Document R20rLqnm41ZQodZ1GxvrNk6Yv

PLAINTIFF'S EXHIBIT MBL-149 Th rapid development of the drywail construction trade in the United States isdescrit It is estimated that some 75,000 U.S. construction workers are currently employed ifl? trad*. The use of a variety of spackle and taping compounds is shown to be associiS with significant asbestos exposure; air samples taken in the breathing zone of dn tapers during sanding of taping compounds show fiber concentrations exceeding^ several times, the maximum levef permitted by United States Government reguiatio These findings are given together with the result of a clinical field survey of dry construction workers demonstrating that asbestos disease may be an important haaS hazard in this trade. Drywail construction and asbestos exposure ALF FISCHBEIN. M.D.. ARTHUR N. ROHL. Ph 0.. ARTHUR M. LANGER. Ph.O. and IRVING J. SELIKOFF. M.O. Environmental Sciences Laboratory. Oaoartmem of Community Medicine. Mount Sinai School of Medicine of the City University of New York. New York. NY T0029 introduction The introduction of new industrial processes widens the range of possible occupational health hazards. During the past two decades, drywail construction has achieved an important place in the United States construction industry. In such construction, taping of joints is required, i.e. applying and finishing spackle material at wallboard joints. S pacSding and drywail taping compounds consist of extremely fine-grained white powders or pre-mixed pastes. Analyses of samples of spaclding, patching and jointing compounds have shown that plaster of Paris is frequently a major constituent but that quartz, talc, micas, clays and ground limestone may occur in many formulations. More significantly perhaps, it has beea shown that some compounds contain asbestos minerals, added as reinforcing agents. Either chrysotile or amphibole asbestos, or both, have been found in 13 out of 13 industrial products; analysis of spackle compounds, which are used in home repair work, has demonstrated that they also frequently contain from 5 to 12% by weight of asbestos minerals, as well as quartz.1" history and tarminology The history of the drywail industry m the United States began at the end of World War U. Thu work w* supported in pan ev mh3 pram ES OOS2S- Cemnw 1979. AsMncw I MJ Returning servicemen and their families i living places rapidly and inexpensively, construction would very soon replace lathingT and plastering. While the plasterer had to! J perform his work alone, the drywail tapers w_- j able to synchronize their work with that jot ~ -j carpenters, electricians, plumbers and other^t building trades, greatly reducing construction time. Low cost housing went into rria^* production in 1947-1948; wallboard sections^ were soon manufactured to fit standard room/' dimensions, enabling a worker to construct; living quarters within a few hours. DryvnuT^. construction was also considered suPeriofj: f because of its adaptation to soundproofing wj ' fire codes. The tapers were to be called "drywfl I workers" as opposed to plasterers or "wet ^ workers." Initially part of the Internationa^ j Brotherhood of Painters and Allied Trades local f unions, workers engaged in this new trade, j tended to be separately defined in the 1960's and . local union groups of drywail tapers and \ painters were constituted with currently a total^ j of some 3.500 members. Furthermore, it'JJ*. . estimated that some 75,000 other construction^ l workers were currently employed in this traded \ the United States, but are not in defined i unions of tapers. (In addition, carpenteflj^ \ electricians, and plumbers may work ifl ^ vicinity of ongoing drywail construction.) Son lm.iai.Hrt 4ooc J. (40) 06251 448020 United States is descrit ) currently employed in^ is shown to be associat breathing zone of drywi cantrations exceeding^ 3 Government reguiatidr cal field survey of dryvva lay be an important heaT s exposure ind nt :Z9 men and their families ne :ly and inexpensively. Dr id ' '' soon replace lathin Wh. the plasterer had ^ alone, the drywall tapers we iize their work with that* ricians, plumbers and otte greatly reducing construcuoa : housing went into mas H7-1943; wallboard section ictured to fit standard rooo ling a worker to construe ithin a few hours. Dr i also considered superior station to soundproofing j :-crs were to be called "dr sed to plasterers or "wet-* iy part of the Intematiou ainters and Allied Trades jo engaged in this new rately defined in the 1960'sj ups of drywall tapers .ju stituted with currently a tob members. Furthermore,' ivj me 75,000 other construction rently