Annals of Occupational Hygiene Advance Access originally published online on February 3, 2009
Annals of Occupational Hygiene 2009 53(2):139-151; doi:10.1093/annhyg/men082
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Retrospective Exposure Assessment for Carcinogenic Agents in Bitumen Waterproofing Industry in Finland and Denmark
1 Finnish Institute of Occupational Health, Topeliuksenkatu 41 A, 00250 Helsinki, Finland
2 Department of Environmental and Occupational Medicine, School of Public Health, Aarhus University, Bartholins Allé 2, bg. 1260, 8000 Århus C, Denmark
* Author to whom correspondence should be addressed. Tel: +358-30-474-8640; fax: +358-30-474-8615; e-mail: eero.priha{at}ttl.fi
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Objectives: The purpose of the study was (i) to identify the carcinogenic agents that may cause confounding when studying the exposure–response relationship between bitumen fume exposure and cancer among roofing membrane-manufacturing workers and roofers and (ii) to assess exposures to the identified carcinogens and bitumen fume in roofing membrane manufacturing and roofing in Finland and Denmark from 1950 to 2005.
Methods: Information on the use of carcinogenic agents and other relevant data were collected through semi-structured interviews of senior employees in the industry. Semi-quantitative exposure assessments were made on the basis of available measurement data and information obtained from the interviews and literature.
Results: Most of the production line workers in roofing membrane plants in Finland were exposed to asbestos until the mid-1970s. Also, some of the mixer operators in the plants were exposed to asbestos in Finland during the 1970s and in Denmark from the mid-1960s to the mid-1980s. In both countries, coal tar pitch was used in roofing membrane manufacturing until the mid-1960s, and consequently, exposure to polycyclic aromatic hydrocarbons (PAHs) in the plants was high in the 1950s and still significant in the early 1960s. Exposure of production line workers to quartz dust was high until the 1980s and is still relatively high compared with current occupational exposure limit values. Bitumen roofers' exposure to coal tar-derived PAHs may have been significant in both countries until the end of 1960s. Roofers' exposure to asbestos and quartz was estimated to have been near background level.
Conclusions: The estimated average annual exposures to asbestos, coal tar-derived PAHs and quartz dust in the bitumen waterproofing industry in Finland and Denmark were significant in the past but have a clear declining trend. Exposure to bitumen fume was found to follow a similar trend.
Keywords: asbestos • bitumen • coal tar • crystalline silica • exposure assessment • roofing • roofing membrane manufacturing • waterproofing
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
The potential carcinogenicity of bitumen fume has been investigated in several epidemiological studies (e.g. Hansen, 1989; Engholm et al., 1991; Stern et al., 2000; Boffetta et al., 2003a,b). Some of the studies suggested an increased risk of lung cancer and other cancers in workers exposed to bitumen fume. The results were, however, inconsistent.
In a meta-analysis of the most informative studies published from the 1970s to the 1990s, an increased risk of lung and stomach cancers was suggested among roofers exposed to bitumen fume (Partanen and Boffetta, 1994). However, most of the studies included in the analysis failed to adjust for confounding by other occupational and non-occupational exposures. In particular, the uncontrolled exposure to coal tar pitch volatiles in roofing activities complicated the interpretation of the results.
Recently, the International Agency for Research on Cancer (IARC) assembled a large multicentre cohort study to clarify the association between bitumen fume exposure and cancer (Boffetta et al., 2003a,b). The study showed that mortality from lung cancer was slightly increased among bitumen workers in comparison to workers in ground and building construction (Boffetta et al., 2003a). In a further, job-exposure matrix-based analysis, lung cancer mortality among workers exposed to bitumen fume was comparable to that of non-exposed workers (Boffetta et al., 2003b). However, among road pavers, a dose–response for lung cancer was suggested for average level of bitumen fume exposure when applying a 15-year lag. Similar results were gained for head and neck cancer. Currently, the IARC is conducting an international case–control study nested in a cohort of European bitumen workers.
The objectives of the present study were (i) to identify the carcinogenic agents that may cause confounding when studying exposure–response relationship between bitumen fume exposure and cancer among roofing membrane-manufacturing workers and roofers and (ii) to assess exposures to the identified carcinogens and bitumen fume in roofing membrane manufacturing and roofing in Finland and Denmark from 1950 to 2005.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
The study design consists of four elements: (i) identification of the carcinogenic agents used in bitumen waterproofing industry, (ii) collection of information on the application of these agents in Finland and Denmark from 1950 to 2005 through interviews and literature search, (iii) collection of exposure measurement data and, finally, (iv) assessment of exposure on the basis of measurement data and information obtained from the interviews and literature.
Identification of carcinogenic agents
Identification of the carcinogenic agents used in the industry was based on preliminary interviews of roofing membrane manufacturers in Finland and on previously published data. Of the identified agents, only those classified by the IARC as known human carcinogens (Group 1) or as substances probably carcinogenic to humans (Group 2A) were included in the exposure assessment. Particular interest was on agents increasing the risk of respiratory tract and lung cancers.
