Annals of Occupational Hygiene Advance Access originally published online on March 17, 2007
Annals of Occupational Hygiene 2007 51(3):249-260; doi:10.1093/annhyg/mem007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Occupational Exposure to Wood Dust in the British Woodworking Industry in 1999/2000
1 Health and Safety Executive, City Gate West Toll House Hill, Nottingham, NG1 5AT, UK
2 Health and Safety Executive, Grove House Skerton Road, Manchester M16 0RB, UK
3 Health and Safety Executive, Rose Court 2 Southwark Bridge, London SE1 9HS, UK
*Author to whom correspondence should be addressed: Tel: +44 (0)115 971 2800; fax: +44 (0)115 971 2802; e-mail: mailto: nigel.black{at}hse.gsi.gov.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
Exposure to inhalable wood dust and compliance with the British Control of Substances Hazardous to Health (COSHH) Regulations 1999 were assessed at a representative cross-section of the British woodworking industry. Median exposures ranged from 1.5 to 2.8 mg/m3 across the selected industry groups, the lowest being in sawmilling and planing of wood. Overall, 27% of values exceeded the maximum exposure limit (MEL) at that time of 5 mg/m3. These results showed that the percentage of exposures above the MEL was less than in a survey carried out 10 years earlier. A wide variation of exposures was identified at different machines and tasks. At least 90% at bandsawing and cross-cut sawing were <5 mg/m3. In contrast, dust emission at circular sawing, sanding, cleaning and a miscellaneous group of activities was poorly controlled. Between 32 and 50% of results from these categories exceeded 5 mg/m3. The lower exposures in sawmills were largely attributable to the low usage of sanders and a group of circular saws, to the high use of bandsaws and moulders and to coarser dust from undried timber. Compliance with the COSHH Regulations was inadequate. Companies that claimed to have some form of written COSHH assessment were generally no more effective at controlling exposure to dust than those without an assessment. Similarly the ability of premises that provided information, instruction and training on the risks to health from wood dust and on measures to prevent or control those risks was not generally greater than in those that did not. Maintenance of local exhaust ventilation systems emerged as essential for achieving good control. Companies that followed both the statutory 14-monthly thorough examination and testing schedule and a weekly check system were more successful in this respect than those that did not. In spite of this, local exhaust ventilation alone provided insufficient dust control at several woodworking activities.
Keywords: compliance • determinants of exposure • wood dust
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
Wood dust is consistently high in the top 10 causes of occupational asthma in the United Kingdom (HSE Statistics, 2005). In addition, rare sino-nasal cancers have been diagnosed in workers exposed to hard wood dust. The British Control of Substances Hazardous to Health Regulations (1999) placed a duty on employers to reduce exposure to wood dust to as far as was reasonably practicable below a limit of 5 mg/m3 (Statutory Instruments, 1999). Even so, this is not a ‘safe’ limit. The European Commission's Scientific Committee on Occupational Exposure Limits has reported that workers exposed to dust from hard or soft wood at concentrations greater than 0.5 mg/m3 have exhibited significant health impairment (European Commission's Scientific Committee on Occupational Exposure Limits, SCOEL/SUM102 B June 2002. Unpublished report).
In 1999/2000 some 16 000 registered work units employing about 219 000 workers manufactured furniture and wooden products or were engaged in sawmilling in the United Kingdom (Office for National Statistics, 2000). Seventy-nine percent employed fewer than 10 workers and 96% less than 50. A similar size distribution profile occurred across all of the woodworking industries defined by the Standard Industry Classification, SIC92. SIC92 is a numerical system for identifying and classifying businesses by their economic activity. It was the United Kingdom's implementation of the European Community's NACE Rev. 1. (Office for National statistics, 1993). The small nature of the enterprises implied that few would have measurements of exposure to wood dust to support their mandatory assessments made under COSHH. Similarly the large number of small premises presented difficulties in ensuring that they received and acted on health and safety information.