employed in this trader t, but are not in defined lo s. (In addition, carpenU pi* ers may work in '.g ti -all construction.) son bn. Ind. Hrt- Xaoc J. (40 in approximately 1975, the mixing was done several times a day. sanding After the taping compound has been applied at the wallboard joint and has dried, the surface is sanded. Two different sanding techniques have been used: pot* sanding This is the most common method at present. The apparatus used consists of a 5-foot pole with a steel plate to which a piece of sandpaper is attached (Figure 1). hand sanding This technique is preferred to pole sanding by some tapers, since it allows better visual and manual control. Hand sanding is also required in those situations in which the space relationships prevent the taper from using pole sanding. Figure 1 -- Pols sanding ot drywall taping compounds. Sovsral applications of taping compounds may be required to make a smooth join between gypsum wail boards, requiring extensive sanding. work procadures Work performed by tapers may be associated with significant asbestos exposure. Tapers spend approximately 5-10% of their working time sanding dried spacide and thus may experience intermittent asbestos exposure. Significant exposure may also occur among painters who occasionally apply or sand spackle and taping compounds. analysis of materials Analysis of taping compounds and spackle materials indicate that a number of bfologicaily active minerals are to be found. The compounds that have been analyzed by x-ray powder diffraction, optical and electron microscopy include both dry and paste types. Of 15 industrial drywall taping and spackling compounds. rnmainrri-ashrsxps (Table 1). In addition, three compounds contained two different types of asbestos, namely chrvsonlc and tremalite. Quartz, and other silicates were also present in some other compounds. The results of these analyses showed that most of the taping compounds used in the United States construction industry contained asbestos in the order of 4-5%. One compound contained 1015% chrysotiie and 8-12% tremoiite. _ mixing Prior to applying the spackle material at the wall board joints, the compound is prepared in the following way: Half of a 25 lb. bag of drywall taping compound is poured into a 5 gallon bucket containing about 2 gallons of water. The toixture is stirred until it has obtained a desired degree of consistency, whereupon the remainder f the dry mix is added. The mixture is again stirred. This operation takes approximately 6050 seconds. Until pre-mix material came into use personal air sampling during drywall taping A series of dust counts showed that asbestos fiber concentration may be considerable. Table II shows fiber concentrations experienced during pole and hand sanding of drywall taping compounds. Air samples taken in the breathing zone of the drywall tapers during sanding of wall board taping showed that fiber concentrations often exceeded, by several times, the maximum level permitted by United States government J**nc*' Indian* Hyptm (vt s/n m. 06. 448021 regulations. Measurements were made uti the standard techniques of the National InStS of Occupational Safety and Health (NIOSI asbestos sampling and analysis. During pole sanding, 7 out of 10 exceeded the then existing threshold limit vaJL (TLV) of 5 fibers/cm3 longer than 5 qm/ffSH TLV has since been reduced to 2 fibers/cmJJ longer than 5 nm). Hand sanding generated fibeH concentrations close to or in excess of the TLV = It may be emphasized that the asbestos fiber^ concentrations generated by sanding were^l similar to those measured in the work 12 environment of asbestos insulation workeixT'* Mixing of drywall taping compounds, procedure which is currently not frequently* used, generated fiber counts 7-10X the threshold limit values. Moreover, the data in Table II show \ that, at various distances from the actual taping - ' operations, detectable fiber concentrations ire-r4 present in adjacent rooms during the varioS cUr procedures. Other building trades would thus - ^ also be exposed. It should be emphasized that'- 3 the fiber concentrations measured by standard { procedures using optical microscopy reflect only. a small proportion of the number of fibers actually present since they fail to detail fibers shorter than 5 tim in length and 0.5 urn in width/HoweveTT electron microscopic analyses ofthese"' air samples showed a large number of these'' smaller fibers too fine to be seen by optics! microscopy (Figure 2). The disease-Lnducint. potential of such fibers may be significant Si U u f re prevalence of asbestos disease among taping workers A clinical