Annals of Occupational Hygiene Advance Access originally published online on July 7, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ann. occup. Hyg., Vol. 48, No. 5, pp. 383-391, 2004
© 2004 British Occupational Hygiene Society
Published by Oxford University Press
Determinants of Exposure to Metalworking Fluid Aerosol in Small Machine Shops
1 School of Occupational and Environmental Hygiene, University of British Columbia, 2206 East Mall, Vancouver, BC V6T 1Z3, Canada; 2 Department of Health Care and Epidemiology, University of British Columbia, 5804 Fairview Avenue, Vancouver, BC V6T 1Z3, Canada
Received 3 July 2003; in final form 30 December 2003; published online on 7 July 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The purpose of this study was to evaluate personal exposure to metalworking fluid (MWF) aerosols in very small machine shops (1–8 machinists per shop) and to investigate workplace factors associated with exposures. A total of 20 willing machine shops in Vancouver, Canada (from 46 eligible shops, 43%) and 88 machinists participated (participation rate for machinists 92%). Most machinists wore two personal sampling trains (an open-faced 37 mm cassette and a PM10 impactor) on each of two full work shifts. Observational data were collected regarding potential determinants of exposure at 15 min intervals throughout each shift. A total of 322 personal samples were taken over 54 days. Mean aerosol exposure was 0.32 mg/m3 (range 0.06–2.19) for the 37 mm cassette samples and 0.27 mg/m3 (range 0.026–3.67) for PM10. Exposures from the two sampler types were highly correlated (R = 0.86). The mean shop-specific ratio comparing exposure from the 37 mm cassette with that from the PM10 sampler was 1.43 and varied significantly across shops, ranging from 0.97 to 2.19. Machine, task and shop characteristics associated with significantly increased aerosol exposure included the proportion of time spent grinding, operating an enclosed computer controlled machine, the presence of welding in the shop for both sampler types and the number of machines using MWF for PM10 samples only. Factors associated with reduced aerosol exposure included machining aluminum, milling, the height (and shape) of the shop roof (for both sample types) and the presence of mechanical shop ventilation (for the 37 mm cassette samples).
Keywords: coolant fluid; determinants of exposure; machinists; metalworking fluid; occupational exposure; small machine shops
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There is growing evidence from the past decade that exposure to metalworking fluid (MWF) aerosols is associated with adverse respiratory health effects and that this risk increases with increasing MWF aerosol exposure (Kennedy et al., 1989; Greaves et al., 1997).
Identifying sources of increased exposure is a vital step in hazard identification and control. Studies of MWF aerosol exposure levels in the automobile industry have identified differences in airborne MWF concentrations among different MWF classes, types of machine tools and local exhaust systems (Hallock et al., 1994; Woskie et al., 1994, 1996a; Hands et al., 1996). Few reports exist on MWF exposures in small equipment manufacturers and machine shops, yet small machine shops employ the majority of machinists in many regions of the world (Piacitelli et al., 2001).
The purpose of this study was to examine characteristics of small machine shops, to evaluate personal MWF aerosol exposure in such shops (using two sampling instruments to collect different size fractions of the aerosol) and to conduct an analysis of workplace factors associated with exposure levels. Such ‘determinants of exposure’ analysis allows effective and rational decision making on optimal control measures.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Machine shop selection and machinist participation
Machine shops were selected from five designated municipal regions within Greater Vancouver, British Columbia, Canada in 1998. Forty-seven shops located in the designated area were identified and contacted to establish whether the business met the minimum requirements for the study (at least two full-time machinists who used MWF for the majority of their day). Site visits were made to machine shops interested in participating to verify the minimum requirements and to recruit machinists to participate. All machinists working on the day shift were invited to participate. Machinists were defined as trained machinists, apprentice machinists, machine tool operators and machinist’s helpers (who machined simple parts and worked around machine tools and MWF).
