TY - JOUR AU - Warren, Nicholas AB - Abstract The aim of this work was to benchmark respirable crystalline silica (RCS) exposures in brick manufacturing and stone working sectors in Great Britain. This will contribute to a larger programme of work, which will be used to better understand the role of health surveillance in preventing the development of further cases of silicosis and chronic obstructive pulmonary disease. This work was undertaken by means of site visits to measure RCS and respirable dust exposures and assess exposure controls. In addition, historic exposure reports from the sites were collated to allow assessment of exposure trends. The survey, which was conducted in 20 sites (10 from each sector), found that in both sectors over 20% of the measured exposures exceeded the UK RCS 8-hour time-weighted averaged workplace exposure limit (WEL) of 0.1 mg/m3. In the stone sector over 40% of the 8 h time-weighted average RCS exposures were above the RCS WEL compared to 20% in the brick manufacturing sector. In the stone sector, 61% of RCS exposures where water suppression was present exceeded the RCS WEL. This indicates that a variety of exposure controls will be required to control RCS exposures, including respiratory protective equipment (RPE). The use of RPE in situations where RCS exposure exceeded the RCS WEL was more prevalent in stone working than in the brick sector. There were differences associated with RPE and the use of other exposure controls in both sectors. The contextual information in historic consultant’s exposure reports was generally limited, with exposure controls either not mentioned or not fully described. This affects the usefulness of exposure monitoring to dutyholders. This work will provide information on exposures allowing construction of lifetime exposure estimates for use in analysis of the health effects data. A second survey to the sites is planned to determine how exposures have altered. brick manufacturing, contextual information, quartz, respirable crystalline silica, silicosis, stone working, water suppression Introduction Crystalline silica is common in rocks and minerals (Tuomi et al., 2014) and inhalation of crystalline silica in the respirable size range may cause adverse health effects. These include lung diseases such as silicosis, lung cancer, and chronic obstructive pulmonary disease (COPD) (NIOSH, 2002). These diseases are debilitating and sufferers will have a substantially reduced quality of life and potentially premature death. Brown et al. (2012) estimated there were 590,000 workers potentially occupationally exposed to respirable crystalline silica (RCS) in Great Britain (GB). Other research estimated that there may be hundreds of cases of lung cancer per year in the UK caused by workplace RCS exposure (Rushton et al., 2012). The work reported here is part of a longitudinal epidemiological study that also includes an intervention element that aims to improve awareness of health risks and good practice to control RCS at the study sites. Further information about the wider study is available from the authors. The phase reported here was conducted to understand historic trends in RCS exposures and associated controls in the brick manufacturing and stone working sectors prior to the intervention phase. An analysis of respirable dust exposures collected at the same time is available in the Supplementary Material (available at Annals of Occupational Hygiene online). The RCS exposure baseline generated by this work will be used to provide information on RCS exposures allowing construction of lifetime exposure estimates for use in analysis of the health effects data. A second survey is planned later in the longitudinal study after the interventions have occurred. It is known that since the visits reported here were conducted, the GB brick manufacturing sector have implemented a control improvement strategy (Private communication). These improvements and their effect on exposures will be examined in the next phase of this work. In toxicology, exposure refers to person’s dose. However, in occupational hygiene, it refers potential exposure as, in the case of personal air sampling, it does not take into account the protection afforded by respiratory protective equipment (RPE). The occupational hygiene definition is in accordance with draft BS EN 689 (BSI, 2018), which indicates that exposure limits are to be compared with measured exposures outside of RPE. Where exposure is mentioned in this paper, it refers to the occupational hygiene definition. Methodology Site selection The brick manufacturing and stone working industry sectors were selected for this work as previous research (Easterbrook and Brough, 2009) had shown significant exposure of RCS and in comparison with other sectors with the potential for RCS exposures (e.g. construction) they have a stable workforce that facilitated recruitment to a longitudinal study. Ethics approval was granted for this work by the UK Integrated Research Application System (IRAS) (IRAS number 12NW0049). In 2013, there were 100 manufactures of bricks and 1180 stone working companies in the UK (ONS, 2013a) with 4100 workers in the brick sector and 6500 in the stone sector (ONS, 2013b). The brick sector has signed the European Network for Silica (NEPSI) European Social Dialogue, ‘Agreement on Workers’ Health Protection Through the Good Handling and Use of Crystalline Silica and Products Containing it’, which sets out appropriate measures for reduction of RCS exposure based on practice in the UK (Private communication) (NEPSI, 2017). Ten brick and 10 stone working sites were recruited in this study. The sample was designed to recruit 400 workers for the health study. Stone sector sites were contacted directly and those using high silica content stone were invited to participate. Recruitment of brick manufacturing sites was through the British Ceramic Confederation. In both sectors, some companies had multiple sites that were included in the survey. Production processes The brick manufacturing sites that participated used a mixture of automated and manual production methods. Each site crushed and mixed clay in automated mills (‘pans’) in segregated areas. Conveyors moved clay to the adjacent brick manufacturing area. In the automated brick manufacturing process, clay was extruded and cut into uncured (green) bricks, which were coated with a parting agent, commonly containing crystalline silica. In the manual process, clay was hand thrown into wooden moulds to form green bricks. These moulds and bricks were coated with the parting agent by hand. Green bricks were dried to remove excess moisture before being fired in a high temperature kiln. Some sites used two separate kilns; one to dry green bricks and a separate kiln for high temperature firing and others used the same kiln for both processes. Loading (setting) the kilns was performed either manually or automatically. For automated production, bricks were loaded onto kiln cars that were sent through the kiln on