 3. Between-worker distribution suggests large exposure variation among measurements for dry wall abatement. Both the AM and GM are numerically different supporting variability of these data. Distribution for between data measurements suggests a non-normality. When all between values were used, including those with three or more measurements and only two measurements, a total of six individuals, variability as represented by the GSD increased to 1.63, which suggests a distribution of non-normality. The AM for six individuals was 0.84fcirf3TWA and is similar to the summary value with four individuals which was 0.92 fern" 3-TWA. Data with six measurements when non-transformed and transformed were non-normal and normal, at 5%, respectively. However, at 1%, non-transformed data with all values (six) were normal and transformed data remained normal. Evaluation of exposure data for dry wall, with and without the outlier, suggest that this population is not a homogeneous group. Concentration range for floor tile/mastic is suggestive of a homogeneous group. Discussion This study evaluated and analyzed personal asbestos airborne concentration data that were collected during abatement (removal) of dry wall material and floor tile/mastic. Personal samples are employed to determine exposure to workers (Lange et al. 1996a, 1998) and evaluate effectiveness of engineering controls and work practices (Corn 1983, 1985, OSHA 1986, 1994, 1995a, 1998a, Burstyn and Teschke 1999). Previous investigations of airborne exposure concentrations have suggested that summary airborne exposure data be represented by AM, GM, median, SD and GSD, with and without outliers (Lange et al. 1996a, 1997, 1998). The presence of an outlier had some influence on summary statistics, but was not as strong an effect as reported in a previous study (Lange et al. 1998a). (Table 1). Exposure concentrations during abatement of dry wall material exceeded the OSHA PEL for airborne asbestos (Tables 1 and 2). Eight samples also exceeded the numerical exposure value of l.Ofcnf 3-TWA for a half mask respirator. The average concentration for dry wall removal did not exceed the upper numerical value of protection provided by a full face mask respirator. However, when individual workers are evaluated one worker (worker A) exceeded the numerical value for a half mask as represented by both the AM and GM (Table 2). These high airborne fiber concentrations are a result of difficulty in wetting the dry wall material. Considerable effort was undertaken to adequately wet the material, but was only feasibly accomplished after breaking the dry wall. Such disturbance, allowing considerable releases of airborne material before wetting, was possible as is evident by the personal airborne concentration values. Work areas associated with dry wall abatement would meet criteria for defining this work location as a regulated area (OSHA, 1998a). The probability of overexposure for dry wall material, using summary data values, for half and full face masks were low. These data suggest that it is not likely that workers will be `highly' exposed to airborne asbestos or 12 Lange and Thomulka receive a large `dose' of airborne asbestos when using either a half or full face mask respirator. However, a probability at 30% for at least 5% of the workers being exposed to a concentration of greater than the protection factor for a half mask must be considered a potential concern. Evaluation from a practical scenario, this concern can be tempered in part, due to the low percentage of asbestos in the material and counting of all fibers by the method of analysis (PCM, HEI-AR 1991, Mlynarek et al. 1996). Use of a full face mask results in low exposure based on the estimated probabilities for al! fibers and asbestos fibers alone. Caution must be applied for calculation of asbestos fibers alone since this determination is dependent on the airborne distribution as a bulk material percentage of asbestos. Floor tile/mastic exposure values were well below the OSHA PEL (Tables 1 and 3). When probability of overexposure is considered, it is unlikely that any worker will be exposed to airborne concentrations above the OSHA PEL. Previous studies (OSHA 1986a, HEI-AR 1991, Environ Corp. 1992, Lange et al 1995b, 3 996a) of exposure from floor tile/mastic have suggested similar results and that the likelihood of exposure above the OSHA PEL is low. These data, in part, support the Resilient Floor Covering Institute (RFCI) abatement methodology for floor tile, as described in a settlement agreement with OSHA (OSHA, 1995a), in that exposure levels during abatement were low and well below the OSHA PEL (Student Manual 1991, Environ Corp. 1992). Both the RFCI suggested practices and OSHA regulations (engineering controls/work practices) refer to removal of floor tile in whole pieces to prevent release of airborne asbestos (Environ Corp. 1992; OSHA 3995a). This study suggests that even when the floor tile are broken fiber levels in the air are low and weil below the OSHA PEL. Breakage of floor tile commonly occurs during its abatement and removal of large quantities of this material is not economically or practicability feasible by a whole piece method. Thus, concern of floor tile removal in whole pieces to prevent `elevated' exposure does not appear to be warranted based on these study data. A similar finding, as reported for floor tile, is also suggested for mastic abatement based on exposure ievels (Environ Corp. 1992). OSHA has categorized shot blasting as an aggressive methodology and is to be employed only when other methods are not applicable (OSHA 1994). Shot blasting removes more completely the mastic as compared to solvent methods and if properly performed appears to result in low exposure levels (Lange et al. 1996a). Shot blast machines that incorporate HEPA filtration function as a mobile mini-containment and have been previously reported to exhibit low airborne concentrations of asbestos during mastic abatement (Lange et al. 1996a). A previous investigation (Lange et al. 1996a) that reported exposure data for both shot blast and solvent abatement of mastic suggested no difference in airborne concentrations of asbestos. OSHA has incorporated into its regulations (OSHA 1998a) a requirement for use of respirators when performing aggressive methods such as shot blasting, without regard to historical or objective exposure data. Based on these exposure results, little or no benefit from exposure will result and respirator use may have a detriment on the worker due to physiological stress (Raven et al. 1979). Exposure levels as reported in this study for floor tile/mastic do not fully support engineering controls and work practices suggested by OSHA (OSHA 1992, 1994). OSHA has categorized abatement of floor tile as a class II `operation' and considers shot blasting of floors to be a methodology requiring pre-described regulatory controls. Conditions and requirements described by OSHA, based on these and other exposure data (Environ Corp. 1992, Lange et al. 1996), suggest legislation of science in regard to abatement of floor tile/mastic. From a practical prospective, control methods should be based on exposure levels, including those of historical nature (objective data), and not regulatory mandates. Thus, based on exposure, floor tile and Airborne asbestos exposure 13 mastic abatement more likely fall within class III operations and re-evaluation of abatement of these materials by OSHA is warranted. Data presented in this study suggest that use of respirators during abatement of dry wall is appropriate. However, the type of respirator can be debated. Based on the summary statistics alone, a half mask respirator would be appropriate. However, evaluation of exposure concentrations above i.Ofcirf 3-TWA, without regard to summary data, a full face respirator should be employed (Lyles et al. 1997, Letters to the Editor 1998). Use of a full face respirator is supported by average exposure concentrations of worker A, which is above the maximum protection value established for a half mask. When the Cl range for both with and without the outlier are considered, use of a full face respirator is supported. The type of respirator to be selected is dependent on risk evaluation (exposure and hazard assessment) and characterization of regulatory compliance (Letters to the Editor 1998). In determining compliance