field survey of 114 taping workers employed in the New York Metropolitan area, was conducted. Examination included a careful life-time occupational history with recording of all previous occupations on a specially designed questionnaire. Interview also included pa** medical history, smoking history and review of chest symptoms, such as chronic bronchitis*- dyspnea and cyanosis. Chronic bronchitis w** recorded as present if the individual gave * history, of phlegm production for more tbs three months per year during a period of tw^ years or longer. Each individual had comprehensive physical examination Xjl //ML Hrt toec V- W Url. Ifl* U P a Lr 3_ ni'wnr 1;? ia s ia ft. 06253 448022 nents were made utilimg ss of the National InsStufi y and Health (NIQSH)"{pj i analysis. ng, 7 out of 10 sampl sting threshold limit valS i3 longer than 5 pm. (Tffi reduced to 2 fibers/ cm ;nd sanding generated fibwl :o or in excess of the TLV[ ed that the asbestos fibo] crated by sanding werej measured in the worn -estos insulation workenj l taping compounds^ currently not frequent counts 7- 10X the threshold sr, the data in Table II shw aces from the actual taping] le fiber concentrations any rooms during the various] milding trades would thus] should be emphasized ti ior tasured by stand ticai microscopy reflect oSy . of the number of fit ce they fail to detail fit . length and 0.5 in widf microscopic analyses of ties i a large number of tbe^ fine to be seen by optica i 2). The disease-indutiz bets may be significant^ aestos disease among rvey of 114 taping worker ew York Metropolitan; amination included a < nal history with recording] itions on a specially desig .erview also included r.f ^ noting history and review.o such -as chronic bronchi osis. Chronic bronchitis :nt if the individual ga**] i production for more yea iring a period of I . Each individual bay physical examination TABLE II Asbestos Fiber Levels During Use of Tsping Compounds Containing Asbestos Minerals Operation* Number of temple* Mm fiber concantratfbn (fiber*/cm') Rang* of fiber concentration* (fibere/em1) Pol* (ending (3-5 ft.) 10 10.0 1.2-19.3 Background (8 ft.) sama room 3 3.8 3.5-19.8 Background 125 ft.) ad|*c*nt room Hand sanding (3-5 ft.) Background (8 ft.) sama room Background (IS ft.) adjacent room 2 11 2 2 4.8 0.7- 8.8 5.3 1.3-18.9 2.3 2.1- 2.5 4.3 1.5- 7.1 Dry mixing (3*5 ft.) Background (10-20 ft.) sama room 2 3 47.2 5.8 35.4-58.0 0.5-13. t Background (16-35 ft.) adtacam room 2 2.8 2.1- 3.1 Swaaotng floor (10*50 ft) 15 mtnutta attar swaaoing 1 41.4 35 mtnuta* attar swaatxng 1 26.4 special emphasis on examination of the chest. Standard 14" X 17" chest roentgenograms were taken; both PA and oblique projections were taken; interpretation of chest films was done by a panel of 5 experienced readers, and a consensus reading was thereafter obtained. Readings were classified according to the ILO U/C International Classification of Radio graphs of Pneumoconioses, and a reading was considered abnormal if fine or moderately thin irregular opacities were present (s or t 5 1/0) *ad/or pleural thickening was present (5 a-1). The age distribution of the examined workers is shown in Table III. Only 17 (15%) were less than 30 years of age, while 75 workers (66%) were between 30 and 49 years. Many young workers had been in the taping trade for several yean; Table IV illustrates the distribution of duration since onset of taping. The majority of the workers had a long history in the trade; for 67 individuals (59%), 15 years or more had passed since the onset of occupational exposure to taping compounds, while 19 workers (17%) had begun exposure 10 years prior to the examination or later. TABLE III Age Diatnbcrbon of 114 Oryweil Taper* Age <30 30-39 40-49 50 - 59 60 + Torn! Ngmi ar 17 31 44 IS _7 114 Ptroanf 149 27 2 38.8 132 .11 1000 TABLE IV Ouretioo Since Onset of Expoaur* of 114 Otywell Taper* Yaers unce onaet of eapoeur* 0- 9 10 14 15- 19 20-24 25 Tote! Numbar 19 28 31 28 10 114 Percent 18.7 24.8 27.1 22.8 8.8 100.0 06254 448023 ,witll [pleural tl Offline wc ^Sreflchynu T^oiaithic. i'tbce onset OBiet was 1 JC . ^cussior lie ident ipackling conjiderab those who occupation jrywaii coi importance the world, this popuh that found workers." ll 0.4 um potato nan examined Figure 2 -- Electron photomicrograph of air sample taken during sanding of taping compound. Numerous chrysotile fibers ara visible, most of which are shorter than 1 ndioiogic: compared micrometer. They would not be visible by optical microscopy. Granular particles are micas, clays and plaster of Paris. demonstra eoosiderai inadequac 'I There were 79 smokers (69%), 13 non-smokers Rhonchi were found in 15 workers (13%),'Jj procedure (11%) and 22 ex-smokers (19%) among the of whom were smokers. 