Exposure Assessment Strategies: Something for Everyone

​​Exposure Assessment Strategies: Something for Everyone

Monday, June 1, 2015, 10:30 AM - 1:30 PM

SR-106-01 On the Accurate Estimation of the Benchmark Dose

T. Mathew, University of Maryland Baltimore County, Baltimore, MD

Objective: In industrial hygiene and environmental applications, the benchmark dose (BMD) is defined as the dose associated with a specified increase in the probability of an adverse health outcome. The probability will depend on relevant covariates and can be modeled using functions such as logistic regression, probit model, multi-stage model, etc. When binary dose-response data are available, the EPA’s benchmark dose software can be used for estimating the BMD, and for computing a 95% lower confidence limit (denoted by BMDL). The BMDL is of particular interest, since it can be used as a point of departure for developing occupational exposure limits (OELs). Likelihood based large sample theory is usually used for computing the BMDL. However, sample sizes are very often small, in which case the likelihood based BMDL will be inaccurate. Our objective is to explore alternative methodology for the accurate computation of the BMDL.

Methods: The methodology that we shall explore for the accurate computation of the BMDL is an old an under-utilized approach in statistics, namely, the fiducial approach. The fiducial concept has undergone a recent revival, and is now recognized to be important in applications. The fiducial method can be easily adopted to the logistic regression model, probit model, etc. The fiducial methodology and its implementation will be explained for computing the BMDL. A bootstrap correction will also be explained for improving the accuracy for small samples.

Results: Simulation results show that the proposed methodology does provide an accurate BMDL. Apart from simulation results, the methodology will be illustrated using the analysis of industrial hygiene data sets. 

Conclusions: A 95% BMDL computed using the likelihood based approach can have coverage probability significantly smaller than 95%, especially for small samples. This will result in a larger BMDL, giving the misleading conclusion that the BMD is larger than what it actually is. The proposed fiducial approach is thus preferable, since the BMDL so computed does provide the correct coverage. 

CS-106-02 Professional Judgment in Exposure Assessment (EA): How Low Do You Go?

T. Morris, Morris Innovative IH & Safety Solutions, Cincinnati, OH

Situation/Problem: IHs rely on many factors to guide us, especially professional judgment. When evaluating materials for an EA, professional judgment typically leads to the rejection of those at low levels (<1%) because it’s assumed there isn’t enough to pose a risk. This was reinforced by the HAZCOM standard which defined an untested mixture’s health hazards as those of the individual components at ≥1% and ≥0.1% for carcinogens. Cut-offs continue in the revised standard but OSHA still includes a requirement that any material/quantity be reported if it could pose a health hazard.

Resolution: The 0.1/1% values have (for many) become de facto limits for SDSs and EAs. This coupled with professional judgment targeting the “primary” contaminants of a process have resulted in missed exposures to other materials. Some may be known components that were ignored due to a low %, unknown materials, or materials whose exposure potential had not been adequately evaluated. These unsuspected exposures can surprise employers during an OSHA inspection and confound the etiology of health effects associated with toxicant/industry exposures.

Results: Spraying a coating containing 0.06% formaldehyde resulted in an exposure 220% of OSHA limits; Cd exposures up to 288% of OSHA limits occurred at four nonferrous foundries using zinc containing 0.0002% Cd; Pb/Cd exposures of 185/15.4 µg/m3 occurred from dumping bags of ZnO containing 0.005% Pb and 0.001% Cd; and plasma cutting bare 304 stainless steel (“0”% Pb) resulted in a 75.1 µg/m3 Pb exposure. All companies were unaware of the contaminants, the exposures, and the associated regulatory requirements.

Lessons Learned: Significant exposures occurred to materials present at levels well below what professional judgment may indicate is relevant. Cd exposures were concomitant with Pb and are a previously unreported finding that must now be considered when evaluating kidney toxicity in nonferrous foundry workers. Professional judgment must be calibrated to new information which, as shown here, includes reporting low level components and incorporating them into the exposure assessment process. This is a challenge that covers all aspects from material composition, detailed task analysis to sampling and analytical method selection.

CS-106-03 Similar Exposure Groups (SEGs) in Hydraulic Fracturing

R. Rottersman, ENVIRON International Corp., Chicago, IL

Situation/Problem: In 2012, it was estimated that over 360,000 people were employed in the drilling/production of shale gas and that this number would increase to over 600,000 by the end of the decade. There have been a number of studies that have evaluated occupational exposures related to the fracturing process. In addition to fracturing, many employees are engaged in other operations beginning with well pad construction through monitoring producing wells. The variety of services and tasks performed in the hydraulic fracturing industry results in diverse occupational exposure profiles. 

