Back to Final Report Metalworking Fluids Standards Advisory Committee
7.1 GENERAL INFORMATION
Exposure assessment was discussed primarily at the seventh, eighth and ninth meetings and was discussed at other meetings. An Exposure Assessment Work Group was developed and the product of their work and an ad hoc work group is found at the end of this chapter. Exposure assessment includes the use of exposure limits, representative sampling strategies, initial and as needed repeat monitoring, statistical treatment of data, provisions for determining the need and termination of monitoring, standardized monitoring procedures with defined accuracy and precision, evaluation of controls and employee notification (Exposure Assessment Work Group, 1999). Exposure assessment can also employ qualitative assessment tools. This chapter focuses on exposure assessment, more on typical exposures can be found in Chapters Two and Three.
Dr. James d'Arcy, General Motors, spoke on his research on comparing methods, the development of the ASTM method, and how the ORC Document addresses exposure assessment (M8:10). Committee member, Dr. Dennis O'Brien provided an overview of air sampling and presented information for his NIOSH colleague, Dr. Robert Glaser (M7:12). Committee member, Dr. Frank Mirer provided a systems approach to exposure assessment (M2:12-13). Greg Piacitelli and Dr. Karl Sieber, NIOSH, discussed sampling methods in their presentations on the NIOSH Small Business Study. Dr. Daniel Goon, Castrol noted some sampling issues during a panel discussion. Dr. Eugene White discussed endotoxin sampling during the ninth meeting. The exposure assessment work group provided a recommendation on quantitative exposure assessment and an ad hoc work group provided a recommendation on qualitative exposure assessment that the committee reviewed. These two items are at the end of this chapter.
The evaluation of the work environment is done using exposure assessment, which can include qualitative and quantitative evaluations of exposure. The committee determined that the critical health concerns were nonmalignant respiratory effects and dermatitis, followed by cancer. As such, the focus of exposure assessment for MWFs is on the inhalation and dermal routes of entry. Air quality can be assessed qualitatively and quantitatively while the dermal route is usually assessed qualitatively.
O'Brien explained that air sampling addresses toxins that enter via the
inhalation route (M7:12). O'Brien noted that air sampling is done to determine the health risk and relate these results to dose-response (M7:12). Air sampling is also done to comply with regulations and monitor the performance of engineering controls (M7:12). Mirer explained that exposure assessment is a trigger to determine if any employer action such as more sampling, improved ventilation or medical surveillance is needed (M8:6). Howell noted that exposure assessment can relieve an employer of other obligations (M8:6). Sampling helps identify the high hazard jobs or tasks so we can determine ways to reduce these exposures (M7:12).
O'Brien stated for most air measurements, a known volume of air is drawn into a collection device by a pump and the contaminant is collected into or onto a medium (M7:12). The medium, such as a filter for aerosols (dusts and mists) or charcoal for gases, is analyzed in the laboratory (M7:12). Other options are available for gases and some direct reading light scattering instruments can be used for aerosols (M7:12). Diagrams and pictures of sampling equipment can be found in handouts (O'Brien, 1998; Piacitelli, 1998).
O'Brien highlighted air sampling rules such as having a defined and Iow limit of detection, collecting the appropriate amount of contaminant, and using the proper medium and correct flowrate (M7:12-13). He noted that some sampling strategies address identifying the maximum risk employee for each work operation (M7:13). He described homogeneous exposure groups (HEG) as those employees who would not have a significant difference from one another in their exposures because of the exposure mechanism (M7:13). Representative sampling can be done on some members of a HEG to represent the group (M7:13). Factors such as work hours and machine operating hours are variables to consider, according to Howell (M8:6).
The best air sampling protocol, according to O'Brien is full-period, consecutive sampling and next best is a full shift single sample (M7:13). Of less quality is partial period sampling, and grab sampling is the least desirable approach (M7:13). Time weighted averages (TWAs) are calculated from full-period, consecutive samples by multiplying each concentration by its time period, summing these and dividing by the total time (M7:13). Mirer explained that for a typical filter, sampling is done for eight hours (M2:12).
According to O'Brien, a ceiling limit is a level that is not to be exceeded at any point during the day (M7:13). A grab sampler or direct reading instrument may be able to detect excursions above a ceiling value with measurements taken during the period of maximum or peak exposure (M7:13). If 11 to 13 random grab samples are taken during the day, one would be 95% sure of collecting at least one sample in the top 20% of the exposure range (M7:13).
O'Brien explained that the frequency of sampling for regulatory purposes is usually based on an action level that is approximately half of the PEL (M7:13). Situations above the PEL may require quarterly sampling, while those between the action level and PEL may be done every 6 months, and below the action level, less frequently, if at all (M7:13). Mirer explained if one random sample on one day is above the action level, it is predicted that on any given day, exposures above the PEL will occur (M7:13).
