Document xjQQ03knXYbKYx4Zmw77avyLg

\ t . f is. WRChuesl "``f&t *4t? ; Environmental Asbestos Exposure and Cancer Mortality / M. NEUBERCER, M.D. M. KUNDI, Ph.D. Institute of Environmental Hygiene University of Vienna Austria H. P. FRIEDL, M.D. Central Statistical Office Vienna, Austria >4% ABSTRACT. From 1970 to 1980, mortality from cancer of the lung and stomach was ana* lyzed in a town with asbestos deposits and in a town with asbestos processing. The popula tions of the entire country, the province, the district, and of all communities of the same size and agricultural index were used as references. In the town with asbestos contamina tion of air and water from natural tremolite deposits and an endemic occurrence of pleural plaques, no increased risk for lung or stomach cancer was found. In the town with asbestos cement production lung cancer rates were lower in males and higher in females, and stomach cancer rates were lower than expected. No significant differences could be | attributed to environmental asbestos exposure. IN COHORT STUDIES on dust workers' and asbestos until 1945, and today asbestos still is found in the soil of workers,2 evidence for increased mortality from lung the fields and vineyards, in the atmosphere (from and stomach cancer was found. Extrapolations of dose- natural erosion), and in the drinking water. In this com response curves from occupational studies and theoret munity an increased prevalence of pleural plaques was I ical considerations questioning thresholds for carcino genic substances have induced investigations on public detected6; present levels of asbestos exposure from air and water have been reported.28 The other study area health risks resulting from low level asbestos expo is located around the oldest asbestos cement factory in sures.3'5 They focus on lung cancer as the most impor the world, which now processes about 90% of all tant cancer risk from asbestos inhalation. Some consid asbestos imported to Austria. Present atmospheric eration was also given to gastrointestinal cancer in rela asbestos concentrations in the near vicinity of this plant tion to environmental asbestos exposure from drinking have been reported.2 water. We analyzed mortality data from official death certifi cates from 1970 to 1980. Table 1 shows the size of the METHODS populations studied, migration,9 and the minimal relative risks to be detected in the two towns and corre Two study populations in Austria were selected for sponding districts, calculated from expected numbers the investigation of cancer risks from environmental in Austria.10" In the smallest community studied, a asbestos exposure. One is located in an area with twofold or higher risk for lung or stomach cancer must natural asbestos deposits. Tremolite was mined there show up after the observation period of 11 years. luly/Augus! 1984 IVol. 39, (No. 4)] 10002781 Table 1.--Populations under Study and Detectable Relative Risks (see test) Town with asbestos deposits District Province Town with asbestos processing District Province Austria Population 1971 1981 3.412 53.471 272.119 10.627 109.663 1,223.444 7,456.403 3.425 54.172 272.274 11.039 114.378 1,270.426 7,555.338 Migration 1971-1981 + 195 ( + 0.4%) - 1.855 (-0.7%) 553 (*0.5%) + 7.981 (+0.6%) + 73.710 (+1.0%) Minimal Relative Risk to be Detected in 1970-1980 Lung Stomach Cancer Cancer 2.04 2.03 1.21 1.22 1.45 1.49 1.16 1.16 Table 2.--Lung Cancer Mortality (1970-1980) in a Town with Asbestos Deposits and a Town with Asbestos Processing Austria Province District Community Size Community Size and Agric. Index OBS EXP SMR EXP SMR EXP SMR EXP SMR EXP SMR Town with asbestos deposits Male Female Total Town with asbestos processing Male Female Total 9 11.8 76 12.0 75 10.7 84 11.5 76 11.9 76 3 2.9 105 2.0 147 1.5 200 3.8 79 4.1 73 12 14.6 82 14.1 78 12.2 98 15.3 78 16.0 75 30 56.2 53t 53.8 56t 51.8 581 47.0 64* 47.1 64* 11 8.6 128 7.0 158 5.2 211* 7.5 146 7.7 142 41 64.8 631 60.8 67* 57.0 72" 54.5 75 54.9 75 NOTES: Observed deaths (OBS), expected values (EXP), and standard mortality ratio (SMR) calculated from different reference populations. P < .05. Pc .01. fig. 1.Referencepopulationsusedforcalculatingexpected mortality. 