PETITIONED PUBLIC HEALTH ASSESSMENT ADDENDUM
SHAFFER EQUIPMENT COMPANY
MINDEN, FAYETTE COUNTY, WEST VIRGINIA
The Shaffer Equipment Company (SEC) site, located in Fayette County, Minden, West Virginia constructed electrical substations for area coal mines from the period 1970 to 1984. The site is approximately one acre and has one building (SEC Equipment Building) that served as both a warehouse and office. Electrical equipment such as transformers, switches, circuit breakers, and capacitors were stored on the site. Dielectric oils that contained polychlorinated biphenyls (PCBs) and 1,2,4-trichlorobenzene were found in on- and off-site soils and sediments. Because PCBs are on site and PCB-contaminated oils reportedly were burned as starter fuel in the warehouse/office building, on- and off-site soil samples and on-site sediment samples were analyzed for polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzo-p-furans (PCDFs). On- and off-site soil and on-site sediment concentrations of PCDDs and PCDFs were not judged by the Agency for Toxic Substances and Disease Registry (ATSDR) to be at levels of public health concern. Sampling for PCDD/PCDF and PCB contamination was not conducted for the warehouse/office building.
There is a great deal of community concern and interest about the SEC site. There have been several studies, surveys and reports on the site regarding excess cancers and other adverse health effects. In addition, the Concerned Citizens to Save Fayette County (CCSFC) arranged to have blood and adipose tissue sampling/analysis for PCBs from residents of the area. ATSDR requested that data from the CCSFC. The information has been included in this public health assessment.
A past, completed PCB-exposure pathway was identified for on-site SEC workers and their families. Routes of exposure may have been: 1) dermal contact (contact with contaminated soil, sediments, occupational exposure from handling dielectric oils, secondary exposures of family members primarily through washing clothes contaminated with dielectric oils), 2) inadvertent ingestion (ingestion of small amounts of contaminated soils, sediments from soiled hands during eating, smoking, drinking and other activities), and 3) inhalation of contaminated airborne dusts. This past, completed pathway does not currently exist because SEC is no longer an active, operating facility.
Several past, potential PCB-exposure pathways were also identified for on-site workers and off-site residences that used PCB-contaminated oils as starter fuel for coal igniting. The routes of exposure may have been: 1) inhalation (incomplete burning of PCBs and formation of PCDD and PCDF compounds), 2) dermal contact (handling of PCB-contaminated oils), and 3) inadvertent ingestion (ingestion of contaminants from soiled hands). Those past, potential pathways no longer exist, as PCB-contaminated oils are no longer available from SEC.
Other potential pathways that have been identified as past, present, and future concerns involve trespassers onto the SEC site, children playing in yards and Arbuckle Creek, on-site workers in the SEC Equipment Building, and persons that eat snapping turtles from the area.
Of special concern are individuals who eat snapping turtles that may have been PCB-contaminated from the SEC site. Snapping turtles may bioaccumulate PCBs in the flesh; therefore, pregnant women and nursing mothers are cautioned not to eat snapping turtles caught in the site vicinity until such turtles are caught, bioassayed, and determined safe to eat. Sensitive subpopulations to chemical contamination are fetuses and breast-fed infants who may receive PCB exposure from across the placental barrier or through mother's milk.
Surface water from Arbuckle Creek is contaminated with fecal coliform bacteria associated with the improper disposal of human wastes. The fecal coliform source has not been identified. If analyses do not find PCB contamination in turtles, aquatic life from Arbuckle Creek should still not be eaten until the source of the fecal coliform bacteria is removed or unless the food is thoroughly cooked. Recreational uses of the creek should be restricted until further testing indicates fecal coliform counts are within public health guidelines.
ATSDR has determined that the SEC site (on-site) continues to pose a public health hazard for trespassing adults and children, and for on-site workers. These groups are at risk of inadvertently ingesting, inhaling, and or having dermal contact/absorption with on-site surface soils and sediments that were highly contaminated with PCBs in the past and where contamination may still exist. Based on recent off-site PCB sampling in the vicinity of SEC, ATSDR believes the SEC site is an indeterminate public health hazard for the general off-site population under current conditions because complete PCB data were not available for off-site surface water and off-site groundwater contamination. There is no significant human exposure evident in regard to off-site soils, sediments and air pathways.
ATSDR has made recommendations to eliminate or reduce the potential for future exposures. Recommendations have also been made for additional on-site characterization, specifically for surface soil sampling. ATSDR's Health Activities Recommendation Panel made recommendations for follow-up investigation and possible retesting of an infant/child showing levels of PCBs in serum, referral of the final public health assessment to the National Institute for Occupational Safety and Health for review, and possible investigation of past worker exposures, and environmental health education for local public health professionals and the medical community.
ATSDR released a draft of this public health assessment, for public comment, from January 25 - February 23, 1993. ATSDR addressed those comments in Appendix 4 of this document. A public meeting is planned to follow the release of the final Shaffer Equipment Company Public Health Assessment.
A. Site Description and History
The Shaffer Equipment Company (SEC) site is in Fayette County, Minden, West Virginia (Figure 1), off Old Minden Road (Route 17). The facility constructed electrical substations for area coal mining industries from 1970 to 1984. SEC stored unneeded, damaged, or outdated transformers, capacitors, switches, and voltage regulation/distribution devices on site. Leaks from that equipment, possible spills, and dumping practices appear to be responsible for the on-site polychlorinated biphenyl (PCB) contamination. Local newspapers reported on a public meeting held by EPA and attended by ATSDR (May 29, 1990) that PCB oils (oils contaminated with PCBs) were burned on site and may have been given away or sold as fuel. ATSDR interviews (1) with several former Shaffer employees also indicated that this may have occurred.
PCBs are commercial compounds that were widely used for insulation and lubrication in electrical cables, transformers, and other equipment. There are 209 individual PCB compounds. Commercial manufacture of PCBs was stopped in the United States in 1977. Use of PCBs was halted because of the persistence of these compounds in the environment and the potential for adverse health effects from exposure to PCBs (2,3).
The site size is approximately one acre and contains one building (SEC Equipment Building) that
served as both warehouse and office (Figure 2). The site is in a valley and is surrounded by hills.
The immediate site is flat and is bordered to the northwest by Arbuckle Creek. Arbuckle Creek
experiences occasional flooding, which is accompanied by sediment migration. Arbuckle Creek
flows to the northeast. The site is bounded on the south by hills, north by Arbuckle Creek, and on
the southwest and northeast by fencing. Although access to the southwest part of the property near
the warehouse/office was restricted by a chain-link fence and locked chain-link gate, the northeast
portion of the property is easily accessed through a cattle gate. Not all of the PCB-contaminated
property is fenced. Most of the area that EPA excavated and reclaimed during emergency removal
operations of PCB-site contamination in 1984-1985 is not fenced. That includes some property of
the Berwind Land Company. There are reports of vandalism and recreational use of the site. Heavy
brush and trees cover most of the site, although some areas are cleared for parking and turn-around.
No concrete or asphalt pavings are present. There are residences across Arbuckle Creek from the
site, both up- and downstream. The Minden wastewater treatment plant 
is about a mile upstream
from the Shaffer site. The plant is permitted to discharge wastewater into Arbuckle Creek.
The West Virginia Department of Natural Resources (WVDNR) visited the site in September 1984. An inspection of the site revealed several hundred transformers and capacitors. A composite surface soil sample and a grab-sample of soil/sediment from a site drainage ditch into Arbuckle Creek indicated PCB concentrations of 26,749 parts per million (ppm) and 1,136 ppm, respectively.