Interviews and literature search
Information on the use of the agents, production conditions, work methods and other relevant issues was collected through semi-structured interviews of senior employees in roofing membrane plants and contracting companies (Table 1). In Finland, two or three senior or retired employees from each of the three existing roofing membrane plants and five senior employees from the roofing field were interviewed, either individually or together. The interviewees (n = 12) were 42–68 years of age and had 20–45 years of experience in the field. In addition, a questionnaire with similar content was sent to seven smaller, long-running roofing companies.
|
In Denmark, two or three senior or retired employees from the two existing roofing membrane plants were interviewed for information on both the plant itself and the roofing contractors associated with the plant. Additionally, an owner of a smaller roofing company was interviewed. The ages of the interviewees (n = 6) were between 45 and 82 years and they had 17–38 years of experience in the field.
To confirm and supplement the information obtained from the interviews, we examined documents such as company histories, constructional histories, handbooks, standards and national regulations.
Collection of measurement data
Available exposure measurement data were gathered from the Register of Occupational Hygiene Measurements of the Finnish Institute of Occupational Health, the measurement database of the National Research Centre for the Working Environment, Denmark, and from the companies and research reports. In order to assess the intensity and determinants of exposure, a database containing all of the compiled data was created. The variables included were measurement year, measured concentration, sample type (personal/area), sampling site or task, company, country and description of the working or production conditions during the measurement.
In addition to measurement data from Finland and Denmark, measurement data from other countries, most importantly from the US, were collected from several research reports and articles. The data from other countries were primarily applied to the assessment of exposure to coal tar pitch volatiles, for which data from Finland and Denmark was lacking.
Exposure assessment
Inhalation exposure to bitumen fume, asbestos, respirable quartz and coal tar pitch volatiles [benzo[a]pyrene (BaP) as an indicator compound] was assessed semi-quantitatively by a panel of four industrial hygienists on the basis of available measurement data and information obtained from the interviews and literature. The mean annual work-related exposure levels were assessed separately for roofing membrane-manufacturing workers and for roofers in 5-year periods from 1950 to 2005.
The collected measurement data was applied to create task-related exposure estimates, i.e. to describe exposure levels in particular tasks (e.g. roofing by pour and roll method) in particular time periods. For the tasks and time periods for which measurement data were available, the arithmetic mean of the data was used as task-related exposure estimate. Task-related exposure levels for time periods for which no measurement data were available were estimated on the basis of the available measurement data for other time periods and information on the timing of important changes made in production conditions and work methods obtained in the interviews.
The mean annual work-related exposure levels were calculated on the basis of the task-related exposure estimates and proportion of working time used for each task in the time period in question. Data on the average proportion of annual working time used for different tasks in different time periods were collected in the interviews. Also, the percentage of exposed workers was estimated for each agent on the basis of information obtained from the interviews.
Statistical modelling was feasible for respirable quartz dust, for which sufficient amount of measurement data was available. The effects of the potential exposure determinants (e.g. company, decade and chippings type) on the exposure level in two Finnish roofing membrane plants were studied with linear mixed-effect models. The potential exposure determinants were treated as fixed effects in the models. The measurement data were best described by log-normal distribution, and, therefore, natural logarithms of the measurement results (n = 105) were used in the analysis. The level of statistical significance was set at P <0.05 (two tailed). All statistical procedures were performed with SAS software, version 9.1 (SAS Institute, Cary, NC, USA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
The carcinogenic agents identified in the bitumen waterproofing industry and their main exposure routes are listed in Table 2. The use of the identified agents in the industry in Finland and Denmark is summarized in Fig. 1. It should be noted, however, that the use of the agents has varied from one company to another. For example, in some roofing membrane plants, asbestos has been used solely as filler in coating bitumen mixtures, whereas in others solely as surfacing material.
|
|
In the following sections, we present exposure measurement data and results of the semi-quantitave exposure assessment for bitumen fume, asbestos, coal tar pitch volatiles and respirable quartz. In addition, we discuss exposures to formaldehyde, diesel exhaust and benzene.
Bitumen fume
Measurement data.
A summary of the available data on occupational hygiene measurements of bitumen fume in roofing membrane plants and on roofing/waterproofing sites in Finland and Denmark is presented in Table 3. Bitumen fume was measured as the benzene-, cyclohexane- or tetrachloromethane-soluble fraction of total particulates. Most of the data on roofing are from personal (breathing zone) measurements, whereas the majority of the data on roofing membrane manufacturing are from area measurements.
|
Exposure to bitumen fume: Finland.