In 1999 the Health and Safety Executive (HSE) commissioned the Health and Safety Laboratory (HSL) to conduct a survey of exposure to wood dust and to assess compliance with COSHH. The survey predated the introduction of workplace exposure limits (WELs) in 2005.
|
METHODOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
Personal air samples were collected at premises classified by SIC92 as manufacturers of wood and wood products or furniture. HSE used local knowledge from each of their seven geographical regions to prepare a shortlist of approximately 60 sites in Great Britain. The proportion of sites in each standard industry class (SIC) mirrored the proportion on the HSE database. This selection aimed to provide a representative cross-section of the woodworking industry in terms of SIC and the size of premises. The choice was not intentionally biased towards either good or poor control of exposure to wood dust. HSL selected sites for the survey from this shortlist.
Employees are also exposed to wood dust in industries, such as forestry, logging and related services (SIC 02), building and repairing of ships and boats (SIC 351) and construction joinery installation (SIC 45420). These occupations were outside the scope of this survey but estimates for them in the United Kingdom based on exposure modelling have been reported as part of the EU/Wood-Risk project QL K4-2000-00573 (Black et al., 2005). Data from the HSE survey was also used in the Wood-Risk project.
Samples of inhalable dust [BS EN 481, (1993)] were taken as described in MDHS 14/3 (HSE Books, 2000). Air from an employee's breathing zone was drawn at a rate of 2 l/min through a 25 mm diameter glass fibre filter held in an IOM cassette and the loading on the filter determined gravimetrically with a limit of detection of 
0.01 mg/m3. Samples were deliberately collected from a cross-section of workers directly exposed to wood dust. These were usually machine operators. Workers who were indirectly exposed to wood dust, such as fork lift truck drivers, were not included in the study. Each survey team used their own judgement to decide which activities to sample and the proportion of employees within each activity group. On average 79% of exposed employees were sampled in premises where less than 25 were directly exposed. At sites where between 25 and 49 were directly exposed this figure was 54%.
Most air samples were collected for between 3 and 6 h during activities judged to be representative of the whole shift. A small number lasted <2 h when discrete machining tasks were done. Exposures were calculated as 8 h time weighted averages (TWAs) for the day of the survey. Samples from a specific activity were normally collected from different workers rather than repeated on the same employee. If an employee undertook different activities that were sampled separately, a TWA was calculated for each task as if it had accounted for the whole of their exposure. Information about the working environment at each site was recorded on a standardized questionnaire.
Occupational exposures are often assumed to follow lognormal distributions that are conveniently characterized by their geometric means and geometric standard deviations. Exposures within a distribution are considered to be from a homogeneous exposure group. While these assumptions may be valid for large datasets it is not necessarily so for smaller subsets. This paper uses non-parametric statistical tests to avoid these assumptions. In addition, the authors use quartiles (Q1, median and Q3) and the 90th percentile to represent data subsets. The interquartile range, Q3-Q1, is a measure of dispersion and the 90th percentile represents the higher end of the data distribution. Unlike the geometric standard deviation and range, neither value is sensitive to extreme measurements for the size of subsets used. Unless otherwise stated the tests were conducted at the 95% confidence level.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
A total of 406 samples were collected at 46 sites. None of the premises were from SIC 20 200 (manufacture of veneer sheets and other boards). Registered sawmills (0.62%) and between 0.16 and 0.32% of the other classes were visited.
Dilworth (2000) provided an overview of results soon after the survey but some samples were later found to have a high non-wood content or insufficient survey information for more detailed evaluation. They were subsequently excluded and an additional two sites added for this appraisal. Table 1 shows the number of sites and samples collected for each SIC.
|
Classes 36 110–36 140 were combined into a single furniture group for data analysis. Classes 20 400 and 20 510 followed similar exposure distributions. They were pooled into an ‘other’ category to improve confidence in statistical tests. For convenience, the descriptors of SIC 20 100, SIC 20 300 and SIC 361 are abbreviated to ‘sawmills’, ‘joinery’ and ‘furniture’ in this paper.