Air sampling protocol
Personal full-shift air samples (6–8 h) were collected from each machinist who volunteered to participate. The target was to collect a full-shift sample from each participant on two randomly selected days separated by a minimum of 8 days, to reduce the chance of correlation between sample days (Buringh and Lanting, 1991).
Each machinist wore two complete sampling trains. One consisted of an open-faced 37 mm plastic cassette (Gelman Sciences, Ann Arbor, MI) loaded with either a polyvinylchloride (PVC) filter (5 µm pore size; Gelman Sciences) or a polytetrafluoroethylene (PTFE) filter (2 µm pore size Teflo PTFE membrane with PMP ring; Gelman Sciences) and attached to a personal air sampling pump (SKC Aircheck Sampler Model 224-PCXR4; SKC Inc., Eighty Four, PA), calibrated to a flow rate of 2 l/min. [Early on in the study we switched to PTFE filters to facilitate measurement of extractable mass (to be reported separately) and to enable a more accurate comparison between the two samplers.] The second sampling train consisted of a single stage impactor sampler (Personal Environmental Monitor PM10; MSP Corp., Minneapolis, MN) loaded with a PTFE filter (5 µm pore size; Gelman Sciences) attached to a SKC sampling pump, calibrated to a flow rate of 4 l/min. Pumps were calibrated at the machine shops before and after sampling with a BIOS DC-1 flow calibrator (BIOS International Corp., NJ). The cassettes and personal impactors were attached to the machinist’s collar to sample air in the breathing zone.
The first sampling train is designed to collect what is traditionally referred to as ‘total’ aerosol mass. Although it has been used as a standard method in North America to approximate total inhalable aerosol, this sampler has been shown to somewhat under-sample the inhalable fraction (Tsai, 1996). The second sampling train is designed to collect the thoracic fraction of the aerosol (or the size range of aerosol particles that are likely to be deposited in the tracheo-bronchial region of the respiratory tract). This sampler uses the PM10 specifications, i.e. a sharp cut-off with a 50% collection efficiency of 10 µm, to efficiently sample aerosols 
10 µm in aerodynamic diameter.
Prior to pre-weighing, filters were conditioned for 48 h in a temperature (29 ± 1.0°C) and humidity (40 ± 5%) controlled weighing room. Before post-weighing, filters were desiccated for at least 24 h and then conditioned for 48 h. All weighing was done on a microbalance measuring to 0.001 mg (Sartorius M3P-000V001; Sartorius GmbH, Weender Landstrasse, Germany). A total of 14 field blanks were collected for the PVC filters, 20 for the PTFE filters used in 37 mm cassettes and 25 for the PTFE filters used in the PM10 samplers. The mass limit of detection was established as three times the standard deviation of the field blanks.
Determinants of exposure
Collection of data regarding potential determinants of exposure was done at several levels: for every machinist (tasks related to the sampling day and fixed characteristics), every machine tool (certain activities, fixed characteristics and MWF factors) and every machine shop space (daily conditions and fixed shop features). These are listed in Table 1. Tasks and characteristics that varied during the day were recorded every 15 min throughout the sampling period. Machinists who smoked were asked to wait until their breaks or to notify the researcher when they planned to smoke a cigarette. Sampling pumps were placed on hold while the machinist smoked so that their filters were not contaminated. Machinists cooperated with this arrangement.
|
The 15 min interval observational data were merged with the machine tool data to generate quantitative estimates of the proportion of the work shift that the machinist spent at each task, using a particular type of tool and type of MWF. For example, if a machinist was observed to be drilling for 14 of 28 15-min recording intervals, then 50% of the shift was assigned as being spent at the drilling task.
To permit analyses of infrequently observed tasks, these were combined with similar tasks: three specific milling tasks were collapsed into one variable, ‘milling’; drilling, tapping, boring and reaming were combined as ‘drilling’; turning and threading were combined as ‘turning’.