Mirer noted the difference between area samples and personal samples, and that area values are usually higher than personal (M2:12). O'Brien recommended taking samples during different seasons (M7:13).
Sampling and analytical methods have different abilities to detect the substance of interest, d'Arcy described the limit of detection (LOD) as three times the standard deviation of the blank values (M7:17). The limit of quantification (LOQ) is ten times the standard deviation of the blank values (M7:17).
O'Brien described the importance of particle size and respiratory deposition in the ultimate toxicity of an aerosol (M7:13). O'Brien described the ISO/ACGIH definitions for particle size (M7:13). Particles that can reach any part of the respiratory tract are defined as inhalable (M7:13; ACGIH, 1999). Those that can enter the trachea and reach the airways and lung are called thoracic (M7:13; ACGIH, 1999). Particles less that can reach the air exchange region are respirable (M7:13; ACGIH,1999). The equations that describe the three particulate mass fractions and tables noting collection efficiencies by size are found in the ACGIH TLV booklet (ACGIH, 1999)
7.4.1 Experiences and Resources Related to Qualitative Exposure Assessment
d'Arcy described qualitative exposure assessment, noting the importance of answering the questions who is exposed, where and how they are exposed and what controls exist (M8:10). Mirer described information gathering at a plant, e.g. previous sampling records, floor plans, ventilation system diagrams and testing results (M2:12).
Dermal exposure can only be assessed qualitatively and work practices are a key component (M8:10). Qualitative observations may lead to quantitative exposure assessment (M8:10). Qualitative exposure assessment is part of the ORC Document's approach, according to d'Arcy (M8:10).
7.4.2 Experiences and Resources Related to Quantitative Exposure Assessment
220.127.116.11 Concepts in Quantitative Exposure Assessment for MWF Environments7.4.3 Measurements done in the Different Studies
NIOSH has used gravimetric and some infrared methods in its earlier studies, according to O'Brien (M7:14).
Piacitelli explained that the NIOSH Small Business Study used a variety of methods to assess the environment including: the NIOSH 0500 method, the ASTM PS 42-97 extraction method, thoracic sampling, impactor sampling, an electrostatic precipitator and a direct reading light scattering device (M4:1). Vapor sampling was not done (M4:1). Piacitelli explained that for the NIOSH 0500 method, they used a Teflon filter (M4:1).
The GM/UAW studies used impactors and gravimetric analysis, according to O'Brien (M7:14). According to Mirer, the studies by Kennedy, Greaves, Robbins and Rosenman as well as Kriebal use total particulate (M7:20).
7.4.4. Additional Information from the NIOSH Criteria Document
Chapter seven of the NIOSH Criteria Document provides information on the sampling and analysis of MWFs (NIOSH, 1998). A compilation of methods used at the time the document was written is found in Table 7-1 of the Criteria Document (NIOSH, 1998). Details about thoracic samplers are found in section 7.2.2 (NIOSH, 1998). A discussion of biases such as sampler inlet bias is found in sections 7.2.3 through 7.2.5 (NIOSH, 1998). Definitions of terms used in the determination of precision and accuracy are found in sections 7.2.6 through the end of the NIOSH Criteria Document chapter seven (NIOSH, 1998). The discussions of the committee used the NIOSH information as a baseline and added new information to the material available on exposure assessment for MWFs.
7.5.1 Size of Business
Mirer recommended that OSHA send out a manual to small and medium size businesses on how to do air sampling and exposure assessment (M7:20). There was concern about the cost of sampling and analysis for small business and this is addressed in Chapter Four, Economic Feasibility. Letters from some small businesses indicate a concern about hiring professionals to do air sampling (PMPA,1999; PMA 1999). Some members noted that many small facilities would not need to do air sampling if a good qualitative assessment tool was available.
7.5.2 Sampling Problems
Each sampling and analytical method has its limitations. O'Brien noted that NIOSH has an accuracy criteria of +/- 25% of the true mean, 95% of the time which translates to a coefficient of variation of less than 12.8% (M7:14).
Mirer explained problems such as particle entry losses and evaporation (M2:12). The cassette method may under-sample according to O'Brien (M7:15). The "total" particulate sampler has been viewed by many aerosol scientists as inadequate to measure inhalable aerosol, according to Sheehan (M7:16). Other inhalable samplers collect more due to better inlet design and weighing of the whole device not just the filter (M7:16). A variety of articles by Vincent and Baron and others addressing the limitations of aerosol sampling are found in the American Industrial Hygiene Association Journal, Applied Environmental and Occupational Hygiene and Aerosol Science and Technology.
Mirer cited Leith's work stating that half the oil evaporates from the filter (M2:12). Mirer noted that Leith's data shows that the vapor phase is about equivalent to the particulate because Leith stated that about half of the total evaporated (M7:20).
O'Brien explained that gravimetric analysis is simple and consistent historically but subject to evaporation and is not specific (M7:14). It discriminates against dry machining methods (M7:14).