An uncharacteristically high TWA of 202 mg/m3 was identified at one furniture manufacturer. It was omitted from graphs where its inclusion could obscure the underlying sample distribution, but was included in statistical tests based on ranks. Four results at or below the limit of detection of about 0.01 mg/m3 were treated similarly. The remaining 98.8% of the data ranged from 0.05 to 101 mg/m3.
The effects of company size on exposure
Eighty percent of sites employed fewer than 25 exposed workers and 89% less than 50. Overall, the activities sampled were representative of the range of tasks undertaken during the survey. Table 2 and Figure 1 summarize how the distribution of exposures to wood dust varied with the number of employees exposed.
|
0.01 mg/m3 or >100 mg/m3are omitted for clarity.
|
Median exposures within the groups ranged from 2.3 to 2.7 mg/m3. A Kruskal-Wallis test showed that the differences between medians were not significant (H = 2.39, P
0.5).
Variation of TWA exposures across industry groups
Table 3 and Figure 2 compare the distributions of exposures across the manufacturing groups. A Kruskal-Wallis test showed that at least one median exposure was significantly different from the others (H = 13.75, P < 0.01). A supplementary multiple comparison test (Conover, 1999) identified sawmilling as the only significantly different group. The median TWA was less than in other classes. In addition, the ninetieth percentile was also lower.
|
|
The influence of wood type on exposure
The surveyed premises used predominantly a mixture of hard and soft woods although a significant proportion of manufacturers of joinery and furniture also used composite woods (medium density fibreboard, chipboard and plywood). Nine establishments used only softwood in contrast to only three sites that worked exclusively with hardwood and four that used only composite woods.
The small number of sites that worked only with hardwood or composites and the low number of hardwood samples collected prevented a full comparison of exposures to hardwood, softwood and composite dusts. Even so, a tentative comparison of exposure to softwood dust alone (45 samples) and composite dust alone (39 samples) did not show significant differences by a Kolmogorov-Smirnov test (Dmax = 0.104, Dcrit = 0.298). Dust from sawmilling was omitted from the comparison so that the differences identified in Figure 2 did not skew the result. See Figure 3.
|
The effects of woodworking machine or activity on exposure
Tasks and woodworking machines were allotted to one of fifteen groups as reproduced in the Appendix Table A1. Circular saws were classified as either circular saws or cross-cut saws since these two groups controlled wood dust emissions differently.
Table 4 and Figure 4 summarize the sampling results.
|
|
Approximately 25% of survey samples were classed as multitasking, in which more than one machine or process was used during the sampling period. There were too few samples in routing, tenon machine operation, planing, lathe operation and cleaning for reliable statistical analysis so these were omitted from Figure 4.
On a relative scale, bandsawing and cross cut sawing provided moderate control. At least 90% of TWA exposures at these machines were <5 mg/m3. Moulding and routing were not so well controlled although 90% of exposures were <5.7 mg/m3. Circular sawing, sanding, cleaning and the ‘other’ categories created the highest values across the exposure range. Their 75th percentiles ranged from 8.4 to 20.7 mg/m3.
The influence of local exhaust ventilation (LEV) on exposure
Not all samples from machines or tasks without LEV were clearly identified but 60% of the samples were assigned as being collected under the influence of LEV at some stage of sampling. Multitasking was included if local exhaust ventilation was applied to at least one of the constituent tasks. Nearly all of the saws, moulders, tenoners and lathes were reported to be fitted with LEV. Half of the multitask, sanding and ‘other’ groups were classed as using it. In contrast, there was little reported use of LEV for cleaning or assembly and no details were recorded for LEV on these activities. Sanding included fixed machinery that was likely to operate under LEV and hand-held tools, only some of which would have it.
There was some form of LEV at each site but often only parts of the extraction system could be shown by dust lamp and smoke tubes to be effective. Typical faults included:
- Poor balance of extraction between machines;
- Hole in filter bag, split joint in ducting, poor connections;
- Captor hood too far away from source to be effective;
- Air movement, the position of the wood or employee influenced performance;
- Dust from saw blown towards operator;
- Smoke blown away when machine switched on;
- Mobile LEV could only be used on one machine at a time.
