Document QgkajdZeDXydgmZQdjZye9jGR

FILE NAME Brakes BRK DATE 1980 Oct 1 DOC BRK113 DOCUMENT DESCRIPTION Journal Article - Malignant Mesothelioma in North America Reprinted from CANCER Vol 46 No. 7 October , 1980. Copyright ' 1980 by the American Cancer Society Inc. J. B. Lippincott Company Printed in U.S.A. Malignant Mesothelioma in North America ALISON D. MCDONALD MD FFCM AND J. CORBETT MCDONALD MD FRCP Ascertainment through 7,400 pathologists of all fatal malignant mesothelial tumors in Canada 1960 75 and the U. S. A. 1972 gave a total of 668 cases 272 in 1972 In Canada the annual number of male cases rose from about 17 in 1966 to 25 in 1972 but the number of female cases remained fairly steady at a much lower level The annual incidence in North America in 1972 was estimated at 2.8 per million males and 0.7 per million females aged 15 years and over Occupational histories were obtained blind for 480 of the 557 cases through 1972 and their matched controls relative risks were as follows insulation work 46.0 asbestos production and manufacture 6.1 heating trades other than insulation 4.4 For nearly half the male cases and for about % of female cases the tumor could be attributed to occupational exposure to asbestos of which a fifth were in shipyards No indication was found of other possible causes including made mineral fibers tobacco smoking or residence near zeolite deposits Four subjects were men who had been employed in Quebec chrysotile mines and 3 were children of employees but no other subject had lived in the mining area The findings remain consistent with a much greater mesothelioma potential for crocidolite and amosite than for chrysotile however further studies of factory workers exposed to chrysotile only are needed to confirm this Mineral fiber analysis of lung tissue from patients and controls is in progress Cancer 1650-1656 1980 A INTERVALS SINCE 1970 we have reported continuing studies of malignant mesothelial tumors throughout Canada and more recently in the United 12-14.17 Complete ascertainment of all fatal cases known to pathologists has been attempted followed by detailed control enquiries The need to monitor this disease and available methods of case finding have also been recently reviewed The purpose of this report is to bring together the information derived from these surveys published and unpublished and to con- sider the epidemiologic interpretation with particular regard for the carcinogenicity of the various mineral types of asbestos fiber and the roles of other possible etiologic factors From the Department of Epidemiology St. Mary's Hospital Medical School London and TUC Centenary Institute of Occupational Health London School of Hygiene and Tropical Medicine London England Supported in part by a grant from the Institute of Occupational and Environmental Health of the Quebec Asbestos Mining As- sociation Address for reprints Dr. Epidemiology St. Mary's Alison D. Hospital McDonald Department of Medical School London W2 1PG England The authors thank pathologists throughout North America State and Provincial Health Departments and their field workers the Canadian Tumour Reference Centre and the UICC Meso- thelioma Panel U. S. A. for their generous collaboration and Dr. David Oakes for performing the statistical analyses of matched control pairs Accepted for publication November 7 1979 Materials and Methods At the end of 1967 we wrote to all members of the main pathologic societies in Canada enquiring about all fatal cases of primary malignant mesothelial tumors diagnosed by means of autopsy or biopsy examination since 1959. These enquiries were continued until a reply was received from every pathologist The procedure was repeated at two to three year intervals and by this means all known cases were ascertained to the end of 1975. In 1973 letters were written to all 6,977 pathologists listed by the United States Armed Forces Institute of Pathology enquiring about any fatal case in 1972.17 Responses were obtained from all but 394 pathologists 311 of whom had worked at a hospital from which a reply had been received so we believe coverage in the U. S. A. was also close to com- plete In each case a physician on our staff visited the hospital reviewed the records with the pathologist and selected as a control a case matched for sex age and year of death in which pulmonary metastases were present from a nonpulmonary malignant tumor Relatives were then visited usually by a public health nurse who without knowledge of the identity of cases and controls completed a detailed page standard questionnaire covering full occupational