5 of these show which use mkroscop examined workers. Chronic bronchitis was parenchymal disease on x-ray, while 2 .h large cum reported by 25 individuals (22%), ail of whom pleural thickening. optical mu were heavy smokers. 10 workers (9%) gave a The roentgenographic abnormalities amol ante disc history of dyspnea on exertion. 9 of whom were the examined workers are shown in Tables V l worker wi heavy smokers; one was an ex-smoker. Chest VI. The prevalence of abnormal chest wy large -f! examination revealed dry rales in 10 individuals increased with longer duration since onset fine to be (9%). 7 of whom were smokers and 3 of whom exposure; among those with 10 years from on were ex-smokers. 6 of these, however, had a of exposure or less, 26% had positive has been positive chest x-ray for parenchymal disease; roentgenographic signs (> S, 1/0); this] none had any visible pleural abnormality. prevalence increased to a high of 51% amonjj eondussr We h*v TABLE V Ro.ntg.norgr.phic Abnormliti** (Parenchymal) Among 110 Orywali Tapare Yv* sine* ons*t of sxposur* 0- 9 10 - 19 20 + Total Total number xammed 19 58 35 110 Abnormal x*ry (>s. 1/01 Number s 22 18 Percent 45 TABLE VI Pleural X-ray Abnormalities Among 109 Orywali Tapsre Duration (yeenl Total number xamtned Number of ibnormalftiea Total 109 8.3 J construct; Our frndi 40$ Am. ind. Hrt Assoc J. (40) MIT. 06255 448024 ng jing ortar man t panicles are ,n 15 workers (13%)j| s. 5 of these shoj n x.-ray, while 2jb rjfl c abnormalities amoa :e shown in Tables Vj abnormal chest, duration since onse -vith. 10 years from oi . 26% had positjj ,ns S, l/G);t_ a high of 51% ainofl LEVI lormalities Among vail Tapers ju"w! ith 20 yean or longer since onset of [ thickening was found in 9 (8%) of the HgSinfe workers, 5 of whom had concomitant fSgjenchynial abnormalities. All individuals with rotund thickening had a history of long duration onset of taping; the mean duration since i was 19.1 years (range 13-22), (Table VII). ^discussion fiThe identification of asbestos in drywall ipackling and taping compounds and the 'considerable prevalence of asbestosis among - those who use these materials have added a new - occupational health perspective to industrial drywail construction, a trade that is of growing importance in the United States and throughout the world. In fact, the prevalence of asbestosis in this population of drywall tapers is similar to that found by others among asbestos insulation workers.121 In these studies, evidence of pulmonary asbestosis was present in 542 of 1,117 examined workers, i.e. in 48.5% (having ndiological change as sole criteria); this is to be compared with- the present study, which demonstrated a prevalence of 40.9%. It is of considerable importance to emphasize the inadequacy inherent in the standard NIOSH procedure for determining asbestos exposures which uses optical microscopy alone. Electron microscopic examination of air samples shows a luge number of fibers too fine to be seen by optical microscopy, but which nevertheless may cause disease. Analysis of the lung of a taping worker who suffered from lung cancer, showed *ery large numbers of fibers, all too small and fine to be seen with the optical microscope. This Msbeen reported elsewhere.111 conclusion We have found that taping workers in drywall construction have risk of exposure to asbestos. On? findings suggest that asbestos disease, as a result, is an important hazard in this trade. It is therefore important and urgent that appropriate measures be taken in order to prevent further asbestos exposure. The availability of asbestosfree taping compounds would be an important step toward this objective. Despite this, however, asbestos-related disease will inevitably be seen in taping workers in the future since asbestos-free materials were not widely introduced into the trade until very recently, and the result of past exposure will be with us for some decades in the future. The question of appropriate long-term surveillance is important and unresolved. acknowledgement* Thanks are due to Dr. Ruth Lilis, Dr. Henry A. Anderson, Dr. Susan M. Daum, Dr. William Lorimer and Dr. William Rom for their participation in the clinical and radiologic studies. We are indebted to Mrs. Janet Kaffenburgh, Mrs. Frances Perez and Mrs. Diane Monahan for their assistance in this study. We wish to thank Mr. John Aifarone, Mr. Frank Aifarone and members of The Drywall Tapers and Painters, Local Unions 1974 of New York, New York and 1976 of Hopeiawn, New Jersey for their cooperation in this project. references 1. RoW. A. N.. A. M. Linger, t. J. Selikoff and W. J. Nicholson: Exposure to Asbosto* in tho Us* of Consumer Sgackting. Patching and Taping Compound*. Set. 759:551-553(1975). 