Data analysis
Analyses were conducted using SAS 8.2 (SAS Institute Inc., Cary, NC). Descriptive statistics were used to summarize the characteristics of the participating shops, the machines, the machinists and their exposures. The distribution of the aerosol exposure was positively skewed, therefore exposure data were log transformed before analysis (Shapiro–Wilks test for normality, P < 0.001 for both total and thoracic aerosol before transformation; P > 0.05 after log transformation).
Determinants of exposure models were constructed using SAS Proc Mixed, with either ‘total’ or ‘thoracic’ aerosol exposure as the dependent variable, the determinants described in Table 1 as the fixed effects, and person as a random effect (using a compound symmetry covariance structure). Prior to model building, correlations between all pairs of independent variables were examined. Simple linear regression modeling was used to identify potential explanatory variables that would be considered for the multiple regression models (P 
0.20). When correlated pairs were identified (Pearson’s correlation coefficient 0.65), only the variable most likely to be associated with exposure (based on previous knowledge) or the variable more strongly associated with exposure from simple linear regression analysis was selected for the determinants of the exposure model.
Two approaches to model building were used. In one, separate models were developed for the variables associated with increased exposure and those associated with decreased exposure. Manual backward stepwise regression was used to eliminate variables, one at a time, with the highest P values (0.10). In each model only variables with P values 0.10 were retained. Finally, the two models were combined and variables were removed until the final determinants of exposure model contained those variables, significant at the P 0.10 level. In the other approach, variables were offered to the models in logical clusters (machinist tasks and characteristics, machine characteristics and activities, MWF characteristics and shop factors), then results from each cluster were combined using backward stepwise regression as described above. Results from both approaches yielded identical final models. Residuals were plotted to identify potential influential data and patterns of unexplained variance in the final model. The squared Pearson correlation coefficient (R2) comparing predicted and observed values was calculated to estimate the proportion of variance explained by each model.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study participation and characteristics of machinists
Of the 47 businesses sent invitation letters, 35 were eligible to participate. Of these 35, 15 agreed to participate (43%), while 20 refused. Two of the participating businesses had three shops and one had two shops, so sampling was carried out at 20 different machine shop locations. There were 96 machinists identified as eligible, but five refused to participate and three signed consent forms but were not available on the testing day (laid off, shift change or vacation), leaving 88 participants (92% participation), of whom 73 participated on each of two sampling days. The 15 workers with only one sampling day were missed on the second day because of sickness or shift changes. Of the 88 participants, 72 were journeyman machinists, 6 were apprentices and 10 were helpers or other operators. The average number of years experience in machining was 16.5 (range <1–42).
Characteristics of machine shops, tools and MWF management
Table 2 shows the characteristics of the shops studied. The businesses represented a range of machining operations: heavy manufacturing, repair and maintenance services, gear manufacturing, prototype/custom services and volume contract work (jobber shops). These were all small shops, with the largest having 8 machinists working on one shift. Welding operations were adjacent to machining areas in 6 of the 20 machine shops.
|
Characteristics of the machine tools in the shops are shown in Table 3. Machinists were observed to work on 173 machine tools ranging from computer numeric controlled (CNC) machine tools to conventional machine tools over 50 yr old. The majority were conventional lathes (46) and conventional vertical mills (31). Few machine tools were equipped with exhaust ventilation. Most machine tools had individual MWF sumps, with soluble MWF the most commonly used fluid. Straight MWF was also often applied by hand as required when using machine tools not equipped with MWF sumps (n = 26).
|
MWF management was not regarded as an important issue in most of the shops, and in most cases fluid management was left to the discretion of the machinist. Machine tools that were not used regularly received little MWF maintenance. Machinists who worked predominantly at one machine were more likely to monitor the condition of the MWF. Only two shops had comprehensive fluid management strategies, including weekly checks of the fluid concentration and continuous or weekly oil removal from MWFs. Shops that had a mixture of older conventional machine tools and modern CNC tools often maintained the MWF for the CNC tools but not the others. The most common MWF management practice was to change the fluid in all or some of the machine tools yearly or every several years, regardless of the condition of the fluid.