with regulatory exposure standards, it has been suggested that summary exposure data, rather than individual measurements from day-to-day, be utilized (Lyles et al. 1997, Letters to the Editor 1998). Employment of summary statistics, particularly GM, for evaluation of exposure and enforcement of compliance is supported by an Appellant Court decision on this issue (United States Court of Appeals 1991, Letters to the Editor 1998). Since asbestos is a chronic disease agent, summary data will better represent the potential of disease occurrence for an occupational population and provide information for epidemiological investigations and control measures (Seixas et al. 1988, Armstrong 1992, Lange et al. 1996a). From only an epidemiological prospective, studies have suggested that the AM is the best measure for prediction of occupational disease (Sexias et al. 1988; Armstrong 1992) The primary issue related to exposure is the risk of disease (HEI-AR 1991, Lange et al. 1996a, Letters to the Editor, 1998). Certainly all activities carry risk and the question to be answered is what is acceptable risk. OSHA has defined acceptable risk to be exposure below the PEL (OSHA 1986, 1994). Establishment of the OSHA PEL for asbestos was `designed' for an exposure period of 40h week-1, 50 weeks years-1 and 45 years in a lifetime (OSHA 1986, Mlynarek et al. 1996). The time period of 45 years provides a life time risk from exposure to airborne asbestos (OSHA 1986). For determination of exposure risk, OSHA employed a linear model (OSHA 1986). If the risk scenario, as described by OSHA is employed, likelihood of elevated diseases for those abating dry wail material when using a half mask respirator and floor tile/mastic with no respirator are within the acceptable published regulatory criteria. Distribution as evaluated by the Shapiro-Wilk W test suggests that exposure concentrations for dry wall, with and without the outlier, and floor tile/mastic are non-normal and can be best represented in a logarithmic form. Sample distribution as determined by SD, difference between AM and GM mean values, similarity between GM and median values, skewness, and GSD support the finding of non-normality. Previous studies that have evaluated concentration distributions of asbestos (Paik et al. 1983, Lange et al. 1996a), dust (Niven et al. 1992, Seixas et al. 1997) and environmental contaminants (Esmen and Hammad 1977) suggest non-normality and a distribution that is best represented by a logarithmic form (Leidel et al. 1977, Lange et al. 1996a, Keyes et al. 1991, Burstyn and Teschke 1999). Within- and between-worker exposure variation for both dry wall and floor tile/mastic abatement were also suggested to be non-normal for distribution (Tables 2 and 3). Skewness for both dry wall and floor tile/mastic removal suggest non-normality and a tail to the right. These characteristics are commonly associated with occupational exposure concentrations (Leidel et al. 1977, Lange et al. 1996a). 14 Lange and Thomulka Sample variation, as represented by GSD values, is elevated for both dry wall and floor tile/ mastic abatement, but is at the lower range generally reported for occupational exposure studies (Esmen 1998, Phillips and Esmen 1999). Occupational studies have reported GSD values that range from about 1.7 to 5.2 (Buringh and Lanting 1991, Lange et al. 1998, Esmen 1998). One study (Buringh and Lanting 1991) suggested that a GSD of 2.7 is an appropriate estimator of most TWA exposure concentrations. This investigation and a previous study (Lange et al. 1996a, 1997) suggest that a GSD of around 2 is applicable for airborne exposure concentrations associated with the abatement industry. However, employment of a GSD of 2 has not been supported in other abatement studies (Lange et al. 1996a, 1998). Buringh and Lanting (1991) suggested that a GSD of 2 is too low and will likely underestimate exposure variability, with a GSD of 2.7 being more appropriate as an estimated representative of variation. Most occupational exposure studies (Kromhout et al. 1993, Scheeper et al. 1995, Hall et al. 1997) have been conducted for non-construction industries and may not have good applicability to the abatement