Mesothelioma Cancer
Risk Assessment of Lung Cancer and Mesothelioma in People Living near Asbestos-Related Factories in TaiwanHo-Yuan Chang INVESTIGATORS HAVE copiously used risk assessment to bridge the intricate relationship between science and policy in the prioritization and control of hazardous materials, especially chronic health hazards and carcinogens.[1] Researchers have implemented four steps before they assess risk: (1) hazard identification, (2) dose-response assessment, (3) exposure assessment, and (4) risk characterization.[2] However, investigators have often questioned the validity of risk estimates in previous studies because there was an incomplete and/or inaccurate assessment of exposure and population at risk, and there was also an uncertain exposure-effect model.[3] Therefore, these difficulties must be resolved if any form of risk assessment can influence poi icy effectively.
Asbestos has been used widely as a raw material in more than 3 000 commercial products.[4] Given that asbestosis, pleural plaques, malignant mesothelioma, and lung cancer are associated with asbestos exposure,[5-7] investigators, during the past four decades, have conducted numerous studies that focused on occupational and nonoccupational exposure to asbestos.[8] Although asbestos concentrations in the general ambient air are relatively low,[9] the potentially greater exposure level to residents who live near asbestos factories is of great importance.
Approximately 30 000 tons of asbestos were consumed annually by the following types of factories in Taiwan:[10] 21 cement, 13 friction (brake production), 3 textile, 2 ground tile, 1 insulation, and 1 refractory. However, the wet-manufacturing process was used by only some of the asbestos cement and refractory factories. The average air concentrations of asbestos in these industries ranged between 2.1 fiber/ml and 6.3 fiber/ml[11]--levels potentially harmful to individuals who live in nearby neighborhoods.
Investigators have already demonstrated that workers in a forging factory[12] and kindergarten children[13] who attended a school located near a lead-recycling factory had significantly increased lead absorption and/or IQ impairment. These results have alarmed individuals who reside near asbestos factories. In addition to a very high population density (i.e., [is greater than] 580 persons/[km.sup.2]), Taiwan has poor zoning regulations between industrial and residential areas. The risk of contracting the two most common types of asbestos-related cancer--lung cancer and mesothelioma--from factory emissions is a critical issue. We therefore sought to determine the risk of lung cancer and mesothelioma among Taiwanese residents who lived near asbestos factories anytime after birth.
Materials and Method
Exposure assessment. We performed samplings in all 41 asbestos factories registered in Taiwan. Investigators visited each factory 1 d prior to sampling to draw schematics of each facility. The Taiwan Environmental Protection Agency, which requested the walk-throughs, asked investigators to also complete a questionnaire that contained information about emission control and waste management. We used satellite maps (1:5 000) of each asbestos factory to locate the sampling sites. The asbestos emission source of each factory was located, and concentric circles with diameters of 200 m, 400 m, and 600 m were plotted. We used a high-volume sampler (Aircheck sampler model 224-PCXR7 [Eighty Four, Pennsylvania]). We collected an 8-h sample at the upwind side--at least 1 000 m from the emission source of each asbestos factory--to ascertain the background level of asbestos exposure in the ambient atmosphere. The background sampling site was chosen on the basis of the following criteria: (a) no asbestos-related factories were within 1 000 m of the sampling site; (b) no observable asbestos material was at the site, including asbestos cement wall, roof, etc.; and (c) traffic intersections were at least 100 m from the site, thus avoiding potential asbestos fiber generated from brake friction. We used mixed cellulose ester filters (pore size = 0.8 [micro]m [Millipore type A]) to collect the asbestos fibers. The flow rate of sampling was set at 3 l/min, and calibration of flow rate was performed both before and after each sampling. We discarded samples if there was a 10% deviation between the flow rate noted before and after each sampling. At least two field blanks were prepared for each factory sampling. We used a cassette sealing band to prevent leakage from the cassette holder. Samples were stored upright in a sealed box, thus avoiding possible contamination during shipment.