Machining task characteristics
A total of 999 h of task observation was recorded. Sampling times ranged from 2 to 8 h with a mean time of 6 h. The proportion of time spent engaged in some tasks differed from the proportion of time spent at machines with the same name (e.g. milling/milling machine), since several task types could be performed on one machine but some tasks were specific to a certain machine tool. Machine tools were seen to be actively cutting metal for a mean percentage of 44% of the total time during machinists’ shifts. The remainder of the shift was normally divided equally between tool setting and planning. Milling and turning were the most commonly performed machining tasks (mean 24 and 27% of time, respectively, for machinists performing these tasks, 8 and 10%, respectively, for all machinists). Splashguards were used whenever machine tools were turned on. The most common metals being machined were plain steel (38% of machining time), aluminum (14%), stainless steel (6%) and copper alloys (4%). When machining, all machinists worked within 2 m of the cutting edge of their machine tools and were in that position for the majority of the shift. The use of compressed air to clean the work area was common, with 60% of machinists observed using compressed air for 18% of their shift on average.
Aerosol exposure levels
A total of 322 personal samples were taken over 54 days of fieldwork spaced over a period of 5 months: 161 from each sampler type. Exposure levels for machinists with two samples on different days were correlated (R = 0.64, P < 0.0001 for the 37 mm cassette samples; R = 0.65, P < 0.0001 for the PM10 samples; log transformed data). Measures from the two sampler types on the same machinist on the same day were also correlated (R = 0.86, P < 0.0001).
Table 4 shows exposure results for both sampler types. Only 4% of the aerosol exposures measured by the 37 mm cassettes were below the limit of detection (0.042 µg/m3), with the minimum exposure detected being 0.06 mg/m3. None of the PM10 samples were below the limit of detection (0.024 mg/m3). Significantly higher average aerosol exposures were measured for machinists in shops with welding (P < 0.05, Student’s t-test).
|
The ratio of ‘total’ to ‘thoracic’ aerosol exposure was calculated for machinists who wore both samplers on the same day, for the same duration (n = 153 data pairs). Only exposures greater than the limit of detection were used. These values are shown in Table 4. Ratios were also calculated for each shop, based on the shop average ‘total’ exposure divided by the shop average ‘thoracic’ exposure. These shop-specific average ratios differed significantly between machine shops (P < 0.01, ANOVA), ranging from a low of 0.97 to a high of 2.19. The difference between ‘total’ and ‘thoracic’ exposure levels was significantly greater than 0 (Student’s t-test, P = 0.02), indicating that the ‘total’ sampler captured significantly more aerosol mass than the ‘thoracic’ sampler.
Determinants of exposure analysis results
Although many task, machine, MWF and shop variables were significantly associated with exposure in univariate analyses, most were no longer significant in multivariable analyses. The final models showing factors associated with aerosol exposure (for each sampler type) are shown in Table 5. The determinants listed accounted for an estimated 77% of the variance in aerosol exposure for the 37 mm cassette and an estimated 70% of the variance in aerosol exposure for the PM10 sampler. The residuals for each model were normally distributed and no individual values were shown to be unduly influential.
|
The final multivariable model retained only two machining tasks: the proportion of the shift spent grinding (comprising time spent grinding at a wet grinding machine and grinding with small dry grinders), associated with increased exposure for both sampler types; milling, associated with reduced aerosol exposure (with percent of shift at a vertical milling machine associated with reduced ‘total’ aerosol and percent of shift at any milling machine associated with reduced ‘thoracic’ aerosol).
Only two machine or metal type variables were retained in the multivariable model: operating an enclosed CNC machine, associated with an increase in both ‘total’ and ‘thoracic’ aerosol exposure; the proportion of the shift spent machining aluminum, associated with a decrease in ‘total’ aerosol exposure.