and residential history and smoking habits For each oc- cupation reported the respondents were system- atically questioned about exposure to dust from 0008-543X 0.85 'American Cancer Society No. 7 MALIGNANT Mesothelioma McDonald and McDonald 1651 cement wood fiberglass asbestos rubber copper or nickel Respondents were also asked whether any member of the household had brought home dusty clothing and if so the nature of the work In Canada 312 cases were registered during the period from 1960 through 1972 and questionnaires completed for % including 281 control pairs In the U. S. A. 245 cases were registered for 1972 and questionnaires com- pleted for % including 209 control pairs A list of occupations associated with asbestos exposure was supplied by Professor I. J. Selikoff This had been used in New York and in the Netherlands and included a series of jobs with which our clerk coded the occupations blind as to the identity of subjects and controls The jobs listed were separately assessed at four centers experienced in asbestos research Table 2 footnote as to the probability that they entailed exposure to asbestos There was good agreement over the definite and unlikely categories but some difference in separating probable from possible Table 2 Occupations were headings G five classified in which under seven an association main with mesothelioma has been recognised A insulation B asbestos production and manufacture C heating trades D shipyards E construction a sixth group F of other listed jobs thought to entail asbestos exposure and a seventh group G of unlisted jobs Place of residence was classified as urban or rural and examined to see whether it was within 20 miles of an asbestos mine and in the U. S. A. of a known zeolite deposit The latter was added because of the recent report of a very high incidence of malignant mesothelioma in certain villages in Turkey with a natural occurence of zeolites.,, The cases in which a biopsy or autopsy specimen could be obtained are being reviewed by the Canadian and American mesothelioma panels of pathologists and specimens of lung tissue from control pairs are being examined by Dr. F. D. Pooley for mineral fibers by means of electron microscopy and ray dispersive analysis The results of this work will be reported later Results Sex and Age and Tumor Site NUMBER 30 SURVEY STARTED 20 10 - MALES FEMALES t FEMALES pe aa aN 1960 61 62 63 64 65 66 67 68 69 70 73 75 FIG 1. Annual number and sex of fatal cases of malignant mesothelioma reported by pathologists in Canada 1960-75 three moving averages males 307 78 were pleural including three pericardial tumors and 88 22 were peritoneal including 14 also affecting the pleura In females 99 61 were pleural including six pericardial tumors and 63 39 were peritoneal including seven also affecting the pleura At the time of death % of male subjects were under 45 years of age 50 were 45-64 years old and % 65 years or older Only one subject a male was under 15 years of age Incidence As malignant mesothelioma is usually fatal within 18 months of first symptoms deaths reflect incidence quite closely The trend in Canada by date of death is presented in Fig , using three moving averages Ascertainment was begun in 1967 and some of the increase before then was probably due to incomplete case finding for the earlier period After 1969 the trends separate as that for males continues to rise and for females to fall back Although the average annual incidence for the period between 1966 and 75 based on the population for 1971 was considerably higher in Quebec than in other parts of Canada the incidence of accepted cases differed little in the three regions Table ) In the U. S. A. the rates for males were highest in the Pacific and Atlantic regions review by the pathology panel reduced that in the Atlantic Rates for females were not related to those for males Of 557 cases reported to the end of 1972 395 71 were male and 162 29 were female In * Canadian Tumour Reference Centre National Cancer Association of Canada at University of Ottawa Ontario UICC Mesothelioma Panel at Barnert Memorial Hospital Paterson New Jersey Work Histories Analysis of probability of occupational exposure to asbestos for patients and controls during the period up to 10 years before death is presented in Table 2. Re- sults based on the four centers gave similar rates for 1652 CANCER October 1 1980 Vol 46 TABLE 1. Annual Incidence of Malignant Mesothelioma per Million Population by Age Sex and Geographic Region and the Result of Review by Pathology Panel Male Incidence per mill 15-44 years 45 years and over 