Document aBVx3GO55oYYv3e1QZLK5JkkX

PLAINTIFF'S EXHIBIT Background Documentation on Evaluation of Occupational Exposure to Airborne Asbestos Joint ACGIH-AIHA Aerosol Hazards Evaluation Committee This review presents background information and literature documentation to supplement the "Recommended Procedures for Sampling and Counting Asbestos Fi bers: Procedures for the Evaluation of Occupational Exposure to Airborne Asbestos'* prepared by the joint ACGIH-AIHA Aerosol Hazards Evaluation Committee. It reviews the nature of the inhalation hazard associated with asbestos fibers, the sampling and analytic methods which have been used, and a rationale for the selection of the membrane filter sampling-optical phase microscope identification and assay methodology which is recommended. Introduction The inhalation of asbestos fi bers is known to cause asbestosis and mesothelioma in man, and to significantly increase the incidence of lung cancer. Most, if not all of the increased mortality has been associated with occupational exposures. As bestos is a nearly ubiquitous material in in dustrialized societies, and has also been found in community air samples and in hu man lungs in most cases where it has been diligently sought. However, the association between community air pollution exposures and increased lung cancer mortality has not been established. One reason is the general difficulty of establishing a clear causal rela tionship between low-level exposures and a possible increase in disease incidence in a large population, especially when other known and suspected carcinogens are also present in the atmosphere. This problem is compounded by the lack of a method for the quantitative assay of airborne asbestos fibers when they are present in much lower concentrations than are other airborne par ticles. The method recommended herein for occupational exposure evaluations depends Reprints of this article are available for purchase from either the American Industrial Hygiene Association. 66 South Miller toad. Akron. Ohio 44313 -or the American Conference of Governmental Industrial Hygienist. P.O. Bo* 1937, Cincinnati. Ohio 45201. The cost is $1.00 per copy. on visual discrimination of fibers in the presence of a background of non-fibrous particles. It cannot be used for community air evaluations, where background particles would usually obscure too large a fraction of each viewing field. Diseases Associated with Asbestos Asbestosis The inhalation of excessive amounts of airborne asbestos fibers over an extended period of years will cause in some individuals a pneumoconiosis, termed asbestosis. The onset of asbestosis probably depends upon the asbestos dust concentration, fiber mor phology, and the length of exposure. Merewether and Price* demonstrated in 1930 that asbestosis posed a threat to workers in the asbestos textile trade. It is now known that in any of the trades in which excessive airborne exposures to asbestos fibers occur, it is likely that some employees will develop asbestosis. Major sources of exposure occur during asbestos mining, milling, textile weav ing, insulation installation and stripping, and during the manufacture and application of friction products and cement products. Har ries2 in 1968 suggested that a worker en gaged in trades in which intense intermittent exposures to asbestos occur, may also devel op asbestosis.. Asbestosis is a progressive disease which 91 \ ASARCO ALV 0002041 92 may develop fully in seven to nine years and may cause death as early as thirteen years after the first exposure.3 Usually, the pneu moconiosis becomes evident 20-40 years after the first exposure to asbestos. Once established, asbestosis progresses even af ter the exposures have ceased, as shown by Lanza3 in 1938. The consistent pathologic change in the lung is interstitial, septal, peribronchiolar or perivascular fibrosis which is eventually vis ualized hy x-ray. It is accompanied by the formation of intra-alveolar ferruginous bod ies. The principal findings in asbestosis in clude reduction in forced vital capacity and forced expiratory volume in one second, as well as reductions in other measures of pul monary function. Selikoff et al.* referred to asbestosis as a monosymptomatic disease with dyspnea being the main complaint. The British Occupational Hygiene Society3 sug gests that basal rales are the first sign of asbestosis. Becklake et al.6 administered a symptom questionnaire to approximately 1,000 employees undergoing pulmonary function tests and found that breathlessness on exertion was probably the most sensitive symptom index related to dust exposure. Neoplasia Bronchogenic carcinoma and mesothe lioma of the pleura and peritoneum are causally associated with asbestos exposures. Selikoff et alP have shown that asbestos in sulation workers who smoke had a higher incidence of lung cancer than those who did not. Excesses of cancer of the stomach, colon, and rectum have also been observed.8 There is a little likelihood of asbestosis or lung cancer arising from non-occupational exposures but mesotheliomas have been found among families of asbestos workers, and in the neighborhoods of asbestos plants not having environmental controls. Accord ing to Selikoff9 it may be unjustified to give any estimate of the magnitude of the risk associated with indirect occupational or fam- *-. i.A.vnr . . . ' - . ; ily exposure. According to Cooper10, the major potential for risk appears to be related to "indirect occupational contact, household contact, or residence in the immediate neigh- borhood of an asbestos source; and even there, the actual risk is poorly defined. It' V: v?,' is not known what range of respirable air-l^* bome asbestos fibers will ultimately be found to have no measurable effects on health." ' All studies which indicate differences'in^-;.? . pathogenicity among various types of as- '. bestos suffer from lack of information about fiber characteristics, lack of quantitative V. data on cumulative exposure and the in- . ' . fluence of trace materials such as nickel, ' ;- r;; chromium or manganese, and various hydro- carbons.1* There is evidence that the devel- , 1 opment of mesothelioma varies between : ;" groups exposed to different types of fi bers.12-*7 The association of exposure to asbestos with bronchogenic carcinoma was first re- ported in 1935 by Lynch and Smith18 but it was not until 1947 that this Was fully rec ognized. Merewether19 demonstrated bron chogenic'carcinoma in 13% of autopsies of persons who had