The air concentration of bitumen fume in roofing membrane plants in Finland was relatively high, on the order of 2 mg m–3, until the early 1980s. The fume level fell significantly when the use of organic base materials requiring bitumen impregnation ceased in the early 1980s. Since the 1990s, the bitumen fume level was in the range of 0.2 mg m–3. The proportion of working time that production line workers were exposed to bitumen fume was estimated at 70% until the mid-1960s and at almost 100% from the mid-1960s onwards. On the basis of these data, the mean annual work-related exposure levels of production line workers were estimated to have been high (1–2 mg m–3) until the end of the 1970s, medium (0.3–1 mg m–3) during the 1980s and low (0.05–0.3 mg m–3) since the beginning of the 1990s (Table 4).
|
In bitumen roofing, a gradual decrease in the use of the traditional pour and roll roofing method and a corresponding increase in roofing with torch-on membranes occurred since the 1970s. The average bitumen fume level in pour and roll roofing was estimated to have declined from 1.5 mg m–3 in the 1950s to 0.5 mg m–3 in the 1990s, mainly due to improvements in the temperature control of bitumen kettles. In torching, the bitumen fume level is on the order of 0.2 mg m–3.
Waterproofing indoors with hot bitumen was common in Finland from the 1950s until the end of the 1970s. In interior waterproofing, bitumen fume exposure was high, especially if ventilation was insufficient; fume concentrations ranging from 0.8 to 28 mg m–3 were reported in two studies from the late 1970s (Ahonen et al., 1977; Priha et al., 1980).
Bitumen workers' mean annual work-related exposure levels to bitumen fumes were estimated to have been high (1–2 mg m–3) from the 1950s to the 1970s and to have decreased considerably during the 1980s due to the cessation of indoor waterproofing and decrease in the use of hot bitumen in roofing (Table 4).
Exposure to bitumen fume: Denmark.
Only limited measurement data were available from the Danish roofing membrane plants and, therefore, the bitumen fume levels in the plants were estimated mainly on the basis of the Finnish data. In Denmark, the use of organic base materials ceased in the 1970s, most probably resulting in a similar decline in bitumen fume concentrations as seen in Finnish plants in the 1980s. Since the 1980s, the improvements in control measures have further reduced the exposure (Table 4).
In bitumen roofing, the shift from pour and roll roofing to roofing with torch-on membranes since the 1970s was more thorough than in Finland. The bitumen fume exposures in pour and roll roofing and in torching were presumed to have been on the same level as in Finland. This assumption was supported by the available measurement data. As indoor waterproofing with hot bitumen was rare in Denmark, it was excluded from the evaluation.
The mean annual work-related exposure levels of Danish bitumen roofers to bitumen fumes were estimated to have been medium (0.3–1 mg m–3) until the end of 1970s and to have decreased to low (0.05–0.3 mg m–3) in the 1980s (Table 4).
Asbestos
Measurement data.
Data on occupational hygiene measurements of asbestos are presented in Table 5. The measurement method consisted of sample collection on membrane filter and fibre count by light microscopy. Most of the data are from area measurements.
|
The measurements conducted in roofing membrane plants were related to the preparation of an asbestos-containing coating bitumen mixture and to the surfacing of roofing membranes with talc, which contained tremolite asbestos as a contaminant. Although the manufacturing of bitumen adhesives and paints was excluded from the study, measurement data on asbestos-containing bitumen adhesive manufacturing was utilized in the exposure assessment for workers in roofing membrane manufacturing.
Exposure to asbestos: Finland.
Asbestos was used as roofing membrane surfacing material in Finland until the mid-1970s. In the plants where asbestos was used, the proportion of asbestos-surfaced membranes ranged from 5 to 30% of the total production. The mean proportion of working time during which the production line workers were exposed to asbestos was estimated at 15%. During asbestos surfacing, the air concentration of asbestos fibres was probably high, especially near the chippings applicator at the middle of the production line. As measurement data on asbestos surfacing were not available, the mean air concentration of asbestos during asbestos surfacing (5 f cm–3) was estimated based on measurement data from other industries (Riala, 1991).
Talc containing a small amount of tremolite asbestos as contaminant was used for roofing membrane surfacing from the mid-1970s to the mid-1980s. The mean proportion of working time during which the production line workers were exposed to asbestos-containing talc was some 20%. The talc contained 5% of fibres, of which only 2% were asbestos fibres, making the air concentration of asbestos during the surfacing <0.1 f cm–3.
Additionally, asbestos was used as filler in coating bitumen mixtures in roofing membrane manufacturing in the 1970s. According to the sparse measurement results available, the exposure of mixer operators during the adding of asbestos was high on the order of 2 f cm–3, and the background level in the mixing area was 0.2 f cm–3. Exposure of production line workers was, on the other hand, insignificant as the asbestos was bound to the bitumen mixture.