O'Brien explained that extraction differentiates MWFs from dust and is historically consistent with some datasets (M7:15). It is limited depending on the solvents used and is more expensive (M7:15). Sheehan and Howell hoped that if a fluid formulator had a fluid that could not be easily extracted by the ASTM PS42-97 method, the manufacturer would indicate this in their literature (M7:21).
According to O'Brien, infrared has calibration problems (M7:15). Impactors cost more and the sample is split, lessening sensitivity (M7:15). Precipitators are not well known (M7:15).
Direct reading instruments cost more and are more complex. Some limitations of light scattering according to O'Brien are: difficulty standardizing the response of the instrument and concerns about size, shape and refractive index effects (M7:14).Teitelbaum thought it may take more skill to assess a peak value than a TWA (M7:14). A well calibrated direct reading instrument may be acceptable, based on Abrams work, according to Mirer (M7:20).
O'Brien noted that people have unrealistic expectations of the accuracy and precision of instruments (M7:15). He noted variability during the day and Mirer thought the variability in exposures is much greater than the variability due to the method (M7:15,21). Mirer thought that the amount collected by any method is about 1/5 to 1/10 what the operator really receives (M4:3).
Lick disagreed with Mirer's estimates (M4:3). Lick stressed the importance of limiting variability (M4:3). A minor shift in results around 0.5 mg/m3 could cost millions of dollars, according to Lick (M4:3). As the exposure limit drops, variability becomes more important (M3:4).
d'Arcy agreed with Lick and noted that if the coefficient of variation is 0.2, it is very difficult to prove you are less than 0.5 mg/m3 (M4:3). d'Arcy showed results of a study he did of the GM contract labs (M7:17). In this study, he submitted 20 blind blanks to each lab (M7:17). The three labs produced LOQs of 0.18, 0.25 and 0.34 mg/m3 while the NIOSH 0500 predicts an LOQ of 0.12 (M7:17). d'Arcy noted that the 0500 method was developed for nuisance dust with a PEL of 15 mg/m3 and that the lowest concentration recommended for the 0500 method is 0.75 mg/m3 (M7:17) By pushing below what the method was designed to do, we cannot expect the same quality, according to d'Arcy (M7:17) The NIOSH 0500 method needs to be re-evaluated for lower concentrations (M7:17)
7.5.3 Background Values
Mirer emphasized the importance of knowing what to measure, and that cross contamination can be a health and measuring problem (M2:12). Mirer explained that assembly areas can receive some of the cross contamination depending on location and workplace layout (M2:12). Assembly areas are not zero and should be measured (M2:12). Outdoor levels are also a concern, according to Burch (M2:12).
Mirer noted that the background levels in the NIOSH Small Business Study were very Iow compared to data collected in auto plants (M4:2). A UAW/GM study showed the average background level of MWFs when only flumes were running was 0.11 mg/m3 as total particulate. Another survey showed workers with exposures of about 0.5 mg/m3 when their machine was not operating and only surrounding machines affected the exposure (M7:20). Another study Mirer showed, measured exposure and then shut off the ventilation (M7:20). It showed that ventilation reduced exposure to background values but that background values were around 0.5 mg/m3 (M7:20). Mirer emphasized the importance of assessing direct exposure plus the background to fully evaluate the effectiveness of controls (M7:20).
O'Brien estimated background as 0.1 mg/m3 if production is off and flumes were off (M7:20). In EPA non-attainment areas, outside particulate levels can be 0.07 mg/m3, according to O'Brien (M7:20). Mirer noted outdoor averages of 0.03 mg/m3 with excursions to 0.07 mg/m3 (M7:20).
Additional information is provided in the handout on Air Sampling for Source Identification (UAW, 1999). The Occupational Exposure Sampling Strategy Manual is an additional resource (Leidel et al, 1977).
7.5.4 Appropriate Metric for the Health Effects
O'Brien explained that if cancer was the basis of a regulation, inhalable full shift samples would be the air monitoring method (M7:11,17). If any regulation was based on respiratory effects, he viewed that thoracic particulate would be the monitoring method (M7:11). d'Arcy agreed, noting that we should also consider the respirable portion (M7:17). O'Brien thought thoracic sampling both as full shift and peak should be considered (M7:15)
Wegman stated it was not clear to him that the respiratory effects that have been identified are linked to respirable or thoracic particulate (M7:11). He stated we do not know how asthma gets triggered and we do not know the route of entry for the agent that causes HP (M7:11).
Sieber noted that total particulate sampling is more common and widely available for use than thoracic (M7:16). d'Arcy thought there were enough thoracic samplers available (M7:18).
Sieber viewed that the health effects were better related to thoracic (M7:16). Previous studies showed a very high correlation between thoracic and total, according to Sieber (M7:16).