tracks. Once fired, bricks were unloaded (dehacking), either manually or automatically, then visually checked and packaged for dispatch. In the automated production process kiln cars were cleaned using an automatic vacuum cleaner prior to loading with green bricks. At one site, bricks were tumbled in a rotary device to remove sharp corners to produce ‘aged’ bricks. Cleaning was performed using hand held tools (shovel, broom), vacuum cleaners or ride-on cleaners. The stone working sites made a mixture of products such as natural stone paving flags, kitchen worktops, and bespoke stone products. Most sites had primary saws that cut boulders (commonly several tonnes in weight) into smaller pieces. These saws operate at slow speeds and had reciprocating saw blades or used a continuous wire. All sites had secondary saws to cut stone slabs to the required dimensions. These consisted of high-speed circular saws with up to four blades. These saws were both automatic and manually operated. Stone was shaped by computer numerically controlled cutting or profile cutting machines. Stonemasons who used hand held power tools to cut and shape stone were present at a number of sites. Some sites used guillotines to make stone setts (cobbles) or house ‘bricks’. At one site, setts were tumbled to remove sharp edges and at another a rotating blade was used to remove the edges. Surfaces were polished using hand held or automated machines. At some sites the stone’s surface was scabbled (roughened) using hand held tools (e.g. propane torches) or automatic machines. Work areas were generally segregated into process areas, such as primary saws, hand-masonry and secondary saws. Cleaning was generally carried out using water hoses to wash debris into drains. Work tasks and common exposure controls in each sector are summarized in Table 1. Table 1 shows the similarly exposed groups (SEGs) based on an observational approach classified by job description or work task (Ramachandran, 2005; HSE, 2006). A minority of the measurements were for workers outside a building. Table 1. Work activities covered by SEGs and common exposure controls observed. SEG description . Tasks included . Common exposure controls observed (dependent on site) . Brick manufacturing sector Clay preparation Working in the clay preparation area. This is where the dry materials are crushed, graded, and mixed to form the clay used to make bricks. LEV, segregation, and half mask RPE. Extruder Manufacturing green bricks and coating surfaces with silica using automatic methods and other related tasks. LEV was used on the brick coating operation. Hand brick making Hand manufacturing green bricks and coating with silica, and related tasks. Half mask RPE was used during this process. Setter manual Manually placing green bricks in kiln for firing. None. Setter automatic Automatically placing green bricks on kiln cars for firing. No exposure controls in use at all the sites.Some sites used LEV when removing bricks from kiln cars and half mask RPE when manually removing bricks. Kiln operator Operating and maintaining brick kilns. Dehacking manual Manually removing bricks from kiln they have been fired. Dehacking Automatic Sorting bricks after they have been removed from the kiln cars and packing them for dispatch. Ancillary Tasks not included in the other SEGs (e.g. drying sand, tumbling bricks, maintenance, etc.). The controls in use varied but included half mask RPE and water suppression. Fork lift truck Driving fork lift trucks. No exposure controls in use, often working outside. Cleaning Cleaning the plant. Half mask RPE and disposable coveralls. Stone working sector Primary saw Saws used to cut large stone boulders to blocks to a suitable size for other processes (e.g. secondary saw). Commonly these are reciprocating or wire saws, operating at slow speeds. Water suppression and half mask RPE. Secondary saw Cutting medium sized stone working blocks using higher speed saws compared with primary saws. These saws are typically high-speed circular saws fitted with up to four blades. Segregation, water suppression, and half mask RPE. Hand mason Cutting stone working used hand held tools including power tools. Segregation, water suppression, LEV, and powered hood RPE were commonly used. At some sites performed outside with no LEV. Polishing Polishing stone working surfaces using hand held or large polishing machines. Water suppression was always in use. Half mask RPE occasionally used. Fork lift truck Driving fork lift trucks. Half mask RPE was occasionally used. Packing Packing stone and other activities related to dispatching products. No exposure controls used. Surface roughening Roughening the surface of stones using propane torches or tumbling stone setts to remove smooth edges. Half mask RPE and/or water suppression. Maintenance Maintenance (electrical and mechanical), of the building and equipment. Half mask RPE for selected maintenance tasks. Manager Manager. No exposure controls in use. SEG description . Tasks included . Common exposure controls observed (dependent on site) . Brick manufacturing sector Clay preparation Working in the clay preparation area. This is where the dry materials are crushed, graded, and mixed to form the clay used to make bricks. LEV, segregation, and half mask RPE. Extruder Manufacturing green bricks and coating surfaces with silica using automatic methods and other related tasks. LEV was used on the brick coating operation. Hand brick making Hand manufacturing green bricks and coating with silica, and related tasks. Half mask RPE was used during this process. Setter manual Manually placing green bricks in kiln for firing. None. Setter automatic Automatically placing green bricks on kiln cars for firing. No exposure controls in use at all the sites.Some sites used LEV when removing bricks from kiln cars and half mask RPE when manually removing bricks. Kiln operator Operating and maintaining brick kilns. Dehacking manual Manually removing bricks from kiln they have been fired. Dehacking Automatic Sorting bricks after they have been removed from the kiln cars and packing them for dispatch. Ancillary Tasks not included in the other SEGs (e.g. drying sand, tumbling bricks, maintenance, etc.). The controls in use varied but included half mask RPE and water suppression. Fork lift truck Driving fork lift trucks. No exposure controls in use, often working outside. Cleaning Cleaning the plant. Half mask RPE and disposable coveralls. Stone working sector Primary saw Saws used to cut large stone boulders to blocks to a suitable size for other processes (e.g. secondary saw). Commonly these are reciprocating or wire saws, operating at slow speeds. Water suppression and half mask RPE. Secondary saw Cutting medium sized stone working blocks using higher speed saws compared with primary saws. These saws are typically high-speed circular saws fitted with up to four blades. Segregation, water suppression, and half mask RPE. Hand mason Cutting stone working used hand held tools including power