asbestosis arising from em- ' ployment in the textile trade. In 1955, Doll20 found that textile workers exposed for at least 20 years during the period between 1922 and 1953 had a frequency of ten times the expected rate. More recently, Williams2* in 1965, and Selikoff et al.n in 1969, re ported a frequency of bronchogenic carci noma among insulation workers much great er than that anticipated. The possibility that asbestos is a co-carcinogen, also, was sug gested by Wright23 in 1969. Cigarette smoking is strongly implicated as a co-carcinogen among asbestos workers. Selikoff et al.1 reported that the incidence of bronchogenic carcinoma among nonsmok ing asbestos workers is not significantly greater than that of non-asbestos workers, while asbestos workers who smoke have a much higher incidence. ... In 1959, Wagner24 reported, 3ptcases of pleural mesothelioma occurring `after ex- ASARCO ALV 0002042 American Industrial Hygiene Association Journal posurc to crocidolite in the Northwest Cape region of South Africa. Additional informa tion supporting association between mesothe lioma and asbestos exposure has been de veloped. Selikoff et al.25 reported 22 deaths from mesothelioma among 532 asbestos workers who were exposed to mixed asbestos insulation materials, compared to the ex pected very low incidence of mesothelioma in the general population. Stumphius and Meyer26 reported 17 cases of mesothelioma in 1968 among shipyard workers, and con cluded that amphibole asbestos minerals, especially crocidolite, may have caused the turnon. McDonald et al.27 reported several cases of mesothelioma among current and former Quebec chrysotile miners and mill operators. They concluded that primary malignant mesothelioma, although rare in Canada, occurred more frequently than ex pected among workers in manufacturing and industrial applications, but not among work ers in mining and milling operations. Wright23 has noted that employees engaged in dusty chrysotile mining and milling op erations have lower incidence of asbestosis, bronchogenic carcinoma and mesothelioma than those working in insulation operations.. In 1971, Stanton and Wrench2* and Wag ner et al.29 postulated that the development of mesothelioma associated with asbestos ex posure may not be due to its chemical com position but to its morphology. Mesothelial responses were initiated in animals by in trapleural implants of durable fibers other than asbestos, such as fibrous glass. The car cinogenicity of asbestos and fibrous glass un der these experimental conditions appeared to be primarily related to the structural shapes of these materials rather than to physiochemical properties. Signs and Symptoms of Asbestos Exposure Pleural Plaques and Pleural Calcifications Localized pleural thickening, which may become radio-opaque through calcification, has often been observed in persons having occupational exposures to asbestos. Calcified 93 plaques also occur in certain geographic regions such as Northern Finland,30 Czech oslovakia31 and Bulgaria.32 It is interesting to note that even large calcified plaques ap pear to have little effect on respiratory func tion. There is no evidence that they result in malignant disease.33 Asbestos Bodies Murray* reported the first recorded death resulting from asbestos exposure and made, perhaps, the first reference to asbestos bod ies. Similar observations were made by Marchand and Riesel as cited by Gloyne34 and by Fahr.35 Cooke36 in 1924 described the first con clusive case of asbestosis in which Cooke and Hill37 demonstrated the presence of "curious bodies". Gloyne38'39 described the asbestos body, stating that its formation was the result of a tissue reaction to a foreign body acting as a benign irritant. Merewether,40 Blount,41 and Suzuki and Churg42 have described its biochemical composition. Pseudoasbestos Bodies--Ferruginous Bodies Prior to the adoption of the more general term "ferruginous body", there was consid erable debate about the origin of atypical asbestos bodies. Thus, they were referred to as "pseudoasbestos bodies" and have been reported forming around rutile needles by Sundius and Bygden,43 brucite and MgO.HaO by Vorwald,44 and ceramic aluminum fibers, cosmetic talc and attapugite by Gross et al.*9 Gross et al.*9 proposed that the contro versy between the term "asbestos" body and "pseudoasbestos" body be ended by adopt ing the more general term "ferruginous" body. Thus, the asbestos body is one kind of ferruginous body. Properties of Asbestos Affecting Lung Retention Timbrell46-49 has discussed the properties of fibers in connection with their deposition in the lungs and bronchial passages. He I ASARCO ALV 0002043 94 showed that deposition depends on fiber diameter, fiber length and fiber shape. He also stated that because of aggregation, par ticularly with chrysotile, it is very difficult to define diameter and length of many fibers. The extreme length of the fibers results in increased bronchial deposition by intercep tion, a mechanism whose effect can usually be neglected for particles of more compact shapes. The fiber shape is important because it affects the orientation which the fibers as sume within the airways; the straighter, more needle-like amphiboles tend to align them selves with the axis of the airway, and pene trate deeper into the lung, while the "curly" chrysotile fibers do not. Characterization of Airborne Asbestos On the basis of the preceding discussion, it appears that the hazards from inhaled as bestos are more closely related to the fibrous nature of the particles than to their chemical composition or surface properties. Thus, the most appropriate index of hazard is fiber concentration, but whether this should be ex pressed by number, surface area, or mass is not yet clear. At present, the only practical way to determine fiber concentration is by counting fibers which have been mounted on a suitable substrate so that they can be identified as individual fibers. Such identifi cations can be made with optical or electron microscopes. Electron microscopy reveals fibers which are too small to be resolved with light microscopy, but is too expensive for most routine evaluations. Furthermore, it is believed that occupationally related as bestos diseases are most closely associated with fibers which are large enough to be resolved optically.50'51 For air pollution ex posures, it is possible that the smaller fibers may also be of concern. Since the inhalation hazard from airborne asbestos is initimately related