The only significant MWF variables in the final models were the proportion of the shift that the machinist worked at a machine tool in which the MWF was changed periodically (significantly associated with reduced ‘thoracic’ aerosol) and an increased number of machine tools using MWF in the shop (associated with an increase in ‘thoracic’ aerosol exposure).
Both increased roof height and a peaked roof were associated with reduced ‘total’ aerosol exposure, as was the presence of any mechanical ventilation. Reduced thoracic exposure was also associated with increased roof height, but not with roof shape or ventilation. Welding in the shop or adjacent was a significant predictor of increased ‘total’ and ‘thoracic’ aerosol exposure.
Neither job title nor years of experience in machining were retained in the final models, although including the job title apprentice in the final model for total aerosol gave a coefficient of +0.246 (SE = 0.16, P = 0.13), but did not change the other estimates in any appreciable way. Additional models were developed using only data from shops where welding was not present. There was very little change in the model components but the variance explained was reduced for both models (estimated R2 = 0.65 and 0.46 for the ‘total’ and ‘thoracic’ aerosol exposure general linear models, respectively).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Recently, Piacitelli described MWF exposures in 79 US machine shops (many of them custom job shops) employing an average of 46 machinists per shop and mainly comprised of machines with individual MWF sumps (Piacitelli et al., 2001). Our results, though from even smaller shops on average, are remarkably similar, perhaps not surprising given the similar nature of the machining operations being carried out in each study. For example, the average aerosol exposure for 40 samples from machinists engaged in mixed machining tasks in Piacitelli’s study (most similar to those studied by us) was 0.27 mg/m3 (compared with 0.32 mg/m3 in our study). These exposures are also similar to those seen in previous studies from large automobile manufacturing shops (Kennedy et al., 1989; Woskie et al., 1994; Robins et al., 1997), despite the considerable difference in the nature of machine tools observed here (i.e. mostly manual machine tools with individual sumps without local exhaust systems or enclosures) compared with those used in large manufacturing.
Previous studies have provided descriptive data on MWF exposures according to different machine types, metals machined and MWFs used (Chan et al., 1990; Hallock et al., 1994; Woskie et al., 1994, 1996b; Hands et al., 1996). However, in small machine shops, where the work is often very dynamic requiring several different machine tools to make a single part and where the products and parts being machined change frequently, it is difficult to assign personal exposures to a specific machine or MWF type. This difficultly was also recognized by Piacitelli et al. (2001).
The strength of this study was that we incorporated detailed observations of every worker sampled each 15 min of the workday to facilitate a detailed ‘determinants of exposure’ analysis that would take into account the complexity of work tasks performed in these small shops. Although collection of task duration information from 15 min observation intervals either by direct observation or self-reports has been reported previously in other sectors (Preller et al., 1995; Burstyn et al., 1998; Teschke et al., 1999), to the best of our knowledge a study like this one has not been done before in this industry sector.
Our results provide parallel support for some of the ‘machine’-specific results found in other studies. For example, in both large (Woskie et al., 1996b) and small shops (Piacitelli et al., 2001) aerosol exposures for grinders were higher than for workers using other machine tools. This is consistent with our results showing the proportion of the shift spent grinding to be a significant determinant of increased exposure. Milling has not been previously identified as associated with reduced MWF exposure. In this study, vertical mills were the least likely machine tools to have MWF sumps. It was common to observe machinists working on vertical mills applying small amounts of straight MWF fluid by hand and then standing back to avoid inhaling the plume of smoke or aerosol produced. This avoidance behavior may partially explain why work at vertical mills was associated with reduced exposure.
In contrast to most other machinists in this study, who performed many different tasks throughout the day, CNC machinists had relatively constant tasks, tending to stand by their machine while the cutting cycle was completed. These machinists, on average, spent more time at their machines than machinists operating manual machines. Therefore, it is not surprising that the ‘best’ predictor of exposure associated with the use of CNC machines was simply the dichotomous variable ‘working at a fully enclosed CNC machine’. This was associated with increased exposure, despite the presence of the enclosure. This finding is consistent with that of Piacitelli and colleagues, who also found that machinists working at machines without enclosures had slightly lower average exposures than those working at fully or partially enclosed machines (Piacitelli et al., 2001). This could well be due to enclosures being supplied mainly for machines expected to produce the highest aerosol levels (Woskie et al., 1996b) or could be related to leaks in the enclosures or excessive exposure when parts are being taken out of the machines.