Overall, the mean annual exposure level with respect to airborne asbestos for exposed production line workers was estimated to have been high, >0.5 f cm–3, until the mid-1970s (Table 6). From the mid-1970s to the mid-1980s, when only asbestos-containing talc was applied, the level of exposure was estimated as low, in the range of 0.01 f cm–3. For exposed mixer operators, the mean annual work-related exposure level was estimated to have been medium (0.1–0.5 f cm–3) during the 1970s.
|
Bitumen roofers have been exposed to asbestos mainly in occasional demolition of asbestos–cement roof slates and, to a lesser extent, in the mounting and demolition of asbestos-containing roofing membranes. Based on measurement data, the air concentration of asbestos fibres in careful demolition of asbestos–cement roofs was on the order of 0.1 f cm–3. Asbestos exposure in the installation and demolition of asbestos-containing roofing membranes was estimated to be two times lower. As the proportion of total working time during which bitumen roofers were in contact with asbestos-containing products was only a few percent, the mean annual work-related exposure levels to asbestos were estimated at <0.01 f cm–3 from the 1950s to the 2000s (Table 6).
Exposure to asbestos: Denmark.
According to our interviewees, asbestos was not used as roofing membrane surfacing material in Denmark. Yet, asbestos was used as filler in coating bitumen from the mid-1960s to the mid-1980s. In the 1980s, however, the production of asbestos-containing roofing membranes was very small.
Talc was commonly used for roofing membrane surfacing in Denmark from the mid-1960s to the mid-1980s. Some of the talc used in the industry may have contained asbestos as a contaminant, depending on the source of the mineral. As the talc's asbestos content was not known, the asbestos content of the Swedish talc used in Finland in the 1970s was chosen as the best estimate, and exposure assessment was based on the available measurement data from Finland.
The mean annual work-related exposure of production line workers to airborne asbestos was estimated as low, in order of 0.01–0.1 f cm–3, from the mid-1960s to the mid-1980s, when talc was used as surfacing material, and ceased in the mid-1980s (Table 6). For the mixer operators, the exposure levels of exposed workers were estimated to have been medium (0.1–0.5 f cm–3) from the mid-1960s to the late 1970s and to have decreased to low (0.01–0.1 f cm–3) in the beginning of 1980s and ceased in the mid-1980s.
As in Finland, the bitumen roofers in Denmark have been exposed to asbestos mainly in the occasional demolition work of old asbestos–cement roofs. As the proportion of the total working time during which bitumen roofers were in contact with asbestos-containing products was only a few percent, the mean annual exposure to asbestos was estimated to be near background level from the 1950s to the 2000s (Table 6).
Coal tar pitch volatiles
Measurement data.
Data on occupational hygiene measurements of coal tar pitch volatiles measured as the benzene- or cyclohexane-soluble fraction of total particulates are presented in Table 7. Measurement data on BaP, the indicator compound for polycyclic aromatic hydrocarbons (PAHs) in coal tar pitch volatiles, are presented in Table 8. In the case of demolition work, the given concentrations refer to the benzene- or cyclohexane-soluble fraction of the dust formed during demolition work and to the concentration of BaP in the dust. Both personal and area measurements are included in the data set.
|
|
Exposure to coal tar pitch volatiles: Finland.
Coal tar pitch was used in roofing membrane manufacturing in Finland until the mid-1960s. In the 1950s, coal tar pitch containing roofing membranes (referred to as tar felts) constituted 30% of the total production of roofing membranes of the plants producing tar felts. At the end of the 1950s, the production of roofing membranes containing coal tar pitch dropped rapidly, and since the beginning of the 1960s, tar felts constituted only 2% of the total production of the Finnish plants.
No measurement data are available on exposure to coal tar pitch volatiles in roofing membrane plants in Finland. The BaP exposure during tar felt manufacturing was estimated to have been in order of 5 µg m–3, based on description of working conditions during tar felt manufacturing and on the available measurement data from the US on roofing with coal tar pitch. Based on the above information, the mean annual exposure level of exposed production line workers to coal tar-derived BaP was estimated to have been high (1–10 µg m–3) until the end of the 1950s, medium (0.1–1 µg m–3) until the mid-1960s and near background level since the mid-1960s (Table 9).
|
In roofing, the mounting of coal tar-based roofing membranes was estimated to have constituted <10% of the total working time in the 1950s. Tar felts were rarely used during the 1960s, and their use was discontinued in the 1970s. According to the interviews and literature, tar felts were usually fastened by nailing or glued with hot bitumen. Although no information on the fastening of tar felts with hot coal tar pitch was obtained, it is possible that hot pitch was occasionally used. Demolition of old tar felt roofs was estimated to have constituted a few percent of roofers' total working time in the 1950s and <1% since the 1970s.
Measurement data on the fastening of tar felts with hot bitumen were not available. Although the warming-up of tar felts during the work may have caused release of vapours and fumes containing PAHs, exposure to BaP was estimated to have been low. Similarly, BaP exposure in the installation of a new roofing membrane roof over an old tar felt roof was presumed to be low. When tar felts were fastened by nailing, the BaP exposure was presumed to have been negligible. However, if hot coal tar pitch was used for the fastening of the felts, the exposure may have been high. In demolition work, the exposure to BaP in the forming dust was estimated to be in order of 1 µg m–3.