2. Selikatf. I. J.. J. Churg and E. C. Hammond: Tha Occurrence of AsbestosisAmong Insulation Workers in the United States. Ann N. Y. Acad. Sci 732:139155 (19851 3. Flachbein. A.. A. M. Linger. Y. Suzuki and I. J. Selitoff: Cancer of the Lung in a Qrywell Taping Worker Reoort of a Case. Presented at the Third Intarnationai Symposium on Oataction and Pravannon of Cancar. New York. April 28-May 1. 1978. Abstract. I7 Number at abnormalities Hyt Asoc J. (*0) wir.'i Indusm* Hyptn* Aaoeutm JOUMM. <4* i/71 '[A 7 06256 448025 y Index il article: seveq or ed title of but it is ade in the Further umber of with the erne. This invention Vll rights id under n may be nutted in iisaon of rer 1 on. i ing of tne is always adverd to the Industrial n WC1H 'ndustnal senpuon S12 X 00). r, British ravistoefc aced with (For the den with Medical' 104A All olishcr m nd single publisher ending at y. British \irfrdght /end, NJ Editorial British Journal of Industrial Medicine 1988:45:505-508 Low level exposure to asbestos: is there a cancer risk? In the early days of asbestos use when its harmful potential was not understood hygiene standards at the workplace were extremely low. Airborne dust concen trations were uncontrolled and fibre levels in excess of 100 per ml were not unusual. This type ofexposure led to the discovery that inhalation of asbestos produced pulmonary fibrosis.' bronchial carctnoma.: and mesothelioma.1 Efforts to reduce concentrations of asbestos fibre began jn the 1930s and. since the 1950s. increasingly stringent limits have been imposed by law. Some countries have banned many uses of asbestos, although it remains almost essential for some special packing and friction products. Various fibre substitutes have been introduced in advanced coun tries. especially for plastic or cement products, though these are generally less durable and more costly. The. latter point is a serious matter in third world countries where asbestos cement products are well-nigh essential for water and drainage pipes and for building con struction. In addition, it now seems possible that some manmade mineral fibres could carry similar risks to health.4 The use of asbestos today in the developed world may only be undertaken under strict control, in most countnes at levels of 1 fibre; ml or less. Exposure levels in buildings and in the general environment of our cities, resulting from the use of asbestos products, are far lower still. The question of whether such low occupational and environmental exposures are free from risk has become a major practical, social, and legal issue. As it is apparent that controlling dust has largely eliminated the risk of pulmonary asbestosis. the main concern is now over cancer and whether or not there is a safe threshold below which asbestos does not cause tumour production. Evidence on this matter is afforded by both experimental studies in laboratory animals and epidemiological studies in human popula tions exposed to asbestos. Both approaches have their limitations. Animals may be exposed to measured doses ofasbestos but their short life span makes direct extrapolation of the results to man difficult. Epidemiological studies, on the other hand, suffer from great uncertainty about the precise nature and intensity of exposure, especially in the dtstant past. Whereas the impossibility of proving a complete negative is accepted, both the epidemiological and expenmental findings provide a basis for rational decisions. Expenmental studies have shown that the path ological effects of asbestos depend not only on the mass of dust deposited in the lung tissues but on fibre length and durability. Wagner el al reported on rats exposed to the main types of asbestos in which a constant mass of dust was administered for periods of one day to two years.1 The degree ofdiffuse pulmonary fibrosis varied proportionately with duration of exposure. The expenence of tats exposed for one day only was similar to that of unexposed control animals. The production of pulmonary tumourswas less clearly related to duration but in general more tumours occurred in animals exposed for long periods. Some animals exposed for a single day developed pulmonary tumours but. again, similar numbers were found in control antmals. Information is needed on the effect of a range of doses administered over a long period but such studies have not been undertaken. The production of mesotheliomas by intratracheal or mtraperitoneal injection of asbestos provides a less expensive method of examining the carcinogenic effects of widely varied doses. Davis reported studies in which reference samples of crocidolite. amosite, and chrysotile had been injected into the peritoneal cavities of rats at doses from 25 mg down to 0 01 mg.