tools. Segregation, water suppression, LEV, and powered hood RPE were commonly used. At some sites performed outside with no LEV. Polishing Polishing stone working surfaces using hand held or large polishing machines. Water suppression was always in use. Half mask RPE occasionally used. Fork lift truck Driving fork lift trucks. Half mask RPE was occasionally used. Packing Packing stone and other activities related to dispatching products. No exposure controls used. Surface roughening Roughening the surface of stones using propane torches or tumbling stone setts to remove smooth edges. Half mask RPE and/or water suppression. Maintenance Maintenance (electrical and mechanical), of the building and equipment. Half mask RPE for selected maintenance tasks. Manager Manager. No exposure controls in use. Open in new tab Table 1. Work activities covered by SEGs and common exposure controls observed. SEG description . Tasks included . Common exposure controls observed (dependent on site) . Brick manufacturing sector Clay preparation Working in the clay preparation area. This is where the dry materials are crushed, graded, and mixed to form the clay used to make bricks. LEV, segregation, and half mask RPE. Extruder Manufacturing green bricks and coating surfaces with silica using automatic methods and other related tasks. LEV was used on the brick coating operation. Hand brick making Hand manufacturing green bricks and coating with silica, and related tasks. Half mask RPE was used during this process. Setter manual Manually placing green bricks in kiln for firing. None. Setter automatic Automatically placing green bricks on kiln cars for firing. No exposure controls in use at all the sites.Some sites used LEV when removing bricks from kiln cars and half mask RPE when manually removing bricks. Kiln operator Operating and maintaining brick kilns. Dehacking manual Manually removing bricks from kiln they have been fired. Dehacking Automatic Sorting bricks after they have been removed from the kiln cars and packing them for dispatch. Ancillary Tasks not included in the other SEGs (e.g. drying sand, tumbling bricks, maintenance, etc.). The controls in use varied but included half mask RPE and water suppression. Fork lift truck Driving fork lift trucks. No exposure controls in use, often working outside. Cleaning Cleaning the plant. Half mask RPE and disposable coveralls. Stone working sector Primary saw Saws used to cut large stone boulders to blocks to a suitable size for other processes (e.g. secondary saw). Commonly these are reciprocating or wire saws, operating at slow speeds. Water suppression and half mask RPE. Secondary saw Cutting medium sized stone working blocks using higher speed saws compared with primary saws. These saws are typically high-speed circular saws fitted with up to four blades. Segregation, water suppression, and half mask RPE. Hand mason Cutting stone working used hand held tools including power tools. Segregation, water suppression, LEV, and powered hood RPE were commonly used. At some sites performed outside with no LEV. Polishing Polishing stone working surfaces using hand held or large polishing machines. Water suppression was always in use. Half mask RPE occasionally used. Fork lift truck Driving fork lift trucks. Half mask RPE was occasionally used. Packing Packing stone and other activities related to dispatching products. No exposure controls used. Surface roughening Roughening the surface of stones using propane torches or tumbling stone setts to remove smooth edges. Half mask RPE and/or water suppression. Maintenance Maintenance (electrical and mechanical), of the building and equipment. Half mask RPE for selected maintenance tasks. Manager Manager. No exposure controls in use. SEG description . Tasks included . Common exposure controls observed (dependent on site) . Brick manufacturing sector Clay preparation Working in the clay preparation area. This is where the dry materials are crushed, graded, and mixed to form the clay used to make bricks. LEV, segregation, and half mask RPE. Extruder Manufacturing green bricks and coating surfaces with silica using automatic methods and other related tasks. LEV was used on the brick coating operation. Hand brick making Hand manufacturing green bricks and coating with silica, and related tasks. Half mask RPE was used during this process. Setter manual Manually placing green bricks in kiln for firing. None. Setter automatic Automatically placing green bricks on kiln cars for firing. No exposure controls in use at all the sites.Some sites used LEV when removing bricks from kiln cars and half mask RPE when manually removing bricks. Kiln operator Operating and maintaining brick kilns. Dehacking manual Manually removing bricks from kiln they have been fired. Dehacking Automatic Sorting bricks after they have been removed from the kiln cars and packing them for dispatch. Ancillary Tasks not included in the other SEGs (e.g. drying sand, tumbling bricks, maintenance, etc.). The controls in use varied but included half mask RPE and water suppression. Fork lift truck Driving fork lift trucks. No exposure controls in use, often working outside. Cleaning Cleaning the plant. Half mask RPE and disposable coveralls. Stone working sector Primary saw Saws used to cut large stone boulders to blocks to a suitable size for other processes (e.g. secondary saw). Commonly these are reciprocating or wire saws, operating at slow speeds. Water suppression and half mask RPE. Secondary saw Cutting medium sized stone working blocks using higher speed saws compared with primary saws. These saws are typically high-speed circular saws fitted with up to four blades. Segregation, water suppression, and half mask RPE. Hand mason Cutting stone working used hand held tools including power tools. Segregation, water suppression, LEV, and powered hood RPE were commonly used. At some sites performed outside with no LEV. Polishing Polishing stone working surfaces using hand held or large polishing machines. Water suppression was always in use. Half mask RPE occasionally used. Fork lift truck Driving fork lift trucks. Half mask RPE was occasionally used. Packing Packing stone and other activities related to dispatching products. No exposure controls used. Surface roughening Roughening the surface of stones using propane torches or tumbling stone setts to remove smooth edges. Half mask RPE and/or water suppression. Maintenance Maintenance (electrical and mechanical), of the building and equipment. Half mask RPE for selected maintenance tasks. Manager Manager. No exposure controls in use. Open in new tab Exposure benchmarking Volunteers were recruited to participate in the associated health survey of this research. Where possible, personal air sampling was carried out to characterize exposures for these workers. Additional personal monitoring was performed to characterize exposures for SEGs where no workers had volunteered for the health study. Sampling periods were all greater than half the shift, except for task-specific