For waterproofing, it seems that bitumen was preferred to coal tar pitch in Finland already in the 1940s. The interviewees, whose careers had begun in the 1950s or 1960s, confirmed that coal tar pitch was not used for waterproofing during their careers. However, liquid coal tar was used by some of the contractors as primer in bitumen waterproofing even in the 1970s.
Based on the sparse data on the use of coal tar pitch in the 1950s, the mean annual BaP exposure level of the roofers exposed to coal tar-based materials was estimated to have been from medium to high (0.1–10 µg m–3) (Table 9). In the mid-1960s, the exposure level was estimated to have been medium, in order of 0.1–1 µg m–3, due to the occasional mounting of tar felts and demolition of old tar felt roofs and to have decreased to low (0.01–0.1 µg m–3) by the beginning of 1970s. The proportion of roofers exposed to coal tar pitch fumes or dust, for instance during occasional demolition work, was estimated to have dropped from 75% in the 1950s to <5% from the 1980s onwards.
Exposure to coal tar pitch volatiles: Denmark.
In Denmark, coal tar pitch was used for roofing membrane manufacturing until 1967. In the 1950s, tar felts constituted 
30%, and in the 1960s 10%, of the annual production of roofing membranes. Based on these data and the estimated BaP level of 5 µg m–3 during tar felt manufacturing, the mean annual exposure levels of production line workers to BaP were estimated to have been high (1–10 µg m–3) until the mid-1960s, medium (0.1–1 µg m–3) until the end of the 1960s and near background level since the 1970s (Table 9).
In roofing, the mounting of coal tar-based roofing felts was estimated to have constituted 10% of the total working time in the 1950s and <5% in the 1960s. According to the interviews, tar felts were usually fastened by nailing. In addition, bituminous roofing membranes containing a few percent of coal tar pitch were used in build-up roofing until the end of the 1970s. Mounting of coal tar pitch containing bitumen membranes was estimated to have constituted 15% of the total working time in the 1950s and to have declined to 5% in the 1970s. As in Finland, demolition of old tar felt roofs was estimated to have constituted a few percent of the roofers’ total working time in the 1950s and <1% since the 1970s.
Measurement data on roofing with coal tar pitch containing bitumen membranes were not available. Although warming-up of the membranes during the work may have released PAH-containing vapours and fumes, exposure to BaP was estimated to have been low. In the mounting of tar felts by nailing, BaP exposure was presumed to have been negligible. If hot coal tar pitch has been used for the fastening of the felts, the exposure may have been high.
Overall, the mean annual work-related exposure levels of Danish bitumen roofers to BaP were estimated to have been medium to high (0.1–10 µg m–3) in the 1950s, medium (0.1–1 mg m–3) in the 1960s and low (0.01–0.1 µg m–3) from the 1970s onwards (Table 9). The proportion of roofers exposed to coal tar pitch fumes or dust was estimated to have declined from 75% in the 1950s to <5% since the 1980s.
Crystalline silica
Measurement data.
A summary of data on occupational hygiene measurements of respirable quartz dust in roofing membrane plants is presented in Table 10. The respirable fraction was separated from filter-collected total particulates by liquid sedimentation, and the quartz content of the fraction was determined by X-ray diffraction. Approximately, one-third of the data are from personal measurements and two-thirds from area measurements. In the interpretation of the data, it should be taken into account that the measurement technique applied in Finland differs from the commonly applied technique based on cyclone separation. The liquid sedimentation separation technique results in concentrations that are 2-fold greater than the concentrations obtained by cyclone separation.
|
Around 10% of the quartz dust measurement results from the 1970s and 1980s were outstandingly high (>2 mg m–3). These results were usually related to exceptional situations, such as malfunction of local exhaust ventilation during sample collection, and may be slightly over-represented in the data.
Exposure to crystalline silica.
Quartz sand has been used for roofing membrane surfacing in all plants in Finland and Denmark continuously from the 1950s to the 2000s. In addition, quartz-rich slates have been frequently used as surfacing material in the plants. As quartz sand was in use continuously on at least one production line in each plant, the proportion of working time production line workers were exposed to quartz dust was estimated at nearly 100%.