Subsequent sampling of soil and sediment by EPA in late 1984 indicated PCB contamination ranging from less than 1 ppm to greater than 200,000 ppm. There were approximately 150 transformers and about 50 capacitors on the site; some showing evidence of oil leakage. PCBs had either migrated from the site during occasional flooding or had been spilled into Arbuckle Creek because PCB concentrations of up to 190 ppm were found in creek sediments.
The Centers for Disease Control (CDC) certified in December 1984 that an imminent threat to
public health existed because of high levels of PCBs in the environmental media (4). EPA approved
funds for an emergency removal of PCBs from the environment and the removal of
PCB-contaminated equipment in late December 1984. PCB-contaminated soils were excavated and
placed in a clay-lined staging area for treatment. An innovative process for extracting PCBs from
soil using freon and methanol was tried without success until November 1985, when the process was
terminated. EPA contracted for the removal of the waste pile (4735 tons) by trucking to a
hazardous waste landfill in Emelle, Alabama. Removal activities began in September 1987 and
ended in November 1987. EPA site operations were declared complete in December 1987.
In April 1989, 19 drums containing paints, solvents, and waste oils were discovered on the site. Those drums were to be previously disposed of by the property owner in an agreement with EPA. The property owner declined responsibility and EPA initiated removal of the drums in May 1989.
On May 24, 1989, ATSDR received a petition from the West Virginia Department of Health (WVDOH) to conduct a public health assessment for the SEC site. ATSDR subsequently accepted the petition.
In March 1990, EPA conducted additional on- and off-site sampling for volatile and semivolatile organic compounds, PCBs, and inorganic chemicals from surface soil, subsurface soil, surface water, sediment, and raw water (water used for public water supplies from the Minden Mine intake and the Rocklick Mine intake). That action was in response to a request from the Agency for Toxic Substances and Disease Registry (ATSDR) for additional off-site sampling. On-site and off-site soil samples were also collected for polychlorinated dibenzo-p-dioxin (PCDD) and polychlorinated dibenzo-p-furan (PCDF) analyses. Approximately 19 on-site and 5 off-site surface soil/subsoil samples, 6 surface water samples, 7 sediment samples, 1 spring surface water sample, and 2 raw water samples were taken.
In May 1990, EPA resampled for verification purposes, a number of the same locations that tested positive for PCBs from the March 1990 soil/sediment sampling event.
In June 1990, EPA performed additional sampling of three areas near the Shaffer Equipment building in response to reports made by the Concerned Citizens to Save Fayette County (CCSFC) at the May 29, 1990 public meeting.
EPA conducted another emergency removal action from November 1990 to January 1991 to remove soil from the remaining areas containing high levels of PCBs. Sampling was conducted before, during, and following the removal actions.
ATSDR's evaluation of the 1990-91 sampling results, which are the most detailed and includes on- and off-site sampling, will be the primary focus of this public health assessment.
ATSDR staff (Lynn Wilder and Donald Joe) visited the Shaffer site on August 25, 1989 and on May 29, 1990. Site conditions were noted including accessibility, demographics, and proximity of any residences to the site. ATSDR met with representatives from the West Virginia Department of Health (WVDOH) and the Fayette County Health Department.
C. Demographics, Land Use and Natural Resource Use
Minden, West Virginia is a small coal mining town of approximately 2,000 inhabitants. During ATSDR's site visit in August 1989, it was noted that the only elementary school in Minden had been closed. There were no schools, parks, playgrounds, nursing homes, or hospitals within a 1-mile radius of Minden. The racial character of Minden was estimated during the site visit to be 90% Caucasian and 10% African-American. There are many elderly people and children in Minden. The average age for the area was estimated to be 30 years. Approximately 65-75 people live within 400-500 feet of the site. In the past, coal mining was prevalent, but many of the coal mines have closed. Homes and yards are well-kept. The general area is economically depressed. Current land use in the vicinity of the site is residential.
In the 1980 census, Fayette County had a population of 57,863. The majority (eighty-four percent) of the population lives in rural areas of the county. Sixty-two percent of those in the county have lived in the same house for at least five years. An additional twenty-three percent have lived in the county for this period but in several different homes (5,6).
Sixteen percent of the county's 19,889 households had annual incomes of less than $5,000. Seventeen percent of the county residents live below the poverty level compared to twelve percent of the residents in the state. The per capita income in the county was $5,295 in 1980. (5,6) Reports from the citizens' group (CCSFC) in Minden indicated that the per capita income in Minden is $4,000.
The racial make-up of the county is predominantly white (93 percent). Eight percent (4,505) of the county residents are under age five and nine percent (4,952) are over the age of sixty-five. In 1980, 15,858 persons over the age of sixteen were employed. Of those, 3,194 (20 percent) were employed in the mining industry. (5,6)
The population figures from 1986 indicated the county's population was falling with a 4.1 percent net loss. In 1984, the percent of persons under age five remained relatively stable at seven percent while the percentage of those over age 65 rose to 14 percent. The birth rate for the county in 1984 was similar to that of the state (12.5 to 12.6 per 1000 population). The crude death rate was 10.6 in the county and the state death rate was 9.8. The infant mortality rate for the county was 7.0 per 1000 live births. This is much lower than the rate of 11.0 per 1000 live births in the state. (7)
The evaluation of health outcome data may give a general picture of the health of a community. Those data may confirm the presence of excess disease or illness. However, elevated rates of a particular disease may not be due to hazardous substances in the environment. Other factors such as socio-economic and personal habits may have a tremendous influence in the potential development of disease. In contrast, even if elevated rates are not found, a contaminant may still have caused illness or disease. ATSDR must depend on previously gathered data to perform a public health assessment. There were several sources of health outcome data available for ATSDR review and these are listed as follows:
ATSDR reviewed the 1980 Census for Minden and the Riggans Cancer Mortality Data for Fayette County, West Virginia. (5,6,7,8)
In 1983, the West Virginia Department of Health conducted an examination of cancer mortality statistics for the Oak Hill, W.V. area. ATSDR reviewed the report, Oak Hill, W.V. Investigation -- Stage I: Initial Assessment. (9)
In 1986, the Appalachian Student Health Coalition of Vanderbilt University assisted the local citizen's group, CCSFC in conducting a health survey of the residents of Minden. In 1989, CCSFC and the Virginia Student Environmental Health Project conducted a similar survey of the residents of Page, West Virginia to use as a cohort/control population to compare with the Minden survey. ATSDR reviewed both of these surveys. (10)
In 1989, a Beckley, West Virginia gynecologist, who was a member of the Health Department Board of Directors, conducted a health survey and provided the results to the ATSDR Division of Health Studies for evaluation. (11)
Several former employees were contacted by ATSDR by phone to obtain information about health concerns. (1)
At a meeting on May 29, 1990, CCSFC informed ATSDR that blood samples from several volunteers had been analyzed for PCBs. Concurrently, ATSDR was also informed that adipose (fat) tissue was in the process of being sampled and analyzed for PCBs under the direction of the CCSFC. ATSDR requested the results of the blood and fat sampling (results were expected to be available near the end of June 1990). A spokesperson for the CCSFC indicated that the members of CCSFC would have to vote to release the sampling results; however, no problems were foreseen in giving ATSDR the information. ATSDR received the results in February 1991. They have been included in this public health assessment.
In December, 1990, CCSFC took another set of blood samples (approximately 43) and analyzed them for PCBs. ATSDR received the results of those analyses in July 1991. They have been included in this public health assessment.
A review and discussion of the above studies, surveys, and reports are contained in the Public Health Implications Section of this public health assessment.