to fiber size and shape, it is necessary to consider the crystalline characteristics of the family of minerals known as asbestos. Asbestos is a generic name for naturally February, 1975 occurring mineral silicates whose crystals are in the form of filaments. Thus, when the bulk material is subdivided, it separates into filaments, some of which are small enough to become airborne. The most widely used asbestos mineral in the U.S. is chrysotile (3MgO.2Si02.2HjO) a fibrous form of ser pentine. Other types of asbestos are known as amphilboles and include amosite (5.5FeO. 1.5Mg0.8Si02.Hj0); crocidolite (Na20. Fe2Os.3FeO.8SiO2.H2O); tremolite (2CaO. 5MgO.8SiO2.H2O); anthophyllite (7MgO. 8Si02.H20); and actinolite (2CaO,4MgO. FeO.8SiO2.H2O). Asbestos Dust in Industry A number of studies have been carried out on composition and fiber size distribution in dust clouds containing asbestos. The com position of the dust depends on the propor tion of the asbestos in the material worked and on other sources of dust. The size distribution measurements49'52-55 that have been carried out on asbestos in dust clouds are not strictly comparable be cause of variations in optical and electron microscope techniques; however, clear in dications of large variations in size distribu tion have been given for different types of asbestos and different manufacturing proc esses and application methods. These studies have shown that: (a) many fibers are smaller than the res olution limit of optical microscopy; (b) the diameter of chrysotile fibers, which occurs in bundles of fibrils, cannot be dearly defined; (c) long chrysotile fibers are frequently curly, the others are mainly straight; (d) other abestoses (crocidolite, amosite, and anthophyllite) frequently occur as single fibers; (e) the minimum diameters for fibers are: chrysotile 0.25 pm, crocidolite 0.06 pm, amosite 0.15 pm, anthrophyllite 0.25 pm; (f) the diameters of individual fibers in the air seldom exceed 5 pm; I ASARCO ALV 0002044 American Industrial Hygiene Association Journal (g) the fiber lengths range from 0.2 /xm to 1000 /im; and (h) large clumps of fibers have been ob served frequently. Threshold Limit Values (TLV's) and Alternative Hazard Indices The asbestos TLV of ACGIH36 for 1974 is 5 fibers longer than 5 jun per cm3 of air for all types of asbestos, using a 400-450 X phase contrast objective lens. As published in- the Federal Register,57 the interim Occu pational Safety and Health Administration (OSHA) standard of 5 fibers/cm3 will be reduced to 2 fibers per cm3 on July 1st, 1976. The British standards for chrysotile5 (1968) and amosite58 (1973) asbestos rec ommend 2 fibers/cm3. The chrysotile stand ard5 gives equivalent threshold limit values using a range of other instruments as shown in Table I. It should be noted that this table refers to conditions in textile factories using chrysotile asbestos and the equivalents may not be aplicable in other places. TABLE I Table of Alternative Indices Used in Great Britain Method Concentration Membrane filter 2 fibers/cm3 (current Brit "Respirable" mass ish Standard) 0.03 mg magnesium/m3 0.04 mg Si02/m3 0.10 mg ash/m3 0.12 mg asbestos/m3 Thermal precipitator 25 particles/cm3 greater than 1 pm after inciner Impinger ation 10 particles/cm3 Royco particle counter 2 particles/m3 greater than 4 jim The fact that many airborne fibers may be less than 5 pm long59 suggests that addi tional data be obtained to quantitative the relative concentration of less than 5 pm fibers. This might provide a basis for esti mating the health significance of fibers which will not be counted using currently recom mended standard procedures.5-56-57-60 It is not practical to extend "high-dry" phase contrast 95 microscopy to counting fibers much less than 5 pm. In all probability this technique would provide unreliable data for fibers much less than 2 to 3 pm in length. The use of electron miscroscopy to provide additional data of this type seems desirable until alternate tech niques for sampling small fibers are devel oped. Various methods of transferring sam ples from membrane filters to electron microscope grids, have been proposed,61-65 as have techniques for direct collection on electron microscope grids.66-67 The scanning electron microscope can, however, be used directly on samples collected on a mem brane with a minimum of preparation.68 It would seem that the main drawbacks to elec tron microscopy are the costs and the need to examine more fields to determine the con centration of long fibers (greater than 5 /am) sufficiently accurately; however, there are possibilities in instrumental identification and assessment of fibers in the presence of other particles.69-70 Although electron microscopy samples may not be practical for routine air monitoring, and cannot be directly compared to the current TLV, they may provide an essential additional basis for a retrospective reevaluation of the inhalation hazard. Review of Sampling and Analysis Sampling Techniques Almost every type of dust sampling in strument has been used for sampling asbestos dust clouds: (a) konimeter,31-70-73 (b) impinger,5-36-71-74-73 (c) thermal precipita tor,5-71-76 (d) membrane filters,5-56-71-76-77 (e) open filters,3-36-71-27-78 (f) electrostatic pre cipitators,78 (g) respirable dust samplers,5 and (h) light scater instruments.5-71 The first four types are assessed by optical micros copy, and (e) to (g) by weighing or analyt ical methods (ash or magnesium determina tion). Of the techniques which have been used for sampling asbestos to estimate potential health hazards,5-56-57-72-77 principle attention in the United States has been directed to wards the use of impingers or membrane fil- ASARCO ALV 0002045 96 February, 1975 ters. Air concentrations of asbestos based on impinger sampling are primarily determined by the concentration of nonfibrous particles present in the air. While a reduction in total particle concentration is probably accom hazard. (3) The lack of knowledge of transloca tion within the body. (4) The size range which ts important.60 panied by a reduction in asbestos fibers, a quantitative relationship between these two components (fibers and other particles) for all processes is not possible.79'80 Since fibers are the etiological contaminant of concern, the currently recommended safe air concen trations and sampling procedures are defined in terms of fiber concentration.5'36'57-60 The determination of fiber concentration involves the collection of essentially all air