The influence of LEV testing on exposure
A significant proportion of exposures at machines that operated under local exhaust ventilation was greater than the exposure limit. Even so, 59% of the companies had carried out either a statutory 14-monthly thorough examination and testing schedule or weekly checks or both. Table 5 shows the distribution of ventilation check intervals. The proportion of joinery manufacturers that conducted both checks and maintenance was less than for the other groups.
|
Median TWA exposures appeared to increase if either the check or thorough examination interval was omitted. Kruskal-Wallis and multiple comparison tests showed that median exposures were significantly lower where both the 14-monthly schedule and weekly checks were in place than where neither was (H = 12.55, P < 0.02). In addition, 78% of exposures were <5 mg/m3 at the former compared with 65% in the latter.
Further analysis showed that the effects of omitting either test interval were significant at the less stringent 90% level. It follows that companies that carried out both 14-monthly thorough testing and weekly checks were more effective at controlling exposures to wood dust than those that did not, but even under these conditions a significant proportion of exposures did not comply with the occupational exposure limit.
Recirculation of filtered air into the workplace
Approximately one-third of furniture manufacturers and sawmills and two-thirds of joinery manufacturers recirculated filtered air into the workplace. Only five of these companies had some form of maintenance schedule specifically for the return air system.
Seventy-four percent of exposures determined where the premises had a recirculated air system were <5 mg/m3, compared to 73% where no recirculation occurred, indicating that the returned air was not a significant source of exposure. BS EN 12779, (2004) recommends that the residual dust content in return air is <0.3 mg/m3 at all individual outlets.
Selection and use of respiratory protective equipment (RPE)
All companies except one sawmill provided respiratory protective equipment but its use was largely at the discretion of employees. They tended to wear it only for very dusty tasks, such as sanding with hand-held power tools, emptying collection bags and dry sweeping. RPE was compulsory at one sawmill when working with hardwoods and at one joinery manufacturer when working with MDF.
Less than one-fifth of the companies provided formal training on RPE. Details were not recorded in the survey. Employees at several sites received no formal training but relied on the RPE product information sheets. In spite of this the surveyors considered that the work force generally wore their RPE correctly at 88% of the premises where RPE was used.
The Approved Code of Practice to the COSHH Regulations requires reusable RPE to be thoroughly examined at suitable intervals and records of the examinations to be kept, but at the seven sites that reused RPE only four examined and tested the equipment at suitable intervals and two kept maintenance records.
Both reusable and disposable respiratory protective equipment should be stored in a clean place, such as a locker, bag or storeroom. Sixty-nine percent of the sites that used RPE complied with this. Common storage failures included RPE left on benches after use, folded and kept in employees' pockets, or hung up in a dusty area. RPE was issued daily at one site but was dirty inside.
Twenty percent of samples were collected while employees wore RPE for all or part of their tasks. Two-thirds of employees who used hand-held sanders and one-third of multitasking employees did so. Although circular sawing produced high dust levels, only 23% of these saw operators used RPE. No employee wore appropriate RPE during cleaning. One used a nuisance dust mask while wet brushing even though nuisance dust masks are unsuitable for dusts at concentrations greater than their exposure limit or for those that may cause asthma.
Approximately half of the employees who wore RPE had 8 h TWA exposures <5 mg/m3. It is likely that most of this use was limited to dusty activities rather than routine. Most employees wore no RPE but an additional 16% of the study group required a respirator, or other measures, to control their exposure to <5 mg/m3. An additional 4% wore respirators that provided insufficient protection or inappropriate items, such as nuisance dust masks or organic vapour respirators. Activities with a high 90th percentile TWA in Table 4 are likely to require RPE or other additional control measures.