ATSDR met with citizens of Minden, West Virginia on May 29, 1990 to discuss community health concerns about the SEC site. Additional community concerns were gathered through interviews with former Shaffer employees (8). The following is a summary of those community concerns:
Residents have repeatedly expressed concerns about the health effects of exposures to PCBs. They are especially concerned about the effects on the liver, kidney, skin and respiratory tract.
Currently and throughout the EPA removal process, CCSFC, the local citizens' group has been actively seeking performance of a health study on previous workers and downstream residents.
The citizens have also expressed a great deal of frustration at the duration and extent of the clean-up process. They feel that not all sources of PCB contamination have been investigated and that local water supplies have been contaminated by PCBs.
These community health concerns will be evaluated and discussed further in the Public Health Implications Section of this public health assessment.
ENVIRONMENTAL CONTAMINATION AND OTHER HAZARDS
The tables in Appendix 2 list contaminants in each medium. Those contaminants are evaluated in subsequent sections of the public health assessment to determine whether exposure to them has public health significance. ATSDR selects and discusses contaminants based upon several factors. They include concentrations on and off site, the quality of the field and laboratory data, sample design, comparison of on- and off-site concentrations to background concentrations (if available), comparison of on- and off-site concentrations to public health assessment comparison values for noncarcinogenic and carcinogenic endpoints, and community health concerns.
The listing of a contaminant in the tables does not mean that it will cause adverse health effects if exposure occurs at the specified concentrations. The listing of contaminants in the tables means that those contaminants are further evaluated in this public health assessment. The potential for adverse health effects resulting from exposure to contaminants without a comparison value, those that exceed the comparison value, and those of health concern is discussed in the Public Health Implications Section.
Comparison values for ATSDR public health assessments are contaminant concentrations in specific media that are used to select contaminants for further evaluation. ATSDR and other agencies developed those values to provide guidelines for estimating the media concentrations of a contaminant that are unlikely to cause adverse health effects, given a standard daily ingestion rate and standard body weight. See Appendix 3 for a description of the comparison values used in this public health assessment.
SURFACE SOIL AND SUBSURFACE SOIL
Sixteen on-site samples and one duplicate on-site sample taken in March 1990 indicated that some PCB contamination remains on site (Table 1). PCB concentrations ranged from not detected to 240 mg/kg in soil. The on-site soils are contaminated, to a minor extent, with other volatile organic compounds (VOCs) and semivolatile organic compounds (SVOCs) (Table 1).
Of the VOCs, no ATSDR soil comparison values were exceeded and therefore VOCs in surface and subsurface soils will not be considered further.
Of the SVOCs, no soil comparison values were found for, 2-methylnaphthalene, acenaphthylene, dibenzofuran, and di-n-octyl phthalate. In addition, comparison values were exceeded for benzo(a)anthracene, benzo(b)fluoranthene, benzo(a)pyrene, and dibenz(a,h)anthracene. These compounds will be retained for further evaluation in the Public Health Implications Section.
In March 1990, thirteen on-site samples and one duplicate on-site sample were analyzed for PCDDs and PCDFs. The analyses included PCDD and PCDF congeners that were assigned a toxicity equivalency factor (TEF) associated with the most toxic dioxin--2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). The TEF was used to calculate the total toxicity associated with all PCDD and PCDF congeners found. ATSDR did not find any PCDD or PCDF levels in the soil samples, except for one sample and its' duplicate sample, that exceeded the ATSDR comparison value EMEG (non-pica child). The sample and its' duplicate were taken from a depression east of the Shaffer Equipment Building. A TEF of 0.128 ug/kg was calculated for the sample and a TEF of 0.801 was calculated for the duplicate sample.
In May 1990, EPA took 14 soil samples (Table 8), four of which were random samples at locations different than those in March 1990. Those random samples showed PCB levels ranging from 0.148 mg/kg to 374 mg/kg. (12)
During a meeting held on May 29, 1990, the CCSFC reported PCB contamination in additional areas of the site. Thirty additional samples were taken near the Shaffer Equipment building and near a shale pile in June 1990. PCB concentrations ranged from "not detected" to 40,300 mg/kg (Table 1A). Four of the 30 soil samples were also analyzed for SVOCs (Table 1B). All of the compounds for which a comparison value was available did not exceed the comparison values and will not be considered further. A comparison value was not available for 2-methylnapthalene; therefore, this compound will be retained for further evaluation in the Public Health Implications Section.
In November 1990, EPA conducted 58 soil samples which were analyzed for PCBs along with an additional three confirmation and four duplicate samples. These samples were taken prior to and during EPA removal actions. The PCB concentrations ranged from 0.3 to 10500 mg/kg (Table 10), with no PCBs being detected in four of the samples using a detection limit of 0.1 mg/kg.
Post sampling was conducted on January 24, 1991, at the excavated areas. The PCB concentrations ranged from 0.44 to 1020 mg/kg (Table 11). The sampling consisted of 53 samples that were separated into 9 related areas and analyzed as composites from the original samples. Four additional samples were taken from areas on site that were believed to be unimpacted by PCB contamination. One of those four samples revealed a detectable PCB concentration at 0.33 mg/kg. A 1-3 foot cover of soil, taken from the unimpacted areas, was used to fill the excavated areas. The major areas of PCB contaminated soil is believed to have been removed and backfilled; however, it has not been determined that all areas of the site are free from surface soil contamination.
ON-SITE SEDIMENTS
Two on-site sediment samples collected in March 1990 indicated some PCB, VOC, and SVOC contamination at the site (Table 2). The highest PCB level of 660 mg/kg was found in the sediments of the excavated area drainage ditch, located in the area where EPA had previously initiated and completed the 1984-1985 emergency removal of PCBs. Sediment comparison values were not available for the compounds 2-methylnapthalene and dibenzofuran. Sediment comparison values were exceeded for the compounds benzo(a)anthracene, benzo(a)pyrene, and PCBs. Those compounds with no available comparison values and those that exceed the comparison values will be analyzed further in the Public Health Implications Section.
Two on-site sediment samples (March 1990) were also analyzed for PCDDs and PCDFs, including congeners. ATSDR determined that PCDD or PCDF levels associated with the sediment samples did not exceed the ATSDR EMEG comparison value.
In May 1990, two sediment samples were taken from the same locations as those collected in March of 1990 and analyzed for PCBs only. The May 1990 sampling revealed PCBs at a concentration of 112.1 mg/kg at the building area drainage ditch and 39.2 mg/kg at the excavated area drainage ditch.
AIR
No air sampling data were available for ATSDR's evaluation.
PCB contamination at the Shaffer site are found primarily in soil. The rate of PCB volatilization from soils (i.e., the migration of PCBs by evaporation from the soil into the air) is very small because of soil binding of PCBs (2). The rate of PCB volatilization from surface water into the air is much higher (2); however, PCBs were not detected in surface water samples at the Shaffer site. Because of these factors, PCBs are not expected to be present in ambient air at levels of health concern. When additional site monitoring was conducted in March 1990, EPA and ATSDR decided not to perform air monitoring for PCBs at the Shaffer site (13). However, EPA scanned the site with a field instrument (HNu) in March 1990 to investigate VOCs and SVOCs in the air at the site. That instrument did not detect any organic vapors (VOCs, SVOCs).
GROUNDWATER
No groundwater samples (other than from an on-site spring discussed below) were taken by EPA or were available for ATSDR's evaluation.
An on-site spring (no known uses) that flows into Arbuckle Creek was sampled for VOCs, SVOCs, and PCBs. Nothing was found above the laboratory measuring instrument's limits of detection. However, the PCB detection limits (0.5 µg/l and 1.0 ug/l) of the measuring instrument was higher than the ATSDR CREG comparison value of 0.0045 ug/l, therefore it cannot be stated with certainty that PCBs are or are not a problem. The spring was not analyzed for PCDDs or PCDFs; however, because PCDDs and PCDFs have a high affinity for soils, sediments and have low solubility in water, they are not expected to be present in groundwater or springwater.