The only other task variable associated with decreased exposure (to ‘total’ aerosol, but not to ‘thoracic’ aerosol) was the percentage of the shift machining aluminum. Aluminum is a soft metal that creates large chips when it is machined. Aluminum was also frequently machined on vertical mills (Spearman rank correlation 0.49 between proportion of time spent on vertical mills and proportion of time spent machining aluminum). It is possible that the reduced exposure may be partly due to the same ‘avoidance’ behaviour observed with vertical milling.
Fixed characteristics of the shops (the presence of welding, physical shop dimensions and general ventilation) were also important predictors of exposure levels. We had expected shop volume to be a more important predictor of exposure, based on dilution ventilation equations, but found that roof height and shape were better predictors. Dilution ventilation equations rely on the assumption that the room is well mixed and this assumption may not be true in the field. Our finding that peaked roofs and shop height were associated with reduced exposure suggests that vertical dispersal of aerosols out of the breathing zone may be more protective than horizontal dispersal.
The presence of mechanical air ventilation in the shop was also associated with reduced ‘total’ exposures, while categorization of unventilated shops into high and low dilution ventilation was not informative. This could have been due to measurement error, since natural ventilation was evaluated subjectively. Quantitative measurements, such as using CO2 as a surrogate for natural ventilation, might work more effectively (Thorne et al., 1996). The presence of local exhaust ventilation was not associated with reduced exposure in the study, possibly because equipped machine tools were rare and inadequate capture velocities and generally poor exhaust ventilation design was observed in all of the systems in place. It is also possible that local exhaust may not be the most effective approach to reducing exposures to MWFs. Machine shops considering installing exhaust ventilation should seek the advice of qualified individuals to ensure they get the most out of their investment.
Welding in machine shops was a significant predictor of aerosol exposure in this study. Welding produces metal oxide fumes and combustion products from welding rod fluxes, electrode coatings, oil from metal treatments and paint. Therefore, it was reasonable to expect some contamination of measured MWF aerosol with welding fume. The advantage of using multivariable modeling to evaluate determinants of exposure, as in this study, is that the effect of welding contamination could be accounted for so that associations between MWF and other factors could be investigated.
It is also important to consider the factors not found to be associated with exposure. Although working at a machine tool where MWF was managed periodically was associated with reduced exposures to ‘thoracic’ aerosol, comprehensive management programs were not associated with aerosol reductions. This may have been due to the fact that comprehensive management was more likely to be carried out on CNC machines. The increased exposure associated with these machines may have masked any benefit of MWF management.
It seemed plausible that increased exposures might be associated with the use of compressed air to clean parts. Compressed air use was found to be weakly associated with reduced exposures in simple regression models, but was not associated with exposure levels after adjustment for other factors in multivariable models.
The measurements taken from the same machinist on each of two sampling days were moderately correlated (for ‘thoracic’ aerosol, although not for ‘total’ aerosol), even though a week or more separated the sampling days. The degree of within-worker correlation was higher than observed in some other occupational settings (Francis et al., 1989; Teschke et al., 1999) and within-worker variance was small compared with between-worker variance. This is likely because machinists observed on different days worked at the same station with the same MWF and appeared to have a consistent pattern for using MWF. The pieces being machined changed but the work area and patterns of work were consistent. It is possible that the week or two between sampling days may not have been sufficiently long to capture the overall work practice variability of these machinists.