We collected a total of 246 samples from 41 factories. Given the manpower and time constraints, we used stratified random sampling to examine only parts of the samples by transmission electronic microscopy (TEM). We followed analytic method number 7402 prescribed by the National Institute of Occupational Safety and Health (NIOSH).[14] The following factories were chosen randomly in each different strata: 5 of 21 cement factories, 3 of 13 friction factories, 2 of 3 textile factories, 2 ground-tile factories, 1 insulation factory, and 1 refractory factory. We measured 69 valid samples obtained from these factories by TEM, and the remainder of the samples were determined with phase-contrast microscopy (PCM).[15] We used TEM (Zeiss D-7082), equipped with an energy-dispersive x-ray analyzer (Kevex, model 3200-0400), and PCM (Olympus, Japan; 10 x 40) to determine airborne asbestos concentrations.
We sent one-tenth of TEM samples to the Environmental Science Laboratory (McGill University, Canada) for cross-check, thus assuring accuracy and precision of our electromicroscopic examinations.
Dose-response model for lung cancer. The dose-response curves for the relationship between asbestos and lung cancer were taken from previous epidemiological studies.[16-20] These curves were based on the following three assumptions: (1) there was no threshold effect, (2) the effect of cumulative exposure remained steady following 10 y of an induction period, and (3) risk was proportional to the product of duration and concentration of asbestos exposure.
Inasmuch as these curves resulted from occupational exposure settings--for which the annual exposure profile is 8 h/d, 5 d/wk, and 48 wk/y--adjustment for residential exposure was necessary. In Taiwan, individuals who lived near the factories were exposed to asbestos 16 h/d, 7 d/wk, for 50 wk/y. Therefore, the conversion factor between occupational and nonoccupational residential exposure was 50 x 16 x 7/(40 x 48) = 2.92. The formula for excess risk of lung cancer attributable to asbestos exposure can be expressed as follows:
D = O - E = E x (b/100) x X x w,
where D = excess deaths from lung cancer; O = observed deaths; E = expected deaths; X = cumulative exposure dose of asbestos (i.e., fibers/y [multiplied by] [cm.sup.3]); w = weighting factor or conversion factor; and b = slope (i.e., fibers/y [multiplied by] [cm.sup.3]), dependent on asbestos factory (asbestos cement: 0.5, asbestos textile = 3). Given the contradictory findings[21,22] and the absence of a model, we did not analyze the elevated risk of lung cancer for friction factories (Fig. 1).
[Figure 1 ILLUSTRATION OMITTED]
Dose-response model for mesothelioma. In human epidemiological studies, investigators have shown that the incidence of mesothelioma might be associated with asbestos fiber types (e.g., crocidolite and amosite cause higher incidence rates than chrysotile). These findings, however, remain inconclusive.[23-26] Moreover, Peto et al.[27] calculated that the incidence rate of mesothelioma in the general population is approximately 1 per million. The cumulative incidence rate of mesothelioma in Taiwan (up to age 74 y)[28,29] was 1.24 x [10.sup.-6]--which was very similar to that reported for mesothelioma in the literature. Given that no substantial difference was found between mesothelioma mortality in Taiwan and mortality reported in the literature, we simply applied the rate to assess the excess numbers of mesothelioma for the individuals in Taiwan. We adopted Peto's model for asbestos-related mesothelioma[30,31]; in that model, mesothelioma is assumed to be independent of smoking and associated to duration since initial exposure (with an order of 3.2-4.0). The model is as follows:
I(t) = K x C ([T.sup.3.2] - [T - D)[sup.3.2])
0 [is less than] T [is less than] D
I(t) = K x C x [T.sup.3.2]
T [is greater than] D,
where I(t) = mesothelioma incidence at time t(per year), C = average asbestos concentration during exposure (fibers/[cm.sup.3]), T = duration since first exposure (year), D = duration of asbestos exposure (year), T [is less than] D = current exposure, T [is greater than] D = ever exposed, and K = constant ([cm.sup.3]/fibers [multiplied by] [y.sup.-4.2]). The constant K varied with type, diameter, and length of asbestos fiber. In Taiwan, asbestos-related factories used chrysotile as their main raw material.[10,11] Crocidolite was found only in an insulation factory, and the factory personnel claimed that the crocidolite was leftover from another factory and was no longer in use. We did not find or identify any crocidolite in the airborne samples in that factory. We used chrysotile in the dose-response estimation, and we, therefore, applied 0.04 x [10.sup.-8] as the K value.