SURFACE WATER
No on-site surface water (except for a spring discussed under on-site groundwater) exists.
SOIL
Off-site (residential) surface soils were found to be PCB-contaminated at a maximum concentration of 15 mg/kg in December 1984. In March 1990, five off-site surface soil (but no subsurface soil) samples (Table 3) were taken. Those samples consisted of one background soil sample, three soil samples from residential yards in the area (from two homes that yielded levels of PCBs during an earlier sampling event and one home that had a potential for PCB contamination), and one sample from the mouth of the old Minden Mine. Again, there was some contamination from PCBs, VOCs and SVOCs. The PCB contamination in the samples taken from the yards of three homes and the Minden Mine area ranged from 0 to 2.1 mg/kg.
Comparison values were not available for the compounds 2-methylnapthalene, acenaphthylene, and dibenzofuran. Comparison values were exceeded for benzo(a)pyrene and PCBs. Those compounds that did not exceed the comparison values will not be considered further. The others with no comparison values available or that exceed comparison values will be further evaluated in the Public Health Implications Section.
Four off-site soil samples taken in March were analyzed for PCDDs and PCDFs and their congeners. No sample exceeded ATSDR's PCDD or PCDF EMEG comparison value.
SURFACE WATER
Five surface water samples (two upstream and three downstream from the site) were collected and analyzed for PCBs, VOCs and SVOCs (Table 4). PCBs were not found in the surface water samples above the level of the detection limit of the laboratory's measuring instruments. However, as under on-site groundwater, the PCB detection limits of the measuring instrument was higher than ATSDR's comparison value. Some VOC and SVOC surface water contamination was found both upstream and downstream of the site. Some of these chemicals (eg., methylene chloride, acetone) are common laboratory artifacts or contaminants introduced by laboratory handling, as evidenced by their presence in laboratory blank samples. The reported levels of all contaminants except methylene chloride were not above comparison values; therefore these surface water contaminants will not be considered further in this health assessment. Methylene chloride is will also not be considered further. ATSDR believes the elevated concentrations were the result of laboratory contamination.
No surface water sampling for PCDD or PCDF was performed during March sampling. Although no PCDD and PCDF analyses were conducted for surface water, those compounds are not expected to be found in surface water in any appreciable amounts due to their physical and chemical characteristics. PCDDs and PCDFs will normally leave surface water and adsorb onto sediment that tightly bind the compounds (14). Because PCDD and PCDF were not found above ATSDR's comparison values in on-site soil or sediment, levels are not expected to be found in surface water.
SEDIMENT
Five off-site sediment samples were collected in March 1990 and analyzed for PCBs, VOCs, and SVOCs (Table 5). PCBs were found above the comparison value in the off-site sediments at a maximum of 5.2 mg/kg. Benzo(a)anthracene, benzo(b)fluoranthene, and benzo(a)pyrene were also found above the sediment comparison value. Comparison values were not available for 2-methylnaphthalene, dibenzofuran, and di-n-octyl phthalate. Those compounds that did not exceed the comparison values will not be considered further.
No off-site sediment sampling for PCDD or PCDF was performed during any sampling. However, because on-site soils and sediments did not contain PCDDs and PCDFs at levels above comparison values, off-site sediments are not expected to contain levels of PCDDs/PCDFs that came from the site above comparison values.
AIR
No off-site air samples were taken during any sampling period.
GROUNDWATER
Two raw water supplies at the Minden Mine intake (West Virginia American Water Company) and the Rocklick Mine intake (Arbuckle Public Service District) were sampled for PCBs, VOCs, and SVOCs (Table 6). This action was taken by EPA in response to concerns by the CCSFC and the community that PCBs were disposed in the mine shafts. All results were below the detection limits except for levels of methylene chloride, which ATSDR considered a result of laboratory contamination ("B" identifier in the Table indicates that methylene chloride was found in the laboratory sample blank). However, the PCB detection limits (0.5 ug/l and 1.0 ug/l) of the measuring instrument was higher than the ATSDR CREG comparison value of 0.0045 ug/l, therefore it cannot be stated with certainty that PCBs are or are not a health concern.
PCBs are rarely found in groundwater because PCBs have an affinity for soils, sediments, and organic matter. PCBs also have low leaching potential under normal soil conditions (2).
No PCDD or PCDF analyses were performed on the raw water samples above. No other groundwater samples were taken during the March sampling.
As discussed in surface water, PCDDs and PCDFs are not expected in any appreciable amounts in groundwater due to their physical and chemical characteristics. These compounds will separate out of groundwater and adsorb onto soil particles that tightly bind the compounds. Because PCDD and PCDF were not found above comparison values in the soil, high levels are not expected to be found in the groundwater.
BIOTA
No data are available on PCB levels in aquatic biota (snapping turtles). It is not certain whether there are snapping turtles in the site vicinity. No PCB sampling of fish has been performed. EPA has stated that there are no edible species of fish in Arbuckle Creek.
Some PCBs are highly resistant to degradation and may bioaccumulate in aquatic species. The concentration of PCBs in fish, shrimp, and oysters can reach 26,000 to 660,000 times their concentrations in water. Area residents have reportedly eaten snapping turtles taken from the area (15,16). There is evidence in the literature that PCBs bioconcentrate in snapping turtles (reptiles)(14).
OTHER SOURCES OF ENVIRONMENTAL SAMPLING DATA
ATSDR received a copy of a report entitled Community Health and PCB Exposure in Minden, West Virginia dated September 1989 (10) that contained additional PCB sampling data. The data are in Table 7 and the sampling locations are stated in Figure 3. Those results agreed closely with the EPA sampling results from March 1990. Reported on-site PCB levels were lower than discovered by EPA (64.4 mg/kg [soil] versus 240 [soil] and 660 mg/kg [sediment]). Off-site PCB concentrations (soil and sediment) in the report were similar to EPA values.
C. Quality Assurance and Quality Control (QA/QC)
The results of this Petitioned Health Assessment are based on an evaluation of the sampling data obtained from the March 1990, May 1990, and June 1990 sampling conducted by EPA and the data obtained from the Community Health and PCB Exposure in Minden, West Virginia report dated September 1989 (10). Some of the data that ATSDR reviewed, showed methylene chloride (Tables 4, 6) and acetone (Table 4) contamination. ATSDR believes detection of methylene chloride and acetone was the result of contamination by the laboratory of samples with common solvents used in preparation of the samples for analysis. The validity of the analysis and conclusions drawn for this Petitioned Health Assessment are determined by the availability and reliability of the referenced information.
In May 1990, EPA resampled several on-site locations previously sampled in March 1990 to verify PCB concentrations in the soil/sediment (Tables 8, 9). Data from each sampling locations are not listed in Tables 8 and 9. Only the minimum, lowest positive, and maximum PCB concentration of all the samples are given.
In comparing the raw PCB data taken in May and March 1990, several discrepancies in PCB concentrations were noted for samples that were taken from the same location. Ideally, samples from the same location should indicate the same concentration of contaminant. However, when the same locations were resampled, anomalies in the sampling results occurred between the two dates. For comparison, on-site soil PCB concentrations (110 mg/kg) identified in the March 1990 sampling was 9.6 mg/kg in the May 1990 sampling. In another comparison, an on-site soil PCB concentration of 17 mg/kg identified in March 1990 was 538.9 mg/kg in May 1990 for the same sampling location. The reasons for the discrepancies are unknown. It is very probable that differences in sampling technique, sampling depth, physical changes because of weather and finding the exact same sampling spot would have had a profound effect on sampling outcome.