15 years and over Total num- ber of cases Pathology panel Number viewed Ac- cepted % Female Incidence per mill 15-44 years 45 years and over 15 years and over Total num- ber of cases Pathology panel Number viewed Ac- cepted % Canada 1966-75 Provinces Ontario Quebec Rest TOTAL U. A. 1972 Census regions New England and South Atlantic Atlantic Northeast Central Northwest Southeast Southwest Central and Mountain Pacific TOTAL 0.7 1.2 0.4 0.8 0.5 0.1 0.2 0.2 0.0 0.2 12.0 3.0 2.6 4.9 2.2 7.0 3.0 5.4 8.0 6.0 4.3 10.3 6.5 2.5 3.6 2.6 1.8 4.0 2.7 71 103 58 232 60 83 49 192 WAW 222 WAW 222 WAW 222 35 37 222 189 125 83 8248 8248 8248 100 69 82 787 787 74 0.4 0.8 0.2 0.4 0.1 0.1 0.0 0.3 0.0 0.1 2727 2727 2727 2727 0.9 1.3 2.9 1.1 1.7 1.6 1.1 22218 2222 60 2.2 22218 2222 4938 0.7 22218 2222 4938 1.3 98 2222 4938 0.4 0.7 1.3 0.6 0.7 0.8 709 3<W 100 709 3<W 57 709 3<W 82 3716 14 67 3716 Gla 75 56 Gla 69 definite and unlikely exposure and for the two intermediate categories taken together Asbestos exposure for both subjects and controls was considerably more frequently recorded in the U. S. A. than in Canada In Canada 45.9 of subjects and 18.9 of controls had worked in occupation groups A and in the U. S. A. 64.8 of subjects and 27.0 of controls Table 3 The five groups were ranked according to the ratio of subjects to controls and made mutually exclusive by entering each man only once in the highest group for which he qualified Relative risks6 were calculated for both countries combined by comparing each group in turn with the two groups F and G taken together Insulation carried by far the greater risk RR = 46.0 followed by asbestos production and manufacture RR = 6.1 and the heating trades other than insulation RR = 4.4 After exclusion of men known to have worked in insulation or heating trades an elevated risk still remained for those who worked in shipyards RR 2.8 and construction RR = 2.6 In all 49 cases involved men who had worked in shipyards including 29 who were insulators TABLE 2. Center Distribution % of 185 Canadian and 159 U. S. Male Control Pairs According to the Probability of Occupational Asbestos Exposure as Classified in Four Centers Definite Cases Controls Asbestos exposure Probable Cases Controls Possible Cases Controls Unlikely Cases Controls control difference for definite and probable exposures Canada 1960-72 1 2 3 4 11.9 11.9 13.5 11.9 ENEE ENEE ENEE ENEE 43.8 20.5 29.7 22.7 31.4 14.6 21.1 13.0 4.3 21.1 18.9 22.7 2.7 15.7 20.5 27.6 40.0 46.5 37.8 42.7 64.9 68.6 55.7 58.4 23.2 16.7 19.4 20.5 U. A. 1972 1 2 3 4 18.2 18.9 24.5 18.9 WOWW WOWW WOWW WOWW 51.6 31.4 34.0 33.3 35.2 18.7 19.5 15.1 * Centers ) Environmental Sciences Laboratory Mount Sinai School of Medicine New York 2 Gesondsheitsorganisatie TNO den Haag the Netherlands 3 Department of Epidemiology and 3.8 20.8 16.4 22.6 1.9 18.6 19.5 27.0 26.4 28.9 25.1 25.2 59.1 58.9 54.1 54.1 30.8 27.8 32.7 33.3 Health McGill University Montreal Canada 4 TUC Centenary Institute of Occupational Health London School of Hygiene and Tropical Medicine London England No. 7 MALIGNANT MESOTHELIOMA McDonald and McDonald 1653 TABLE 3. Employment in Occupational Groups G Ten or More Years Before Death for 344 Male Cases and Their Matched Controls Canada 1960 7U 2 . A. 1972 Jobs Occupational group A. Insulation B. Asbestos production and manufacture Mining and milling chrysotile Manufacture Asbestos cement products Factory using asbestos C. Heating trades excl insulation Job necessitating protective clothing Installing or repairing furnaces or boilers Steamfitter Boiler maker Plumbing and heating Welder D. Shipyards E. Construction industry Building trades Building demolition Painting Sheetrock spackling F. Other listed jobs excl men in groups A Cases 27 4237 4237 4237 4237 8323 8323 8323 8323 8323 14 49 59 933- 933933- Controls I 222222222222- 7921 7921 7921 7921 7921 8 17 0230 0230 0230 0230 G. None of the above TOTAL * Men are listed once only in the group ranked highest see text for definition of groups Men Cases 27 25 Controls 1 7 70 27 24 30 45 30 55 90 101 176 344 344 Relative risk 46.0 6.1 4.4 2.8 2.6 1 or pipe or boiler fitters The remaining 20 men comprised 7 electricians 4 metal workers 2 painters 2 ship fitters 1 carpenter and 4 whose jobs were not specified For subjects briefly exposed to crocidolite asbestos the interval between first exposure and death from mesothelioma was 25-35 years risks were therefore recalculated for this period This slightly increased the relative risk from 2.6 to 3.2 for heating trades and from 1.6 to 2.0 for work in shipyards The distribution of occupations for 156 