Robins explained that both the bacterial particulate and thoracic particulate fit the health effects well in his study (M5:9). O'Brien noted that bacterial testing or endotoxin testing may be a way to also determine the effectiveness of a fluid management program (M5:9). Robins viewed the endotoxin testing as very expensive and difficult (M5:9).
O'Brien explained that the values in the NIOSH Small Business study for area and personal samples were not that different (M7:15). Sieber explained that these values were not statistically significantly different (M7:16). O'Brien opined that the mist size may make exposure homogeneous in a shop (M7:15). Burch noted that OSHA regulates personal exposure, not plant area levels (M7:14). O'Brien viewed area sampling as a useful supplement to personal sampling (M7:15).
Mirer viewed that there was no health basis for using extractables (M7:20). There is no health basis for concern about vapor, according to Mirer (M7:20). Howell agreed with Mirer that all the epidemiological studies used thoracic particulate without extraction (M7:88).
O'Brien noted that in the Small Business study, a peak of 2 mg/m3 corresponded to a TWA of 0.5 mg/m3 (M7:14). O'Brien explained that peak values may be associated with respiratory responses (M7:14).
Who does the sampling was discussed in the context of exposure assessment as well as cost. More on this topic is in Chapter Four on economic feasibility. Recommendations were made to have industrial hygienists set up a sampling program and workers at a plant could conduct sampling. Burch was concerned that OSHA had a bias against employers doing their own sampling (M7:21). Teitelbaum noted that if people other than industrial hygienists were taking samples, training would be essential (M7:25).
Sampling strategies are discussed below. Additional information on sampling statistics is provided in Leidel, 1977.
Mirer viewed any standard as feasibility limited not analytically limited (M7:20). Mirer stated it did not matter what method was used just that the method is consistent (M4:3). Mirer thought if the thoracic is used, the analysis should be for total mass since it represents what actually gets to the lung target (M4:3). Mirer thought due to the relationship between total and extractable, if a standard is based on extractable it should be adjusted down (M7:20). Mirer viewed the extractable method as appropriate if other particle sources are present (M8:24). Mirer noted the historical value of the total particulate and thought the thoracic sampler was too expensive (M8:24).
Mirer thought having a statistically valid air sampling program could be an innovative compliance issue (M7:21). If an employer had good reason to believe he was in compliance, but had one or two OSHA samples out of compliance, that this could be taken into account in the enforcement (M7:22).
Howell recommended giving the widest choice possible to allow users to choose how to get the best information (M7:22). Mirer did not have a problem with this (M7:22). Howell noted the importance of using air sampling to evaluate the effectiveness of controls (M7:21). As a result, Howell viewed the ASTM method as best practice because it allows measurement of total but also helps define the source of the problem by measuring extractable (M7:21).
There was concern noted in other sections of this report about the ability of any method to effectively measure an action level of 0.25 mg/m3. O'Brien thought that with a limit of quantification usually less than 0.2 mg/m3, an action level of this value or 0.25 mg/m3 could be measured (M7:19).
Mirer explained that an air sample is an index of exposure, not a complete exposure determination (M7:20). Howell, Teitelbaum and Mirer agreed that at best, any method is just an indicator because of the complex, dynamic nature of the fluid (M8:26).
In response to the various concerns of the committee, the Exposure Assessment Work Group presented a draft recommendation (M8:23). Their draft provided the opportunity to use one of four sampling and analytical methods: total, thoracic, extractable or a direct reading instrument (M8:23). Ratios were given to show the relationships between different methods (M8:23). As proposed, an employer would be in compliance if 95% of samples in an exposure group are within the PEL (M8:23). Substantial changes in production would be a trigger for additional monitoring (M8:23).
The draft document provided recommended relationships between a PEL, action level and STEL (M8:23). Wegman, Howell and Anderson viewed there was not enough health data to support a STEL and that simply using a multiplier times the PEL was not appropriate (M8:23).
Mirer noted that the work group's recommendation was flexible (M8:24). Frederick recommended a data call in from OSHA in increments of five years, this could be used to determine the effectiveness of an OSHA standard (M8:24). Sherman requested that the exposure assessment group provide a rationale for a 30 year recordkeeping requirement (M8:25).
Note - the remaining part of this page is blank, slightly different formatting is used on the remaining pages of this chapter.
Recommendations for best practices quantitative exposure assessment and exposure monitoring were proposed to the full committee by an exposure assessment working group. A written text was reviewed by the full committee at the May meeting.
The importance of fluid management and the potential difficulties for small business to do exposure assessment were appreciated by the group. As a result, an ad hoc work group drafted a qualitative observational assessment to integrate with quantitative requirements.
The full committee unanimously endorsed the written recommendation as a best practice for exposure monitoring for workers exposed to metal working fluids, with some reservations regarding the action level recorded in the transcript and summarized below. The full committee unanimously endorsed the observational checklist. The committee did not separately vote on this proposal as an enforceable OSHA regulatory text.