De Klerk et al.[32] reported that the earlier equation could be expressed as follows:
Log I(T) =
Log (K) + 1 Log (C) + 1 Log (D) + 3 Log (T) - 1.5 (D/T).
In 1991, Finkelstein used a conditional logistic-regression model[33] to test this model with real-life mesothelioma cases among asbestos cement workers. He found that the calculated values were 1.3 for the multiplier of Log (D), 3.5 for the multiplier of Log (T), -3.9 for the multiplier of (D/T), and 0.4 for the multiplier of Log (C). In this study, Finkelstein also applied the calculated multipliers to estimate the incidence rate of mesothelioma.
Estimation of population. We estimated the population within our designated concentric circles (i.e., 200-m, 400-m, and 600-m diameters from each asbestos-related factory). We counted the actual number of people who resided in the area on the satellite map, and we verified this information with local policemen and census officers.
Results
In general, asbestos concentrations around asbestos-related factories were low and appeared inversely related with the distance from the factories (Table 1). The considerably large geometric standard deviations (i.e., 1.23-2.77 and 2.22-3.49 for asbestos examined by TEM and PCM, respectively) evidenced unevenly distributed concentrations for the same asbestos-manufacturing process. The concentrations near refractory industries were lower than near other asbestos-related factories (i.e., even below the detection limit of TEM), thus indicating a negligible release of asbestos fiber during the wet, clay-like manufacturing process. In textile and ground-tile factories, there was a higher asbestos exposure than in other asbestos-related factories. Perhaps the dry and relatively open manufacturing process in these two factories contributed to the higher levels of asbestos.
Table 1.--Asbestos Concentrations in Ambient Air around Asbestos Factories in Taiwan
Asbestos concentrations
(fibers/ml)
Distance from factory
200 m
Factory No. Identification
type fac- method GM GSD(*)
tories
Cement 5 TEM
Nonasbestos 0.004 2.403
Asbestos 0.006 1.230
PCM 0.01 3.49 ([dagger])
Friction 3 TEM
Nonasbestos 0.014 2.232
Asbestos 0.008 2.441
PCM 0.01 3.22 ([dagger])
Textile 2 TEM 0.018 2.544
Nonasbestos
Asbestos 0.012 2.221
PCM 0.02 3.21
Ground 2 TEM
tile Nonasbestos 0.005 2.432
Asbestos 0.033 1.412
PCM 0.4 3.21
Insulation 1 TEM
Nonasbestos 0.006 2.665
Asbestos 0.012 2.321
PCM < 0.01
Refractory 1 TEM
Nonasbestos 0.014 2.221
Asbestos < 0.0001
PCM < 0.01
Overall 14 TEM 0.0015 1.943
PCM 0.06 3.29 ([dagger])
Asbestos concentrations
(fibers/ml)
Distance from factory
400 m
Factory
type GM GSD
Cement
0.014 2.327
0.007 1.487
0.01 2.91 ([dagger])
Friction
0.022 2.363
0.008 1.978
0.02 2.88 ([dagger])
Textile
0.072 2.373
0.020 1.432
0.02 3.33
Ground
tile 0.0033 1.873
0.021 1.421
< 0.01
Insulation
0.080 2.591
0.020 2.210
< 0.01
Refractory
0.006 2.342
< 0.0001
< 0.01
Overall 0.0011 2.022
0.01 3.121 ([dagger])
Asbestos concentrations
(fibers/ml)
Distance from factory
600 m
Factory
type GM GSD
Cement
0.0032 3.107
0.006 1.301
< 0.01
Friction
0.014 2.563
0.002 2.221
< 0.01
Textile 0.120 2.057
0.006 1.756
< 0.01
Ground
tile 0.012 3.212
0.025 2.321
0.01 2.21
Insulation
0.006 3.021
0.006 2.773
< 0.01
Refractory
0.018 2.411
< 0.0001
< 0.01
Overall 0.007 2.221
0.01 2.21 ([dagger])
([double dagger])
Notes: TEM = transmission electromicroscopy, PCM = phase contrast microscopy, GM = geometric mean, and GSD = geometric standard deviation. The authors plotted concentric circles around the various factories, the diameters of which were 200 m, 400 m, and 600 m.
(*) The concentrations shown here were corrected with the corresponding concentration of reference sample.