borne particles on a submicron pore size membrane filter. The filter is rendered trans parent by the application of immersion oil, and the fibers with lengths greater than 5 pm are subsequently counted using phase con trast microscopy procedures. While it may not be completely appropriate to exclude fi bers shorter than 5 pm in length when esti mating potential risk to the worker, this restriction has been adopted to minimize variations due to microscopy technique.5'60 Fibers are defined as particles having a length to diameter (aspect) ratio of at least 3 to 1. Analytical Techniques In many situations, separate determination of asbestos in the presence of other dusts is essential to determine the health hazard and has led to studies into analytical procedures. The techniques reported are: (a) optical mi croscopy (fiber counting),(b) optical microscopy (fiber counting) using phase con trast illumination,5'71'61- (c) electron mieroscopy,68'81 (d) optical sizing of airborne par ticles,72 (e) ash determination,27'7s>6t (f) magnesium determination,7J-7S"M (g) infrared spectroscopy,61'82 (h) x-ray diffraction,61 (i) trace constituents,78 (j) orientation in a mag netic field.81 Microscopy offers ready discrimination between fibrous and other particles. How ever, it has been found that more and more fibers can be seen as the resolution is in creased through the optical mkroscope range down into the electron microscope range, and a consistent reasonable definition of smallest size assessed is essential. Phase con While the membrane filter method is gen erally considered the best available at this time, it should be remembered that it is far from ideal. One problem with the method is the inherent difficulty in distinguishing as bestos fibers from a much greater back ground of other types of dust. The other limitations derive from our conitnued ignorance of critical factors affect ing the hazard: (1) The lack of knowledge of the deposi tion site and criteria (i>., size ag gregation and curliness) for deposi tion of fibrous particles in the respi ratory tract system.46-4* (2) The lack of knowledge as to which physical property of asbestos (/.., number, surface area or mass) is most closely related to the health trast illumination improves the visibility of fine fibers in optical microscopy. Optical sizing of single particles by light scatter has been suggested as a method of assessing airborne fibrous dust on the basis . that the majority of particles which scatter more light than a 4 ^m-diamettr sphere are fibers. This technique has been used routine ly in textile factories,72 but requires calibra tion for each particular situation. Ash determination on gravimetric samples have been used78 as a means at assessing as bestos when it is used mixed with organic fibers in some textile manufacture.^ The determination of magnesium,61 by atomic absorption spectrophotometry, can be used for chrysotile asbestos fibers if there are no other components containing appreci able amounts of magnesium. The technique is very sensitive. lS ASARCO ALV 0002046 f American Industrial Hygiene Association Journal Infrared spectroscopy and x-ray diffrac tion analysis of asbestos have been used suc cessfully in some instances but interferences can readily occur and more experience is needed. The orientation in a magnetic field and as sessment by optical methods using a laser beam shows promise. Its sensitivity is al ready in the microgram range. Keenan and Lynch61 have produced a useful review of fibers identification tech niques. All analytical methods have their limitations, and these must be borne in mind. It is apparent that, at present, microscopy is the only assessment that can be used to distinguish asbestos fibers from other dust and fibers in all types of industry and it is therefore desirable that microscopic assess ment continue to be the basis for standards. It is clear that as the resolution of a micros cope system is improved, more fibrous par ticles can be seen. Without pertinent data on the relationship between fiber size and the health hazard, it is necessary to set some lower size limit on the basis of instrumental capability and repeatability. In the absence of health data it is difficult to justify the ex tra cost of electron as opposed to otical mi croscopy. A lower fiber length limit of 5 lm has been suggested as the minimum size that can be resolved by standard microscope optics with reasonable reproducibility as is discussed elsewhere in the report Phase contrast illumination is used to en hance the visibility of the fibers against the background of the translucent filter material and is also useful in identifying various types of fiber by their differences in refractive in dex. Membrane Sampling with Phase-Contrast Assessment Sample Size Visual microscope assessment systems are prone to large subjective differences. To minimize these it is necessary to optimize all the variables fh the system such, as sample density and particle visibility. The variables 97 should be set to give the most accurate and convenient assessment when sampling a dust cloud containing asbestos fibers at the threshold limit value. While accuracy will diminish for dust clouds of lower and higher concentrations, this is not usually of much importance in assessing the health hazard. Selection of Counting Fields The Guide recommends that counting fields within the wedge shaped section of the membrane filter prepared for microscopy should be selected at random along a radial line. This assumes that fibers are randomly collected at any given angular orientation from the center of the sampling filter, but may vary as a function of distance from the filter center. Although experimental data which describe the variation of particle col lection per unit filter area as a function of specific location on the filter surface have not been published, geometric constraints im posed by the filter holder suggest that such variations are possible. One might expect relatively lower fiber concentration per unit filter area close to the filter edge compared with the filter center, due to the shroud and outlet configurations, (See Figure 1 of Guide) and the distribution of air flow through the filter. Statistical reliability requires that mini mum criteria be set for selection of the num ber and fiber concentration in the counting fields. The Asbestos Criteria Document60 recommends that microscopy fields "... con taining over 20 fibers not be counted because in addition to the fibers