industry. Studies involving the abatement industry that have been considered to employ good engineering controls generally reported GSD values of around 2 (Lange et al. 1996, 1997). Comparison of within and between measurements for workers suggest that the predominant `source' of exposure to airborne asbestos is from the process rather than individual practices. A previous investigation (Lange et al. 1999) supports this finding that process is the predominate factor for exposure source. These study data suggest that workers performing dry wail abatement are not a homogeneous group. This would suggest that a single exposure measurement, as commonly employed for determination of compliance with the OSHA PEL, may not be representative of actual exposure (Rappaport 1991). When a group is determined to be non-bomogeneous, a `larger' number of samples for determination of exposure are suggested due to this variability (Spear and Selven 1989, Gardiner 1985). Floor tile/mastic exposure data suggest that this activity (abatement) provides a homogeneous exposure group. This homogeneity is likely a result of the low exposure concentrations. Regardless, since floor tile/mastic abatement has been suggested to have low exposure values (Lange et al. 1996a) and appears to be a homogeneous worker group, limited measurements during this activity will likely represent exposure to this group. Historical data from this and other studies (Lange et al. 1995a, 1996a) for floor tile/mastic abatement do not support a requirements of frequent sampling during these work practices since the likelihood of exceeding either the AM or GM numerical value for the OSHA PEL is low. Based on exposure data in this study, no doubt exists as to the necessity for use of respirators during abatement of dry wall material. However, use of respirators during abatement of floor tile/mastic is suggested to be inappropriate. This is based on the low fiber concentrations reported in this study and others during abatement of floor tile and mastic (Lange et al. 1995 a, 1996a) and respirator use has been suggested to result in a physiological stress (Raven 1979, Jones 1991, Seliga et al. 1991, OSHA 1998b). In balancing exposure and physiological stress, for floor tile/mastic the risk of injury from respirator usage is likely to greatly out-weigh any potential protection from asbestos-related diseases. The suggestion of inappropriate respirator use for floor tile/mastic must be considered and that use may create a hazard is discussed in the OSHA respiratory regulations and preamble to regulations [29 CFR 1910.134 (c)(2)(i)] (OSHA, 1998b). As related to a hazard from `improper respirator use', OSHA states `An employer may provide respirators at the request of employees or permit employees to use their own respirators, if the employer determines that such respirator use will not in itself create a hazard.' Although medical surveillance examinations are required, under many circumstances, for asbestos Airborne asbestos exposure 15 abatement workers, these medical evaluations often provide little actual suggestion of physiological stress during respirator use (Lange 1993, Barbanel and McCunney 1995) and may provide little information on `fitness for duty' (Kales et al. 1998). Most medical examinations for abatement workers serve only as a `medical' work permit for those in this industry (Lange et al. 1991, Lange 1992, 1993, Barbanel and McCunney 1995). The high percentage of smokers in the asbestos industry will likely further magnify stresses from respirator use in abatement workers (Lange et al. 1987, Lange 1992, Lange et al. 1993). Conclusion These data suggest that elevated exposure concentrations of asbestos fibers can occur during abatement of asbestos-containing dry wall, while low exposure occurs during abatement of floor tile/mastic. Distribution for both types of abatement are suggested to be non-normal and best be represented by a logarithmic form. Summary data of exposure are suggested to be best represented by AM, GM, median, GSD, SD and range. Testing for outliers is suggested for description of the sample population. Based on comparison for within-and between-worker variability the process