cases and matched controls in group F excluding those who qualified for groups A 188 is shown in Table 4. There were small excesses of some occupations but none in the larger notably garage maintenance and transport work Among other occupations unlisted G an excess was found only in horticulture 4 to 0 and railway work 5 to 2 Work with asbestos was recorded for only 2 female subjects One woman had sewed amosite blankets for naval insulation and the other packed gasmasks probably with crocidolite a substantial quantity was found in lung tissue at autopsy However 42 subjects and 26 controls had been employed outside the home 20 through 40 years before death in work of an in- dustrial nature Home Exposure Two male and 6 female subjects but no matched control had been exposed at home to the dusty clothing of an asbestos worker 2 controls were so exposed and the paired cases were not x= 3.03 1 DF P = 0.08 Of the 8 subjects 5 had been ex- TABLE 4. Employment in Other Listed Jobs of the 156 Male Cases Who Had Not Worked in Occupational Groups A and Their Matched Controls Occupation Garage Factory making rubber products Maintenance man or building superintendent Oil or chemical refinery Carpentry Factory making plastic products Paint factory Lathing Dry cleaning Sheet metal Transport industry Paper factory TOTAL * See text for definition of groups Cases 11 3 13 M39022 M39022 M39022 M39022 2--5315 2-5315 2-5315 2--5315 2--5315 2--5315 Controls 12 1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 3283-1032T1 156 1654 CANCER October 1 1980 Vol 46 TABLE 5. Occupational Group and Site of Tumor in Male Cases and Controls Pleura Peritoneum Occupational group Cases Controls Cases Controls By group A. Insulation B. Asbestos manufaa nd cmat nuu facr tue re C. Heating trades excl A D. Shipyards excl A C E. Construction excl A C D |= 222 222 222 222 10.9 44.2 A All other TOTAL 221 221 265 55.1 44.9 100 * See text for definitions Pleura includes pericardial tumors and peritoneum includes 1 16 2.3 29.1 5 7 21 9 11 20.4 1 24.1 22 9 60 205 265 22.6 77.4 100 2212 53.2 2212 46.8 2212 100 tumors of both peritoneum and pleura |0 26 26 2 8 18 61 79 2.5 20.3 22.8 77.2 100 posed in childhood and the remaining 3 plus 2 controls as adults In 3 cases and for 1 control the clothing was that of a Quebec chrysotile production worker and in 5 cases and for 1 control an employee engaged in insulation or factory work In the U. S. A. survey an additional question was asked about exposure to asbestos in work or hobbies in the home Among the 156 men who had not worked in occupational Groups A and their matched controls affirmative answers were given for 5 cases and 2 controls Neighborhood Exposure Excluding those with occupational or domestic exposure no subject but 2 controls had lived within 20 miles of a Canadian chrysotile mine In the U. S. A. 1 subject and 2 controls had lived within 20 miles of a chrysotile mine in California The places of residence of women were examined for the period 20 through 40 years before death Of 96 control pairs in Canada 19 subjects and 21 controls had lived only in rural areas of 50 control pairs in the U. S. A. 5 subjects and 10 controls had done so In Canada 67 subjects and 66 controls and in the U. A. 15 subjects and 13 controls had lived in urban areas only Site of Tumor and Occupation Of 265 male subjects with pleural tumors 55.1 had worked in one of the occupational Groups A to E compared with 22.6 of their matched controls Table 5 The corresponding percentages of 79 males with peritoneal tumors and their matched controls were similar 53.2 and 22.8 respectively However 29.1 of peritoneal cases and 10.9 of pleural cases compared with 2.3 and 2.5 in the respec- tive control groups had worked as insulators or in asbestos factories Other Materials Relative risks in male control pairs are shown in Table 6 for exposure to dust from fiberglass cement rubber copper nickel and wood Only for fiberglass was the relative risk increased after allowing for exposure to asbestos this fell to 1.62 95 confidence limits 0.75 and 3.53 respectively Seventeen subjects and 12 controls had lived for 20 through 40 years before death within 20 miles of reported zeolite deposits22 in Western U. A. A paired analysis gave a relative risk of 1.83 reduced to 1.60 75 confidence limits 0.58 and 4.93 respectively when allowance was made for occupational exposure to asbestos Smoking Findings concerning cigarette smoking habits were similar to those previously reported No association with mesothelioma was found for male cases Fe- male subjects tended to be rather heavier