EPA, in May 1990, also resampled several on-site locations sampled in March 1990 to verify PCB concentrations in the sediments (Table 9--note that only the minimum, lowest positive and maximum concentrations are given). Again, several discrepancies were noted. A PCB concentration (660 mg/kg) identified in the March 1990 sampling was 39.2 mg/kg in May 1990. Again, the reasons for the discrepancies are unknown.
ATSDR health assessors saw no apparent physical hazards at the site.
TOXIC RELEASE INVENTORY
In order to identify other possible facilities that might contribute to the release of contaminants into the environment near Minden, ATSDR searched the most recent data in the Toxic Chemical Release Inventory (TRI). EPA developed TRI from the chemical release (air, water, and soil) information provided by specified industries. For any of the 300 - plus toxic chemicals in TRI, EPA requires that the manufacturing industry report annual estimated releases to the environment. TRI data are those releases from manufacturing, processing, and otherwise use of toxic chemicals. The database does not distinguish spills or unintentional releases from routine releases. TRI showed no facilities in the Minden area that released PCBs to the environment.
To determine whether people are exposed to contaminants released from the SEC site, ATSDR evaluated the environmental and human components that lead to human exposure. This pathways analysis consists of five elements: 1) source of contamination, 2) environmental medium in which the contaminants may be present or may migrate, 3) points of human exposure, 4) routes of human exposure such as ingestion, inhalation, or dermal absorption, and 5) receptor population. ATSDR identifies exposure pathways as completed, potential, or eliminated. A completed exposure pathway exists in the past, present, or future if all five elements of an exposure pathway link the contaminant source to a receptor population. Potential pathways, however, are defined as situations in which at least one of the five elements is missing, but could exist. Potential pathways indicate that exposure to a contaminant could have occurred in the past, could be occurring now, or could occur in the future. Pathways are eliminated when at least one of the five elements is missing and will never be present. Completed and potential pathways may also be eliminated when they are unlikely to exist, or to be significant. All completed, potential, and eliminated exposure pathways at SEC are presented in Table 14 of Appendix 2.
A. Completed Exposure Pathways
A past, completed exposure pathway for the on-site workers was identified at the SEC site. While SEC was operating, plant workers could have been exposed to PCBs during the handling, recycling, refurbishing, and removal of liquid and filling of PCB-type transformers, capacitors, and switchgear. However, the concentration of PCBs reaching the workers is unknown because no environmental sampling was done during active facility operations. Indirectly, the workers may have exposed their families to PCBs by returning home with PCB-soiled clothing.
ON-SITE WORKERS
SEC employees were probably exposed to PCBs during plant operation via inhalation, dermal contact, and inadvertent ingestion. Interviews by ATSDR with several former SEC employees indicated that safe handling practices of PCB oils were lax (1). Some employees' hands were soaked in PCB oils while they changed transformer taps. Other employees stated that by the end of the day their clothes were soaked with PCB oils and that washing did not always remove the oils. Washing PCB-soaked clothes with other family members' clothes would contaminate those clothes, exposing other family members to PCB oils primarily through dermal contact and inadvertent ingestion. Those employees worked at the SEC site for an estimated average of 5-7 years. ATSDR believes employees and their families are the people most likely to have been exposed to significant levels of PCBs, although the actual duration and magnitude of exposure are not known.
OFF-SITE SURFACE SOILS, SEDIMENTS
Off-site (residential) surface soils were PCB-contaminated at a maximum concentration of 15 mg/kg in December 1984. Later sampling results in March 1990 indicate a maximum of 2.1 mg/kg of PCBs in residential soils. PCBs adhering to surface soils can migrate or be transported by surface-water runoff and erosion from on-site sources. PCBs found in the sediments of Arbuckle Creek (190 ppm, November 1984; 5.2 ppm, March 1990) and in the off-site soil of flood-prone areas downstream (2.1 ppm, March 1990) of the Shaffer site indicate that migration of PCBs has occurred. The concentrations of PCBs in sediments appeared to have decreased over time as transport mechanisms took effect. Arbuckle Creek flows at an average rate of 3000 gallons per minute. Heavily contaminated sediments were probably washed downstream to the New River, reducing PCB concentrations in the remaining sediments. The decrease in PCB concentrations from the November 1984 samples to the March 1990 samples indicate this possibility.
A potential exists for PCB exposure to children playing in contaminated residential soils and the contaminated sediments of Arbuckle Creek. PCB concentrations in both the residential soil and sediments appear to have decreased with time. Children playing in PCB-contaminated soils or sediments may be exposed through inhalation of airborne dusts (soil), dermal contact (soil and sediment), or inadvertent ingestion from soiled hands (soil and sediment).
B. Potential Exposure Pathways
OFF-SITE USE OF PCB-CONTAMINATED OIL AS FUEL
PCB-contaminated oils were also reported to be given away or sold as starter fuel to the general population (1). If so, exposure could occur from handling PCB-contaminated oils and possibly inhaling the PCB, PCDF, and PCDDs generated from incomplete combustion of the PCBs. Inadvertent ingestion might also occur if PCB-soiled hands were used in eating, smoking, or drinking. No data exist on the amounts of PCB-contaminated oils that were sold or given away, who burned the PCBs, who were possibly exposed to PCBs, PCDFs and PCDDs, or the severity of the exposure. Because these events took place in the past, it would be difficult, if not impossible, to quantify PCB, PCDF and PCDD exposure.
ON-SITE USE OF PCB-CONTAMINATED OIL AS FUEL IN SEC EQUIPMENT BUILDING
PCB-contaminated oils were also reported by former employees to be used in the past as a fuel to start coal fires in the Shaffer Equipment Building (1). As above, exposure to on-site workers in the Shaffer Equipment Building could have occurred from handling PCB-contaminated oils and possibly inhalation of the PCB, PCDD, and PCDDs from incomplete combustion of the oils. Inadvertent ingestion could occur if PCB-soiled hands were used in eating, smoking, or drinking. No documentation exists on the quantity of PCB-contaminated oils used as starter fuel, the duration of use, or the amount of worker exposure that may have occurred.
Early reports from the community stated that residents might have consumed snapping turtles in the area (15,16). Snapping turtles have not been analyzed for PCBs, PCDDs, and PCDFs.
The EPA does not believe that snapping turtles are consumed from this area. At a public meeting on May 29, 1990, EPA asked the audience if snapping turtles were being eaten; there was no response. However, because of limited attendance, a lack of response did not necessarily indicate that turtles were not eaten. Because of the uncertainty surrounding the use of turtles for food and the lack of data regarding turtle contamination, ATSDR could not evaluate if exposure to these contaminants are actually occurring by ingestion. If turtles are being eaten and without data to negate contamination, their ingestion would be considered a potential human exposure pathway.
ON-SITE SURFACE SOILS, SEDIMENTS
Although a fence (part chain-link, part cattle) surrounds part of the site, children and adults may easily gain access to the site. Some parts of the fence prevent only large animals from entry. On-site workers and trespassers may be exposed to PCBs from dermal contact with on-site soil and sediment, inhalation of contaminated airborne particulates from soil, and from inadvertent ingestion of on-site soil and sediment from soiled hands by playing, eating, smoking, and other activities.