On the basis of abundant measurement data from Finnish plants, the mean annual level of exposure to respirable quartz dust among the production line workers was estimated to be very high (>0.5 mg m–3) until the mid-1980s (Table 11). By the mid-1980s, improvements in dust control and work methods reduced the mean exposure levels to 0.1–0.5 mg m–3. Based on the interviews, the exposure levels in Denmark may have been somewhat lower than in Finland. However, as measurement data from Danish plants were not available, the exposure level in Denmark was estimated to have been the same as in Finland (Table 11).
|
No statistically significant differences in respirable quartz dust levels between the two Finnish plants were found in the statistical analysis (Table 12). It should be noted, however, that the data from the companies are from different years which is likely to affect the results. Generally, the respirable quartz levels in the plants were considerably lower in 1990–2005 compared to 1970–1980. Also, the levels were found to be different in different parts of the production line. The type of chippings was also shown to contribute to the quartz dust levels; most importantly, the levels were significantly higher when unwashed quartz sand was used compared to washed quartz sand currently in use.
|
In roofing and waterproofing, exposure to quartz dust was estimated as insignificant. Some of the roofers may have been exposed to quartz dust in bridge waterproofing, during which quartz sand is occasionally used for sandblasting and in epoxy sealing work. Bridge waterproofing was, however, excluded from the assessment due to the rareness of the exposing work phases.
Other agents
Formaldehyde measurements were carried out in one roofing membrane plant in Finland in 2003. The mean formaldehyde concentration in the plant was 0.04 mg m–3, with a range of 0.02–0.05 mg m–3 (n = 4). In a prior Finnish study on roofing work with torch-on roofing membranes, formaldehyde concentration was found to be below the detection limit of 0.05 mg m–3 in all measurements (n = 4) (Degerth et al., 1991). In one occupational hygiene measurement report form a Finnish roofing membrane plant in the early 1970s, a pungent smell of formaldehyde was reported when a particular glass fibre fabric was used as base material for roofing membranes. The air concentration of formaldehyde was not measured, but in laboratory analysis, the glass fibre base was found to contain 0.26 mg g–1 of free formaldehyde. One similar case was reported in Denmark during the interviews.
Based on these observations, formaldehyde concentrations may occasionally have been high in the roofing membrane plants before the 1980s, when urea–formaldehyde resin-containing glass fibre fabrics were used as base material in the roofing membranes. However, as measurement data and detailed data on the composition of the fabrics are lacking, the mean annual exposure levels of the production line workers could not be evaluated. Currently, phenol–formaldehyde resin, which releases very little formaldehyde, is mainly used as a binding agent in the glass fibre fabrics. Based on the sparse measurement data from the 1990s and the 2000s, formaldehyde exposure in the industry today is insignificant.
Exposure to diesel exhaust was found to be insignificant both in roofing membrane manufacturing and in roofing, as diesel-powered equipments were not used in the plants or in roofing work.
Benzene was used as diluent for coal tar pitch at least in one Danish roofing membrane plant until the mid-1960. The amount of benzene used, annual length of work-related exposure and the proportion of companies that used benzene could not be traced, and, consequently, semi-quantitative exposure estimation for benzene could not be carried out. According to the information received, benzene was not used in Finnish roofing membrane plants.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Not many exposure studies have been conducted on roofing membrane manufacturing and bitumen roofing and waterproofing, and very few of them have dealt with other exposures than bitumen fume (Priha et al., 1980; ABSC, 1983; Degerth et al., 1991; SiD, 1998; NIOSH, 2001, 2006; Burstyn et al., 2003; McClean et al., 2003; Rühl et al., 2006). In the present study, the focus was on a retrospective exposure assessment of carcinogenic agents in roofing membrane manufacturing and roofing in Finland and Denmark from 1950 to 2005.
The information on production conditions, work methods and materials applied in the industry during the time of the exposure assessment is mainly based on responses from the five existing roofing membrane plants in Finland and Denmark, contracting companies associated with these plants and a few smaller contracting companies. How representative these companies are is not known and already closed roofing membrane plants cannot be validated. However, the persons who were selected as interviewees had decades of working experience in the field and had followed the development of the field closely for a number of years. Also, many of the interviewees had been employed by more than one company during their career.
For the data collection, we preferred semi-structured interviews to a questionnaire in order to collect all relevant information, including previously unknown viewpoints, to avoid misinterpretation of the questions and to enable us to expand on the answers with additional questions when needed. This approach was found very useful: our conception of the use of carcinogenic agents in the industry deepened during the interviews, and we were able to focus and expand on the most relevant issues.
Exposure measurement data on roofing membrane manufacturing and roofing were sparse both in Finland and Denmark. The estimated average exposure levels were, therefore, based both on available measurement data and on the determinants of exposure levels, such as technological changes in the plants. For the measurement data, the sampling strategy may have caused bias in the results: in many cases, the purpose of air monitoring at the workplace was not known. Thus, part of the measurement results may not be representative of the actual exposure level. Furthermore, due to the scarcity of the data, the exposure assessments needed to be based on combined personal and area measurement data, although measurement results based on area sampling are generally expected to be somewhat higher than those based on personal sampling.
Statistical analysis of measurement data on respirable quartz dust showed that the workers' exposure in roofing membrane plants is not equal in different parts of the production line. The air concentrations are high at sites where raw materials are mixed or applied: for example, quartz concentration is highest at the site where it is scattered on the surface of the membrane. However, due to the varying tasks of the workers and the fact that the production lines are located in an open hall, the exposure intensities of all the production line workers were considered to be similar in the exposure assessment.