being counted, there are also present a number of grains, which interfere with the accuracy of the count." While this statement is generally correct, it should be used only as a guide for selecting the sampling time and not as a justification for discarding high counts. A better index of potential risk to the worker will be pro vided if counting fields are randomly se lected. If any fields show more than 20 fi bers, this information should be provided as part of the analyst's report with an indica- ASARCO ALV 0002047 98 tion of how seriously this condition nur have interfered with the estimation of fiber con centration. -To minimize the counting error resulting from variations in fiber concentration across the membrane filter surface, and the statis tical distribution of fibers within the micros copic counting field, crtieria must be defined regarding the minimum number of fields, and fibers to be counted. These criteria must be consistent with practical limitations, and with inherent errors due to other aspects of the sampling procedure. A minimum of 20 counting fields has been recommended40-77 for the 37 mm diameter Millipore Filter Field Monitor. This sampling system has an effective filter sampling area of 855 mm1 or a radius of 16.5 mm. Obtaining 20 random counting fields along a radial line would provide an average of approximately 800 fim between counting fields which are 50 to 70 (im in length. This probably provides a reasonable estimate of the total fiber con centration sampled. However, a statistical evaluation of the error introduced by vary ing the number of counting fields has not been reported in the literature. Estimation of Errors It is worthwhile to consider the various errors introduced into the final calculated fiber concentration by the several sampling and analytical procedures. Some of these can be estimated quantitatively, while others or major concern can only be discussed qualitatively. Fiber concentration(F), in fibers/cc air sampled, will normally be cal culated as follows: P _ Asr.rtf Avr.nvrJ.Q where: Asf -- Effective filter area, cm1 Avr -- Area of viewing field, cm2 rtf -- Net number of fibers counted nvF -- Number of fields counted t = Sampling time, minutes Q -- Sampling flow rate, cm3/min February, 1975 A statistical estimate of the error intro duced by counting a limited number of fibers can be developed. It is expected that the Poisson distribution defines the variation in particle (or fiber) concentration when look ing at a selected viewing field on a sampling filter.40 While this has been verified experi mentally,14 other investigators have noted that the distribution of dust particles counted in the microscope field is more disperse than indicated by Poisson statistics.*5-44 For a Poisson distribution, a minimum fiber count of 100 would have a standard deviation of 10% of the mean. For a more disperse distribution, the standard deviation would be greater, perhaps as much as : 15%. The sampling flow rate can usually be calibrated to be within 10%, based on a reasonably careful laboratory calibration using a one-liter burette, wet test or dry gss meter, or spirometer. Because of the importance of air density on flowmeter response, this calibration must simulate pres sure drop, barometric pressure and tempera ture to be found under field conditions, so that air flow measurements will be sufficient ly accurate. Since flow rates through many personal samplers pulsate, calibration of the rotometers of such samplers under steady flow conditions will introduce errors. If the sampling filter becomes heavily loaded, caus ing a significant increase in filter pressure drop, the laboratory calibration will be in correct, and it may be necessary to recali brate the sampling system after use to pro vide an estimate of the average flow rate during the sampling period. Under these loading conditions, particulate concentra tions on the filter may make fiber counting extremely difficult. The error (1 standard deviation) in measuring the area of the view ing field (VF) or of the sampling filter (SF) is on the order of 4%. Based on the estimated errors for the in dividual parameters and assuming no error in rtf or t, part of the potential error in the reported fiber concentration can be calcuhad using the law of propagation of error.87 ASARCO ALV 0002048 American Industrial Hygiene Association Journal This assumes that the error for each sam pling parameter is independent of the other parameters. Using the previously defined errors, the error in fiber concentration is -- 15% for one standard deviation. To provide 95% confidence limits for sampling results, one should consider a potential error of - 2 standard deviations or - 30%. This type of analysis is limited to those variables for which errors can be quanti tatively approximated. Several other major sources of error exist which can only be discussed in qualitative or semi-quantitative terms. Variations in the fiber-cloud concen tration as a function of time and sampling location are difficult to define. The best estimates of hazard to the worker are ob tained when using personal samplers located in the immediate vicinity of the individual's breathing zone. General room samplers pro vide an estimate of average fiber concentra tion, and can be used to monitor overall performance of control procedures, but they are not adequate to estimate exposure to the individual workers. Experimental studies have quantitated variations in the concen tration of particulates released from a point source as a function of distance from the release point.88'89 These studies indicate that there are major differences in air concentra tion associated with small changes in loca tion close to the source, and provide a basis for estimating the magnitude of errors inherent in estimating worker exposure based on general room samplers. Field data com paring air concentrations defined by per sonal samplers and general samplers for glove box work areas90 show similar varia tions. While these work situations are not directly comparable to those involving as bestos operations, the results indicate the difficulties associated with relating general room air sampling data to individual ex posures. Variations in fiber concentration as a function of time during the work day will depend on the specific operations involved. To date quantitative estimates of this varia 99 tion have not been developed, and the in dividual collecting air samples must care fully observe the work cycle in