SEC EQUIPMENT BUILDING
It is not known if the SEC Equipment Building is currently used for a business or commercial activity, although some activity has been noted at the site. Both PCDD/PCDF and PCB sampling have not been conducted inside the building. On-site workers, if any, may be exposed to PCBs via inhalation, dermal contact, or inadvertent ingestion from soiled hands by working in the SEC Equipment Building, if the building is contaminated. Because PCB-contaminated oils were reportedly burned as a starter fuel in the building (1), incomplete combustion may have generated PCDD and PCDF compounds.
C. Eliminated Exposure Pathways
REMEDIAL WORKERS
Remedial workers could have been, are, or will be exposed to contaminants in a variety of environmental media while conducting on-site activities in the past, present, and future, respectively. It is unlikely that such exposures would be at levels of concern however, provided appropriate work practices, as defined by the state or federal regulatory or permitting authorities, such as the Occupational Safety and Health Administration (OSHA), are followed. Those include worker education, certification, supervision and training, and use of personal protective equipment.
FOOD CHAIN
Because no farm animals or vegetable gardens were seen in the site vicinity, human exposure should not be occurring through consumption of contaminated plants or farm animals. EPA has documentation stating there are no game and edible species of fish in Arbuckle Creek. Except for snapping turtles, the food chain (fish, vegetable gardens, and farm animals) was eliminated as an exposure pathway.
Adults and children who trespass can gain access to the Shaffer site, and may be exposed to polychlorinated biphenyl compounds (PCB) through dermal contact, dust inhalation, and inadvertent ingestion of contaminated soils. Exposure to contaminated on-site soils would be intermittent, and therefore, represents a minimal potential exposure. Children who played in off-site residential yards and those who played in Arbuckle Creek may have been exposed to contaminated soils and sediments. Past PCB sediment concentrations were high. Nothing is known about the frequency and duration of exposure. Therefore, ATSDR could not evaluate the public health implications of this exposure. Current concentrations of PCBs in sediment and residential soil appear to be low and decreasing with time; therefore exposures are thought to have minimal public health consequences. Remedial and other workers on the site have the same potential routes for exposure; however, remedial workers normally follow appropriate work practices and use personal protective equipment. There are some concerns regarding possible exposure through the ingestion of snapping turtles from Arbuckle Creek. Past on-site workers and their families were probably exposed to levels of PCBs; however, no data exist to quantify exposure. The public health implications of this exposure cannot be assessed.
There was vandalism and recreational use of the site prior to EPA emergency removal operations. Activities may have resulted in children and adults being exposed to high levels of PCBs; however, the extent, duration, and magnitude of exposure cannot be quantified. It would be difficult for ATSDR to assess public health implications of this past exposure without quantitative data.
ATSDR has received reports that PCB-contaminated oils were used in the past as a starter fuel to ignite coal for heat at the SEC Equipment Building and for off-site residential use. Therefore, the equipment building and residences that used the PCB-contaminated oil may be contaminated with PCBs, PCDDs, and PCDFs. Sampling inside the building was not performed. Contamination inside the building would most likely affect employees and would best be assessed by the Occupational Safety and Health Administration (OSHA) and the National Institute of Occupational Safety and Health (NIOSH). For residences that used the PCB-contaminated oils, no environmental data were available and the extent, duration, and magnitude of exposure could not be quantified. Without this information, these past exposures cannot be assessed by ATSDR.
At this time, there is not enough evidence to state that PCBs are carcinogenic in humans. EPA has classified PCBs as probable human carcinogens and recommends that all commercial PCB mixtures be considered to have similar carcinogenic potential. The International Agency for Research on Cancer (IARC) has classified PCBs in Group 2B based on sufficient evidence in animals, inadequate evidence in humans, and inadequate evidence for mutagenicity. IARC places chemicals into groups based on their evaluation of the cancer risk to humans from exposure to the chemical. Group 2 are those chemicals felt to be probably carcinogenic for humans. Group 2 is divided into two subgroups A and B with A having stronger evidence to indicate carcinogenicity than those in group B. In addition, NIOSH has recommended that PCBs be regarded as potential human carcinogens in the workplace (2,17).
Studies in animals and occupationally exposed groups have indicated that the liver and skin are the major target organs. Increased serum levels of liver enzymes have been seen in some studies of workers (2,18,19). Other studies have not shown a large number of workers with high enzymes (20,21,22). Increased levels of liver enzymes detectable in the serum are an easily measured marker of liver damage. Dermatological effects such as chloracne, pigmentary changes, rashes, and swelling or thickening have been described (2,20,21,23,24,25). Chloracne is a chronic skin condition produced by environmental exposures to certain compounds containing chemicals such as chlorine or bromine. There are two predominant skin lesions seen in chloracne cases: the chloracne cyst and the comedo. The chloracne cyst is a skin colored sac with a central opening. The comedo is a blackhead. Most studies in workers indicated that those with chloracne also have some evidence of liver injury (22,26).
PCBs have been shown to cross the placenta, thus there may be PCB exposure to fetuses of women who have increased PCB exposure. There have been several reports of effects on development in children of mothers who were occupationally exposed to PCBs during pregnancy or were consumers of PCB-contaminated fish during pregnancy. Those effects have included decreased birth weight, gestational age, and cognitive functioning (2,27,28). One subpopulation sensitive to PCBs are nursing infants. PCBs are fat soluble so they tend to concentrate in the areas of the body with high fat content. Breast milk has a high fat content so PCBs are excreted through the milk. Infants may maintain higher levels of PCBs in their bodies. There have not been many studies to determine the effects of exposure from PCB-contaminated breast milk. One study that was conducted by Gladen et. al. did not demonstrate any effect on infant psychomotor responses associated with exposure through breast feeding (29). During ATSDR's site visit, it was noted that there were many children in the vicinity of the site. The number of women in the area who breast-feed their children is not known. It must be emphasized that ATSDR does not know if fetal exposure to PCBs through breast milk or during pregnancy is occurring and that such exposures are speculative. For those exposures to occur, mothers must eat contaminated turtles or breast-feed after eating contaminated turtles. ATSDR has not been able to confirm or deny the hypothesis that contaminated turtles have been eaten from the site vicinity.
There have not been any studies of PCB levels in the general population, but there have been several small studies of specific groups (see references 2 and 22 for reviews). While differences in methodology make it difficult to directly compare these studies, they do give an indication of the ranges of PCB levels that have been seen. Tables 10 and 11 give a summary of some of these data which will be used in the discussion of health outcome data evaluation in this public health assessment.
The polynuclear aromatic hydrocarbons (PAHs) benzo(a)anthracene, benzo(b)fluorene, benzo(a)pyrene and dibenzo(a,h)anthracene were found at levels exceeding comparison values. PAHs are common contaminants at many sites, often associated with burning of combustible materials. These compounds have little acute effect but have been shown to cause cancer. Long-term exposure to levels exceeding comparison values may result in a slight increased risk of cancer. However, exposure to these contaminants at SEC is mostly infrequent and sporadic, and is probably insufficient to have any effect.
Appropriate comparison values could not be determined for a 2-methylnapthalene and acenaphthylene, dibenzofuran, and di-n-octyl phthalate. Therefore, we are unable to determine if the concentrations present are of public health concern. Comparison values may be unavailable for a number of reasons, including uncertainties and conflicting results in animal studies, lack of human studies, and scientific controversies. ATSDR is working to establish comparison values for compounds commonly found at contaminated sites.
Acenaphthylene and 2-methylnaphthalene are PAHs. No comparison values are available for these two PAHs. However, PAHs, in general, have little acute toxicity especially at the relatively low concentrations found at SEC. These two PAHs are not thought to cause cancer.
Dibenzofuran and di-n-octyl phthalate also have no comparison values, largely because the toxicity of these compounds has been little studied. Phthalates as a group have low acute toxicity, but some have been shown to be carcinogenic by non-genetic mechanisms. Di-n-octyl phthalate itself has not been adequately evaluated for carcinogenicity. Dibenzofuran is of low acute toxicity, while its long term effects are uncertain.