Asbestos, coal tar pitch and crystalline silica (quartz) were identified as important occupational carcinogens in the industry with respect to cancers of the respiratory tract. All these agents are classified as known human carcinogens (Group 1) by the IARC. Asbestos was applied in roofing membrane manufacturing in Finland and Denmark before the mid-1980s, and the exposure of manufacturing workers was particularly high in Finland, where asbestos was used as roofing membrane surfacing material until the mid-1970s. The four diagnosed cases of occupational asbestos disease among Finnish roofing membrane-manufacturing workers in 1993–2002, recorded in the Finnish Register of Occupational Diseases, confirm that exposure to asbestos in the plants has been significant in the past.
The most uncertain estimates of the exposure variables are the average proportion of working time the roofing membrane-manufacturing workers and roofers were exposed to coal tar pitch volatiles in the 1950s and early 1960s and the actual exposure levels. We have information confirming that coal tar-based roofing membranes were produced both in Finland and in Denmark until the mid-1960s, but accuracy of the estimated proportion of annual working time manufacturing workers and roofers worked with coal tar pitch may be poor. As measurement data from Finland and Denmark were lacking, estimates on the exposure levels were based mainly on sparse data from the US. Also, as a consequence of poor hygienic conditions, dermal and hand-to-mouth exposure of PAHs was probably significant when coal tar and coal tar pitch were handled. The level of exposure by these routes could not, however, be evaluated as measurement data are lacking.
Respirable quartz dust levels in roofing membrane plants have decreased in the last decades but in comparison to current occupational exposure limit values, the levels are still high. During the past years, the increased risk of lung cancer from quartz dust has become more evident (IARC, 1997). Therefore, exposure limits of respirable quartz have been lowered in many countries from 0.2 to 0.05 mg m–3. Also, according a new EU agreement, exposure to respirable quartz dust should be controlled by the industry (EU, 2006). Unfortunately, the sampling and analysis techniques for crystalline silica differ between countries, and there is a need for better standardization of the techniques.
Workers in the bitumen waterproofing industry, especially roofing membrane-manufacturing workers, were, in addition to bitumen exposure, concurrently exposed to several carcinogenic agents. In addition to the effects of the agents themselves, the possible additive or synergistic effects caused by the combined exposure to these agents and to other factors such as smoking should be taken into consideration when adjusting for confounding exposures in epidemiological studies. For example, the combined effect of asbestos exposure and smoking is suggested to increase the risk of lung cancer 50-fold compared to that of a population of non-smokers with no exposure to asbestos (WHO, 1999). Similar effects may be expected for combined exposure to asbestos and PAHs (Fournier and Pezerat, 1986; Loli et al., 2004).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Our results show that the estimated average annual exposures to asbestos, coal tar-derived PAHs and quartz dust in the bitumen waterproofing industry in Finland and Denmark were significant in the past but have a clear declining trend. Exposure to bitumen fume in the industry was found to follow a similar trend. This trend is mainly due to the lesser use or ban of hazardous agents, such as asbestos and coal tar pitch, and to more automated production lines and improved work methods.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
Bitumen Waterproofing Association, UK.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
The authors would like to thank all of the interviewed persons in the bitumen waterproofing industry for providing us with valuable information and the Confederation of Finnish Construction Industries, the Finnish Roofing Association and the Danish Roofing Advisory Board for their helpful co-operation during the study. We are also grateful to Dr Ritva Luukkonen from Finnish Institute of Occupational Health for her help in statistical analyses.
Received April 15, 2008; in final form September 17, 2008
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSFUNDINGACKNOWLEDGEMENTSREFERENCES |
|---|
ABSC. Arbetsmiljøet i tagpapbranchen: En undersøgelse foretaget hos tagpapdækkere [Work environment study in roofing field] (1983) Silkeborg, Denmark: Asfaltbranchens Bedriftssundhedscentre. (in Danish).
Ahonen K, Engström B, Lehtinen PU. Rakennustyökartoitus: työhygienia [Construction work survey: occupational hygiene] (1977) Helsinki, Finland: Finnish Institute of Occupational Health. (in Finnish).
Boffetta P, Burstyn I, Partanen T, et al. Cancer mortality among European asphalt workers: an international epidemiological study. I. Results of the analysis based on job titles. Am J Ind Med (2003a) 43:18–27.[CrossRef][Web of Science][Medline]
Boffetta P, Burstyn I, Partanen T, et al. Cancer mortality among European asphalt workers: an international epidemiological study. II. Exposure to bitumen fume and other agents. Am J Ind Med (2003b) 43:28–39.[CrossRef][Web of Science][Medline]
Burstyn I, Boffetta P, Kauppinen T, et al. Estimating exposures in the asphalt industry for an international epidemiological cohort study of cancer risk. Am J Ind Med (2003) 43:3–17.[CrossRef][Web of Science][Medline]
Degerth R, Könni U, Olkinuora P, et al. Hitsattavat kumibitumit kattojen vedeneristyksessä: Työhygieeninen ja ergonominen selvitys [Torch-on roofing membranes in roofing: occupational hygiene and ergonomics] (1991) Helsinki, Finland: LEL Työeläkekassa. (in Finnish).