The exposure assessment group accepted the NIOSH criteria document Chapter 7, "Sampling and Analytical Methods," as the starting point for determining best practices for exposure assessment. This information was supplemented by extensive additional work including the NIOSH small business study, and the development of ASTM-PS42. The NIOSH criteria document recommends a limit based on the thoracic fraction analyzed by gravimetric methods. NIOSH proposes that the thoracic size fraction would be 80% of the total particulate fraction.(1)
The text for exposure monitoring consisted of "boiler plate" OSHA standard language, derived from the formaldehyde standard (29 CFR 1910.1048), with new ideas for changes from the standard practice highlighted in the committee draft and in the attached text. These issues included: omission of a STEL or excursion limit; statistical compliance evaluation; definition of qualitative observations which would constitute objective evidence for initial determination of the need for initial monitoring; and provision for multiple air sampling and analytical methods as appropriate.
The written recommendation is attached.
The discussion and vote did not directly address the value for the PEL.
Key issues in the recommendation follow.
7.7.1. Method of Analysis. The Committee recommends that an employer could rely on samples analyzed by extraction methods to demonstrate compliance with the PEL, or to determine the need for additional engineering or other controls. Extraction analysis would be permitted where the employer could demonstrate there were sources of particulate other than MWF processes in the work area where the sample was collected. The Committee also agrees that it would be desirable to allow employers to use the cheaper gravimetric method, or direct reading instruments to demonstrate compliance. The Exposure Assessment group noted that the extraction method would permit the employer to stop the process at the weighing (gravimetric stage) if these results demonstrated compliance.
The Full Committee and the Exposure Assessment Group maintain a range of views on the appropriate method of analysis.
The consensus of the exposure working group was that side by side samples analyzed by extractable methods would not provide a higher exposure value than the gravimetric samples, and likely would provide a lower exposure value than gravimetric samples. The extent of the deviations is not known, but the consensus was that it could be substantial in areas where dry grinding or other exposure sources were present. Some argued that the respiratory effects studies used for exposure response assessment were based on exposures measured by gravimetric analysis (of thoracic samples) and so a health based standard should reflect the analytical method. In addition, gravimetric analysis was known to be substantially cheaper than extraction methods. Others argued for the specificity of extraction methods, and possible improved analytical precision of the method. They noted that extraction methods could also provide a total gravimetric result.
The consensus of the exposure assessment working group was that exposure assessment using a properly calibrated direct reading instrument (real-time aerosol monitor) was highly desirable and yielded data on both short term exposures and sources which would not be provided by standard filter sampling methods. Such measurements were to be encouraged. However, the group was divided on the availability of equipment and persons trained to interpret the results, and felt it could not require such measurement methods.
7.7.2. Size Selective Sampling. The Committee recommends that the PEL be stated as either a total particulate sample, or a thoracic fraction sample at 80% of the total particulate level.
The NIOSH criteria document recommends a limit based on the thoracic fraction for protection against respiratory effects. NIOSH proposes that the thoracic size fraction would be 80% of the total particulate fraction(2). Thus, the recommendation was 0.5 mg/m3 total particulate and 0.4 mg/m3 thoracic.
The consensus of the full committee was that a thoracic fraction was more appropriate for a standard to prevent respiratory effects, and that an inhalable fraction would be more appropriate for a standard to prevent cancer effects. However, neither thoracic nor inhalable sampling devices are commonly used in current industrial hygiene practice. Thoracic sampling would be more costly. The quantitative relationship between total and thoracic will be variable, depending on exposure circumstance. Principally, close to exposure sources with large sized particulate emissions, inhalable and total samples will exceed thoracic by a larger amount than samples collected farther away.
7.7.3 Action Limit. The majority of the Committee recommends an action limit of 1/2 the PEL for the purposes of triggering continuing monitoring.
The rationale for the action limit for exposure assessment, which is common to all OSHA chemical exposure standards, lies in the variability of exposure levels in all studied industrial processes. A random sample as high as ½ the PEL, collected from a work area, predicts that exposures over the PEL will occur. Continuing air sampling would be needed to determine with certainty that PEL compliance was achieved. In addition, continuing exposure surveillance in a work area where exposures are close to the PEL should be maintained to insure that controls do not deteriorate and exposures increase to exceed the PEL.
A minority of Committee felt that an action limit of 0.25 mg/m3 would impose a very large amount of continuing air sampling, and that this value posed analytical accuracy problems by gravimetric and extraction methods, but not when using direct reading instruments.
7.7.4. Short Term Exposure Limit. The committee as a whole did not recommend a short-term exposure limit. Discussion was divided on this issue, although the ultimate vote was unanimous.
The prevailing view was that existing sampling methods would not support short-term exposure measurements based on a 30 minute STEL as high as 2.0 mg/m3 In addition, it was felt that studies directly showing additional adverse effects of short term high exposures were not sufficient to support such a recommendation. Further, it was argued that evidence was insufficient to support a conclusion that a STEL would furnish additional feasible protections beyond those of a TWA exposure limit.