B. Health Outcome Data Evaluation
ATSDR has reviewed the results of PCB analysis of adipose (fat) tissue in 20 residents and former employees. In addition, the results from the 1986-1987 serum (blood) sampling was also evaluated (30,31). Those results are provided in the following table:
PCBs in Minden Residents
| SAMPLE | NUMBER | AVERAGE | RANGE |
| Adipose |
20
|
0.564 ppm
|
0.100 - 0.970 ppm |
| Serum |
10
|
7.7 ppb
|
5 - 14 ppb |
ppm = parts per million
ppb = parts per billion
The range and mean values seen in the Minden group are consistent with the levels seen in groups without occupational exposure (Table 12). A comparison with the above table and Table 12 does not indicate increased exposure in Minden residents compared to other populations.
Regarding Table 12, the median PCB levels in blood serum of the general population of the United States without occupational exposure has been found in most studies to be less than 10 ng/mL (ppb), but sex and age of the population must be taken into account (32). Mean serum levels were usually between 4 and 8 ng/mL (ppb), and 95 percent of individuals had levels less than 20 ng/mL (ppb) (2). There have been fewer studies utilizing adipose samples in the population. While there is variation in groups studied, PCB levels in adipose tissue are between 100 and 200 times the levels in blood serum (22). Studies attempting to correlate serum and adipose levels with health effects have had inconsistent results (2).
Follow-up serum samples were drawn on 43 individuals in Minden in 1990. The results from those samples are consistent with other results. There were 30 persons with no detectable levels of PCBs in their sera (Detection limit was 3 ppb). The range of levels seen in persons with detectable PCBs was 3.2 - 11.0 ppb (Mean = 5.0 ppb) (33). The levels are consistent with those seen in the general population.
One infant/child (age unknown, but listed as "0" years when the blood sample was taken) in the 1990 sampling indicate a sera PCB level of 11.0 ppb. This PCB level is unusual for an infant and the accuracy of the test is suspect and should be verified. If the results are verified, further tests and investigation may be necessary to determine and eliminate the source of the PCB exposure.
There have been a few studies of populations surrounding hazardous waste sites in which exposure assessment has been done (Table 13). PCBs in Minden sera are not significantly higher than the ranges of PCBs seen in comparison populations.
A review of the census data for the area demonstrate the factors that may influence exposure and the risk from exposure. The census data indicate that the population is stable and the majority lives in rural areas. A large portion of the population are employed in the mining industry. Industry exposures must be considered when conducting health studies in the area. The per capita income for the Minden area is less than $4,000. Income influences access to health care and the types of illnesses that the population experiences.
Several health studies were conducted in the Minden area by individuals or organizations. The West Virginia Department of Health (WVDOH) conducted a cancer mortality study of Oak Hill (Minden is on the outskirts of Oak Hill) for the period 1979-1981 (9). The study concluded that there was a greater rate of respiratory cancers in males in the Oak Hill population than in the United States. An evaluation of the study by ATSDR indicated several weaknesses that may have affected the conclusions of the study. One of the concerns is that the study chose to use the U.S. standardized mortality rates rather than West Virginia mortality statistics. Because many males in the area have a history of working in the coal mines, they have a higher risk for developing respiratory cancers. That additional exposure would not be reflected in the rates for the United States. The results of ATSDR's evaluation of the study are found in Appendix 3. Another aspect of the cancer mortality information comes from data on the Riggans mortality tapes produced by EPA and National Cancer Institute (NCI). Review of the data indicates that Fayette county has had an increased rate of respiratory cancer deaths when compared to the U.S. The increase has been present since the decade of the sixties so factors such as age and occupation that may be associated with the increase would have existed at that time. This increase existed before SEC was operating; therefore, other factors (e.g., age, smoking status, socioeconomics) that could lead to this increase must be examined (8).
In the summer of 1986, the Concerned Citizens to Save Fayette County (CCSFC) sponsored a health registry of random sampling of Minden and Rock Lick residents. The study was conducted by Vanderbilt University's Appalachian Student Health Coalition and is commonly known as the Vanderbilt Study. In 1989, there was a survey of residents in Page, WV that was intended to provide a control group for the 1986 Vanderbilt Study. The survey in Page was part of a Virginia Student Environmental Health Project Intern Report (10). The reports indicated that certain health conditions such as shortness of breath, unexplained weight loss, persistent cough, and urinary tract infections occur at a higher rate in Minden than in Page, WV. The increased incidence of health problems described in the report for the Minden community does not correlate well with current PCB literature regarding symptoms and health effects associated with PCB exposure. ATSDR's Division of Health Studies reviewed the Vanderbilt study and noted the strengths and weaknesses of the surveys conducted in Minden (1986) and Page (1989). The full results of that review are in Appendix 3. ATSDR concluded that a clear relationship could not be established between PCB contamination and the significantly different observed rates of the four symptoms (described above). In addition there were variables such as age, smoking history, and history of working in the mines that may affect symptom prevalence. Those variables were not addressed in the two surveys. Socioeconomic and nutritional status should also be considered. When such variables are not controlled for, one cannot determine what accounted for the findings (Appendix 3). Although the surveys could not establish a definitive link between PCB contamination and specific health conditions, the surveys did serve to identify diseases of concern for the community such as those described earlier (shortness of breath, unexplained weight loss, persistent cough and urinary tract infections).
Another study was conducted by a Beckley, West Virginia gynecologist (11) who submitted a health survey to ATSDR in 1989 for an evaluation. Although the survey did indicate areas of health concern similar to those mentioned previously, the survey did not encompass the Minden area. The design and methodology of the study were obscure. Therefore, ATSDR was not able to draw any conclusions for this public health assessment.
Past occupational exposures to high concentrations of PCBs have been reported at the Shaffer site. A number of former Shaffer employees were interviewed by ATSDR by telephone to evaluate past work exposure to PCBs, and past and current health (1). One of six employees interviewed reported getting a skin rash while working with PCB oils. Most employees rated their current health as generally good and could not relate any adverse health effects to working with PCBs. One of the employees participated in the CCSFC serum and adipose (fat) sampling in May 1990.
The area around Minden, West Virginia is economically depressed. Large portions of the population are employed in mining. Such factors must be taken into account when attempting to determine health consequences from possible exposure to PCBs. Previous health studies conducted in the area have identified health outcomes of concern (shortness of breath, unexplained weight loss, persistent cough, and urinary tract infections), but have not found outcomes that have been consistently linked to PCB exposure. The results of the serum and adipose tissue did not indicate body burdens of PCBs that are significantly different than those seen in other non-occupationally exposed populations. The review of the toxicological information and the health information do not indicate high levels of exposure to PCBs at the present time or increases in health outcomes that can be linked to PCB exposure.
C. Community Health Concerns Evaluation
A discussion of PCB effects on the liver and skin are contained in the Toxicological Evaluations portion of the Public Health Implication Section of this document. The effects of PCBs on the kidney and respiratory system occur mostly at exposures considerably higher than found at this site. PCBs may cause kidney damage (hydronephrosis) including swelling and structural damage. Little effect is seen in the lung, although the lung may be a major route of absorption. PCBs can damage lung membranes, and may lower tolerance to infection, but these effects are unlikely at the exposures seen at this site.