Resolution: Defining similar exposure groups (SEGs) followed by qualitative and possibly quantitative assessment is a way to allocate and prioritize assessment resources. The large and diverse employment demographics within the hydraulic fracturing industry is well suited for this risk based type of assessment strategy. A model based upon the American Industrial Hygiene Association’s Exposure Assessment Strategy was implemented to evaluate exposure risks from SEGs including silica transport, roustabouts performing well construction, vacuum truck operators and well monitors/pumpers who gauge tanks in producing wells. 

Results: Findings of the assessment identified a number of scenarios related to non-routine tasks or changes in operational conditions that could result in exposures that might be overlooked during a routine industrial hygiene assessment. These include potential exposure to hydrocarbons or hydrogen sulfide during tank gauging, biocides in tank cleaning operations, traffic safety/driver fatigue, and noise exposure for those working in the vicinity of vacuum trucks. While these exposures might not exceed occupational exposure limits, they should be considered in exposure assessments. 

Lessons Learned: Occupational exposures assessments in the hydraulic fracturing industry often focus on the fracturing process or risks related to physical safety while drilling. Using an approach based on the AIHA strategy can help identify other SEGs and potential exposures that might otherwise be overlooked. 

CS-106-04 Health Risk Analysis and Characterization of Similar Exposures Groups of Municipal Solid Waste Employees Based on Historic Monitoring Event

A. Rohr Daniel, ENVIRON International Corporation, San Francisco, CA; P. Harper, ENVIRON International Corp., Phoenix, AZ; C. Torres CIH, CSP, ENVIRON International Corporation, Atlanta, GA; R. Wills, Waste Management, Houston, TX

Situation/Problem: Similar exposure groups (SEGs) are defined by the American Industrial Hygiene Association (AIHA) as groups of employees with similar exposure profiles. SEGs are employees that are classified similarly due to the work environment, environmental agents (i.e., contaminants of concern (CoCs)), and exposure factors which include job tasks. Characterizing SEGs is a critical component of an occupational exposure assessment. Health risk is a function of the toxicity of what workers are exposed to, and the risk of exposure to those CoCs. For this study, the SEGs among municipal solid waste employees were characterized and analyzed for relative health risk due to multiple CoCs based on a Bayesian Decision Analysis (BDA) of historic exposure data from 62 facilities across North America.

Resolution: The potential health risk to municipal solid waste employees was evaluated. Personal monitoring data collected from 1998 to 2014 consisted of 1,526 data points from 62 facilities in two countries have been reviewed and tabulated. Of the various tasks and job descriptions 19 SEGs and 203 CoCs were determined. CoCs with ≥ 3 data points for each SEG in each LOB were reviewed, and BDA was used to determine the AIHA exposure category.

Results: The study points out the shortcomings of traditional execution of exposure assessment programs. Despite the large number of samples, only a small handful of SEG/COC combinations were capable of characterization with statistical significance. Results indicating either high statistical variability or too few data points provide insight into current knowledge gaps and the potential to mischaracterize health risks to SEGs.

Lessons Learned: Performing the BDA allowed further understanding that, although numerous monitoring events had been conducted, the lack of planned assessments left many SEGs uncharacterizable and requiring additional data to be collected. The analysis of the BDA facilitated the design of future monitoring plans based on the quality and quantity of the historic data. BDA is a widely utilized tool in the industrial hygienists’ tool belt that proved to be an invaluable asset in characterizing the potential health risks to municipal solid waste employees.

SR-106-05 Characterization of Exposure to Diesel Particulate Matter in an Underground Metal Mine

D. Stephenson, Boise State University, Boise, ID; E. Zamzow, Boise State University, Boise, ID

Objective: Elevated exposure to diesel particulate matter (DPM) has been linked to adverse health outcomes including respiratory irritation, cardiovascular disease, immune dysfunction, and lung cancer. Due to the nature of their work in confined spaces and in close proximity to diesel powered equipment, underground miners experience amongst the highest exposures to DPM of any occupation. The overall goal of this research is to use realtime and time-integrated sampling methods to evaluate environmental and occupational exposure to DPM in an underground metal mine.

Methods: Using underground miners placed in high, medium, and low exposure categories and results of environmental monitoring, correlations between DPM measurements acquired using MSHA-approved methods and realtime measurements of diesel exhaust combustion byproducts were determined. Co-location of DPM impactors with realtime monitors allowed for regression analysis between measurements. Samples collected using DPM impactors were subjected to analytical methods that determined the elemental carbon (EC) and organic carbon (OC) fractions of total carbon (TC). The resulting data provides important information related to using the EC faction to estimate TC in the presence of airborne carbon interferences.