smokers than controls perhaps related to the higher proportion employed outside the home Overall it seems improbable that tobacco smoking is an etiologic factor of any importance in this disease Discussion The list of occupations provided by the Environmental Sciences Laboratory Mount Sinai School of Medicine proved a satisfactory method of classifying occupations thought to entail asbestos exposure With minor modifications the list could improve the comparability of control surveys in different regions and countries Greater discrimination was achieved between cases and controls by selecting occupations reported to have been associated with mesothelioma than by assigning probabilities of asbestos exposure No. 7 MALIGNANt MesotheLIOMA McDonald and McDonald 1655 to all occupations listed It is reassuring however that when the latter method was used the order of difference between cases and controls remained similar The very high relative risk in insulators bears out the results of earlier cohort studies Chrysotile has been used for insulation in the U. S. A. since the beginning of the century Amosite introduced in the 1930s and employed extensively for marine construc- tion in the World War 20 became the predominant fiber for insulation in the 1950s.3 Insulators who died in 1972 were thus exposed to amosite as well as chryso- tile some 25-35 or more years before a mesothelial tumor developed Since the observed risk of mesothelioma has been high after work with amosite only21 but not with chrysotile only circumstances tend to incriminate the amosite component In subjects who had done insulation and asbestos processing work a high proportion of the tumors were peritoneal This was also true of two large series of insulators amosite factory workers crocidolite gasmask workers and factory workers exposed to crocidolite amosite and chrysotile This contrasts with the pattern in chrysotile miners and millers among whom all cases so far observed have been pleural However the principal difference between chrysotile production workers and persons exposed to mixed mineral types of asbestos in processing or application lies in the absolute level of risk Of 4,247 deaths among chrysotile miners and millers 0.3 were due to mesothelioma,,compared with 8% in the other groups mentioned It remains uncertain however whether the risk continues to be low when chrysotile is processed Two relatively small studies of factory workers suggest that this may be so and in the present survey no increase in risk was found in garage workers certainly exposed to chrysotile from brake linings Borow et al reported 72 cases of mesothelioma 58 in employees of a large asbestos factory in New Jersey.2 Chrysotile only was manufactured from 1918 until the 1930s when crocidolite was introduced for production of large asbestos cement pipe and amosite for some other products The first case of mesothelioma occurred in 1951 and there were 3 more cases before 1960. In the first half of the 1960s the a number of cases had increased by factor of 8 and in the second half by a factor of 20. There was only 1 case of mesothelioma in the large female work force employed in the textile department in which chrysotile only was used This contrasts with the 8.9 death rate from mesothelioma reported by Newhouse for workers in a textile factory in which chrysotile crodicolite and amosite were used.18 TABLE Exposure to Certain Materials in Male Control Pairs Exposed to Fibre glass Cement Rubber Copper Nickel Wood Case only SEXTAN SEXTAN SEXTAN SEXTAN SEXTAN SEXTAN Control only 233326 233326 233326 233326 233326 233326 Both 3 7 4 6 I 19 Neither 302 256 300 267 312 200 Relative risk 2.25 1.13 0.60 1.03 1.38 0.98 95 Confidence limits 1.14 4.44 0.73 1.75 0.32 1.14 0.65 1.64 0.68 1.40 0.69 1.40 The association of mesothelioma with exposure to contaminated clothing in the home was again confirmed Cases have not been found in the neighborhood of the Quebec chrysotile mines and mills in our studies or in those of Theriault and Grand but as the number of cases among the workers themselves was few this is not surprising The absence of meso- thelioma in the neighborhood of the New Jersey amosite factory is more reassuring evidence.8 In 1972 our inquiry from all North American pathologists yielded a total of 274 fatal cases of meso- thelioma From detailed analysis of data available for California we concluded elsewhere,,that our survey probably underestimated the true incidence by about 30 On the other hand this is balanced by an almost similar proportion of cases rejected by the pathology panels It has been argued that