([dagger]) p [is less than] .05, Wilcoxon rank-sum test between asbestos concentrations, by TEM and PCM.
([double dagger]) Detection limits: 0.01 fibers/ml for PCM and 0.0001 fibers/ml for TEM.
The asbestos concentrations obtained from PCM examinations were significantly higher than concentrations obtained from TEM. In the same factory, asbestos concentrations were generally lower than concentrations of nonasbestos fibers. In some of the samples, the sum of the asbestos fiber and nonasbestos fiber-like chemicals in TEM were similar to results obtained from PCM.
Most of the asbestos concentrations at the downwind sites were higher than concentrations at the upwind sites. However, the difference was not significant statistically for most nontextile asbestos factories (Table 2).
Table 2.--Comparison of Asbestos Concentrations between Samplings of Upwind and Downwind Sites
Asbestos concentrations
(fibers/ml)
Distance from factory
200 m
Identification Upwind
Factory type method GM GSD(*)
Cement TEM 0.005 1.130
PCM 0.01 3.22
Friction TEM 0.009 2.271
PCM 0.01 3.32
Textile TEM 0.011 1.821
PCM 0.02 2.71
Ground tile TEM 0.031 1.628
PCM 0.5 2.66
Insulation TEM < 0.013 2.361
PCM < 0.01
Refractory TEM < 0.0001
PCM 0.01
Asbestos concentrations
(fibers/ml)
Distance from factory
200 m 400 m
Downwind Upwind
Factory type GM GSD GM GSD
Cement 0.006 1.332 0.005 1.242
0.01 3.53 0.01 2.44
Friction 0.006 2.601 0.008 2.181
0.01 3.11 0.02 2.98
Textile 0.013 2.45 0.010 1.592
0.01 3.32(*) 0.02 2.55
Ground tile 0.038 1.174 0.021 1.214
0.3 3.32 0.01
Insulation < 0.010 2.162 0.02 2.710
< 0.01 < 0.01
Refractory < 0.0001 < 0.0001
< 0.01 < 0.01
Asbestos concentrations
(fibers/ml)
Distance from factory
400 m 600 m
Downwind Upwind
Factory type GM GSD GM GSD
Cement 0.009 1.613 0.005 1.521
0.02(*) 3.66 < 0.01
Friction 0.008 1.842 0.002 2.241
0.01 2.72 < 0.01
Textile 0.032 1.166(*) 0.005 1.625
0.02 3.11 < 0.01
Ground tile 0.022 1.554 0.017 2.261
< 0.01 0.01 1.82
Insulation 0.02 2.021 0.006 3.110
< 0.01 < 0.01
Refractory < 0.0001 < 0.0001
< 0.01 < 0.01
Asbestos concentrations
(fibers/ml)
Distance from factory
600 m
Downwind
Factory type GM GSD
Cement 0.006 1.277
< 0.01
Friction 0.002 2.032
< 0.01
Textile 0.007 2.123
< 0.01
Ground tile 0.035 1.981(*)
0.01 2.660
Insulation 0.007 2.121
< 0.01
Refractory < 0.0001
< 0.01
Notes: TEM = transmission electromicroscopy, PCM = phase contrast microscopy, GM = geometric mean, and GSD = geometric standard deviation. The authors plotted concentric circles around the various factories, the diameters of which were 200 m, 400 m, and 600 m.
(*) p [is less than] .05, Wilcoxon rank-sum test.
Estimates of excess deaths from lung cancer and incidence cases were calculated (Table 3). Most of the attributable ratios showed that there was no large discrepancy (0-80%) between the estimates of excess numbers of lung cancer and mesothelioma, indicating that selection of sampling site did not influence considerations of exposure assessment in this study.