The minority view was that a STEL would be needed to protect workers if an exposure limit of 0.5 mg/m3 were adopted, because material impairment to health has been observed among workers exposed to levels less than the TWA. They argued that available studies did not measure peak exposures within that TWA, but that other evidence supported conclusions that the peaks were there. A substantial fraction of the observed respiratory effects of MWF appear to be reactive airway responses. By analogy to many other substances, and also in the experience of respiratory physicians, it would be likely that peak exposures or excursions were important in causing these effects. Proponents of the STEL noted that the ACGIH recommends excursion limits as a general practice:
Excursions in worker exposure may exceed the TLV-TWA for no more than 30 minutes during a workday, and under no circumstances should they exceed the TLV-TWA, provided that the TLV-TWA is not exceeded.
7.7.5 Statistical Compliance. The Committee recommends a statistical compliance scheme as a departure from existing OSHA regulatory practice.
The committee intended to address a feasibility concern with this new approach. The concern is whether a few outlier samples would trigger installation of additional engineering controls in a work area which was generally well controlled. It is intended that an employer could rebut an OSHA citation for a PEL (or Action Level) violation by showing that 95% of samples within a homogeneous exposure group were in compliance. This would apply to the most highly exposed homogeneous exposure group in the work area. This is largely a quantification in regulatory text for a practice which already exists in the field. It was suggested that OSHA take this concept in to account in feasibility determination.
The consensus did not intend that OSHA would have to show that 5% of samples were out of compliance.
7. 7.6 Exception from Initial Monitoring
The committee recommends that a clear meaning be given to the existing allowance for use of "objective data" to determine that initial monitoring not be required. The committee recommends a checklist approach in which qualitative observations of the production process and systems approach could be used to predict whether exposures above the action level or PEL are foreseeable.
The committee believes that many small, Iow volume, or well-ventilated machining operations do not create exposures above an appropriate exposure limit. By providing detailed observations which could be used to support a determination, the committee believes that employers who maintain such operations will be able to avoid the effort of collecting air samples to confirm that no further action is necessary. This provision would primarily benefit small employers with few professional industrial hygiene resources. However, it will also benefit employers with isolated machining operations which support other production activities.
The exposure assessment and ad hoc qualitative assessment groups developed a checklist which is provided after the quantitative exposure assessment document. The committee believes that this checklist represents the general consensus of the MWF community on the observations which would trigger the need for quantitative exposure assessment. It is supported by the observations and experience of the team which conducted the NIOSH Small Business study. The committee concedes that this checklist should be validated in the field, and that alternative weighting of responses may be plausible. Issues of concern for some members of the committee included: wording, weighting, validity and recordkeeping requirements. The committee further notes that an employer would not be required to use this checklist as the sole objective evidence that quantitative exposure assessment was not needed.
The quantitative exposure assessment best practice and the qualitative checklist are found on the following pages.
Best Practices Description for Monitoring and Analysis of MWF Exposures(1)
(c) Permissible Exposure Limit (PEL).(2)
(1) TWA. The employer shall assure that no employee is exposed to an airborne concentration of MWF which exceeds 0.X mg MWF per cubic meter of air (0.X mg/m3) (or 0.8 [0.X ]MG MWF PER CUBIC METER OF AIR (0.X MG/M3 ) MEASURED AS THORACIC FRACTION) as an 8-hour TWA.
(3)ACTION LEVEL. THE ACTION LEVEL FOR THE PURPOSE OF INITIATION OR TERMINATION OF MONITORING WILL BE ½ THE PEL AS A TWA.(3)
(d) METHOD OF ANALYSIS
(1) THE EMPLOYER MAY MEASURE EMPLOYEE EXPOSURE USING THE FOLLOWING METHODS:
(i) TOTAL GRAVIMETRIC;
(ii) TOTAL EXTRACTABLE IF THE EMPLOYER CAN DEMONSTRATE THAT A WORK AREA HAS SOURCES OF PARTICULATE EMISSIONS UNRELATED TO MWF EMITTING PROCESSES;
(iii) THORACIC GRAVIMETRIC; OR
(iv) REAL TIME AEROSOL MONITORING CONSISTENT WITH PARAGRAPH (d)(2) OF THIS SECTION.
(2) REAL TIME AEROSOL MONITOR. (i) THE EMPLOYER MAY RELY ON SAMPLING RESULTS FROM A REAL TIME AEROSOL MONITOR IF THE INSTRUMENT IS CALIBRATED AGAINST SAMPLES COLLECTED AND ANALYZED BY ONE OF THE OTHER METHODS DESCRIBED IN PARAGRAPH (C) OF THIS SECTION IN THE WORK AREA WHERE THE SAMPLES ARE COLLECTED.