is suggested to be the source of exposure. This supports engineering controls as an important factor in maintaining a low exposure concentration in the work area. Respirator use appears to be appropriate for dry wall abatement, but not for floor tile/mastic activities. Re-evaluation of OSHA requirements for abatement of floor tile/mastic should be undertaken. Use of respirators during abatement of floor tile/mastic may result in unnecessary physiological stress and provide no protective benefit to the wearer. Data provided can be used in establishment of historical exposure data for abatement that involves dry wall and floor tile/ mastic asbestos-containing materials. Previous asbestos abatement studies support these findings. Additional investigation of exposure levels during asbestos abatement is warranted. Armstrong, B.G. (1992) Confidence intervals for arithmetic means of lognormaily distributed exposures. Am. hid. Hyg. Assoc. J. 53, 481-5. ASTM (1990a) Standard practice for rotameter calibration. Standard D 3195-90, Philadelphia, PA. ASTM (1990b) Standard guide for air sampling strategies for worker and workplace protection. E 1370-90, Philadelphia, PA. ASTM (1993) Standard practice for collection of airborne paniculate lead during abatement and construction activities. Standard E 1553-93, Philadelphia, PA. Barbanel, C.S. and McCunney, R.J. (1995) Environmental surveillance of respiratory disorders: the hazardous waste site as an example. In Environmental, respiratory diseases (E.M. Cordasco, S.L. Demeter and C Zenz, eds) pp. 479-503. New York: Van Nostrand Reinhold. Bozzelli, J.W. and Russell, J.F. (1982) Airborne asbestos levels in several schools before and after bulk asbestos removal. Int. J. Environ. Stud. 20, 27-30. Bresnitz, E.A., Gilman, M.J., Gracely, E.J., Airoldi, J., Vogel, E. and Gefter, W. (1993) Asbestos related radiographic abnormalities in elevator construction workers. Am. Rev. Respir. Dis. 147, 1341-4. Buringh, E. and Canting, R. (1991) Exposure variability in the workplace: its implications for the assessment of compliance. Am. Ind. Hyg. Assoc. J. 52, 6-13. Burstyn, I and Teschke, K. (1999) Studying the determinates of exposure: a review of methods. Am. Ind. Hyg. Assoc. J. 60, 57-72. Chessen, J,, Hatfield, J., Schultz, B., Dutrow, E. and Blake, J. (1990) Airborne asbestos in public buildings. Environ. Res. 45, 179-203. 16 Lange and Thomulka Com, M. (1983) Assessment and control of environmental exposure. J. Allergy Clin. Immun. 72, 231-41. Com, M. (1985) Strategies of air sampling. Scan. J. Work Environ. Health 11, 173-80. Corn, M., Crump, K,, Farrar, D.B. Lee, R.J. and McFee, D.R. (1991) Airborne concentrations of asbestos in 71 buildings. Reg. Tox. Pharm. 13, 99-114. Daniel, W.W. (1991) Biostalistics: a foundation for Analysis in the Health Sciences. New York: Wiley and Sons. Dement, J.M., Harris, R.L., Symans, M.J. and Shy, C. (1982) Estimates of dose-response for respiratory cancer among chrysotile asbestos workers. Ann. Occup. Hygiene 28, 869-87. Environ Corporation (1992) Evaluation of worker exposure to airborne fibers during the removal of resilient floor coverings and asphaltic cutback adhesives using recommended work practices. (1 May 1992) Prepared for Resilient Floor Covering Institute and Armstrong World Industries, Inc., Arlington, VA. Environmental Protection Agency. (1987) Asbestos-containing materials in schools: final rule and notice. 40 CFR 763 appendix A to subpart E. Federal Register, 30 October 1987, volume 52: pp. 41825-905. Washington, DC. Environmental Protection Agency. (1992) Statistical analysis of ground-water monitoring data at RCRA facilities (draft). Office of Solid Waste, Permits and State Programs Division, Washington, DC. Environmental Protection Agency. (1999) National Emissions Standards for Hazardous Air Pollutants. Title 40 Code of Federal Regulations 61, subparlM, Washington, DC. Esmen, N.A. and Hammad, Y. (1977) Log-normality of environmental sampling data. J. Environ. Sci. Health. A12, 29-41. Esmen, N.A. (1998) Exposure estimation in four major epidemiological studies in the acrylonitrile industry. Scand. J. Work. Environ. Health 24, 63-70. Ganor, E., Fischbein, A., Brenner, S. and Froom, P. (1992) Extreme airborne asbestos concentrations in a public building. Br. J. Indust. Med 49. 