* A clear dose response was found with all three types of asbestos fibre, but whereas crocidolite did not produce tumours at the lowest dose level, chrysotile and amosite did. With each fibre type, the relation was linear when plotted on a log scale: however, with all three, both the time to first tumour and the mean induction period increased progressively with reduc ing dose. Tumours produced by the lowest doses in general occurred dunng the last few months of the rat's life, suggesting that lower doses stiil had insufficient time to produce any tumours. It also appears that not all fibres are equally harmful and evidence is accumulating that very short fibres may be innocuous. Both Stanton and Wrench and Pott and Friedrichs reported that dust samples which produced the most mesotheliomas in rats were those that contained the highest number of long and thin fibres.'* This work has been progressively refined, and it is now evident from injection studies that the most dangerous fibres are those with a length > 8 urn and a 505 06257 448026 506 diameter of < 0-25 /un.* When any fibres longer than8 /an are considered, the lowest injected doses (0-01 mg) which produced any tumours contained about 410 000 fibres of chrysouie and 110 000 fibres of amostte: a dose containing 60 000 fibres of crocidolite produced none. At all dose levels chrysouie was the most carcinogenic dust followed by amosite. with crocidolite the least so. Long term inhalation studies have recently been conducted using short fibre preparations of amosite and chrysouie.10" IrTtHese experiments the effect of short fibre dust was compared with a normal dust cloud of the same batch of asbestos containing many long fibres. The amosite preparauon was extremely short with almost no fibres greater than 5 /im in length; the short fibre chrvsotilc was less satisfactory in that some long fibres remained. Whereas the long fibres of both chrysotile and amosite produced large amounts of pulmonary fibrosis and a considerable number of pulmonary tumours, the short fibre amosite prepara tion produced neither fibrosis nor tumours. Short fibre chrysotile produced a small amount of fibrosis and some tumours, roughly in proportion to the number of long fibres present. There remains the question of whether, to cause disease, asbestos fibres need be retained in the lung for long periods: a low retained dose might well produce little disease regardless of the number of fibres inhaled: In this respect chrysotile and the amphiboie varieties differ appreciably. After varying inhalation periods at high dose levels, rats have been shown to retain up to ten times more amphiboie than chrysotile.1" The recent studies of Davis et al have extended these findings to include fibre length."0 Fewer long than short fibres were present at the end of dusting for both chrysouie and amosite: the elimination of short fibres was faster than that of long but. regardless of size, chrysouie was cleared from lung tissue much faster than amosite. Amphiboie fibres are extremely durable even in aggressive chemical environments whereas chrysotile undergoes decompositon readily, even in water. Davis et al reported that m rats. 18 months after inhalation of chrysotile at high doses for one year, no fibres remained in pulmonary macrophages or in direct contact with any of the other pulmonary cells." Only a few fibres were detected and these in old acellular fibrous tissue or in thickened basement membranes, chemical backwaters where fibre dissolu tion is retarded. The rapid removal of chrysotile and the durability of amphilbole fibres have also been noted in several studies of human lung ussues.'