262 Standard mortality ratios (SMRs) were calculated'113 on the basis of population census results from 1961, 1971, and 1981 (corrections for births, deaths, and mobility were used for the intervening years), and an nual mortality data, both grouped according to sex and age (5-yr age groups). Expected numbers of deaths from lung and stomach cancer were calculated using five populations as references: (1) the Austrian popula tion, (2) the respective province population to which the study population belongs, (3) the population of the respective district, (4) the subpopulation of Austria liv ing in communities belonging to the same community size class, and (5) the subpopulation living in communi ties with the same agricultural index as the communi- ties under Study (Fig. 1). 10002982 Archives of Environmental Health LUNG CANCER STOMACH CANCER deposits males processing town deposits 28 processing 79 district province Austria females 105 128 town 66 district 64 province Austria Fig. 2. Standard Mortality Ratios: 100 x OBS/EXP (Austria) in 1970-1980 for lung and stomach cancer in town, district and province with asbestos deposits or asbestos processing. Arrows indicate significant differences (P < .05) RESULTS Lung cancer mortality in districts and towns with as bestos deposits or asbestos processing was lower than expected (Table 2). In the town with asbestos deposits, the SMR was 76 in males and 105 in females (Fig. 2). After adjusting for community size and agricultural in dex, both males and females showed lower, but not statistically significant, SMRs (Table 2). In the town with asbestos processing, the SMR was significantly lower in males and insignificantly higher in females. This result remained essentially unchanged after adjusting for community size and agricultural index (Table 2). Stomach cancer mortality for both males and females (Fig. 1) was higher in the districts and lower in the towns with asbestos deposits or asbestos processing (Table 1 and Fig. 1). In the town with asbestos deposits, the SMR (with or without adjustments for community size or agricultural index) was significantly lower in uly/August 1984 [Vol. 39, (No. 4)] males, and in the town with asbestos processing the SMR was significantly lower in females (Table 3). DISCUSSION While we found an increase in risk from cancer of the lung and stomach in highly exposed asbestos workers, '* no evidence of cancer excess from environmental as bestos exposure could be detected in this study. Simi larly, the U.S. Cancer Institute reported no increase of cancer mortality rates in U.S. counties with asbestos deposits, but did not draw definite conclusions from these findings because of lack of exact information on exposure and because of the dilution of exposure in studying the population of counties.'5 Our study covered smaller populations, but was based on better data on exposure and cause of death. In the town with asbestos deposits, measurements in 1975-1979 pro vided evidence that practically all inhabitants there are 10002983 263  ----- r - --~*-- --- .exposed to tremolite asbestos via air and drinking water, so that a dilution of the effect from dilution of ex posure can be disregarded. Because of the small size of this community, however, a small excess in cancer deaths could not be ruled out, and only relative risks greater than 2 (for lung or stomach cancer) would have been detected. As a population risk marker for asbestos cancer we have also looked for the geographical distribution of 161 malignant mesotheliomas in Austria1 and found no clustering of this tumor around asbestos deposits or processing. In a nationwide survey on pleural mesothe lioma in 1969-1974,M'6 no case was detected in the district with asbestos deposits (only 0.9 cases, however, would have been expected there under the assumption of no increased risk). Since then no case of mesotheli oma has been reported from the town with asbestos deposits. In the population of this town, however, we found an increased prevalence of pleural plaques. Pleural plaques have been suggested as risk markers for asbestos cancer.'7 Our findings suggest that low levels of (environmental) asbestos exposure can lead to an endemic occurrence of pleural plaques without in creasing the cancer rates. This hypothesis is consistent with the results of Kiviluoto et