sampling (e.g. cleaning). Task-specific samples were included with the longer samples in the calculation of 8 h time-weighted average (TWA). Keen et al. (2012) indicated that the conducting an exposure survey does not significantly affect workplace exposures, based on collecting urine samples over five working days, with an exposure survey being conducted only on the first day. Personal air sampling was used to measure exposure to respirable dust using PVC filters mounted in respirable dust samplers aspirated at 2.2 l/min. Respirable dust was measured and analysed using MDHS 14 (HSE, 2000) and RCS was analysed using X-ray diffraction (HSE, 2005). The measured concentrations were converted to 8-h TWA exposure. Great Britain has set a workplace exposure limit (WEL) to RCS as 0.1 mg/m3 as an 8 h TWA (HSE, 2011). In accordance with BS EN 689, the measured exposures do not take into account any protection offered by RPE. Typical RCS limit of detection was 0.02 mg/m3. Contextual information on the work process (e.g. working practices and exposure controls) was obtained using a questionnaire when conducting exposure measurements. The exposure controls in place were examined and their effectiveness assessed. Historic exposure data Historic exposure data were acquired from 18 of the 20 sites visited. This was in the form of occupational hygiene or ‘results only’ reports. Reports were mainly written by external consultants and in all cases sample analysis was conducted by external laboratories. These reports covered the period from 1997 to 2013. There were 45 reports obtained from the brick sector and 32 from the stone working sector. Historic exposure measurements were evaluated using the decision tree developed by Vincent and Werner (2003). This sets out criteria related to representative sampling periods, sampling procedures, tasks performed, statistical analysis, exposure controls, and consideration of other exposure routes. For the purposes of this study, it was decided that the RCS exposure data should meet the level three criteria with one exception that related to statistical design of the exposure survey. Also, the exposures had to be reported as 8 h TWAs. Any exposure report not meeting these criteria was excluded from the statistical analysis. There is no agreed classification system for contextual information in reports. For this paper, it was decided to use Tielemans et al. (2002) criteria to grade the contextual information. Contextual information in historic reports was classified using criteria described by Tielemans et al. (2002), which rates the information as good, moderate, poor, or unacceptable. The classification rules are presented in Table 2. The stone working sector included a HSE Buxton Laboratory report, which was excluded from the contextual analysis as it was written to assess the practicality of using high flow respirable samplers for RCS measurement. Table 2. Tielemans et al. 2002, rules for contextual information and findings from the analysis of the historic exposure reports. Tielemans criteria . . Tielemans categorization requirements a . This surveyb . . . Good . Moderate . Poor . Unacceptable . No. of historic reports meeting criteria . Premises Name of premises Y Any of the poor indicators missing 73 Address Y 69 Economic activity Y Y 56 Size Y Y 1 Workplace Department Y 6 Work area Y 24 Process name Y Y 14 Work activity Profession Y Y Y 0 Task Y Y Y 72 Product Product identifier Y 76 Chemical agent Chemical name Y Y Y 76 Exposure modifiers Exposure patternc Y 42 Pattern of controld Y Y 0 RPE used Y 32 Confinement Y 11 Measurement strategy Random, worse case, etc. Y Y 17 Measurement procedure Sampling date Y Y year 74 Sample ID Y 68 Sampling device Y Y 57 Type of sample Y Y Y 71 Sampling times Y 22 Sample duration Y Y Y 56 Exposure duration Y 14 Analytical methods Y Y 65 Results Measured conc. Y Y Y 76 Units used Y Y Y 76 Sample status Y Y Y 56 Number of reports (out of 76b) 0 0 56 20 Tielemans criteria . . Tielemans categorization requirements a . This surveyb . . . Good . Moderate . Poor . Unacceptable . No. of historic reports meeting criteria . Premises Name of premises Y Any of the poor indicators missing 73 Address Y 69 Economic activity Y Y 56 Size Y Y 1 Workplace Department Y 6 Work area Y 24 Process name Y Y 14 Work activity Profession Y Y Y 0 Task Y Y Y 72 Product Product identifier Y 76 Chemical agent Chemical name Y Y Y 76 Exposure modifiers Exposure patternc Y 42 Pattern of controld Y Y 0 RPE used Y 32 Confinement Y 11 Measurement strategy Random, worse case, etc. Y Y 17 Measurement procedure Sampling date Y Y year 74 Sample ID Y 68 Sampling device Y Y 57 Type of sample Y Y Y 71 Sampling times Y 22 Sample duration Y Y Y 56 Exposure duration Y 14 Analytical methods Y Y 65 Results Measured conc. Y Y Y 76 Units used Y Y Y 76 Sample status Y Y Y 56 Number of reports (out of 76b) 0 0 56 20 aTo be categorized as good, moderate or poor all the criteria marked Y must be met. bOne report was excluded as it had been written by HSE in another research project. cReports calculated 8 h TWA RCS exposure, but did not state the exposure pattern. The use of the calculation implies that the exposures are constant. dReports did not discuss the pattern of control, except where high exposures were measured. Open in new tab Table 2. Tielemans et al. 2002, rules for contextual information and findings from the analysis of the historic exposure reports. Tielemans criteria . . Tielemans categorization requirements a . This surveyb . . . Good . Moderate . Poor . Unacceptable . No. of historic reports meeting criteria . Premises Name of premises Y Any of the poor indicators missing 73 Address Y 69 Economic activity Y Y 56 Size Y Y 1 Workplace Department Y 6 Work area Y 24 Process name Y Y 14 Work activity Profession Y Y Y 0 Task Y Y Y 72 Product Product identifier Y 76 Chemical agent Chemical name Y Y Y 76 Exposure modifiers Exposure patternc Y 42 Pattern of controld Y Y 0 RPE used Y 32 Confinement Y 11 Measurement strategy Random, worse case, etc. Y Y 17 Measurement procedure Sampling date Y Y year 74 Sample ID Y 68 Sampling device Y Y 57 Type of sample Y Y Y 71 Sampling times Y 22 Sample duration Y Y Y 56 Exposure duration Y 14 Analytical methods Y Y 65 Results Measured conc. Y Y Y 76 Units used Y Y Y 76 Sample status Y Y Y 56 Number of reports (out of 76b) 0 0 56 20 Tielemans criteria . . Tielemans categorization requirements a . This surveyb . . . Good . Moderate . Poor . Unacceptable . No. of historic reports meeting criteria . Premises Name of premises Y Any of the poor indicators missing 73 Address Y 69 Economic activity Y Y 56 Size Y Y 1 Workplace Department Y 6 Work area Y 24 Process name Y Y 14 Work activity Profession Y Y Y 0 Task Y Y Y 72 Product Product identifier Y 76 Chemical agent Chemical name Y Y Y 76 Exposure modifiers Exposure patternc Y 42 Pattern of controld Y Y 0 RPE used Y 32 Confinement Y 11 Measurement strategy Random, worse case, etc. Y Y 17 Measurement procedure Sampling date Y Y year 74 Sample ID Y 68 