Table 3.--Excess Deaths from Lung Cancer and Mesothelioma, Estimated via Asbestos Levels at Upwind and Downwind Sites, Measured by Transmission Electromicroscopy
Nos. excess deaths and
attributable ratios
Distance from factory
200 m
Factory Populations AR(*)
type and diseases Upwind Downwind (%)
Cement Population (n) 610.5
([dagger])
Lung cancer 0.074 0.089
Mesothelioma 0.001 0.001 20
Friction Population (n) 313.5
Lung cancer --
Mesothelioma 0.001 0.001 -33
Textile Population (n) 114
Lung cancer 0.182 0.215
Mesothelioma 0.000 0.001 18
Ground Population (n) 15
tile Lung cancer 0.011 0.014
Mesothelioma < 0.001 0.000 23
Insulation Population (n) 10
Lung cancer 0.003 0.002
Mesothelioma < 0.01 < 0.001 23
Refractory Population (n) 32
Lung cancer < 0.0001 < 0.001
Mesothelioma < 0.0001 < 0.001 0
Total Population (n) 1095
Lung cancer 0.270 0.320
Mesothelioma 0.003 0.003 18
Nos. excess deaths and
attributable ratios
Distance from factory
400 m
Factory AR(*)
type Upwind Downwind (%)
Cement 2110.5
0.255 0.459
0.004 0.007 80
Friction 1320
--
0.004 0.004 0
Textile 226
0.328 1.049
0.001 0.003 220
Ground 25
tile 0.013 0.013
< 0.01 < 0.001 5
Insulation 14
0.007 0.01
< 0.01 < 0.001 0
Refractory 131
< 0.001 < 0.001
< 0.01 < 0.001 0
Total 3827
0.603 1.528
0.009 0.014 154
Nos. excess deaths and
attributable ratios
Distance from factory
600 m
Factory
type Upwind Downwind AR(*)
(%)
Cement 4550.5
0.550 0.550
0.009 0.009 0
Friction 4208.5
--
0.003 0.003 0
Textile 480
0.348 0.487
0.001 0.001 40
Ground 1100
tile 0.452 0.931
0.007 0.015 106
Insulation 34
0.005 0.006
< 0.01 < 0.001 17
Refractory 254.5
0.001 < 0.001
< 0.001 0.001 0
Total 10682
1.356 1.974
0.020 0.028 46
(*) AR = attributable ratio = (up/down)/down, up/down = geometric mean concentration of asbestos in upwind divided by that of downwind site.
([dagger]) The authors assumed that the populations were distributed evenly at the upwind and downwind sites.
Asbestos factories in Taiwan were overly allocated in western regions. Population distributions around 41 asbestos factories were also uneven. The populations that lived near the only 3 insulation-type factories were the smallest (i.e., 20, 28, 68, respectively) in the entire study.
The model predicted that living near a textile or cement factory imposed the greatest risk at either an upwind or downwind site (Table 4). The estimate of incidence cases of mesothelioma between two exposure-effect models showed that the difference, in general, was not too large (AR ranges from 0 to 100%), except at a point 600-m away from the friction plant (Table 5). Overall, we could attribute 5 excess deaths from lung cancer and less than 1 death from mesothelioma to the emissions from nearby factories (Table 4).
[TABULAR DATA 4 NOT REPRODUCIBLE IN ASCII]
Table 5.--Number of Mesothelioma Cases Estimated from Two Models
Distance from factory
Plant Peto's Klerk et al's AR(*)
Factories no. model(*) modell([dagger]) (%)
Cement 21 0.003 0.003 0
Friction 13 0.002 0.002 0
Textile 3 0.001 0.001 0
Ground tile 2 < 0.001 < 0.001 0
Insulation 1 < 0.001 < 0.001 0
Refractory 1 < 0.001 < 0.001 0
Total 41 0.006 0.006 0
Peto's Klerk et al's AR(*)
Factories model(*) modell([dagger]) (%)
Cement 0.011 0.012 9
Friction 0.008 0.007 -12
Textile 0.004 0.002 -25
Ground tile < 0.001 < 0.001 0
Insulation < 0.001 < 0.001 0
Refractory < 0.001 < 0.001 0
Total 0.024 0.022 -8.3
Peto's Klerk et al's AR(*)
Factories model(*) modell([dagger]) (%)
Cement 0.021 0.024 14
Friction 0.006 0.018 200
Textile 0.002 0.003 50
Ground tile 0.005 0.006 20
Insulation < 0.001 < 0.001 0
Refractory 0.002 0.001 100
Total 0.035 0.052 49
Notes: The authors plotted concentric circles around the various factories, the diameters of which were 200 m, 400 m, and 600 m.
(*) Peto's model described in reference 29.