at.,'* who found no lung cancer excess in persons with pleural plaques from low level exposure to another type of amphibole asbestos. Only when plaques, combined with lung fibrosis, were detected as a sign of higher (occupational?) exposure, were lung cancer rates increased, too. In the town with asbestos cement processing the out door concentrations have been measured near the vicinity of the plant only.7 A dilution of exposure in the population under study can lead to an underestimation of cancer risk related to environmental asbestos ex posure. Conversely, occupationally exposed persons included in the population under study can lead to an overestimation of this risk. But lung cancer rates in males, though partly occupationally exposed, were lower than expected. Lung cancer rates in females were slightly higher than expected; however, this could be a chance finding and cannot be related to environ mental asbestos exposure (overadditive combined ef fects with smoking should lead to increased lung cancer rates in men earlier than in women). There is evidence that smoking habits for males in the town with asbestos deposits are not different from those in the reference population, since relative fre quencies of male smokers show only marginal varia tions according to estimations from microcensus data: " 41.1% smokers in the general male population, 39.5% smokers in the male population of the province, and 39.7% smokers in the population living in communities with the same size and agricultural index. The variations in the female population seem to be of importance, however, since only 9.1% of female smokers would be estimated from the province popula tion, whereas 16% are inferred from the population liv ing in communities with the same size and agricultural index. Unfortunately, no data are available on smoking habits of the communities under study so that no clear conclusions could be drawn. It might be, however, that the slightly increased lung cancer mortality in females is due to smoking habits. For the town with asbestos processing, estimations of smoking habits according to different stratification cri teria reveal for both males and females only slightly dif ferent values, which are well within the sampling error. The microcensus data on smoking habits in 1972 and 1979 showed no essential differences, but major changes during the past decades which might affect regional lung cancer differences cannot be ignored. Table 3.--Stomach Cancer Mortality (1970-1980) Austria Province District Community Size Community Size and Agric. Index OBS EXP SMR EXP SMR EXP SMR EXP SMR EXP SMR Town with asbestos deposits Male Female Total Town with asbestos processing Mate Female Total 2 7.3 28* 9.1 22* 8.2 24* 9.5 21* 9.5 21* 5 7.6 66 9.9 51 9.4 52 11.1 45 10.4 48 7 14.9 47* 19.0 37t 17.6 40* 20.6 34t 20.0 35t 27 34.2 79 39.7 68* 43.3 62* 32.9 62 32.4 83 14 21.9 64 28.5 49t 30.1 4't 26.9 52* 27.4 51* 41 56.1 73* 68.2 60t 73.4 56t 59.7 69* 59.8 69* NOTES: Observed deaths (OBS), expected values (EXP), and standard mortality ratio (SMR) calculated from different reference populations. P < .05. t P < .01. 10002984 264 Archives of Environmental Health Nutrition as a confounding factor for stomach cancer has changed essentially during the past decades30 and could not be accounted for in this investigation. A bias from population mobility seems to be negligible (Table 1). Occupation as a confounding factor would have been expected to increase, rather than decrease the cancer rates in the towns under study. In the study on pleural mesothelioma'416 in 19691974, no case was detected in the district with asbestos cement processing (1.6 expected); since that time 2 oc cupationally exposed mesotheliomas have been found. This contrasts with the observations in the surroundings of shipbuilding and other industries, where not only oc cupational but also environmental asbestos exposure was related to mesothelioma incidence.3' We conclude that asbestos concentrations not only within these in dustries, but also in their vicinities, must have been higher than in our study (or that occupational and paraoccupational cases in the vicinity of shipyards, etc. could not be distinguished precisely from environmen tal ones). Because we did not find any indication of cancer ex cess from environmental asbestos exposure in our study, we cannot extrapolate this result to other countries with higher levels of en\ironmental asbestos exposure. Recently, Siemiatycki" reported that cancer mortality in the comparably high-exposed general population of the Canadian mining areas was not higher than ex pected. In this study an increased lung cancer rate in men was attributed to occupational exposure rather than to combined effects of smoking and environmen tal exposure. Submitted tor publication May 30. 