The COSHH Regulations were revised in 2002 and amended in 2004. The Approved Code of Practice to these Regulations introduced respirator fit testing as a means of checking the suitability of close-fitting respirators. The survey predated this introduction. Consequently the questionnaire did not address it. None of the companies were reported to have assessed respirator face fit.
Summary of compliance with COSHH Regulations
The survey aimed to put the exposure data into context by including information on compliance with the COSHH Regulations. Companies that compiled a comprehensive risk assessment, informed the workforce about those risks and how to reduce them, used and maintained the control measures effectively and provided effective health surveillance were expected to control the risks well. Table 6 summarizes compliance with the COSHH Regulations across the industry groups.
|
The influence of a written COSHH assessment
Twenty-five sites claimed to have written a COSHH assessment and another two had risk assessments that addressed wood. The survey reported whether these assessments covered production, maintenance, cleaning and skin issues but did not assess their quality. In practice, concepts of a COSHH assessment varied from data sheets alone to a document covering all these issues.
Table 7 summarizes the relationship between COSHH assessments and exposure within the industry groups. Directional Mann–Whitney U-tests showed that in both furniture manufacture and the ‘other’ group there was no significant difference in median exposures between sites with written COSHH assessments and those without. Joinery manufacturers with written assessments did control exposures significantly better but, in contrast, performance in sawmills was worse. Overall, the claim to have written a COSHH assessment was not a significant factor in controlling exposure to wood dust.
|
Information, instruction and training
Exposure to wood dust has been associated with occupational asthma and nasal cancer. The survey assessed whether employers had provided sufficient information, instruction and training so that employees knew the risks to health and also the precautions to be taken to prevent or reduce exposure. Table 8 summarizes the extent of information, instruction and training given and their distribution across the industry groups.
|
Kruskal-Wallis and multiple comparison tests showed significant differences in the five information groups (H = 19.35, P < 0.0001) but these were traced to four furniture manufacturers that provided risk-only or control-only information. Companies in which training addressed both risks to health and control measures, and those that considered neither accounted for 72% of surveyed premises and 71% of the samples. No significant differences in median exposures were identified in these two groups at either the 95% confidence level or at the less stringent 90% level. The 90th percentiles of these two groups were also similar.
Information, instruction and training should lead to the correct use of measures to prevent or control exposure. Seventy-three percent of the sites that provided information, instruction and training on both health risks and prevention were judged to be using the available control measures correctly. However, exactly the same proportion of those companies that provided no information also used controls correctly. This implied that employees learned through experience. The survey did not make subjective assessments about the quality of the formal information, instruction and training. Even so it did not show that what companies provided conferred additional benefits in terms of controlling exposure to wood dust.
Use of control measures
The survey looked briefly at whether the measures taken by each company to control exposure were used correctly and whether good hygiene standards were adopted. General assessments were made for each site but not at individual sample level. Correct usage was reported at 74% of sites, but with some qualifications. For example LEV at one site was reported to be used as intended but was otherwise inadequate. At other sites the type of RPE supplied was worn correctly but was unlikely to provide a sufficiently high standard of protection.
Instances of bad hygiene practice were also recorded (Dilworth, 2000). For example, even though 79% of sites had a vacuum cleaner, brush cleaning was observed at 96% of them and the use of compressed airlines to clean machinery or contaminated clothing was seen at 64%. Good practice was also noted. Eighty-five percent of premises had a policy forbidding smoking near woodworking machinery although it is probable that this was because of the fire risk.
Health surveillance
Only furniture manufacturers provided health surveillance. The survey identified five sites out of nineteen. They tended to be larger companies with at least forty and generally more than one hundred employees. At three sites surveillance included nasal cancer screening. The other two sites used lung function tests and one of these included a questionnaire and skin inspection. Two additional companies undertook pre-employment health screening without subsequent health surveillance. The British Furniture Manufacturing Association carried out the health surveillance and kept the records for at least two of the sites.
|
OTHER STUDIES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
Comparison with the 1989/1990 British hardwood dust survey
One objective of the survey was to assess whether exposures had decreased in the 10 years since a similar-sized HSE survey of exposure to hardwood dust in 1989/1990 (HSE, 1990). However several differences between the two surveys limited direct comparison. In particular the earlier study did not analyse exposures by industry. The proportion of furniture manufacturers was higher in the older survey (59% versus 41%) and the proportion of sawmills significantly less (2% versus 15%), which would tend to produce higher results. Overall 60% of the 363 exposures in the earlier study were <5 mg/m3 (95% CI 54–64%) compared with 71% of 339 samples from the later evaluation (95% CI 66–76%). Sawmill data from the latter were excluded to improve comparability.