Sampling device Y Y 57 Type of sample Y Y Y 71 Sampling times Y 22 Sample duration Y Y Y 56 Exposure duration Y 14 Analytical methods Y Y 65 Results Measured conc. Y Y Y 76 Units used Y Y Y 76 Sample status Y Y Y 56 Number of reports (out of 76b) 0 0 56 20 aTo be categorized as good, moderate or poor all the criteria marked Y must be met. bOne report was excluded as it had been written by HSE in another research project. cReports calculated 8 h TWA RCS exposure, but did not state the exposure pattern. The use of the calculation implies that the exposures are constant. dReports did not discuss the pattern of control, except where high exposures were measured. Open in new tab Statistical analysis The RCS exposure data were right skewed and the measurements were (natural) log transformed prior to further analysis. Mixed effects models for RCS 8 h TWA exposures in the stone working and brick sectors were fitted in a Bayesian framework (Winbugs) to allow non-detects to be treated as left-censored data. Fixed effects were used to model time trends, job-specific offsets, and systematic differences between historic and HSL measurements. Variation between work sites was modelled as a random effect, and repeated measures on some individuals allowed the remaining residuals to be modelled as a sum of between- and within-worker variation. The model used can be expressed as: ln(Yi,j,k,l)=Μk+α×Timei,j,k,l+β×I(source)+δl+γj+εi,j,k,l where Yi,j,k,l is the ith measurement on the jth worker in SEG k at site l, I (source) indicates whether the measurement was HSL or historic data and Time was expressed in years relative to 2013. Variations in exposure between sites (δl), between workers (γj) and between measurements on the same workers (εi,j,k,l) were represented through random effects. Measurements less than the limit of detection were treated as left censored. Details of the prior distributions and the model implementation are presented in the Supplementary Material (available at Annals of Occupational Hygiene online). Factors for the presence or absence of exposure controls were not included because there is insufficient data to allow the causal effect of controls on exposure to be differentiated from the effect of targeted use of controls (see Discussion). Furthermore, exposure control usage was not consistently recorded in historic consultant reports. Results Exposure data As part of this study, HSL visited 20 sites in 2012 to 2013 and made 110 exposure measurements in the stone working sector and 159 in the brick manufacturing sector. Seventy-seven historic exposure measurement reports were obtained. Historic reports were written by 18 different authors. Only three authors (eight reports) listed formal qualifications in occupational hygiene. Twenty-four reports were ‘results only’ and the remainder included additional contextual information. Based on the modified Vincent and Werner criteria, 52 reports were identified whose exposure data were suitable for inclusion in the data analysis. The main reason for rejection was the use of short-term task-specific or static sampling. One rejected report did not directly measure RCS, but inferred exposures from measuring respirable dust and the percentage of crystalline silica in bulk samples. Although 8 h TWAs were cited in reports, the exposure pattern was rarely stated and so it is unclear whether these were representative of patterns of continuous or intermittent exposures. Sampling covered routine production tasks in most of the reports. There were only a few instances of task-specific sampling of potential high-exposure short-term tasks. Where task-specific sampling occurred the reports did not state if these results were included in the 8 h TWA calculations. Thirty-six reports described the use of personal cyclone samplers (designed to measure airborne respirable dust) to simultaneously measure both the respirable and inhalable fractions. Dust deposited inside the sampler grit pot was included with the respirable dust collected to provide an estimate of the inhalable concentration, with one such report written in 2013. This has not been tested against the inhalable convention so its validity is uncertain, however this does not affect the reliability of the reported respirable dust and RCS exposures. Analysis of the reports found that the descriptions of the exposure controls were unclear or absent. Application of the Tielemans classification scheme for core contextual information is shown in Table 2. None of the reports was rated as good or moderate. This low classification may not affect the usefulness of the exposure data to the sites, as they will have contextual information from other sources. Use of exposure controls, such as RPE, local exhaust ventilation (LEV), and water suppression (WS) was not consistently recorded in the historic consultant’s exposure reports. RCS exposure levels The pooled exposure data comprise 699 measurements of 8 h TWA RCS exposure levels. The HSL measurements were collected specifically as part of the work reported here, and included all routine activities with exposure potential. The number of repeat measurements per worker varied considerably with 357, 89, 32, and 17 workers having 1, 2, 3, and 4 measurements, respectively. Summary statistics of 8 h TWA RCS exposures are presented in Tables 3 and 4, stratified by SEG in descending order of median exposure. The fraction of measurements below the limit of detection (0.01–0.035 mg/m3 for HSL measurements, 0.01–0.15 mg/m3 for industry measurements) was 15% and 24% in the stone and brick sectors, respectively. Empirical descriptive statistics were calculated after setting measurements below the limit of detection (LOD) to half the LOD. This was considered reasonable because measurements below the LOD are treated as left-censored data in the mixed effects model. In general, exposures in the stone sector were higher than those in brick manufacturing. A supplement to this paper presents a similar summary for respirable dust and for RCS exposures reported by HSL (i.e. with the industry data excluded). Table 3. Eight-hour TWA RCS exposures in stone working sector sites from 2003 to 2013, with measurements below the LOD set to half the LOD. SEG . No. of measurements/ workers/sites . 8 h TWA RCS, mg/m3 . . . Geo standard deviation . % >RCS WEL . % RCS WEL . % RCS WEL . % RCS WEL . % RCS WEL . % RCS WEL . % RCS WEL . % RCS WEL . % 0.1 31 29 94 7 23 19 61 Total 110 44 40 12 11 67 61 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . No. using WS . % using WS . 0–0.02 20 3 15 1 5 7 35 0.02–0.05 41 4 10 2 5 28 68 0.05–0.1 18 8 44 2 11 13 72 >0.1 31 29 94 7 23 19 61 Total 110 44 40 12 11 67 61 Frequencies of RPE and LEV use are calculated from HSL data only. Open in new tab Table 6. Stone working sector: number of measurements taken by HSL (stratified by RCS exposure) and the number for which exposure controls were in use. 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . No. using WS . % using WS . 