The 1998 NIOSH Criteria for a recommended standard: occupational exposure to metalworking fluids included recommendations for exposure limits of 0.5 mg/m3 for ‘total’ aerosol and 0.4 mg/m3 for ‘thoracic’ aerosol (National Institute for Occupational Safety and Health, 1998). We found 32 ‘total’ samples (19.9%) and 33 ‘thoracic’ samples (20.5%) above these guidelines, and 8 of the 15 companies involved in this study had workers with exposures exceeding these recommendations, including all 6 shops where welding was present.
The NIOSH report also recommended that if a thoracic sampler was not used for sampling, a factor of 1.25 could be used to adjust exposures measured with a ‘total’ sampler to ‘thoracic’ exposure, based on an analysis of data from three automobile parts manufacturing shops (Woskie et al., 1994). Data in our study indicated a significant difference in the ‘total’ to ‘thoracic’ ratio from shop to shop, suggesting that particle size distributions vary among machine shops and that this correction factor should be applied with caution.
The following limitations should be considered when interpreting the results of this study. Sampling was carried out mostly in the warmer months of the year when doors were more likely to be open. Machinists at several shops commented that conditions were often worse in the winter when all the doors were closed. The participation rate for shops was only 43%. It is possible that the participating shops were biased in some way related to aerosol exposure or machining conditions. We expect that this is unlikely to be the case because of the large variability in exposures and conditions observed, because only two shops had previously carried out exposure monitoring and because of the similarity of our results to those of other investigators. The participation of machinists within the shops was excellent.
In summary, our findings suggest that grinding and enclosed CNC machine operations should be targeted for exposure controls and that increasing the height of shop roofs and mechanical ventilation may help to control exposures in small machine shops. The large proportion of variance in exposure explained by the models suggests that factors associated with machine tools and machine shops could be used for estimating exposures in risk analyses or epidemiological studies.
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed at: School of Occupational and Environmental Hygiene, University of British Columbia, Room 360A, 2206 East Mall, Vancouver, BC V6T 1Z3, Canada. Tel: +1-604-822-9577; fax: +1-604-822-9585; e-mail: kennedy{at}interchange.ubc.ca
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Buringh E, Lanting R. (1991) Exposure variability in the workplace: Its implications for the assessment of compliance. Am Ind Hyg Assoc J; 52: 6–13.[Web of Science][Medline]
Burstyn I, Teschke K, Bartlett K, Kennedy SM. (1998) Determinants of wheat antigen and fungal alpha-amylase exposure in bakeries. Am Ind Hyg Assoc J; 55: 313–20.
Chan TL, D’Arcy JB, Siak J. (1990) Size characteristics of machining fluid aerosols in an industrial metalworking environment. Appl Occup Environ Hyg; 5: 162–70.
Francis M, Selvin S, Spear R, Rappaport SM. (1989) The effect of autocorrelation on the estimation of workers’ daily exposures. Am Ind Hyg Assoc J; 50: 37–43.[Web of Science][Medline]
Greaves IA, Eisen EA, Smith TJ, Pothier LJ, Kriebel D, Woskie SR, Kennedy SM, Shalat S, Monson RR. (1997) Respiratory health of automobile workers exposed to metal-working fluid aerosols: respiratory symptoms. Am J Ind Med; 32: 450–9.[CrossRef][Web of Science][Medline]
Hallock MF, Smith TJ, Woskie SR, Hammond SK. (1994) Estimation of historical exposures to machining fluids in the automotive industry. Am J Ind Med; 26: 621–34.[Web of Science][Medline]
Hands D, Sheehan MJ, Wong B, Lick HB. (1996) Comparison of metalworking fluid mist exposures from machining with different levels of machine enclosure. Am Ind Hyg Assoc J; 57: 1173–8.[Web of Science][Medline]
Kennedy SM, Greaves IA, Kriebel D, Eisen EA, Smith TJ, Woskie SR. (1989) Acute pulmonary responses among automobile workers exposed to aerosols of machining fluids. Am J Ind Med; 15: 627–41.[Web of Science][Medline]
National Institute for Occupational Safety and Health. (1998) Criteria for a recommended standard. Occupational exposure to metalworking fluid. Cincinnati, OH: DHHS (NIOSH).