Comparison was further constrained by adoption in the 1999/2000 survey of a different sampling method that, on average, collected slightly more dust (Kenny et al., 1997; Kenny and Thompson, 1998).
The more recent survey did not identify significant differences in exposure to softwood dust compared to composite wood dust. It was also unable to show quantitative differences between the amounts of dust generated by machining hardwoods and other types of wood. However Chung et al. (2000) have reported that sanding MDF produced more dust than sanding pine or oak although there were no significant differences due to sawing.
In the absence of other factors the differences in wood and sampling methods used in the later study would tend to produce results that were at least as high as in the hardwood dust survey. This indicates that the higher proportion of exposures <5 mg/m3 in the 1999/2000 study represents a real and significant improvement over the previous decade. The hardwood dust report made general statements about the need for employers to understand published guidance on dust control better, to implement it and to train employees in its use. These issues also arose in the 1999/2000 survey, but the earlier report lacked sufficient detail as a basis for identifying the reasons for the observed improvements.
Other EU surveys
Liukkonen et al. (2005) and Kauppinen et al. (2006) have compared exposure to wood dust in several European countries. Where necessary, measured concentrations were transformed to an equivalent inhalable dust concentration by using published conversion factors to take the effects of different sampling methods into account. The proportion of exposures that exceeded 5 mg/m3 varied across the industry classes. The average was 25% in Great Britain, 10% in Denmark, 9% in Finland, 16% in Germany and 18% in France. Liukkonen et al. classified their data as near-field samples collected close to sources of wood dust, far-field, occasionally exposed and control room. Nearly all data from the British survey were taken at machining operations and would be classed as near field. This might explain some but not all of the higher British exposures. Even so, it follows that inadequate exposure control is also an issue in other parts of the European Union.
Good working practices
The British survey took an overall view of whether control measures were used correctly but it was impractical to report at the individual sample level. Consequently it is not clear from individual results whether failure to contain dust emissions resulted from poor adjustment of LEV, unsatisfactory machine design or connection to an inadequate extraction system. Similarly, the data were unable to provide a statistical basis for identifying working practices that significantly increased or decreased exposure to wood dust. However, Mikkelsen et al. (2002) have listed influences in the Danish furniture industry where exposures were well controlled. They proposed the following preventive measures:
- Automate woodworking machines/processes, in particular when manual sanding is included;
- Ensure effective local exhaust ventilation at all woodworking machines;
- Balance general or local ventilation by the intake of supplementary fresh air;
- Clean production areas daily, preferably by professionals;
- Do not use compressed air for cleaning machines or workpieces. Use a vacuum system. Clean workpieces finally by brush;
- Encourage the introduction of safety representatives.
Small companies in Britain are unlikely to adopt cleaning by professionals as a preventive measure. In addition, cleaning workpieces by brush may be a commonplace but brushing could increase exposure unless the airborne dust that it creates is removed by good ventilation. It is not recommended as general practice. Note, however, that although HSE recommends vacuum cleaning, HSL has identified concerns about the standards of cleaners used in the woodworking industry (Wake, 2003).