*'" Whereas epidemiological studies in man can attempt to measure risk directly their capacity to do so becomes increasingly unreliable as the level of exposure falls. Partly, this is due to deficiencies in exposure data but mainly it is because no reference population is ever more than roughly comparable with the exposed study group. Thus in not one of the ten or so eligible cohort studies of lung cancer mortality in asbestos workers does the fitted line for exposure response pass through the origin unless forced to do so. * Even the remarkably good approximation to linearity observed in these data sets is possibly deter mined more by duration than intensity--these two components of exposure not having yet been separately assessed. Valid though the results of well designed observational surveys are. for these reasons alone any estimate of risk carries a wide margin of possible error. In chrysotile mine and mill workers, for example, no statistically significant increase in lung cancer mortality would have been detected in nearly 2000 men with at least 20 years employment at an average airborne fibre concentration of about 20 fibres/cc. though the linear exposure response hypoth esis would imply that them was one.1' Even in the far more hazardous environment of asbestos texule fac tories the same would probably be true up to levels of about 2 fibres/cc." Such levels are as high or higher than current occupational limits and orders of mag nitude above those causing concern in public build ings. It also appears that in amosite asbestos factory workers the lower the estimated dose, the longer it took for related mortality to become evident and the smaller its magnitude.10' The epidemiological data on exposure response are much more tenuous for mesothelioma than for lung cancer. Such quantitive information as there is points to a direct relation between the risk of this disease and some combination of duration and intensity of exposure, but strongly modified by fibre type. There is now considerable evidence that pure chrysotile per sc seldom causes mesothelioma in man and that most cases are due to amphiboie fibres--crocidolite and amosite in particular--but increasing attenuon is now being directed at fibrous tremoiite. Tremolite asbestos has little or no commercial value but commonly contaminates many mineral deposits, some of which, such as chrysouie. talc, and vermiculite. are widely used industrially, and others not. The potential impor tance of ihe iremoiite factor, noted by Pooley and Rowlands t a/, was direedy shown in vermiculite miners and millers whose only exposure to mineral fibres was from contaminating tremolite." a:i The lung cancer and mesothelioma experience of this cohort was. dose for dose, many times worse than that ofQuebec chrysotile miners and millers. The contribu tion of fibrous tremolite to mortality from these causes had also been suspected in New York State talc miners many years earlier." Although the proportion of tremolite in asbestos dust in the Quebec mines and mills is less than 1%. most fibres found at necropsy in the lungs of both mine workers and chrysouie textile 06253 448027 comparable with : one of the ten or ncer mortality in ne for exposure less forced to do pproximation to is possibly deter* nsity--these two tving yet been te results of well for these reasons wide margin of mill workers, for increase in lung elected in nearly tpioyment at an in of about 20 response hypoth* ' Even in the far 'estos textile fac ie up to levels of i high or higher orders of magn public buildsbatos factory e. ** longer it ev and the Jure response are na than for lung as there is points this disease and nd intensity of are type. There is chrysoule per se and that most crocidolite and ttenuon is now moiite asbestos out commonly some of which, lite. are widely lOtentiai impor. by Pooley and n in vermiculite asure to mineral moiite.11311 The perience of this s worse than that rs.Thecontnbu`rom these causes State talc miners e proportion of aebec mines and nd at necropsy in chrysntile textile factory workers are of this type, not chrysotile.13 It thus seems probable that even the relatively low risk ofmesothelioma in individuals reputedly exposed only to chrysoule may largely result from tremoiite contamination. Further insight into the magnitude of the risk of cancer from low level exposure may be obtained from national trends in mesothelioma incidence and mor tality. Dunng the past 20 to 30 years the pattern has been remarkably similar in the U mted States, Canada. Britain, and probably Scandinavia/' The data suggest that, until the 1950s. the annual mortality from mesothelioma was similar in men and women, at about 1-2 per million. Since ihen the trends have separated: mortality in men has nsen steeply in parallel with using asbestos some 30 to 40 years earlier whereas in women there has been no significant increase. The female trend-is particularly important since it is known that" cases have occurred in women who have worked with asbestos or lived in the household ofasbestos workers. It would be expected, moreover, that during the years in question increased awareness of what was once considered a rare tumour might lead to improved ascertainment. If. despite this, there has been no