Engholm G, Englund A, Lindner B. Mortality and cancer incidence in Swedish road pavers and roofers. Health Environ (1991) 1:62–8.
EU. Agreement on workers health protection through the good handling and use of crystalline silica and products containing it (2006) Eur Off J; Notice No. 2006/C 279/02.
Fournier J, Pezerat H. Studies on surface properties of asbestos. III. Interactions between asbestos and polynuclear aromatic hydrocarbons. Environ Res (1986) 41:276–95.[Medline]
Hansen E. Cancer incidence in an occupational cohort exposed to bitumen fume. Scand J Work Environ Health (1989) 15:101–5.[Web of Science][Medline]
Herwin RL, Emmett EA. NIOSH health hazard evaluation report no. 75-194-324 (1976) Cinnicati, OH: National Institute for Occupational Safety and Health.
IARC. IARC monographs on the evaluation of carcinogenic risks to humans. Polynuclear aromatic compounds. Part 4: bitumens, coal-tars and derived products, shale-oils and soots (1985) 35. Lyon, France: International Agency for Research on Cancer.
IARC. IARC monographs on the evaluation of carcinogenic risks to humans. Silica, some silicates, coal dust and para-aramid fibrils (1997) 68. Lyon, France: International Agency for Research on Cancer.
Loli P, Topinka J, Georgiadis P, et al. Benzo[a]pyrene-enhanced mutagenesis by asbestos in the lung of lambda-lacI transgenic rats. Mutat Res (2004) 553:79–90.[Web of Science][Medline]
McClean MD, Rinehart RD, Sapkota A, et al. Dermal exposure and urinary 1-hydroxypyrene among asphalt roofing workers. J Occup Environ Hyg (2003) 4(Suppl 1):118–26.[CrossRef]
NIOSH. Asphalt fume exposures during the manufacture of asphalt roofing products. Current practises for reducing exposures (2001) Cincinnati, OH: National Institute for Occupational Safety and Health.
NIOSH. Asphalt fume exposures during the application of hot asphalt to the roofs. Current practises for reducing exposures (2006) Cincinnati, OH: National Institute for Occupational Safety and Health.
Partanen T, Boffetta P. Cancer risk in asphalt workers and roofers: review and meta-analysis of epidemiologic studies. Am J Ind Med (1994) 26:721–40.[Web of Science][Medline]
Priha E, Koistinen P, Liius R, et al. Vedeneristysalalla käytettävät bitumituotteet ja niiden terveydelliset haitat [Bitumen products applied in the waterproofing field and their heath hazards] (1980) Helsinki, Finland: Finnish Institute of Occupational Health. (in Finnish).
Riala R. Altisteet työssä: Asbesti. [Exposures in the work environment: asbestos] (1991) Helsinki, Finland: Finnish Institute of Occupational Health. (in Finnish).
Riala R, Pirhonen H, Heikkilä P. Asbesti purku- ja huoltotöissä [Asbestos in maintenance and demolition work] (1993) Helsinki, Finland: Finnish Institute of Occupational Health. (in Finnish).
Rühl R, Musanke U, Kolmsee K, et al. Vapours and aerosols of bitumen: exposure data obtained by the German Bitumen Forum. Ann Occup Hyg (2006) 50:459–68.
Sawicki E. Airborne carcinogens and allied compounds. Arch Environ Health (1967) 14:46–53.[Web of Science][Medline]
Sawicki E, Fox FT, Hauser TR, et al. Polynuclear aromatic hydrocarbon composition of air polluted by coal-tar pitch fume. Am Ind Hyg Assoc J (1962) 23:482–6.[Medline]
Sheehan P, Mowat F, Wiedling R. Simulation of asbestos release from asphalt-based roofing products. Abstract (2006) Chicago, IL: American Industrial Hygiene Association Conference and Exposition.
SiD. Asfaltbekendtgørelsen: En arbejdsmiljøundersøgelse af asfaltbekendtgørelsens virkning blandt tagdækkere [Work environment study in roofing field] (1998) Copenhagen, Denmark: National Union of General Workers SiD. (in Danish).
Stern FB, Ruder AM, Chen G. Proportionate mortality among unionized roofers and waterproofers. Am J Ind Med (2000) 37:478–92.[CrossRef][Web of Science][Medline]
WHO. Environmental health criteria 211: health effects of interactions between tobacco use and exposure to other agents (1999) Switzerland, Geneva: World Health Organization.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||