486-8. Garcia-Closa.s, M. and Christiani, D.C. (1995) Asbestos-related diseases in construction carpenters. Am. J. Indus!. Med. 27, 115-25. Gardiner, K. (1995) Sampling strategies, In Occupational Hygiene (J.M. Harrington and K. Gardiner, eds). Cambridge, MA: Blackwell Science. Grubbs, F.E. (1950) Sample criteria for testing outlying observations. Ann. Math. Stat. 21, 27-58. Grubbs, F.E. and Beck, G. (1972) Extension of sample sizes and percentage points for significance tests of outlying observations. Technometrics 14, 847-54. Hall, G.S., Rice, C.H., Lockey, J.E., Lemasters, G.K. and Gartside, P.S. (1997) A comparison of exposures to refractory ceramic fibers over multiple work shifts. Ann. Occup. Hyg. 41, 555-60. Health Effects Institute - Asbestos Research. (1991) Asbestos in public and commercial buildings: a literature review and synthesis of current knowledge. Cambridge, MA. Jaffery, S.A.M.T., Burden, G.J. and Rood, A.P. (1988) An investigation of airborne asbestos concentrations in two U.K. buildings: before, during and after removal of asbestos. Int. J. Elnviron. Studies 32 169-80. Jones, J.G. (1991) The physiological cost of wearing a disposable respirator. Am. Ind. Hyg. Assoc. J. 52, 219-25. ' Kales, S.N., Aldrich, J.M., Poiyhronopoulos, G.N., Gassert, T., Hu, H., Kelsey, K., Sweet, C. and Christiani, D.C. (1998). Fitness lor duty evaluations in hazardous material firefighters. J. Occup. Elnviron. Med. 40, 925-31. Keyes, D.L., Chesson, J., Ewing, W.M., Faac, J.C., Hatfield, R.L., Hays, S.M., Longo, W.E. and Millette, j.R. (1991) Exposure to airborne asbestos associated with simulated insulation above a suspended ceiling. Am. Ind. Hyg. Assoc. J. 52, 479-84. Koskinen, K., Zitting, A., Tossavainen, A., Rinne, J., Roto, P., Kivekas, J., Reijula, K. and Huuskonen, MS. (1998) Radiographic abnormalities among Finnish construction, shipyard and asbestos industry workers. Scand. J. Work Environ. Health 24, 109-17. Airborne asbestos exposure 17 Kromhout, H., Symanski, E. and Rappaport, S.M. (1993) A comparison evaluation of within- and betweenworker components of occupational exposure to chemical agents. Ann. Occup. Hyg. 17,253-70. Kumagai, S., Kusaka, Y. and Goto, S. (1997) Log-normality of distribution of occupational exposure concentration to cobalt. Ann. Occup. Hyg. 41, 283-6. Lange, J.H. (1992) A survey of smoking patterns during an initial asbestos training program. Jnt. J. Environ. Studies B42, 73-9. Lange, J.H. (1993) A study of asbestos abatement workers for respiratory usage, respirator fit-testing and occupational medical surveillance: evaluation of safety practice relationships and occupational health prevention measures. Int. J. Occup. Med. Toxicol. 2 65-74. Lange, J.H. (1999) A statistical evaluation of asbestos air concentrations. Indoor Built Environ, in press. Lange, J.H., Weyel, D.A., Rosato, L.M., Tucker, D., Malek, D.E., Mayernick, J.A. and Ryan, L. (1987) Preliminary results of smoking patterns for workers attending an asbestos abatement course. Scand. ./. Work Environ. Health 62, 495. Lange, J.H., Rosato,`P.A., and Spence, P.L. (1991) A medical surveillance program for hazardous waste activities and asbestos abatement operations for a consulting engineering firm. J. Environ. Sci. Health A26, 953-70. Lange, J.H., Kaiser, G. and Tbomulka, K.W. (1994) Environmental site assessments and audits: building inspection requirements. Environ. Management 18, 151-60. Lange, J.H., Thomulka, K.W., Lee, R.J., Dunmyre, G.R. and Schwerer, F.C. (1995a) Surface and deposition sampling in a mechanical room that contains pipe and boiler asbestos insulation. Toxicol. Environ. Chem.. SO, 51-6. Lange, J.H., Thomulka, K.W., Lee, R.J. and Dunmyre, G.R. (1995b) Evaluation of lift and passive sampling methods during asbestos abatement activities. Bull Environ Contain. Toxicol 55, 325-31. Lange, J.H., Lange, P.R., Reinhard, T.K. and Thomulka, K.W. (1996a) A study of personal and area airborne asbestos concentrations during asbestos abatement: a statistical evaluation of fibre concentration data. Ann. Occup. Hyg. 40, 449-66. Lange, J.H., Kelly, D.F., Roller, K.S., Komorek, K.M., Mamros, C.T., Mosbacher, E.M., Polon, T.M. and Thomulka, K.W. (1996b) Characteristics of asbestos abatement projects in public buildings in Erie County, Pennsylvania, USA: 1992-3994. Indoor. Built. Environ. 