Inclusion of a performance indicator in LEV ductwork is additional good practice that assists weekly checking. BS EN 12779 (2004) recommends that an indicator be provided at each main duct to demonstrate that the specified extraction requirement is achieved. This standard also recommends continuous monitoring for returned-air systems with an air flow capacity greater than 10 000 m3/h. A residual dust level of 0.3 mg/m3 should trigger an alarm and divert the dust to the atmosphere.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
The survey showed marked differences across the industry groups in terms of the types of woodworking machine used, the ability to control exposure to wood dust and compliance with the COSHH Regulations. Overall, compliance was poor although control of exposure to <5 mg/m3 was greater than in 1989/1990. The numbers of employees exposed and the type of wood used did not appear to significantly influence exposure.
Sawmills
Sawmills did not process composite materials. Exposure profiles in sawmilling were less than in other groups and the differences in median exposures were significant. The lower exposures were attributed to the proportion of band saws, moulder and unspecified ‘other’ activities. In addition, sawing undried wood is expected to produce coarser, damper dust than dried wood. Relative to the other groups, sawmills produced a higher proportion of written COSHH assessments and maintained the LEV more frequently, but were poor at providing information, instruction and training and at implementing health surveillance. Approximately 20% of exposures in sawmills were >5 mg/m3.
Other groups were also characterized by differences in machine and wood use although their median exposures were similar.
Joinery manufacturers
Multitasking was the dominant activity in joinery manufacturers but the proportion of sanders was also high. Joinery manufacturers were particularly ineffective at maintaining LEV and at informing employees about risks to health and control measures. Even so, this group was better than average at using control measures correctly. The proportion of exposures >5 mg/m3 was similar to that in sawmills.
Furniture manufacturers
Furniture manufacturers were better than average at informing employees about risks to health and control measures. They were the only group that provided health surveillance. In spite of these positive factors, correct use of control measures was below average. More than 30% of exposures exceeded 5 mg/m3. Sanding, multitasking and ‘other’ activities were dominant in furniture manufacturer.
‘Other’ group
The ‘other’ group appeared to be comparatively good at putting LEV test programmes in place and providing information, although the small number of sites could have influenced these observations. The main activities were multitasking, circular sawing and assembly. The proportion of exposures exceeding 5 mg/m3 was also >30%.
Although the survey examined the prevalence of written COSHH assessments and some of their contents it was outside its remit to assess suitability, whether they were up to date or how well companies implemented them. Concepts of a COSHH assessment ranged from data sheets only to comprehensive documentation. Even so there was no simple explanation of why sawmills without COSHH assessments controlled exposures better than those with them or why the converse was observed in joinery manufacturers. The main quantitative difference between these two groups was that a higher proportion of assessments from joineries included cleaning and skin issues. Regular maintenance of LEV was shown to be a significant factor in improving control of exposure, but only one joinery manufacturer with a written COSHH assessment carried out both a 14-monthly thorough examination and weekly checks.
The survey identified gaps between the purpose of a COSHH assessment and its implementation. However separate studies would be necessary to identify and remedy the causes. Similar limitations and conclusions apply to the interpretation of data on information, instruction and training.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
The survey highlighted significant weaknesses in compliance with the COSHH Regulations. Fundamental to this were the low proportion of written COSHH assessments and the general lack of impact of these assessments on a company's ability to control exposures.
Companies in the wood industry rely heavily on local exhaust ventilation as a means of controlling exposure. This is a potentially complex subject and is dependent on the adequacy of system design, commissioning, maintenance and correct use. Employers and employees need to be more aware of its limitations and of how to use dust extraction equipment effectively. They should also understand the importance of weekly LEV checks in addition to the statutory thorough examination and testing requirements.
Some processes require the use of suitable respiratory protective equipment in addition to well-designed extraction equipment. Although the survey did not examine respirator fit testing it is probable that the wood industry has a poor awareness of the need for it, particularly in small businesses.
Failure to control exposure to wood dust adequately could also be partly the result of not recognizing the risks to health. Stakeholders, such as enforcing authorities, insurance companies, trade unions and trades associations each have a role in raising awareness of these issues, in promoting good practice and in improving standards of protection against wood dust.
|
Appendix |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
|
Received February 24, 2006; in final form January 17, 2007
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSOTHER STUDIESDISCUSSIONCONCLUSIONSAppendixREFERENCES |
|---|
Black N, Kauppinen T, Vincent R, et al. (2005) Occupational exposure to wood dust in the United Kingdom. EU/Wood-Risk Project QLK4-2000-00573. Finnish Institute of Occupational Health, Institut National de Recherche et de Securite, Helsinki and Nancy.