order to evaluate peak concentrations in addition to developing an estimate of the average work day exposure. In all likelihood, the most significant errors associated with airborne asbestos concentration assays are those associated with microscopy and the variations between counters. The quality of microscopic optics, maintenance and alignment of the micro scope, care in mounting the filter sample, concentration of fibers and particles, and attitude of the microscopist will significantly affect the fiber count reported. Recom mended optical quality77 can easily be ex ceeded in many recently made microscopes. If the sample contains a significant number of fine fibers, a high quality, well maintained microscope may make it possible to resolve up to twice as manv fibers as compared to a microscope satisfying minimal specifications. Variations between different well trained counters (using the same microscope on the same samples) range from 20 to 40% in terms of a one standard deviation co efficient of variation.91 When different fa cilities, employing different microscopes and counters are compared, variations of several fold can be expected. Sample concentration is another variable affecting fiber count. If particle or fiber con centration is high, overlap and difficulty in distinguishing fibers through the particle background will occur. The significance of these effects has not been quantitated, al though experience suggests that at concen trations below 5 fibers, or 100 particles per microscopic viewing field, there will be no significant interference with the count pro cedure. Preliminary data have also suggested that as fiber concentration increases, the effort expended by the microscopist in look ing for the finer fibers by continually re focusing and scanning the viewing area de creases. Unfortunately those potential errors have not been quantitated. t ASARCO ALV x;w 100 This extended discussion of the potential errors associated with the fiber counting technique is designed to place the sampling results obtained in proper perspective. The technique is certainly superior to the previ ous impinger procedure which did not moni tor the contaminant of concern, i.e., the fiber concentration. However, the accuracy of the present technique should be improved in order to provide a more accurate estimate of worker exposure. To minimize the error associated with this estimate, sampling re sults should include the following informa tion: ( 1) Calculated fiber concentration (f/cc). ( 2) Number of viewbg fields with more than 20 fibers, and an estimate of effect on calculated fiber concentra tion. (3) Estimated flow rate error. ( 4) Total and background counts. ( 5) Number of fields counted and area of viewing field. (6) Any data (quantitative or qualita tive) regarding presence of fibers less than 5 /xm in length. (7) Sampling time. (8) Sampling flow rate. (9) Active operations during sampling. (10) Type(s) of fibers handled during sampling (i.e,, types of asbestos, and other fibrous materials pres ent). (11) Types of nonfibrous materials pres ent during sampling (i.e., quartz, talc, etc.) and semi-qualitative esti mate of particulates (nonfibrous background) observed while count ing fibers. (12) Optics used. Interpretation of Fiber Concentration For a disease such as asbestosis, the de velopment depends on long term exposure, and the object of sampling is to show that the long term exposure is not likely to ex ceed the threshold limit. If a large number of samples are taken, the actual cxptKure'-''^^ can be determined with reasonable accuracy, but from a few samples with a mean dose ' * to this threshold limit value, it is not possible to state whether the exposure is above or" below the threshold limit. -.V.- ..;'.^S: There have been a number of studies*1-3* 'i'v' on sampling statistics for various hazards. :; j MOne of these approaches is the National Coal /V-.-jlV1 Board Study (U.K.)96 which shows that for 1- an approved limit of 700 particles/cm' 'v (ppcc) one sample indicating less than 370 . ppcc or two successive samples less than 580 ppcc are sufficient to approve the work- .' ing place. Approved implies that the limit of 700 ppcc will not be exceeded on more than one in ten shifts. The principle behind this approachi is applicable to all TLV sampling problems ., but the actual values used depend on the variability of the concentration and the measuring technique. In no case can one sample be used to justify the statement that there is no hazard from the contaminant in question unless the concentration found is very much less than the TLV and that the sample has covered the operation(s) likely to emit the given substance. References 1. Merewether, E. R. A., and C. W. Price: Re port on the Effects of Asbestos Dust on the Lungs and Dust Suppression in the Asbestos Industry. H. M. Stationery Office, London (1930). . v, , 2. Harries, P. G.: Asbestos Hazards in Naval Dockyards. Ann. Occup. Hyg. 29:222 (1968). 3. r an?a, A. J.: Silicosis and Asbestos, Oxford University Press, New York (1938). 4. Selikoff, I. 3., S. Churg, and E. C. Hammond: Relations between Exposure to Asbestos and Mesothelioma. New Eng. J. Med. 272:560 (1965). ,, -c 5. Lane, R. E. (Chairman), J. M. Barnes. D. E. Hickish, J. G. /ones, S. A. Roach, and E. King (Secretary): Hygiene Standards for ChO'sotile Asbestos Dust. Ann. Occup. Hyg. II:47 (19681. 6. Becklake, M. R., G. Fournier-Massey. C. fc. Rossiter, and I. C. McDonald: Lung Function in Chrysotile Asbestos Mine and Mill Workers of Quebec. Arch. Env. Hl/h. 24:401 (*972). 7. Selikoff, I. J.. E. C. Hammond. and J. Churg: Asbestos Exposure, Smoking and Neoplasm. /. A. M. A. 204:106 (1972). ASARCO ALV 0002050 [ American Industrial Hygiene Association Journal 8. Champion. P.: Two Cases of Malignant Mes othelioma after Exposure to Asbestos. Am. Rev. Resp. Dis. 103:%21 (1971). 9. Selikoff, 1. J.: Discussion of paper by I. Web ster: Asbestos Exposure in South Africa, in Shapiro, H.A. (Ed.) Pneumoconiosis: Proceed ings of the International Conference Johannes burg 1969. Cape Town, Oxford U. Press: p. 214. 0. Cooper. W. C. (Chairman): Asbestos: The Need for and Feasibility of Air Pollution Con trols. National Academy of Sciences. Wash ington (1971). 1. Gibbs. G. W.: Qualitative Aspects of Dust Ex posure in the Quebec Asbestos Mining and Milling Industry, in Walton, W. H. (Ed.) In haled Particles III, Vol. II. p. 783, Unwin Bros., Old Woking, Surrey, (1971). 