5, 76-81. Lange, J.H., Sites, S.L.M. and Thomulka, K.W. (1997) Personal sample measurements of airborne lead during abatement procedures. Bull. Environ. Contain. Toxicol. 58, 659-66. Lange, J.H., Sites, S.L.M. and Thomulka, K.W. (1998) Airborne lead exposure to workers during activity as part of a building demolition project. Indoor Built Environ. 7, 210-16. Leidel, N.A., Busch, K.A. and Lynch, J.R. (1977) Occupational Exposure Sampling Strategy Manual. DEHW (NIOSH) Publication Number 77-173, National Technical Information Service Number PB-274-792. Cincinnati, Ohio: National Institute for Occupational Safety and Health. Letters to the Editor (1998) A log-normal distribution-based exposure assessment method for unbalanced bias. Ann. Occup. Hyg. 42, 413-22. Lumley, K.P.S., Harries, P.G. and O'Kelly, F.J. (1971) Building insulation with sprayed asbestos: a potential hazard. Ann. Occup. Hyg. 14, 255-7. Lyles, R.H. Kupper, L.L. and Rappaport, S.M. (1997) A lognormal distribution-based exposure assessment methods for unbalanced data. Ann. Occup. Hyg. 41, 63-76. Massey, D.G. and Fournier-Massey, G. (1987) Asbestos removal from buildings: a review. Hawaii Med. J. 46, 153-7. Mlynarek, S,, Corn, M. and Blake, C. (1996) Asbestos exposure of building maintenance personnel. Regul. Toxicol. Pharmacol. 23, 213-24. Mossman, G.T. and Gee, J.B.L. (1989) Asbestos-related diseases. New Engl. J. Med. 320 1721-30. Mossman, G.T., Bignon, J. Corn, M,, Seaton, A. and Gee, J.B.L. (1990) Asbestos: scientific developments and implications for public policy. Sci. 247, 294-301. 18 Lange and Thomulka Nicholson, W.J., Perkel, G. and Selikoff, I.J. (1982) Occupational exposure to asbestos: population at risk and project mortality - 1980-2030. Am. J. Ind. Med. 3, 259-311. Niven, R.McL., Fishwick, D., Pickering, C.A.C., Fletcher, A.M., Warburton, C.J. and Crank, P. (1992) A study of the performance and comparability of the sampling response to cotton dust of work area and personal sampling techniques. Ann. Occup. Hyg. 36, 349-62. Occupational Safety and Health Administration (1986) Occupational safety and Health Standards, Title 29 Code of Federal Regulations 1910 and 1926. Occupational Exposure to Asbestos, Tremolite, Anthophyllite and Actinolite; Federal Register 51: 22611-790. Washington, DC: US Department of Labor. Occupational Safety and Health Administration (1992) Administration. Standards Interpretation and Compliance Letters, Sampling of resilient floor covering. Memorandum from Ellen I. Roznowski, Acting Director Office of Compliance Assistance, dated 1/6/92, Salt Lake Technical Center, UT. Internet http://www.OSHA.gov. Occupational Safety and Health Administration (1994) Occupational safety and Health Standards, 29 CFR 1910, e: al. Occupational exposure to asbestos; Final Rule, Federal Register 59: 40963-1162, Washington, DC: US Department of Labor. Occupational Safety and Health Administration (1995a) Directives, CPL-2-2.63 - Inspection procedures for occupational exposure to asbestos final rule 29 CFR 1910.1001, 1926.1101, and 1915.1001, information date 11/03/95, includes settlement agreement for abatement of floor tile. Washington, DC: Department of Labor. Internet http://www.OSHA.gov. Occupational Safety and Health Administration. (1995b) Standards Interpretation and Compliance Letters, Clarification of mechanical chipping device and floor tile removal, Information date 11/01/95, Salt Lake Technical Center, UT. internet http://www.OSHA.gov. Occupational Safety and Health Administration. (1998a) Occupational Safety and Health Standards, Title 29 Code of Federal Regulations 1910 and 1926. Occupational Exposure to Asbestos, Tremolite, Anthophyllite, and Actinolite. Washington, DC: US Department of Labor. Occupational Safely and Health Administration. (1998b) Respiratory Standard. 