(ii) IF THE EMPLOYER CAN DEMONSTRATE THAT A WORK AREA HAS SOURCES OF PARTICULATE EMISSIONS UNRELATED TO MWF EMITTING PROCESSES, THE EMPLOYER MAY RELY ON EXTRACTION ANALYSES TO CALIBRATE REAL TIME AEROSOL MONITORS.
(3) REPRESENTATIVE SAMPLING. (i) AN EMPLOYER SHALL BE DEEMED IN COMPLIANCE WITH THE REQUIREMENTS OF PARAGRAPH (C) IF THE EMPLOYER CAN DEMONSTRATE BY A STATISTICALLY VALID REPRESENTATIVE AIR SAMPLING SCHEME THAT 95% OF AIR SAMPLES WITHIN EACH HOMOGENEOUS EXPOSURE GROUP ARE WITHIN THE PEL OR ACTION LEVEL. HOMOGENEOUS EXPOSURE GROUP MEANS EMPLOYEES PERFORMING ESSENTIALLY IDENTICAL TASKS AT ESSENTIALLY IDENTICAL WORK STATIONS.(4)
(ii) When an employee's exposure is determined from representative sampling, the measurements used shall be representative of the employee's full exposure to MWF.
(iii) Representative samples for each job classification in each work area shall be taken for each shift unless the employer can document with objective data that exposure levels for a given job classification are equivalent for different work shifts. A WORK AREA FOR THE PURPOSES OF THIS SECTION IS THE IMMEDIATE VICINITY OF METALWORKING PROCESSES AND ASSOCIATED EQUIPMENT, WHERE PRODUCTION ACTIVITIES, MAINTENANCE, SERVICE, AND IN PROCESS INSPECTION ARE PERFORMED, AS WELL AS IMMEDIATELY ADJACENT AREAS WHICH ARE NOT SEPARATED FROM DIRECT EXPOSURE BY PHYSICAL BARRIERS TO MOVEMENT OF AIR.
(e) Exposure monitoring.
(1) General. (i) Each employer who has a workplace covered by this standard shall monitor employees to determine their exposure to MWF.
(ii) Exception. Where the employer documents, using objective data, that the presence of MWF EMITTlNG PROCESSES or MWF-releasing products in the WORK AREA cannot result in airborne concentrations of MWF that would cause any employee to be exposed at or above the action level under foreseeable conditions of use, the employer will not be required to measure employee exposure to MWF. THE EMPLOYER MAY RELY ON THE METHODS AND OBSERVATIONS IN THE NON-MANDATORY APPENDIX TO DETERMINE WHETHER INITIAL MONITORING IS NEEDED.(5) (2) Initial monitoring. (i) The employer shall identify all employees who may be exposed at or above the action level and accurately determine the exposure of each employee so identified.
(ii) Unless the employer chooses to measure the exposure of each employee potentially exposed to MWF, the employer shall develop a representative sampling strategy and measure sufficient exposures within each job classification for each workshift to correctly characterize and not underestimate the exposure of any employee within each exposure group.
(iii) The initial monitoring process shall be repeated each time there is a SUBSTANTIAL change in production, equipment, process, personnel, or control measures which may result in new or additional exposure to MWF.
(iv) If the employer receives reports of signs or symptoms of respiratory or dermal conditions associated with MWF exposure, the employer shall promptly monitor the affected employee's exposure. Consideration should be given to other than airborne mechanisms of exposure.
(3) Periodic monitoring. (i) The employer shall periodically measure and accurately determine exposure to MWF for employees shown by the initial monitoring to be exposed at or above the action level.
(ii) If the last monitoring results reveal employee exposure at or above the action level, the employer shall repeat monitoring of the employees at least every 6 months.
(4) Termination of monitoring. The employer may discontinue periodic monitoring for employees if results from two consecutive sampling periods taken at least 7 days apart show that employee exposure is below the action level. The results must be statistically representative and consistent with the employer's knowledge of the job and work operation.
(5) Accuracy of monitoring. Monitoring shall be BY A METHOD AT LEAST AS ACCURATE, AS NIOSH 0500 OR ASTM PS-42 at the 95 percent confidence level, to within plus or minus 25 percent for airborne concentrations of MWF at the TWA and the STEL and to within plus or minus 35 percent for airborne concentrations of MWF at the action level.(6)
(6) Employee notification of monitoring results. Within 15 days of receiving the results of exposure monitoring conducted under this standard, the employer shall notify the affected employees of these results. Notification shall be in writing, either by distributing copies of the results to the employees or by posting the results. If the employee exposure is over the PEL, the employer shall develop and implement a written plan to reduce employee exposure to or below the PEL, and give written notice to employees. The written notice shall contain a description of the corrective action being taken by the employer to decrease exposure.(7)
(7) Observation of monitoring. (i) The employer shall provide affected employees AND their designated representatives an opportunity to observe any monitoring of employee exposure to MWF required by this standard.