BS EN 481. (1993) Workplace atmospheres—Size fraction definitions for measurement of airborne particles.
BS EN 12779. (2004) Woodworking machines. Chip and dust extraction systems with fixed installation. Safety related performance and safety requirements.
Chung KY, Cuthbert RJ, Revell GS, et al. A study of dust emission, particle size distribution, formaldehyde concentration during machining of medium density fibreboard. Ann Occ Hyg (2000) 44:455–66.
Conover WJ. Practical nonparametric statistics. (1999) New York/Chichester/Weinheim/Brisbane/Singapore/Toronto: John Wiley and Sons, Inc. ISBN 0-471-16068-7.
Dilworth M. (2000) Wood dust survey report 1999/2000. JS20001349. Health and Safety Laboratory, Harpur Hill, Buxton Derbyshire.
HSE. (1990) Technical Development Survey Report: Exposure to hardwood dust. HSE internal report.
HSE Books. (2000) Methods for the determination of hazardous substances (MDHS) 14/3: general methods for sampling and gravimetric analysis of respirable and inhalable dust. HSE Books, PO Box 1999, Sudbury, Suffolk CO12 2WA.
HSE Statistics. (2005) Asthma and other respiratory diseases. Available from http://www.hse.gov.uk//statistics/causdis/asthmaindex.htm.
Kenny LC, Aitken R, Chalmers C, et al. Outcome of a collaborative European study of personal inhalable sampler performance. Ann Occup Hyg (1997) 41:135–53.
Kenny LC, Thompson J. (1998) Overview of findings from workplace comparisons of inhalable samplers. R42.85 final report. Health and Safety Laboratory, Harpur Hill, Buxton Derbyshire.
Kauppinen T, Vincent R, Liukkonen T, et al. Occupational exposure to inhalable wood dust in the member states of the European Union. Ann Occ Hyg (2006) 50:549–61.
Liukkonen T, Vincent R, Grzebyk M, et al. (2005) Exposure measurements of wood dust in the European Union. EU/Wood-Risk Project QLK4-2000-00573. Finnish Institute of Occupational Health, Institut National de Recherche et de Securite, Helsinki and Nancy.
Mikkelsen AB, Schluenssen V, Sigsgaard T, et al. Determinants of wood dust exposure in the Danish furniture industry. Ann Occ Hyg (2002) 46:673–85.
Office for National Statistics. (1993) The standard industrial classification of economic activities 1992. Available from http://www.statistics.gov.uk/methods_quality/sic/default.asp.
Office for National Statistics. (2000) Size analysis of UK businesses. Available from http://www.statistics.gov.uk/StatBase/Product.asp?vlnk=933&Pos=&ColRank=1&Rank=240.
Statutory Instrument. (1999) No 437. The Control of substances hazardous to health Regulations. Available from http://www.opsi.gov.uk/si/si1999/19990437.htm.
Wake D. (2003) A review of vacuum cleaners with emphasis on ones designed for hazardous dust that could be used in the woodworking industry. Report IR/EC/03/03. Health and Safety Laboratory, Harpur Hill, Buxton Derbyshire.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. S. Galea, M. Van Tongeren, A. J. Sleeuwenhoek, D. While, M. Graham, A. Bolton, H. Kromhout, and J. W. Cherrie Trends in Wood Dust Inhalation Exposure in the UK, 1985-2005 Ann. Hyg., October 1, 2009; 53(7): 657 - 667. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Scarselli, A Binazzi, P Ferrante, and A Marinaccio Occupational exposure levels to wood dust in Italy, 1996-2006 Occup. Environ. Med., August 1, 2008; 65(8): 567 - 574. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||