29 CFR 1910.134. Washington, DC: Department of Labor. Internet http://www.OSHA.com. Oliver, L.C., Sprice, N.L. and Greene, R. (1991) Asbestos related disease in public school custodians. Am. J. Indust. Med. 19, 303-16. Paik, N.W., Walcott, R.J. and Brogan, P.A. (1983) Worker exposure to asbestos during removal of sprayed materials and renovation activity in buildings containing sprayed material. Am. hid. Hyg. Assoc. J. 44, 428-32. Pennsylvania Department of Labor and Industry. (1991) Occupations, Accreditation and Certifications Act number 1990-194, Harrisburg, PA. Peretz, C., Goldberg, P., Kahan, E. and Goren, A. (1997) The variability of exposure over time: a prospective longitudinal study. Ann. Occup. Hyg. 41, 485-500. Phillips, M.L. and N.A. Esmen (1999) Computational methods for ranking task-specific exposures using multi-task time-weighted average samples. Ann. Occup. Hyg. 43 201-13. Rappaport, S.M. (1991) Assessment of long-term exposures to toxic substances in air. Ann. Occup. Hyg. 35, 61-121. Raven, P.B., Dodson, A.T. and Davis, T.O. (1979) The physiological consequences of wearing industrial respirators: a review. Am. Ind. Hyg. Assoc. J. 40, 517-34. Sawyer, R.N. (1977) Asbestos exposure in a Yale building, analysis and resolution. Environ. Res. 13, 146-69. Sawyer, R.N., Rohl, A.N. and Langer, A.M. (1985) Airborne fiber control in buildings during asbestos material removal by amended water methodology. Environ. Res. 36, 46-55. Scheeper, B., Kromhout, H. and Boleij, S.M. (1995) Wood-dust exposure during wood-working processes. Ann. Occup. Hyg. 39, 141-54. Seixas, N.S., Robin, T.G. and Moulton, L.M. (1988) The use of geometric and arithmetic mean exposures in occupational epidemiology. Am. J. Indust. Med. 14, 465-77. Airborne asbestos exposure 19 Scixas, N.S., Heyer, N.J., Welp, E.A.E. and Checkoway, H. (1997) Quantification of historical dust exposures in the diatomaceous earth industry. Ann. Occup. Hyg. 41, 591-604. Seliga, R., Bhauacharya, A., Succop, P,, Wickstrom, R., Smith, D. and Willeke, K. (1991) Effect of work load and respirator wear on postural stability, heart rate, and perceived exertion. Am. Ind. Hyg. Assoc. J. 52, 417-22. Seiikoff, I.J., Hammond, E.C. and Seidman, H. (1979) Mortality experience of insulation workers in the United States and Canada: 1943-1976. Ann. New York Acad. Sci. 330, 91-116. Shapiro, S.S. and Wilk, M.B. (1965) An analysis of variance test for normality. Biometrika 52, 591-611. Spear, R.C. and Selvin, S. (1989) OSHA's permissible exposure limits: regulatory compliance verse health risks. Risk Anal. 9, 579-86. Student Manual (1991) Removal of resilient floor coverings: a training course developed by the resilient floor covering industry on the removal of resilient Boor covering which may contain asbestos. Rockville, MD: Resilient Floor Covering Institute. Taylor, J.K. (1990) Statistical Techniques for Data Analysis. Boca Raton, FL: Lewis Publishers, Timko, J. and Downie, J. (1992) Statistics on Software. Culver City, CA: Software Labs. Waage, H.R, Johnson, E.S., Hilt, B. and Langard, S. (1994) Asbestosis and plural changes as risk factors for asbestos-induced lung cancer. Int. J. Occup. Med. Toxicol. 3, 319-27. Weill, H. and Hughes, J.M. (1986) Asbestos as a public health risk: disease and policy. A. Rev. Public Health 7, 171-92. US Filter Blaslrac (1998) Model 1-10D & 654 Dust Collector, portable blast cleaning system. Oklahoma City, OK. United States Court of Appeals (1991) American iron and steel Institute and Bethlehem Steel Corporation vs. Occupational Safety and Health Administration. Federal Reporter Second Series (DC Cir.), 939, 975-1010. Appendix: Acronyms ACM AM Cl DLI EPA fern-3 FT GM GSD HEI-AR N NESHAP NIOSH OSHA PCM PEL PLM PPE RFCI SD TWA Asbestos-containing materials Arithmetic mean Confidence interval Pennsylvania Department of Labor and Industry US Environmental Protection Agency Fibers per cubic centimeter Floor tile Geometric mean Geometric standard deviation Health Effects Institute - Asbestos Research Number National Emissions Standard for Hazardous Air Pollutants National Institute of Occupational Safety and Health Occupational Safety and Health Administration Phase contrast microscopy Permissible exposure limit Polarized light microscopy Personal protective equipment Resilient Floor Covering Institute Standard deviation Time-weighted average