2. Enterline, P. E., and V. Henderson: Type of Asbestos and Respiratory Cancer in the As bestos Industry. Arch. Env. Hlth. 27:312 (1973). 3. Newhouse, M. L.: A Study of the Mortality of Workers in an Asbestos Factory. Brit. J. Ind. Med. 26:294 (1969). 4. McDonald, J. C., A. D. McDonald, G. W. Gibbs, 1. Siemialycki, and C. E. Rossiter: Mortality in the Chrysolite Asbestos Mines and Mills of Quebec. Arch. Env. Hlth. 22:677 (1971). 5. Webster, I.: Malignancy in Relation to Crocidolite and Amosite. Biological Effects of As bestos. p. 195, WHO Int. Agency for Res. on Cancer, Lyon (1973). 6. Selikoff, I. J., C. E. Hammond, and 1. Churg: Carcinogenicity of Amosite Asbestos. Arch Env. Hlth. 25:183 (1972). 7. Meurman, L. D., R. Kiviluoto, and M. Hakama: Mortality and Morbidity Among the Working Population of Anthophyllite Asbestos Miners in Finland. Brit. J. Ind. Med. 31:105 (1974). 8. Lynch, K. M., and W. A. Smith: Carcinoma of Lung in Asbestos-Silicosis. Amer. J. Cancer 14:56 (1935). 9. Merewether, E. R. A.: Asbestosis and Carci noma of the Lung. Annual Report of the Chief Inspector of Factories for the year 1947. H. M. Stationery Office, London (1949). 0. Doll R.: Mortality from Lung Cancer in As bestos Workers. Brit. J. Ind. Med. 12:81 (1955). 1. Williams, W. J.: Asbestos and Lung Cancer. Arch.'Env. Hlth. 10:44 (1955). 2. Selikoff, L J., E. C. Hammond, and J. Churg: Mortality Experience of Asbestos Insulation Workers, 1943-1968, Pneumoconiosis Op. Cit, p. 180(1970). 3. Wright, G. W.: Asbestos and Health in 1969. Amer. Rev. Resp. Dis. 100:461 (1969). 4. Wagnef, J. C.. C. A. SIeggs, and P. Marchund: Diffuse Pleural Mesothelioma and Asbestos Exposure in the North-Western Cape Province. Brit. J. Ind. Med. 17:260 (1960). 5. Selikoff, I. J., E. C. Hammond, and H. Seid- 101 man: Cancer Risk of Insulation Workers in the United States. Biological Effects of As bestos. p. 209, WHO InL Agency for Res. on Cancer. Lyon (1973). 26. Stumphius, J., and P. B. Meyer: Asbestos Bodies and Mesothelioma. Ann. Occupy Hyg. II:283 (1968). 27. McDonald, A. D., A. Harper, O. A. El Attar, and J. C. McDonald: Epidemiology of Primary Malignant Mesothelial Tumors in Canada. Cancer 26314 (1970). 28. Stanton, M. F., and G. Wrench: Mechanisms of Mesothelioma Conduction with Asbestos and Fibrous Glass. J. Nat. Cancer Inst. 48:791 (1972). 29. Wagner, J. C., G. Berry, and V. Timbrell: Mesotheliomata in Rats after Inoculation with Asbestos and Other Materials. Brit. J. Cancer. 28:113 (1973). 30. Kiviluoto, R.: Pleural Calcification as a Roentgenoligic Sign of Non-Occupational Endemic Anthophyllite-Asbestosis. Acta Radiologica Suppl. 194:1 (1960). 31. Rous, V., and J. Studeny, Jr.: Aetiology of Pleural Plaques. Thorax 25:210 (1970). 32. Burlikov, T,, and L. Michailova: Asbestos Con tent of the Soil and Endemic Pleural Asbes tosis. Environ. Res. 3:443 (1970. 33. McDonald, J. C., M. R. Becklake, G. W. Gibbs, A. D. McDonald, and C. E. Rossiter The Health of Cyrysotile Asbestos Mine and Mill Workers of Quebec. Arch. Environ. 28:61 (1974). 34. Gloyne, S.: A Method of Staining the Asbes tosis Bodies Found in the Sjxitum of Asbestos Workers. J. ind. Hyg. 13.85 (1931). 35. Fahr, T.: Aerztlicher Verein in Hamburg. Muenchener Medizinische Wochenshrijt 61:624 (1914). 36. Cooke, W. E.: Fibrosis of the Lungs Due to the Inhalation of Asbestos Dust Brit. Med. J. 2:147 (1924). 37. Cooke, W. E., and C. F. Hill: Pneumoconiosis Due to Asbestos Dust. J. Roy. Micr. Soc. 47:232 (1927). 38. Gloyne, S.: The Presence of the Asbestos Fibre in the Lesions of Asbestos Workers. Tubercle 10:404 (1929). 39. Gloyne, S.: Pathology, in Lanza A. J. (Ed.) Silicosis and Asbestosis p. 198, (1938). 40. Merewether, E. R. A.: The Occurrence of Pulmonary Fibrosis and Other Pulmonary Affections in Asbestos Workers (Concluded). J. Ind. Hyg. 12:139 (1930). 41. Blount, M., R. F. Holt, and A. A. Leach: The Protein Coating of Asbestos Bodies. Biochem. J. 101:204 (1966). 42. Suzuki, Y., and J. Churg: Formation of the Asbestos Body: A Comparative Study with Three Types of Asbestos. Environ. Res. 3:107 (1969). 43. Sundius, N.. and A. Bygden: Der Staubhalt einer Asbestoisilunge und die Beschaffenheit der sugenannten Asbestosiskorperchen. Archiv. I i i I j \ f ASARCO ALV 0002051 102 fur Gewerbepathologie un Gewerbehygiene 8:26 (1937). 44. Vorwald, A. J., T. M. Durkan, and C. Pratt: Experimental Studies of Asbestosis. A. M. A. Arch, of Ind. Hyg. &. Occ. Med. J:1 (1951). 45. Gross, P,, T. P. DeTreville, L. J. Cralley, and J. M. C. Davis: Pulmonary Ferruginous Bodies. Arch. Path. 85:539 (1968). 46. Timbrell, V., and J. W. Skidmore: The Effect of Shape on Particle Penetration and Reten tion in Animal Lungs. In Inhaled Particles III Op Cit, p. 49. 47. Timbrell, V.: The Inhalation of Fibrous Dusts. Ann. N. Y. Acad. Sci. 132:255 (1965). 48. Wagner, I. C,, and J. W. Skidmore: Asbestos Dust Deposition and Retention in Rats. Ann. N. Y. Acad. Sci. 132:75 (1965). 49. Timbrell, V.: The Inhalation of Fibres. In Pneumoconiosis Op Cit, p. 3. 50. Gross, P.: Is Shon-Fiberetl Asbestos Dust a Biological Hazard. Arch. Environ. Health 29:115(1974). 51. Timbrell, V.. and S. Holmes: Suggestions on Criteria for Sampling Asbestos Dust, tn Pneu moconiosis Op Cit, p. 610. 52. du Toit, R. S.: Dust in South African Asbestos Factories and Fiberizing Plants. In Pneu moconiosis Op Cit, p. 13. 53. Rendall, R. E. G.: The Data Sheets on the Chemical and Physical Properties of the UICC Standard Reference Samples. In Pneumoconi osis Op Cit, p. 23. 54. Timbrell, V.: Characteristics of the Interna tional Union Against Cancer Standard Refer ence Samples of Asbestos. In Pneumoconiosis Op Cit, p. 28. 55. Timbrell, V., et of.: Characteristics of Respir able Asbestos Fibres In Pneumoconiosis Op Cit, p. 120. 56. Threshold Limits Committee: Threshold Limit Values of Airborne Contaminants and In tended Changes for 1974. American Confer ence of Governmental Industrial Hygienists, Cincinnati (1974). 57. Federal Register, 1972, June 7th, 37, No. 110, Washington, D.C. 58. Committee on Hygiene Standards: British Oc cupational Hygiene Soc., Hygiene Standards for Airborne Amosite Asbestos Dust, Ann. Occ. Hyg. 16:1 (1973). 59. Lynch. J. R,, and H. E. Ayer: Measurement of Dust Exposure in the Asbestos Textile In dustry. Amer. Ind. Hyg. Assoc. 1. 27:431 (1966). 60. Criteria Document: Recommendations for an Occupational Exposure Standard for Asbestos. Nat'l Inst, for Occ. Safety and Health. Rock ville, MD (Feb. 1972). Keenan, R, G.i and I. R. Lynch: Technique for the Detection, Identification and Analysis of Fibers. Amer. Ind. Hvg. Assoc. J. 31:587 (1970). 