([dagger]) Klerk et al's model described in reference 30; this model was modified in the current study in accordance with the multipliers proposed by Finkelstein.(33)
Discussion
Although investigators have considered risk assessment to be a useful tool,[2] one must address the uncertainties and assumptions used during this process to prevent erroneous conclusions and decisions. One of the major problems of risk assessment for asbestos is the determination of asbestos concentrations--especially in nonoccupational settings. In previous investigations, researchers used conventional PCM for asbestos measurement, and they tended to overestimate the real concentration (Table 1). The magnitude of such overestimations seems quite significant when the concentration of nonasbestos fibers in the ambient air is high (i.e., around textile and friction factories) or if the true concentration of asbestos is very low (e.g., observed around a refractory factory). We, therefore, recommend that a more-accurate measurement method, such as TEM (10 000x) equipped with the energy-dispersive x-ray system (TEMEDX), be used to identify and determine the asbestos fiber types and concentrations for all ambient air measurements during risk assessment.
Given the limited manpower and resources, we were unable to examine all samples, thus limiting our ability to accurately determine exposure. Nonetheless, we made efforts to cover all different types of manufacturing processes, and we applied random-sampling procedures simultaneously to select those processes used in more than 1 factory, thus obtaining a reasonably representative sample. Asbestos is a fibrous, curly mineral easily capable of flight; therefore, distance from the pollution source is of great concern. In our study, we determined that concentrations of asbestos were generally inversely related to the distance from the emission source (Table 1). The effect of wind direction might have changed several times, especially on an island such as Taiwan, during the sampling period and might have ameliorated the effects of asbestos exposure. We found different levels of exposure in different types of industries; dry and more-mechanical operations (e.g., textile manufacturing) had higher concentrations of asbestos fibers in their vicinity than other operations (Tables 1 and 2). Therefore, we recommend that in future risk assessments of ambient asbestos exposure, investigators account for different types of asbestos factories and the distance the factories are from the communities.
Investigators have frequently used exposure data, by single measurements, to assess general exposure status. In our study, we showed that risk estimates from geometric and arithmetic means may be substantially different (Table 4). The magnitude of difference may even exceed 2-fold. The data we collected on airborne samples evidenced that asbestos distribution in the same-type manufacturing process is closer to a log-normal curve than a normal curve (not shown). This result indicates that more samples are needed if we are to obtain a representative sample for exposure monitoring of ambient air.
With respect to most risk assessment and environmental epidemiology research, investigators usually find it difficult to obtain an accurate count of the exposed populations. Satellite maps and local police officers assisted us in our counts, which represented the exact numbers of individuals who lived in the concentric-circled areas under study. Therefore, we minimized inaccuracy that may have resulted from miscounting the affected populations.
Selection of an appropriate exposure-effect model is an important issue in risk assessment. In our study, we used equations, in which a linear relationship between cumulative asbestos exposure and health effect was assumed, to estimate excess deaths from lung cancer. We obtained the slopes (potency factors) from seven occupational epidemiological studies, which may not have been suitable for a direct linear extrapolation of the risk to a nonoccupational group.[34,35] The estimates for lung cancer in our study may have been a conservative approximation inasmuch as Sterling et al.[34] reported the possible underestimation of lung cancer if linear extrapolation is applied in the low-exposure level.
To our knowledge, no study has been conducted to date on the dose-response model for both lung cancer and mesothelioma among ethnic Chinese. Thus, in our study, we mainly used models that were generated from Western epidemiological studies. We assumed that carcinogenesis of asbestos exposure was not too different for various ethnic groups. Given the important role of durable nature and fiber size on carcinogenesis, we believe that the above assumption is relatively tenable.
The mesothelioma estimates we calculated from the two exposure-effect models were somewhat similar (Table 5). This similarity indicates that the model selection for mesothelioma is not a critical issue at such low levels of exposure.
In this study, lack of time and resources dictated that we could not consider the possible synergistic effect of smoking. Selikoff et al.[36] found a multiplicative effect between asbestos exposure and smoking on lung cancer, although other investigators found the elevated risk of lung cancer to be lower than the multiplicative relationship.[37] Nonetheless, cigarette smoking and asbestos act synergistically to produce lung cancer; therefore, our estimate of lung-cancer risk may have been low.