(ii) When observation of the monitoring of employee exposure to MWF requires entry into an area where the use of protective clothing or equipment is required, the employer shall provide the clothing and equipment to the observer, require the observer to use such clothing and equipment, and assure that theobserver complies with all other applicable safety and health procedures.
The purpose of this qualitative self assessment is to determine the need for quantitative air monitoring of employee exposures to metalworking fluid mist. The questions in this checklist are organized by three classifications. The answers to the "A" questions are critical. The answers to the "B" questions are important and the answers to the "C" questions denote recommended actions. In summary:
"A" questions refer to critical elements. For example: Has responsibility for MWF management been assigned?
"B" questions are important elements. For example: Are machine enclosures maintained in operating condition?
"C" questions denote recommended practices. For example: Are machines maintained in clean condition?
The results are calculated as follows:
For "A" questions, an answer of "no" to any of the questions results in a requirement to conduct representative quantitative air monitoring of employee exposures to metal working fluid mist.
For "B" questions, if at least seventy five percent (75%) of the answers to the "B" questions are answered "yes" there is no need for employee exposure monitoring. A response rate of fifty to seventy four percent (50-74%) "yes" on the important questions is "marginal" and quantitative employee exposure monitoring is recommended for a representative number of workers in the areas reflective of the deficient elements. A response rate of less then fifty percent (<50%) "yes"on the important questions indicates that representative employee exposure monitoring for metal working fluid mist should be conducted.
"Yes" answers to the "C" questions are often good indicators of the quality of the overall metal working fluid exposure conditions in the facility. Answers to these questions are not used to determine the need for employee exposure monitoring.
[Note: Much of the material included in this checklist was adapted from similar checklists in the ORC Guide to Controlling the Metal Removal Fluid Environment.]
Plant: ____________________ Operation: ____________________
Department: ____________________ Bay/Column: ____________________
Completed by ____________________
Instructions: Place a check in the appropriate boxes below. If "No" is checked, make comments and recommend corrective actions if possible. If an item does not apply to the plant/department/operation, check the "NA" box.
Detail corrective actions or comments in the space provided.
Footnote 1 The exposure assessment group noted that "total" particulate is actually a misleading term. "Total" means a sample collected with a 37 mm closed face filter cassette, or equivalent. The cover is now known to exclude large particles, and in some cases may radically understate the true exposure. Paradoxically, "inhalable" particulate collected with an open faced sampler may exceed "total" by 3-fold in some environments, which "total" exceeds "thoracic" by only 20%. (Back to text)
Footnote 2 The exposure assessment group noted that "total" particulate is actually a misleading term. "Total" means a sample collected with a 37 mm closed face filter cassette, or equivalent. The cover is now known to exclude large particles, and in some cases may radically understate the true exposure. Paradoxically, "inhalable" particulate collected with an open faced sampler may exceed "total" by 3-fold in some environments, which "total" exceeds "thoracic" by only 20%. (Back to text)
Footnote 1 Note: Text was derived from the OSHA Formaldehyde standard. Appropriate substitutions of MWF and mass levels were made in the Formaldehyde text. CHANGES FROM THE ORIGINAL STANDARD TEXT ARE IN CAPS. Deletions from the original standard text are not shown.
The PEL is arbitrarily described as "O.X mg/m3" without stating the PEL. The STEL found in the Formaldehyde standard has been omitted based on discussions in the Committee Recommendations. This proposal assumes that there might be different action levels (or triggers by another name) for training, medical surveillance or other purposes than for monitoring exposure.
This proposal allows for 3 different analytical methods, and for the use of a properly calibrated real time aerosol monitor. The proposal retains the 80% conversion factor for closed face total to thoracic fraction proposed by NIOSH. The proposal permits employers to rely on total aerosol analyzed by extraction methods if they can demonstrate there are sources of particles in the work area unrelated to MWF emissions. (Back to text)
Footnote 2 Effective dates are not contained in this section. Phase in periods might be different for new and existing equipment, or for different sized workplaces. (Back to text)
Footnote 3 Medical surveillance or training might be triggered by different exposure levels than the action level for monitoring. The need for an action level for exposure monitoring is derived from knowledge of time and analytical variation of exposure measurements. (Back to text)
Footnote 4 Employees within the homogeneous exposure group would be deemed to have the same exposure as the representative sample. (Back to text)
Footnote 5 Use of the checklist would relieve an employer from citation for failure to conduct initial monitoring, and periodic monitoring, if OSHA were to enter the workplace and collect samples which exceeded the action level. However, the employer could still be cited for violation of all other requirements of the standard triggered by exposure level if the action level or PEL were exceeded. (Back to text)
Footnote 6 The Committee believes it is better to specify methods which meet the generally applied criteria for accuracy of monitoring, rather than state statistical criteria which are not readily understood by employers and employees. (Back to text)
Footnote 7 Where representative sampling is used, each employee within a homogeneous exposure group shall be provided the notifications applicable to that group. (Back to text)