62. Ettinger, H. I., and S. Posner: Evaluation of Particle Sizing and Aerosol Sampling Tech February, 1975 niques. Amer. hid. Hyg. Assoc. J. 26:17 (1965). 63. Ortiz, L. W,, B. Tomb, C. I. Fairchild, and H. J. Ettinger. Improved Techniques for As sessing Airborne Asbestos Using Optical and Electron Microscopy. Presented at Amer. Ind. Hyg. Conf., Miami, May, 1974. 64. Aerosol Studies Section, Industrial Hygiene Group: Aerosol Research and Development Related to Health Hazard Analysis, July I through December 31, 1973. Los Alamos Sci entific Laboratory, New Mexico, LA-5555-PR, April. 1974. 65. Aerosol Studies Section, Industrial Hygiene Group: Aerosol Research and Development Related to Health Hazard Analysis, January 1 through June 30, 1973. Los Alamos Sci entific Laboratory, New Mexico, LA-5359-PR, luly, 1973. 66. Air Sampling Instruments Committee: Air Sampling Instruments for Evaluation of At mospheric Contaminants--4th Edition. Amer. Conf. of Gov't. Ind. Hyg., Cincinnati (1972). 67. Lauterbach, K. E., R. H. Wilson, S. Laskin. and D. W. Meir: Design of an Oscillating Ther mal Precipitator, Report UR 199, Univ, of Rochester Atomic Energy Project (Apr. 1952). 68. Beckett, S. T.: The Evaluation of Airborne Asbestos Fibres Using a Scanning Electron Microscope. Ann. Occ. Hyg. 16:405 (1973). 69. Harness, I.: Airborne Asbestos Dust Evalua tion. Ann. Occ. Hyg. 16:397 (1973). 70. Walter, E.: 'Untersuchungsergebnisse an Staubprofen aus asbestbetrieben' (Results of Investigations Carried out on Dust Samples from Asbestos Plants), Staub, 27:481 (1967). Translation. Canada Dept, of Secretary of State, Bureau of Translation, Ottawa No. 0297a, 1968/ 71. Addingley, C. G.: 'Asbestos Dust and Its Measurement'. Ann. Occ. Hyg. 9:73 (1966). 72. Walter, E.: `Zur frage der auswertung und beurieitung von staubmessungen in asbestfabriken mit textiler fertigung', Staub 26:422 (1966). (The Problem of Processing and Evaluating Dust Measurements in Asbestos Plants for the Production of Textiles), Trans lation Dept, of Energy, Mines and Resources, FMP, MR 67/23, Ottawa 1967. 73. Resting, A. M.: 'Asbeststaubmessungen in asbestweberein und spinnereien', Staub 26:419 (1966). (Dust Measurements in Asbestos Weav ing and Spinning Mills). Translation Dept, of Energy, Mines and. Resources, MR 67/22. Ottawa, 1967. 74. Lynch. J. R., and H. E. Ayer: Measurement of Asbestos Exposure. J. Occ. Med: 10:21 (1968). 75. Ayer, H. E., and J. R. Lynch: Motes and Fibers in the Air of Asbestos Processing Plants and Hygienic Criteria for Airborne. Asbestos. In Davies C. N. (F.d.) Inhaled Particles and Vapours II, p. 511, Oxford. Pergamon Press, England (1966). 76. Holmes, S.: Developments in Dust Sampling f ASARCO ALV 0002052 American Industrial Hygiene Association Journal and Counting Techniques In the Asbestos In dustry. Ann. N. Y. Acad. Sci. 752:288 (1965). 77. Edwards, G. H., and J. R. Lynch: The Method Used by the U. S. Public Health Service for Enumeration of Asbestos Dust on Membrane Filters. Ann. Occ. Hyg. 11:1 (1968). 78. Simecek, I.: `Beitrag zur asbeststaubmessung". Staub 27:484 (1967). Contribution to the Meas urement of Asbestos Dust. Translation Canada Dept, of Secretary of State, Bureau of Trans lation, No. 297b, Ottawa, 1968. 79. Ayer, H. E,, J. R. Lynch, and J. H. Fanney: A Comparison of Impinger and Membrane Filter Techniques for Evaluation Air Samples in Asbestos Plants. Ann. N. T. Acad. Sci. 752:274 (1965). 80. Lynch, J. R-, H. E. Ayer, and D. L. Johnson: The Interrelationship of Selected Asbestos Exposure Indices. Amer. tnd. Hyg. Assoc. J. 57:598 (1970). 81. Rickards, A. L.: Estimation of Submicrogram Quantities of Chrysotile Asbestos by Electron Microscopy. Anal. Chem. 45:809 (1973). 82. Gadsden, J. A., J. Parker, and W. L. Smith.: Determination of Chrysotile in Airborne As bestos by an Infra Red Spectrometric Tech nique. Atmos. Env. 4:667 (1970). 83. Timbrell, V.: Alignment of Asbestos Fibres in Air Samples by Magnetic Fields. Presented at Amer. Ind. Hyg. Conf., Miami, May, 1974. 84. Reist, P. C., S. B. Van Camerik, and G. E. Chabot: A Further Note on the Reliability of Membrane Filter Dust Sample Evaluation, by Microscope Counting. Ann. Occ. Hyg. 75:201 (1970). 85. Sniegowski, A.: A Note on Reliability of Mem brane Filter Dust Sample Evaluation by Mi croscope Counting. Ann. Occ. Hyg. 9:65 (1966). 86. Lynch, J. R., K. J. Kronoveter, and N. Leidel: The Validity of Poisson Distribution in Dust 103 Counting. Unpublished Report, National In stitute for Occupational Safety and Health (1972). 87. Mandel, J.: The Statistical Analysis of Ex perimental Data. 72-77. Interscience Publishers. 88. Gonzales, M.: Sampling for High Specific Ac tivity Particulates. Master of Science Thesis, University of Arkansas, 1972. 89. Gonzales, M., H. Ettinger, R. Stafford, and C. Breckinridge: Relationship between Air Sampling Data from Glove Box Work Areas and Inhalation Risk to the Worker, presented at the 1972 Health Physics Society Meeting, June 1972. 90. Schulte, H. F.: Personal Air Sampling and Multi-Stage Sampling: Interpretation of Re sults from Personal and Static Air Samplers, ENEA Symposium, p. 495, Stockholm (1967). 91. Ettinger, H., C. Fairchild, O. Moss, and L. Oritz: Aerosol Research and Development Related to Health Hazard Analysis, Quarterly Progress Report LA 5016 PR (July, 1972). 92. Beadle D. G.: An. Investigation of the Per formance and Limitations of the Konimeter. J. Chem. Met. and Min. Soc. of S. A. 57:261 (1951). 93. Walton, W. H.: The Airborne Dust Problems in Coal Mines in Great Britain. Mining Eng. 12691 (1966). 94. Tomlinson, R. C.: Sampling Programmes and Sampling Instruments. Instrum. Tract. 77:578 (June 1957). 95. Tomlinson, R. C.: A Simple Sequential Pro cedure to Test whether Average Conditions Achieve a Certain Standard. Appl. Statist. (5:198 (1957). 96. Anon.: The Sampling of Airborne Dust for the Testing of 'Approved Dust Conditions'. National Coal Board, F3837, London, (Oct. 1, 1965). Joint A1HA-ACG1H Aerosol Hazards Evaluation Committee AIHA Morton Lippmann, Pb.D., Chairman J. LeRoy Balzer, Ph.D. Douglas K. Craig, Ph.D. Graham W. Gibbs, Ph.D. William C. Janes William H. Krebs, Ph.D. Carl A. Mongold Eric B. Sansone, Ph.D. Marvin Tillery * Thomas F. Tomb ,l; Russell W. VonHouten Wesley R. VanPelt, Ph.D. Donald L. Webster ACGIH Howard E. Ayer George Carson, Ph.D. Harry J. Ettinger Murray Jacobson Geoffrey Knight Jeremiah R. Lynch G. Major Owen Moss Milton Scheinbaum Glen W. Sutton ft I ( I ASARCO ALV 0002053