0–0.02 20 3 15 1 5 7 35 0.02–0.05 41 4 10 2 5 28 68 0.05–0.1 18 8 44 2 11 13 72 >0.1 31 29 94 7 23 19 61 Total 110 44 40 12 11 67 61 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . No. using WS . % using WS . 0–0.02 20 3 15 1 5 7 35 0.02–0.05 41 4 10 2 5 28 68 0.05–0.1 18 8 44 2 11 13 72 >0.1 31 29 94 7 23 19 61 Total 110 44 40 12 11 67 61 Frequencies of RPE and LEV use are calculated from HSL data only. Open in new tab Table 7. Brick sector: number of measurements taken by HSL (stratified by RCS exposure) and the number for which exposure controls were in use. 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . 0–0.02 36 0 0 0 0 0.02–0.05 56 0 0 1 2 0.05–0.1 35 3 9 1 3 >0.1 32 9 28 4 13 Total 159 12 8 6 4 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . 0–0.02 36 0 0 0 0 0.02–0.05 56 0 0 1 2 0.05–0.1 35 3 9 1 3 >0.1 32 9 28 4 13 Total 159 12 8 6 4 Frequencies of RPE and LEV use calculated from HSL data only. Open in new tab Table 7. Brick sector: number of measurements taken by HSL (stratified by RCS exposure) and the number for which exposure controls were in use. 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . 0–0.02 36 0 0 0 0 0.02–0.05 56 0 0 1 2 0.05–0.1 35 3 9 1 3 >0.1 32 9 28 4 13 Total 159 12 8 6 4 8 h TWA mg/m3 . No. of measurements . No. using RPE . % using RPE . No. using LEV . % using LEV . 0–0.02 36 0 0 0 0 0.02–0.05 56 0 0 1 2 0.05–0.1 35 3 9 1 3 >0.1 32 9 28 4 13 Total 159 12 8 6 4 Frequencies of RPE and LEV use calculated from HSL data only. Open in new tab RPE was available at every site visited, with evidence that RPE was targeted at tasks with the potential for higher RCS exposures (Tables 6 and 7). In the brick sector, RPE use was often at the discretion of the user, except for tasks identified as having a high-exposure potential (e.g. cleaning). A critical difference between the sectors was the use of RPE where exposures exceeded the RCS WEL. In the brick sector, RPE was worn in just 28% of such cases, while it was used in 94% of such instances in the stone sector. Four of the ten stone working sites used powered hood RPE. At the other stone sites and the brick sector, it was half mask RPE with either P3 or P2 filters. Face fit testing had not been performed at 5 of the 20 sites (20%). The half mask RPE was often used for periods over an hour. RPE was generally stored in clean environments at the end of the shift, though there were instances where it was stored on workbenches. RPE was occasionally removed in heavily contaminated areas for prolonged periods (e.g. stonemasons doing preparatory work in areas where power tools have previously been used). The stone working sector targeted LEV to tasks with the potential for higher exposures, though 23% of the measurements using LEV had exposures above the RCS WEL (Table 6). The LEV often did not capture RCS dust. Table 6 shows that in the stone working sector the use of WS was above 60% in all except the lowest exposure band (<0.02 mg/m3), and was present in 61% of cases where the RCS WEL was exceeded. WS was used for specialized tasks (e.g. tumbling) in the brick sector. One site used mobile water bowsers fitted with sprinkler to damp road areas to control dust levels. At other sites, the use of similar water misting systems to control dust from roads was not observed. Some sites did not have baffles on secondary saws to capture water mist produced by the WS system. Even when present, they were often in poor condition. Even when the baffles were appropriately maintained, the mist was not captured effectively, with mist drifting into the general workplace. Although these saws were frequently contained within safety enclosures, these were of an open construction that would not segregate contamination. Segregation of dusty processes was observed in both of the sectors studied. In the brick sector, clay preparation areas were segregated and in the stone sector, hand masons tended to be segregated. The brick sector used brooms and shovels to clean debris under machines. The debris was damped prior to cleaning, though this was not observed to happen consistently and the water is unlikely to penetrate the dust pile. Exposure controls included RPE and disposable coveralls, though RPE was not always used. Cleaning in other areas included electric ride-on floor sweepers and industrial vacuum cleaners. When these cleaning methods were used RPE was generally not worn. When emptying an electric ride-on floor sweeper a large dust cloud was observed, but RPE was not worn. Discussion Exposures Despite the well-recognized health risks from RCS, this study identified inadequate exposure control in the brick manufacturing and stone sectors. Most of the SEGs had personal exposures above the RCS WEL. The highest exposures were found in the stone working sector where the means were above the RCS WEL for three SEGs (surface roughening, hand mason, and secondary saw). In the brick manufacturing sector one, SEG (clay preparation) had mean exposures at the RCS WEL. These results demonstrate that there is considerable scope for improvement in control. Advice on control improvements was provided in the feedback to the site and during the intervention phase of the wider project. The effect of the advice will be assessed in the follow-up exposure survey. Exposure trends The mixed effects model found a time-related downward trend in RCS exposures in stone working sites (6% per year), but no significant reduction in exposures for brick sites. These findings suggest that the monitoring conducted at these sites is not being used to review risk assessments and improve exposure controls. Essentially, exposure monitoring is not being used correctly in the widely accepted ‘Plan-Do-Check-Act’ approach to risk management (HSE, 2013a). Yassin et al. (2005) reported a non-statistically significant decline in RCS exposure of 10% per year between 1988 and 2003 in the USA across all industry sectors. Peters et al. (2011) found a reduction in RCS exposures of 6% per year in their review of worldwide RCS exposures. In the European quarry industry, Kromhout et al. (2013) found that RCS exposures had fallen between 2000 and 2009, but had subsequently increased. They attributed this increase to the economic downturn resulting in smaller workforces with more diverse tasks and delayed investments in control measures and new machinery. Other researchers have found that technological changes in production processes, availability and introduction of improved equipment, response to new legislation, and follow-up inspections, together with global economic trends were reasons for exposure reduction in various industry sectors (Creely et al., 2007; Galea et al., 2009). More specifically for this study, the reduction of the RCS WEL in 2006 from 0.3 mg/m3 to 0.1 mg/m3 and interventions undertaken by HSE in the stone working sector were potentially significant. Exposure controls The stone sector were generally following HSE COSHH Essentials exposure control guidance (HSE, 2017), though they were not always correctly followed (e.g. working