Results: Regression analysis of results from environmental monitoring suggest fair to poor correlations between realtime measurements of particle concentration and NO2 and measurements of TC (R2 values of 0.75 and 0.076, respectively). Concentration results of personal exposures to TC generally followed a miner’s low, medium, or high exposure category and ranged from 21µg/m3 to 132µg/m3 with an average of 55µg/m3 (SD = 22µg/m3). With respect to miner exposure categories, acquired EC/TC concentration ratios ranged from 2.4 to 7.9 with an average of 3.9 (SD = 3.8).

Conclusions: Study results suggest that there is a fair correlation between realtime measurements of particle concentration and TC, but that similar realtime measurements of NO2 serve as poor surrogates for this MSHA compliance measure. Personal exposure measurements of DPM were below MSHA compliance thresholds validating the use of appropriate ventilation and diesel powered equipment controls. Large OC fractions in DPM samples are unexplained and appear to indicate the presence of interferences in the ambient air of the underground environment.

CS-106-06 Personal Breathing Zone Exposure Monitoring and Surface Contamination Study of Lead at an Indoor Firing Range

P. Kondragunta, International Safety Systems, Vadodara, India

Situation/Problem: Testing of Kevlar material is carried out by firing bullets, loaded with gunpowder, in an enclosed firing range. Bullets (having Lead based primer) are loaded into the barrel of the rifle and the bullet is shot by the user, from outside the firing range using a switch connected to the rifle. The firing pin of the rifle hits the back of the cartridge, activating the shock-sensitive primer, which ignites the gunpowder, forcing the bullet down the barrel of the rifle and on its path. The heat and pressure within the cartridge vaporize the lead present in the primer. The heat of this explosive reaction and subsequent cooling results in the condensation and formation of tiny lead particles at the source, throughout its path and at the bullet trap.

Resolution: Full shift/eight-hour TWA employee exposure monitoring for lead particulates was carried out using modified NIOSH 7300 method. Surface contamination study for lead particulates was also carried out, before and after firing of bullets using modified NIOSH 9102 method. ACGIH prescribed eight-hour TLV-TWA of 0.05 mg/m3 was used as evaluation criteria for employee exposure. OSHA recommended level of acceptable lead loading for non-lead work areas of 0.2 μg/cm2 was used as evaluation criteria for surface contamination study.

Results: The eight-hour TWA employee exposure to lead particulates was below 10 % of the ACGIH TLV-TWA limit of 0.05 mg/m3. The surface contamination level before firing of bullets were above the OSHA recommended level of 0.2 μg/cm2. The surface contamination level increased by 10 folds, after firing of bullets.

Lessons Learned: The indoor firing range lacked proper ventilation and hence, users were exposed to lead dust from the primers. Although the exposure through inhalation was below the OEL, potential for ingestion risk exists due to surface contamination. Avoiding contact with all surfaces and using two pairs of nitrile gloves was recommended to reduce the risk of exposure through ingestion route.

CS-106-07 Challenges of Exposure Assessment for Hexavalent Chromium Exposure in the Automotive Parts Manufacturing Industry (Sponsor: Exposure Assessment Strategies Committee)

N. Lee, Tenneco, Lake Forest, IL

Situation/Problem: Identify the exposure risk to hexavalent chromium for welding operation in an auto parts manufacturing setting using qualitative and quantitative exposure assessment. When engineering control settings are modified frequently, making exposure assessment strategy and management can be challenging. Therefore, the exposure assessment tools were applied to assist with professional judgments in order to manage workplace/workers. The tools included the Checklist, the manufacturing internal qualitative assessment, and the Industrial Hygiene Data Analysis (IHDA). The results were compared between the three assessment tools for appropriate assessment.

Resolution: OSHA ID-215 was used to collect and analyze welding fume samples for hexavalent chromium. An air pump was attached to a 37 mm PVC filter cassette at 2.0 L/min. The sampler cassette was placed outside the employee’s welding helmet at the breathing zone. A pre-calibration and a post-calibration were performed before and after sampling. After post-calibration, all samples were sealed and transferred to a single laboratory for analysis. The Checklist and the IHDA were applied for qualitative and quantitative exposure assessment respectively. All results were compared with the internal assessment tool used for priority of management.

Results: The Checklist includes the options of toxicity and engineering control. The internal qualitative assessment has additional information that includes frequency/duration, generate amount, observed dustiness, and exposure potential. The qualitative assessment from the Checklist indicated the hexavalent chromium level was at exposure category 4 (exposure >100% of OEL) when capturing at points of emissions. Meanwhile, the quantitative assessment from the IHDA concluded that exposure categories were varied, from category 3 to 4, regardless the type of controls in place. The internal assessment tool indicated mixture conclusions, a range from category 1 to 4.