Given all the above uncertainties, risk estimates of lung cancer and mesothelioma for residents who lived around asbestos-related factories in Taiwan were approximately 5 and less than 1, respectively (Table 4). More importantly, we also found that the 3 asbestos textile factories appeared to account for more than half of such risks, whereas the ground-tile and cement factories contributed an additional half. In an effort to mitigate the potential hazard, the Environmental Protection Agency of Taiwan advised textile factories to replace asbestos with other less-durable man-made mineral fibers of a larger diameter and to relocate the factories to less-populated areas. Fortunately, 3 of the factories followed this advice 2 y after the study was completed. In addition, enforcement of safety procedures for the handling of asbestos and for the use of only the wet process were also taken, thus minimizing the exposures both inside and outside all factories.
Submitted for publication February 4, 1998; revised; accepted for publication September 17, 1998.
Requests for reprints should be sent to Ho-Yuan Chang, Sc.D., Department of Environmental and Occupational Health, National Cheng Kung University, 38 Sheng-Li Road, Tainan, Taiwan 70428.
This study was supported by the Environmental Protection Agency of Taiwan, EPA 77-005-19-116.
We are indebted to the Environmental Science Laboratory of the McGill University School of Occupational Health for providing the necessary training of personnel and the double-check of asbestos measurement by TEM.
References
[1.] Valazquez SF, Schoeny R, Cogiliano VJ, et al. Cancer risk assessment: historical perspectives, current issues, and future directions. In: Fan AM, Chang LW. Toxicology and Risk Assessment: Principle, Methods, and Applications. New York: Marcel Dekker, 1996; pp 219-43.
[2.] Committee on the Institutional Means for Assessment of Risks to Public Health. National Research Council. Risk Assessment in the Federal Government: Managing the Process. Washington, DC: National Academy Press, 1983.
[3.] Maynard RL, Cameron KM, Fielder R, et al. Setting air quality standards for carcinogens: an alternative to mathematical quantitative risk assessment (discussion paper). Human Exp Toxicol 1995; 14:175-86.
[4.] Michaels L, Chissick SS. Asbestos: Properties, Application and Hazard. Belfast, UK: John Wiley, 1979; p 3.
[5.] Huncharek M. Asbestos and cancer: epidemiological and public health controversies. Cancer Invest 1994; 12:214-22.
[6.] Brody AR. Asbestos-induced lung disease Environ Health Perspect 1993; 100:21-30.
[7.] Council on Scientific Affairs. A physician's guide to asbestos-related disease. JAMA 1984; 250:2593-97.
[8.] Mossman BT, Bignon I, Corn M, et al. Asbestos: scientific development and implications for public policy. Science 1990; 247:294-301.
[9.] Chatfield EJ. Measurement of Asbestos Fibre Concentrations in Ambient Atmospheres. Ontario, Canada: The Royal Commission on Matters of Health and Safety Arising from the Use of Asbestos in Ontario, 1983.
[10.] Chang HY, Wang JD, Chang JW, et al. Industrial hygiene survey for asbestos-related factories in Taiwan. J Natl Public Health Assoc (ROC) 1988; 8(3):161-71.
[11.] Chang HY, Wang JD, Chang JW, et al. Airborne asbestos concentration in asbestos-related factories in Taiwan. J Natl Public Health Assoc (ROC) 1988; 8(1):28-35.
[12.] Chao KY, Wang JD. Increased lead absorption caused by working next to a lead recycling factory. Am J Ind Med 1994; 26:229-35.
[13.] Wang JD, Jang CS, Hwang HY, et al. Lead contamination around a kindergarten near a battery recycling plant. Bull Environ Contam Toxicol 1992; 49:23-30.
[14.] National Institute for Occupational Safety and Health (NIOSH). NIOSH Manual of Analytical Methods. 7402 4th ed. Cincinnati, OH: NIOSH, 1994.
[15.] National Institute for Occupational Safety and Health (NIOSH). NIOSH Manual of Analytical Methods. 7400 4th ed. Cincinnati, OH: NIOSH, 1994.
[16.] Hughes JW, Weill H. Asbestos exposure-quantitative assessment of risk. Am Rev Respir Dis 1986; 133:5-13.
[17.] Weill H, Hughes J, Waggenspack C. Influence of dose fiber type on respiratory malignancy risk in asbestos cement manufacturing. Am Rev Respir Dis 1979; 120:345-54.