outside the influence of LEV). The brick sector were generally not following the HSE or NEPSI control guidance (e.g. WS was not widely used, LEV not installed on sand coating operations). Available brick sector control guidance sheets do not include all SEGs, such as automatic setting or automatic dehacking. The half mask was used for long periods making it uncomfortable to use, it was not always face fit tested and RPE was removed in contaminated areas These factors will compromise its effectiveness. Determining an individual’s actual inhalation exposure is difficult under such circumstances as the extent of protection across the working day is almost impossible to estimate. Cleaning can lead to high dust exposures, even from ride-on cleaners (Chung and Moss, 1988). Different cleaning methods (e.g. H-Type vacuum cleaner) can reduce inhalation exposure, as would use of RPE. Researchers have noted that LEV in the brick sector was not universally installed and was not effective (Easterbrook and Brough, 2009). The same observation was found in this study. Baffles fitted to saws are often not maintained and even if present they are not effective at capturing RCS containing mist. Cooper et al. (2015) reported that a combination of WS and LEV reduced the RCS concentration by a factor of 10, compared to WS alone. Exposure controls were not included in the mixed effects models because the study does not follow an experiment design: there were not enough measurements of the same task at the same site before and after controls were introduced. Moreover, we do not expect the causal effect of controls (i.e. reduction of exposure) to be reflected in the data, because controls may be targeted at workers performing high-exposure tasks, as suggested by the pattern of results in Tables 6 and 7. Presumably, the high-exposure tasks using LEV would have been even higher if LEV was not used (Fransman et al., 2008), but including LEV as a fixed effect would, contrariwise, suggest that LEV usage increases exposure. Furthermore, exposure control usage was not consistently recorded in the historic reports. Quality of historic industry exposure data There were deficiencies associated with the exposure monitoring reports supplied by the sites visited with a number of them comprising measurements only. Contextual information in these reports was generally limited. This reduces the value of these reports in tracking improvements in exposure control over time and affects their usefulness in informing the dutyholder on how the controls are utilized and how they may be improved. It is important that relevant contextual information is included in exposure reports to enable exposure trends to be properly understood. Minimum standards for contextual information in occupational hygiene reports have been identified by HSE (HSE, 2013b), the British Occupational Hygiene Society (BOHS, 2001), and others (Rajan et al., 1997; Brederode et al., 2001; Tielemans et al., 2002). This and other research studies continue to find insufficient information in exposure reports (Dost et al., 2000; De Vocht et al., 2005; Creely et al., 2007). The Tielemans criteria for grading contextual information could be refined and developed. For instance, the assessment of exposure controls could be expanded to include other exposure controls such as water suppression and procedural controls (e.g. job rotation, segregation). There is currently no assessment of a report’s discussion and recommendations in Tielemans criteria. Conclusions Intelligence gathered from this study will be used in assessment of the effects of interventions at the sites during the repeat exposure assessment. The quantitative exposure data obtained will be used in the analysis of the health data obtained from the volunteer workers. In the stone sector, over 40% of the 8 h TWA RCS exposures were above the RCS WEL compared with 20% in the brick manufacturing sector. There is clear potential to reduce potential exposures in both sectors by better implementation of existing exposure control solutions, including segregation, enclosure, LEV, and water suppression. Where 8 h TWA exposure exceeds the RCS WEL, RPE was more widely used in the stone sector than in brick manufacturing. In the stone sector, exposure to RCS had fallen by an average 6.2% per year over 10 years. There was no evidence of a similar decline in brick manufacturing despite annual exposure monitoring performed at most sites. This suggests that the exposure monitoring programme is not being used effectively as part of a systematic ‘Plan-Do-Check-Act’ approach to risk management. Better reporting of contextual information in exposure monitoring reports is needed to understand trends in exposure control in both sectors, and improve the usefulness of exposure monitoring reports to the companies. Given that participation in this survey was voluntary, and taking into account the small sample size the findings should not be taken to be representative of the brick and stone sectors as a whole. However, they represent the best intelligence on current GB exposures in the public domain. Funding Funding for this project was provided by HSE. The authors declare no conflict of interest relating to the material presented in this Article. Its contents, including any opinions and/or conclusions expressed, are solely those of the authors and do not necessarily reflect HSE policy. Acknowledgements We thank other HSE staff who contributed to this work in some way. Particular thanks are given to Stephen Cottrell, Vincent Sandys, and Steve Stagg for their assistance in the exposure measurement; Jade Sumner and Leon Lewis for their work in recruiting volunteers to the study, Ed Robinson for co-ordinating the ethics approval process, and Professor David Fishwick for design of the overarching study. We are indebted to the brick and stone companies who provided access to their premises and to the individual workers who participated in the exposure assessments. We also acknowledge the contribution of the British Ceramics Confederation (BCC) in co-ordinating liaison with the brickmaking industry. References BOHS ( 2001 ) Clear and concise report writing: guidance for occupational hygienists . http://www.bohs.org/reportwritingguidance. Accessed 15 December 2014 . Brederode D , Linker F, Marquart H et al. . ( 2001 ) Recording of data of individual measurements of occupational exposure: guideline of the Dutch Society of Occupational Hygiene (October 1999) . Appl Occup Environ Hyg ; 16 : 122 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Brown T , Cherrie JW, Van Tongeren M et al. ( 2012 ) HSE Research Report RR858 - The burden of occupational cancer in Great Britain Lung cancer , http://www.hse.gov.uk/research/rrpdf/rr858.pdf. Accessed 15 December 2014 . BSI ( 2018 ) BS EN 689. Workplace exposure. Measurement of exposure by inhalation to chemical agents. Strategy for testing compliance with occupational exposure limit values . UK : British Standards Institute (BSI) , ISBN 978 0 580 90536 0. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chung KYK and Moss RFJ ( 1988 ) A survey of industrial floor sweeping machines . J Aerosol Sci ; 19 : 1467 – 70 . Google Scholar Crossref Search ADS WorldCat Cooper JH , Johnson DL, Phillips ML. ( 2015 ) Respirable silica dust suppression during artificial stone countertop cutting . Ann Occup Hyg ; 59 : 122 – 6 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Creely KS , Cowie H, Van Tongeren M et al. . ( 2007 ) Trends in inhalation exposure–a review of the data in the published scientific literature . Ann Occup Hyg ; 51 : 665 – 78 . Google Scholar PubMed OpenURL Placeholder Text WorldCat De Vocht F , Straif K, Szeszenia-Dabrowska N et al. . ( 2005 ) A database of exposures in the rubber manufacturing industry: design and quality control . Ann Occup Hyg ; 49 : 691 – 701 . Google Scholar Crossref Search ADS PubMed WorldCat Dost AA , Redman D, Cox G. ( 2000 ) Exposure to rubber fume and rubber process dust in the general rubber goods, tyre manufacturing and retread industries . Ann Occup Hyg ; 44 : 329 – 42 . Google Scholar Crossref Search ADS PubMed WorldCat Easterbrook A and Brough P ( 2009 ) Silica baseline survey. HSE Research Report RR689 . http://www.hse.gov.uk/research/rrpdf/rr689.pdf. Assessed 15 December 2014 . Fransman W , Schinkel J, Meijster T et al. . ( 2008 ) Development and evaluation of an exposure control efficacy library (ECEL) . Ann Occup Hyg ; 52 : 567 – 75 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Galea KS , Van Tongeren M, Sleeuwenhoek AJ et al. . ( 2009 ) Trends in wood dust inhalation exposure in the UK, 1985-2005 . Ann Occup Hyg ; 53 : 657 – 67 . Google Scholar PubMed OpenURL Placeholder Text WorldCat HSE ( 2000 ) Methods for the Determination of Hazardous Substances 14/3 (MDHS14/3), General methods for sampling and gravimetric analysis of respirable and inhalable dust . UK : HSE Books , ISBN 0717617491. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC HSE ( 2005 ) Methods for the Determination of Hazardous Substances 101 (MDHS101) Crystalline silica in respirable airborne dust . UK : HSE Books , ISBN 0717628973. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC HSE ( 2006 ) HSG 173 - Monitoring strategies for toxic substances . UK : HSE Books , ISBN 9780717661886. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC HSE ( 2011 ) EH40/2005 Workplace exposure limits . UK : HSE Books , ISBN 9780717664467. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC HSE ( 2013a ) HSG 65 - Managing for health and safety . UK : HSE Books , ISBN 9780717664566. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC HSE ( 2013b ) Control of substances hazardous to health (Sixth edition) The Control of Substances Hazardous to Health Regulations 2002. Approved Code of Practice and guidance . UK : HSE Books , ISBN 9780717665822. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC HSE ( 2017 ) COSHH Essentials direct advice sheets . http://www.hse.gov.uk/coshh/essentials/direct-advice/index.htm. Accessed 21 December 2017 . Keen C , Coldwell M, McNally K et al. ( 2012 ) A follow up study of occupational exposure to 4,4’-methylene-bis(2-chloroaniline) (MbOCA) and isocyanates in polyurethane manufacture in the UK . Toxicol Lett ; 213 : 3 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Kromhout H , Vlaanderen J and Houba R ( 2013 ) Trends in exposure to respirable dust and quartz over a 12-year period in the industrial minerals industry . Inhaled Particles XI conference , Nottingham, UK . Google Scholar OpenURL Placeholder Text WorldCat NEPSI ( 2017 ) European Network for Silica . http://www.nepsi.eu/home. Accessed 12 December 2017 . NIOSH ( 2002 ) Health effects of occupational exposure to respirable crystalline silica . Cincinnati, OH : Department of Health and Human Services, National Institute for Occupational Safety and Health . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC ONS ( 2013a ) The UK Office of National Statistics (ONS) publication ‘Annual Business Inquiry - Workplace analysis by industry’ https://www.ons.gov.uk/businessindustryandtrade/business/businessservices/methodologies/annualbusinesssurveyabs. Accessed 15 December 2014. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC ONS ( 2013b ) ONS statistics report ‘GB and UK level employment (thousands) by 2, 3 and 5 digit SIC 2007 (full-time/part-time and public/private sector split)’ . https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/datasets/industry235digitsicbusinessregisterandemploymentsurveybrestable2. Accessed 15 December 2014. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Peters S , Vermeulen R, Portengen L et al. . ( 2011 ) Modelling of occupational respirable crystalline silica exposure for quantitative exposure assessment in community-based case-control studies . J Environ Monit ; 13 : 3262 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat Rajan B , Alesbury R, Carton B et al. ( 1997 ) European proposal for core information for the storage and exchange of workplace exposure measurements on chemical agents , Appl Occup Environ Hyg ; 12 : 31 – 39 . Google Scholar Crossref Search ADS WorldCat Ramachandran G ( 2005 ) Occupational exposure assessment for air contaminants . USA : CRC press , ISBN 1 566 706 092. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rushton L , Hutchings SJ, Fortunato L et al. . ( 2012 ) Occupational cancer burden in Great Britain . Br J Cancer ; 107 ( Suppl 1 ): S3 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat Tielemans E , Marquart H, De Cock J et al. . ( 2002 ) A proposal for evaluation of exposure data . Ann Occup Hyg ; 46 : 287 – 97 . Google Scholar Crossref Search ADS PubMed WorldCat Tuomi T , Linnainmaa M, Väänänen V et al. . ( 2014 ) Application of good practices as described by the NEPSI agreement coincides with a strong decline in exposure to respiratory crystalline silica in Finnish workplaces . Ann Occup Hyg ; 58 : 806 – 17 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Vincent JH , Werner MA. ( 2003 ) Critical evaluation of historical occupational aerosol exposure records: applications to nickel and lead . Ann Occup Hyg ; 47 : 49 – 59 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Yassin A , Yebesi F, Tingle R. ( 2005 ) Occupational exposure to crystalline silica dust in the United States, 1988-2003 . Environ Health Perspect ; 113 : 255 – 60 . Google Scholar Crossref Search ADS PubMed WorldCat © Crown copyright 2019. This article contains public sector information licensed under the Open Government Licence v3.0 (http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/). © Crown copyright 2019. TI - Exposure to Respirable Crystalline Silica in the GB Brick Manufacturing and Stone Working Industries JF - Annals of Work Exposures and Health (formerly Annals Of Occupational Hygiene) DO - 10.1093/annweh/wxy103 DA - 2019-02-16 UR - https://www.deepdyve.com/lp/oxford-university-press/exposure-to-respirable-crystalline-silica-in-the-gb-brick-T6pgHpNvqQ SP - 184 EP - 196 VL - 63 IS - 2 DP - DeepDyve ER -