very intensive questioning is needed preferably while the subject is still alive in order to elicit a full history of asbestos exposure in cases of mesothelioma While this may be true the same argument would apply of course to the controls but perhaps to a lesser degree In casecontrol studies of the type described here the propor- tion attributable to asbestos exposure may therefore be somewhat underestimated The upward trend in in- cidence in the male population seen in Canada is likely to be paralleled in the U. S. A. and if due to the increased use of asbestos will continue to rise for 20 to 30 years at least because of the long latency period The proportion of all cases attributable to asbestos must also therefore be expected to increase Even so the recent estimate from the National Institutes of Healthof a yearly average for the next 30-35 years of 10,000 deaths from mesothelioma in the U. A. due to asbestos alone is not supported by our findings REFERENCES 1. Baris UI Artvinli M Sahin AA Environmental mesothelioma in Turkey Ann NY Acad Sci 1979 423 432 2. Borow M Conston A Livornese LL Schalet N. Mesothelioma following exposure to asbestos a review of 72 cases Chest 1973 641 646 3. Bridport K Decouffle P Fraumeni JF , et al Estimates of the 1656 Cancer October 1 1980 Vol 46 fraction of cancer in the United States related to occupational factors Report prepared by N.C.I. N.I.E.H.S. N.I.O.S.H. 1978 4. Cochrane JC Webster I. Mesothelioma in relation to asbestos fibre exposure review of 70 serial cases S African Med J 1978 279 281 5. Cooper WC Balzer JL Evaluation and control of asbestos exposures in the insulating trade In Proceedings of the International Conference on Biological Effects of Asbestos Dresden April 22 25 1968 Deutches Zentralinstitut f^...r Arbeitzmedizin Berlin DDR 151-160 6. Doll R. Retrospective and prospective studies In Medical Surveys and Clinical Trials L. J. Witts Ed London Oxford Univ Press 1964 pp 84-5 7. Elwood PC Cochrane AL A follow up of workers from an asbestos factory Br J Ind Med 1964 304-307 8. Hammond EC Selikoff IJ Garfinkel L Nicholson WJ Mortal- ity experience of residents in the neighbourhood of an asbestos fac- tory Ann NY Acad Sci 1979 417 422 9. Jones JSP Pooley FD Smith PG Factory populations exposed to crocidolite asbestos continuing survey IARC Scientific Publications INSERM 1976 117 120 10. Jones JSP Smith PG Pooley FD Berry G Sawle GV The consequences of asbestos dust exposure in a wartime gas mask factory IARC Symposium on Biological Effects of Mineral Fibres Lyon Sept 25-27 1979 proceedings in press 11. McDonald AD Mesothelioma registries in identifying as- bestos hazards Ann NY Acad Sci 1979 441 454 12. McDonald AD Harper A El Attar OA McDonald JC Epidemiology of primary malignant tumours in Canada Cancer 1970 914-919 13. McDonald AD McDonald JC Epidemiologic surveillance of mesothelioma in Canada Can Med Assoc J 1973 359 362 14. McDonald AD Magner D Eyssen G. Primary malignant meso- thelial tumours in Canada 1960-1968 Cancer 1973 869 876 15. McDonald AD McDonald JC Mesothelioma in persons ex- posed to crocidolite in gas mask manufacture Environ Res 1978 340-346 16. McDonald JC Exposure relationships and malignant meso- thelioma In Glen HW ed Proceedings of the Asbestos Sym- posium Johannesburg 1977. Randburg National Institute for Metallurgy 67 90 17. McDonald JC McDonald AD Epidemiology of mesothelioma from estimated incidence Prev Med 1977 426 446 18. Newhouse ML Berry G. Predictions of mortality from mesothelial tumours in asbestos factory workers BrJ Ind Med 1976 147 151 19. Rohl AN Langer AM Wolf MS Weisman I. Asbestos exposure during brake lining maintenance and repair Environ Res 1976 12 110 128 20. Selikoff IJ Hammond EC Churg J. Mortality experiences of asbestos insulation workers In Shapiro HA ed Proceedings of the International Conference on Pneumoconiosis Johannesburg 1969. Cape Town Oxford Press 1970 180-186 21. Selikoff IJ Hammond EC Multiple risk factors in environmental cancer In Proceedings of Conference Persons at High Risk of Cancer December 10-12 1974 Key Biscayne Florida 467-483 22. Shephard RA Zeolites in sedimentary deposits of the United States review Molecular sieve zeolites I. RF Gould Ed Adv in Chem Ser 101 Amer Chem Soc 279-310 1971 23. Theriault G Grand L. Mesothelioma and asbestos in the Province of Quebec 1969 7A2rch Environ Health 1978 15-19 24. Weiss W. Mortality of a cohort exposed to chrysotile asbestos J Occupat Med 1977 737 740 25. Zielhuis RL Versteeg JPJ Planteydt HT Pleural meso- thelioma and exposure to asbestos Int Arch Occupat Environ Health 1975 1-18