Piacitelli GM, Sieber WK, O’Brien DM, Hughes RT, Glaser RA, Catalano JD. (2001) Metalworking fluid exposures in small machine shops: an overview. Am Ind Hyg Assoc J; 62: 356–70.
Preller L, Heederik D, Kromhout H, Boleij JS, Tielen MJ. (1995) Determinants of dust and endotoxin exposure of pig farmers: development of a control strategy using empirical modelling. Ann Occup Hyg; 39: 545–57.
Robins T, Seixas N, Franzblau A, Abrams L, Minick S, Burge H, Schork MA. (1997) Acute respiratory effects on workers exposed to metalworking fluid aerosols in an automotive transmission plant. Am J Ind Med; 31: 510–24.[CrossRef][Web of Science][Medline]
Teschke K, Demers PA, Davies HW, Kennedy SM, Marion SA, Leung V. (1999) Determinants of exposure to inhalable particulate, wood dust, resin acids, and monoterpenes in a lumber mill environment. Ann Occup Hyg; 43: 247–55.
Thorne PS, DeKoster JA, Subramanian P. (1996) Environmental assessment of aerosols, bioaerosols, and airborne endotoxins in a machining plant. Am Ind Hyg Assoc J; 57: 1163–7.[Web of Science]
Tsai PJ. (1996) Worker exposure to nickel-containing aerosol in two electroplating shops: comparison between inhalable and total aerosol. Appl Occup Environ Hyg; 11: 484–92.
Woskie SR, Smith TJ, Hallock MF, Hammond SK, Rosenthal F, Eisen EA, Kriebel D, Greaves IA. (1994) Size-selective pulmonary dose indices for metal-working fluid aerosols in machining and grinding operations in the automobile manufacturing industry. Am Ind Hyg Assoc J; 55: 20–9.[Web of Science][Medline]
Woskie SR, Virji MA, Kriebel D, Sama SR, Eberiel D, Milton DK. (1996a) Exposure assessment for a field investigation of the acute respiratory effects of metalworking fluids. I. Summary of findings. Am Ind Hyg Assoc J; 57: 1154–62.[Web of Science][Medline]
Woskie SR, Virji MA, Kriebel D, Sama SR, Eberiel D, Milton DK, Hammond SK, Moure-Eraso R. (1996b) Exposure assessment for a field investigation of the acute respiratory effects of metalworking fluids. I. Summary of findings. Am Ind Hyg Assoc J; 57: 1154–62.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. A. Virji, S. R. Woskie, M. Waters, S. Brueck, D. Stancescu, R. Gore, C. Estill, and M. Prince Agreement between Task-Based Estimates of the Full-Shift Noise Exposure and the Full-Shift Noise Dosimetry Ann. Hyg., April 1, 2009; 53(3): 201 - 214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Park, P. A. Stewart, and J. B. Coble Determinants of Exposure to Metalworking Fluid Aerosols: A Literature Review and Analysis of Reported Measurements Ann. Hyg., April 1, 2009; 53(3): 271 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Lillienberg, A. Burdorf, L. Mathiasson, and L. Thorneby Exposure to Metalworking Fluid Aerosols and Determinants of Exposure Ann. Hyg., October 1, 2008; 52(7): 597 - 605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Suuronen, M.-L. Henriks-Eckerman, R. Riala, and T. Tuomi Respiratory Exposure to Components of Water-Miscible Metalworking Fluids Ann. Hyg., October 1, 2008; 52(7): 607 - 614. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. PETERS, W. A. HEITBRINK, D. E. EVANS, T. J. SLAVIN, and A. D. MAYNARD The Mapping of Fine and Ultrafine Particle Concentrations in an Engine Machining and Assembly Facility Ann. Hyg., April 1, 2006; 50(3): 249 - 257. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||