[18.] McDonald AD, Fry JS, Woolley AJ, et al. Dust exposure and mortality in an American factory using chrysotile, amosite, and crocidolite in mainly textile manufacture. Br J Ind Med 1983; 40:361-67.
[19.] Henderson VI, Enterline PE. Asbestos exposure factors associated with excess cancer and respiratory disease mortality. Ann NY Acad Sci 1979; 330:117-26.
[20.] McDonald JC, Liddell FDK, Gibbs GW, et al. Dust exposure and mortality in chrysotile mining, 1910-1975. Br J Ind Med 1980; 37:11-24.
[21.] Berry G, Newhouse ML. Mortality of workers manufacturing friction materials using asbestos. Br J Ind Med 1983; 40:1-7.
[22.] McDonald JC. Dust exposure and mortality in an American chrysotile asbestos friction products plant. Br J Ind Med 1984; 41:151-57.
[23.] Timbrell V. The inhalation of fibrous dust. Ann NY Acad Sci 1965; 132:255-73.
[24.] Seidman H, Selikoff IJ, Hammond EC. Short-term asbestos work exposure and long-term observation. Ann NY Acad Sci 1979; 330:61-89.
[25.] McDonald AD, McDonald JC. Epidemiology of malignant mesothelioma. In: Antman K, Aisner J (Eds). Asbestos-Related Malignancy. Orlando, FL: Grure & Stratton, 1987; pp 31-55.
[26.] McDonald AD, McDonald JC. Mesothelioma after crocidolite exposure during gas mask manufacture. Environ Res 1978; 17:340-46.
[27.] Peto J, Heidman H, Selikoff IJ. Mesothelioma mortality in asbestos workers: implication for model of carcinogenesis and risk assessment. Br J Cancer 1982; 45:124-35.
[28.] Department of Health, Executive Yuan, Republic of China. National Cancer Registry Program, 1989-1990. Taipei, Taiwan: Department of Health, Executive Yuan, 1980-1997.
[29.] Rothman KJ. Modern Epidemiology. 1st ed. Boston, MA: Little, Brown and Co., 1986; pp 23-34.
[30.] Becklake Mr. Asbestos-related disease of the lung and pleura. Am Rev Respir Dis 1983; 126:187-94.
[31.] Peto J. Problems in dose-response and risk assessment: the example of asbestos. In: Hoel DG, Merrill RA, Perera FP (Eds). Risk Quantification and Regulatory Policy. New York: Cold Spring Harbor Laboratory, 1985; pp 89-101.
[32.] de Klerk NH, Armstrong BK, Musk AW, et al. Cancer mortality in relation to measures of occupational exposure to crocidolite in Wittenoom Gorge in Western Australia. Br J Ind Med 1989; 46:529-36.
[33.] Finkelstein MM. Analysis of the exposure-response relationship for mesothelioma among asbestos-cement factory workers. Ann NY Acad Sci 1985; 643:85-89.
[34.] Sterling TD, Collett CW, Rosenbaum, WL, et al. Comments on the health effects of institute-asbestos research (HEI-AR) report: "asbestos in public and commercial buildings," with emphasis on risk assessment methods used. Am J Ind Med 1993; 24:767-81.
[35.] Finkelstein MM. Potential pitfall in using cumulative exposure in exposure-response relationships: demonstration and discussion. Am J Ind Med 1995; 28:41-47.
[36.] Selikoff IJ, Hammond EC, Churg J. Asbestos exposure, smoking and neoplasia. JAMA 1968; 204:106-12.
[37.] Berry G, Newhouse ML, Antonis P. Combined effect of asbestos and smoking on mortality from lung cancer and mesothelioma in factory workers. Br J Ind Med 1985; 42:12-18.
HO-YUAN CHANG Department of Environmental and Occupational Health National Cheng Kung University Tainan, Taiwan
CHENG-REN CHEN Department of Internal Medicine National Cheng Kung University Tainan, Taiwan
JUNG-DER WANG Graduate Institute of Occupational Medicine and Industrial Hygiene College of Public Health National Taiwan University Taipei, Taiwan
COPYRIGHT 1999 Heldref Publications
COPYRIGHT 2000 Gale Group
|