apparent increase in female mor tality in a combined population of some 300 million subjects observed for at least 15 years the risk attributable to non-occupattonal exposure must be extremely small. What conclusions may be drawn from a synthesis of these experimental and epidemiological findings? We cannot say. and it may be impossible lo prove or disprove that at very low levels ofexposure to asbestos the risk of cancer is zero. None the less, further statistical analyses are needed to test the linearity of risk in relation to intensity of exposure. So far as lung cancer is concerned, clearly any increase in risk at exposure below today's control limits, even in men who smoke more than they now do. would not be detected. The same could be said for mesothelioma in relation to commercial chrysoule but with less con fidence for the commercial amphiboles. crocidolite. and amosite. If the tremoiite contamination could be removed from chrysoule the safety margin would be wider. Plausible explanations for these assertions have been shown in the laboratory. Chrysoule is poorly retained and rapidly cleared from the respiratory tract: short fibres, which predominate in low level pollution, have little or no carcinogenic potential: with decreas ing dose tumour latency may be prolonged beyond the normal life span. General toxicological experience, and the probability of defence mechanisms, would also suggest the existence of an effective threshold, or at least a sigmoid form of exposure-response. The adverse effects of asbestos have probably been more extensively studied than any other environ 507 mental hazard: and manmade mineral fibres may well require the same attention. It seems unlikely that much more scientific information on asbestos can be expec ted than is now available. Rational decisions will depend on questions ofcost, benefit, and the apprecia tion of risk. It will be for society, not science, to decide what priority should be given to the reduction of unsubstantiated and undetectable risks, even from such dreaded effects as mesothelioma and lung cancer. Institute of Occupational Medicine. Edinburgh J M G DAVIS Department of Clinical Epidemiology. Brampton Hospital. London SW36HP. I C MCDONALD Zcftim 1 Cooke WE. Pulmonary asbestosis. 3r MtdJ I927:ii: 1024-5. 2 Lynch KM. Smuh WA Pulmonary asbestosis in. Caranoma of ' the lung in asbestos uIkosis. Am 1 Cancer 1935:24:36-64. 3 Wagner JC. Slegp CA. Marchand P Diffuse pleural mesothelioma and asbestos exposure in the North Western Cape Provinee. Be J led Med 1960:17:260-71. 4 Conference report. International symposium on man-made mineral fibres in the working environment. Copenhagen. 28?29 October NH6 Am Ofcup Hyg 1987:31.99-102. 5 Wagner JC. Bern G. Skidmore JW. Timbrell V The effects of ihe inhalation of asbestos in rats. Sr J Canctr 1974:29:252-69 6 Dans JMG. A review of recent experiments on the mechanisms of asbestos pathogenicity In: Proceedings of Vih international " colloquium .n Just measuring ternnnjue and strategy. Johmttihure Asbestos Imernaiional Association 1983:25-35. 7 SiamonMF W rench C. Mechanisms of mesothelioma induction with asbestos and fibrous glass. Be J Cancer 1972:48:797-821. 8 Pott F. Fnednchs KH Tumours in rats after inlrapcnioneal injection of asbestos dusts. Satunisstnschaften I972J9; J18-32. 9 Davis JMG. Addison J. Bolton RE. Donaldson K. Jones AD. Smnh T. The pathogeneeny oflong versus short fibre samples of amottta asbestos administered lo rats by inhalation and imrapentoneal nueclion. Be J Exp Pathol 7986.47:4)5-30. 10 Dans JMG. Jones AD. Smith T Comparisons ofthe pathotenuitr of tong and ihurr fibre .amplet of chrvsvttle ushevas in rats. Edinburgh. Inwiiuie of Occupational Medicine. 1987 (Tech memo TM 87 08 i 11 Middleton AP. Beckett ST. Davies JMG. A study of the short term retention and clearance of inhaled asbestos using UICC standard reference samples. In: Walton WH. ed. tnhaied particles IV Otford Petgamon Press. 1977-247-57 12 Oans JMG. Solton RE. Brown D. Tully HE. Experimental lesions in rats corresponding lo advanced human asbestosis. Exp Hoi Pathol 1986.44 207-21 IJ Pooley PD An examination of ihe fibrous mineral conient of asbestos m lung tissue from ihe Canadian cfirysoule mining industry Enrxron Research 1976.12:281-98 14 Gylseth 8. Mow* G Wannag A Fibre lype and concentration in the lunpofworkers m an asbestos cement factory. Br J Ind Med 19*3.4*375-9 15 Sebsstien P Begin R. Case 8W. McDonald JC Inhalation of .hmotik dusa In Wagner JC. ed. Btulngical rlfeets ot ttrstntde Philadelphia JB Lippicott Co. 1987 19-29. :* McDonald JC. McDonald AD Epidemiology of asbeslos-relaied 448028