1983: accepted for publication October 5. 1983. Requests for reprints should be sent to Dr. Manfred Neuberger, In stitute of Environmental Hygiene, University of Vienna, Kinderspitalgasse IS. A-1090 Wien. Austria. REFERENCES 1. Neuberger, M.. Kund' M.: Haider, M.; and Grundorter, W. 1982. Cancer mortalitv of dust workers and controls--results of a prospective study. Presention oi Occupational Cancer. Occupa tional Safety and Health Series 46, pp. 235-41. HO Geneva, Switzerland International Labour Office. 2. Clemmesen, J,, and Hialgrim-Jensen, S. 1981. Cancer incidence among 5686 asbestos-cement workers followed from 1943 through 1976. Ecotoi Environ Safety 5: 15-23. 3. Bundesgesundheitsamt Berlin. 1982. Beurteilung der Krebs gefahren dutch Asbesl. Internat. Symposium. Berlin. FRG (Federal Office of Health). 4. Advisory Committee on Asbestos. 1979. Final Report. Vol. 1. pp. 57-60. London. United-Kingdom: Her Majesty's Stationary Of fice. 5. Zielhuis. R. L.. Ed. 1977. Public Health Risk of Exposure to Asbes tos Report of a working group of experts prepared for the Com mission of the European Communities. Luxembourg: Pergamon Press. 6. Neuberger. M.: Grundorfer, W.; Haider. M.; Komgshofer. R.: Muller. H. W.: Raber, A.: Riedmuller. G.: and Schwaighofer, B. 1978. Endemic pleural plaques and environmental factors. Zbl Bakl Hyg I Abt Orig B 167: 391-404. 7. Felbermayer. W.; Haider. M.: Neuberger, M.; and Ussar, M. 1981. Umweltbelastung durch Asbesl--Forschungsberichte Beitrage Umvveltschutz 2. 1-151. Vienna. Austria: Bundesmimsterium fur Gesundheit und Umvveltschutz 8. Muller, H. W.; Neuberger, M.; Riedmuller. G.; and Schwaighofer, B. 1982. Mineralogical analysis on dusts in connection with endemic pleural calcifications. Staub Remhal Lull 42: 415-18. 9. Osterreichisches Statistisches Zentralamt. 1983. Volkszahlung 1981. Wohnbevolkerung nach Gememden. Beitrage zur Osterreichischen Statistik 630. 1A. 10. Cutler. S.).: Schneiderman, M. A.: and Greenhouse. S. W. 1954. Some statistical considerations in the study of cancer in industry. Am I Public Health 44: 1159-66. 11. Breslow. N. 1975. Analysis of survival data under the propor tional hazards model. Int Statist Review 43: 45-58. 12. Kilpatrick. S. J. 1962. Occupational mortality indices. Population Studies 16: 1975. 13. Holl. A. B. 1971. Principles of medical statistics. 9th ed. England: Oxford University Press. 14. Neuberger. M. 1981. Asbestos: Epidemiological Contributions to Risk Evaluation, pp. 17-56. Vienna, Austria. Facultas. 15. Fears. T. R. 1976. Cancer mortality and asbestos deposits. Am I Epidemiol 104: 523-26. 16. Neuberger, M.; Raber. A.; and Friedl. H. P. 1982. Epidemiologie asbestassoziierter Pieuraerkrankungen in Osterreich. 16. Tagung der Osterreichischen Gesellschaft fur Lungenerkrankungen. pp. 127-130. Vienna. Austria: Hoffmann. 17. Hillerdal. G., and Non, E. 1979. Occupation and bronchial car cinoma. Scand I Respir Dts 60: 76-82. 18. Kiviluoto. R.: Meurman, L. O.: and Hakama. M. 1979. Pleural plaques and neoplasia in Finland. Ann \'Y Acad Science 330: 31-33. 19. Friedl. H. P. 1980. Rauchgewohnheiten der Osterreichischen Bevolkerung. Ergebnisse des Mikrocensus September 1979 Osterr Statist Zentralamt Stahstische \achrichten pp. 9-11, 508-09. 584-86. 642-44 20. Friedl. H. P. 1977. Epidemiologie der Krebskrankheit und Umweltsituation. Osterr Ztschr t Onkologie 2/3: 46-52. 21. Hain. E.; Dalquen, P.; Bohlig. H.: Dabbert. A.; and Hinz, I. 1974. Katamnestische Untersuchungen zur Genese des Mesothelioms. Int Arch Arbeilsmed 33: 15-37. 22. Siemiatycki,). 1982. Mortality in the general population in asbes tos mining areas. World Symposium on Asbestos, Session 5, Mon treal. Canada. luly/August 1984 |Vol. 39, (No. 4)] 10002985 265