Lessons Learned: Before quantitative assessment, if only limited information is applicable, the Checklist can function as a primary assessment tool to narrow down the scope. Furthermore, the internal assessment tool with more details can be used as secondary assessment to make decisions for management use.

CS-106-08 The Use of an Activity Based Study (ABS) as a Novel Method for Tracing Mercury Contamination in Schools and Industrial Settings

J. Callahan, J. Kind, K. Scribner, Center for Toxicology and Environmental Health, North Little Rock, AR

Situation/Problem: Spilled liquid mercury is easily spread via contaminated shoes and clothing, which can lead to widespread contamination in buildings with sizeable populations (e.g. schools/industrial facilities). When a spill occurs, mercury evaporates and becomes an invisible, odorless vapor, which may pose a long-term health risk. In a recent school spill, breathing zone air concentrations were below the Agency for Toxic Substances and Disease Registry (ATSDR) recommended action levels, yet accumulation of mercury on shoe soles was evident upon entry into the building. This indicated persistent mercury contamination on surfaces that was not identified by air monitoring.

Resolution: The ATSDR recommends that headspace readings for personal belongings in the range of 3-6 µg/m3 be considered protective of human health. Using these guidelines, we developed an activity based study (ABS) to examine floor contamination. Workers donned clean footwear, entered the school, and walked for a specified time period in classrooms and hallways. Afterward, footwear was sealed in air-tight bags and heated passively in the sun to an external temperature of ≥85°F for one hour. Headspace analysis was performed, and footwear with mercury vapor readings >3 µg/m3 were considered indicative of the continued presence of mercury contamination in the area.

Results: Using the ABS, we were able to characterize the migration of mercury throughout the school. The initial assessment showed the highest shoe headspace was 48.7 µg/m3, which is 8-10 times the level considered protective of human health. This allowed for targeted clean-up efforts prior to re-entry of students to prevent further contamination of personal belongings and migration off-campus.

Lessons Learned: In large buildings, air monitoring may not be sufficient to detect areas of mercury surface contamination. Even though ambient air concentrations may remain below levels recommended by regulatory agencies, the potential may exist for mercury contamination on personal belongings that could produce a source of vapor when stored in cars, closets, or homes. The use of an ABS in these circumstances provides a detailed examination in order to establish appropriate remediation efforts for the continued safety of students and workers.

CS-106-09 Case Study: Fipronil Exposure from Misapplication of Insecticide during Extermination

B. Fontaine, The Windsor Consulting Group, Inc., South River, NJ

Situation/Problem: Fipronil is a commonly used insecticide in American homes. Limited data are available on the human toxicity. Analyzing pesticide illness surveillance data from 2001 to 2007, there were 103 acute illness cases of fipronil exposure. Findings show reporting of acute illness was relatively uncommon. This case serves as sentinels to warn hygiene practitioners of the need to reinforce the importance of precautionary measures to prevent fipronil exposure and subsequent adverse health effects due to thyroid toxicity and potential for carcinogenicity.

Resolution: After reviewing the sample results, toxicity of fipronil and its mammalian metabolites and the age of the family members and routes of exposure (dermal absorption, ingestion, and inhalation); a qualitative risk assessment was done to determine the next course of action. The results clearly suggest the adults (parents) could cohabitate with the residues on the various building surfaces but the health risk was intolerable for very young children. A recommendation was made to keep the children out of the house until building surfaces could be decontaminated and collaborate with the family pediatrician on biomonitoring thyroid function of the children.

Results: The air and surface sample results for cypermethrin were below the limit of analytical detection. However, the surface samples for fipronil indicated residues in several locations throughout the residence. In particular, 4.9 µg/100cm2 was detected on the wood flooring. Wipe samples collected on the living room floor indicated 3.4 µg/100cm2, 4.1 µg/100cm2 in the master bedroom, 11µg/100cm2 in the children’s bathroom tile, and 2.2 µg/100cm2 on the basement windowsill. Fipronil was not detected in the surface samples collected in the dining room, master bathroom, hall outside the children’s bedroom, or the painted cement basement floor.

Lessons Learned: Fipronil is one of the most commonly used insecticides in American homes. However, limited data are available on the toxicity of this pesticide in humans. Analyzing pesticide illness surveillance data from 2001 to 2007, there were 103 acute illness cases associated with fipronil exposure. The findings of acute illness was relatively uncommon due to underreporting and most cases were residential exposures. This case serves as sentinels to warn hygiene practitioners of the need to reinforce the importance of precautionary measures to prevent fipronil exposure and subsequent adverse health effects.​