As part of ATSDR's Planned Public Health Actions, ATSDR will contact the National Institute for Occupational Safety and Health (NIOSH) regarding concerns for worker health and provide NIOSH with a copy of the Shaffer Equipment Company Public Health Assessment in order that they may be advised of the potential for past PCB occupational exposures at Shaffer. Follow-up worker health studies, if deemed necessary, are under the purview of NIOSH. Because the latest sampling results indicate low levels of PCBs in residential soils (2.1 mg/kg, March 1990) that are not considered to be a public health hazard, and the results of the two PCB serum tests indicate levels of PCBs in Minden residents not significantly higher than those seen in comparison populations, a health study of downstream residents does not appear to be beneficial or justified at the present time. Contaminants other than PCBs are not likely to present a health concern at the exposures likely to occur in the future or to have occurred in the past.
Additional on- and off-site sampling was conducted by EPA in March 1990, May 1990, and again in June 1990 in response to reports made by the CCSFC. Further on-site sampling and removal actions were conducted between November 1990 and January 1991. EPA is also investigating other sites that were allegedly used to dispose of PCBs from the SEC. These other sites are distinct and separate from the SEC site and will not be addressed in this public health assessment. Local raw water supplies from the Minden Mine intake and the Rocklick Mine intake were sampled in March 1990 by EPA in response to reports by the CCSFC that PCBs were dumped into the mine shafts. All results were below detection limits except for levels of methylene chloride, which ATSDR considered a result of laboratory contamination. However, the PCB detection limits of the measuring instruments were higher than ATSDR's comparison values. Therefore, additional sampling is needed to determine if these raw water sources could affect public health.
A. Recommendations and HARP Statement
RECOMMENDATIONS
HARP STATEMENT
In accordance with the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) of 1980, as amended, the data and information associated with the Shaffer Equipment Company Site have been evaluated for appropriate public health actions. There is currently no indication that people are being exposed to contaminants associated with the SEC site at levels of public health concern. Although exposures may have occurred in the past, there is insufficient information to document exposure duration or levels. ATSDR's Health Activities Recommendation Panel (HARP) has determined that an environmental health education program is recommended to advise public health professionals and the local medical community of the nature and possible consequences of exposure to contaminants at the Shaffer Equipment site. The value of obtaining a complete and accurate exposure history will be stressed as part of the program. In addition, information that is provided on PCBs may include, but not be limited to, the physical nature of the contaminant, potential exposure pathways (i.e., soil, water, air, food) and exposure routes (i.e, inhalation, ingestion, dermal), potential health effects, symptoms of exposure and testing and treatment.
In addition, HARP recommends that the public health assessment be forwarded to the National Institute for Occupational Safety and Health for their review and possible investigation of past worker exposure to PCBs at the Shaffer Site.
The Shaffer Equipment Company data that has been evaluated for follow-up health activities included copies of laboratory results from the Pacific Toxicology Laboratories and the National Health Laboratories, Inc. Based on this information, HARP recommends that the results of the PCB level of the infant/child with a PCB level of 11 mcg/l be evaluated for accuracy. If this result is correct, the child should have a repeat test and a case investigation to determine possible exposure routes to PCB materials, including in-utero exposures.
ACTIONS PLANNED
(listed alphabetically)
Jeffrey A. Church, REHS
Environmental Health Scientist
Community Health Branch
Donald Y. Joe, P.E.
Environmental Engineer
Community Health Branch
Virginia Lee, M.D., M.P.H.
Medical Officer
Federal Facilities Branch
Thomas Umbreit, Ph.D.
Toxicologist
Community Health Branch
ATSDR Regional Representative:
Charles J. Walters
Public Health Advisor
Region III
1. Telephone interviews with former Shaffer Employees. Donald Joe, ATSDR. June 1, through June 21, 1990.
2. ATSDR. Toxicological Profile for Selected PCBs (Aroclor -1260, -1254, -1248, -1242, -1232, -1221, and -1016). June 1989.
3. Sittig M. Handbook of Toxic and Hazardous Chemicals and Carcinogens. 2nd edition. Park Ridge, NJ: Noyes Publications, 1985.
4. Superfund Record of Communications from C. J. Walters to R. Caron (EPA) dated December 11, 1984.
5. U.S. Bureau of the Census. 1980 Census of Population: Volume 1: Characteristics of the Population: Chapter B: General Population Characteristics. Department of Commerce. Issued August 1982.
6. U.S. Bureau of the Census. 1980 Census of Population: Volume 1: Characterisitcs of the Population: Chapter C: General Social and Economic Characteristics. Department of Commerce. Issued August 1983.
7. U.S. Bureau of the Census. County and City Data Book 1988. Pages 562-587.
8. Riggans WB, Van Bruggen J, Acquavella JF, et. al.. U.S. Cancer Mortality Rates and Trends, 1950-1979: Vols. I, II, III. Research Triangle Park, N.C.: U.S. Environmental Protection Agency and National Cancer Institute, 1983. EPA publication number (EPA) 600/1-83-015a.
9. Environmental Epidemiology Task Force, West Virginia Department of Health, Oak Hill, W.V., Investigation--Stage One: Initial Assessment, January 11, 1983.
10. Hirschman D and S Walsh. A Virginia Student Environmental Health Project Intern Report: Community Health and PCB Exposure in Minden, West Virginia. September 1989.
11. Letter from Charles W. Merritt, M.D. to J.A. Lybarger, M.D.. March 16, 1989.
12. Eastern Scientific Laboratories, PCB data analysis sheets of 16 soil samples collected at SEC site on May 22, 23, 1990., Purchase order number #0454 dated June 7, 1990.
13. Joe, Don. Comments on the NUS Sampling Plan to the site investigation officer of the USEPA.
14. ATSDR. Toxicological Profile for 2,3,7,8-Tetrachlorodibenzo-p-dioxin. June 1989.
15. Superfund Record of Communications. Telephone call from L.Haddy to C.J. Walters, June 25, 1990.
16. Environmental Protection Agency Region III. Federal On-Scene Co-ordinator's Report; Shaffer Equipment Site, Minden, West Virginia.
17. Shalat SL, True LD, Fleming LE, and PE Pace. "Kidney Cancer in Utility Workers Exposed to Polychlorinated Biphenyls (PCBs)." British Journal of Industrial Medicine 1989; 46:823-824.
18. Lawton RW, Ross MR, Feingold J, and JF Brown. "Effects of PCB Exposure on Biochemical and Hematological Findings in Capacitor Workers." Environmental Health Perspectives 1985; 60:165-184.
19. Fischbein A. "Liver Function Tests in Workers with Occupational Exposure to Polychlorinated Biphenyls (PCBs): Comparison with Yusho and Yu-Cheng." Environmental Health Perspectives 1985; 60:145-150.
20. Fischbein A, Wolff MS, Lilis R, Thorton J, and IJ Selikoff. "Clinical Findings among PCB-Exposed Capacitor Manufacturing Workers." Annals of the New York Academy of Sciences 1979; 320:703-715.
21. Reggiani G and R Bruppacher. "Symptoms, Signs, and Findings in Humans Exposed to PCBs and Their Derivatives." Environmental Health Perspectives 1985; 60:225-232.
22. Kimbrough RD. "Human Health Effects of Polychlorinated Biphenyls (PCBs) and Polybrominated Biphenyls (PBBs)." Annual Review of Pharmacology and Toxicology 1987; 27:87-111.
23. Emmett EA, Maroni M, Schmith JM, Levin BK, and J Jeffreys. "Studies of Transformer Repair Workers Exposed to PCBs: I. Study Design, PCB Concentrations, Questionaire, and Clinical Examination Results." American Journal of Industrial Medicine 1988; 13:415-427.
24. Fischbein A, Wolff MS, Berstein J, Selikoff IJ, and J Thornton. "Dermatological Findings in Capacitor Manufacturing Workers Exposed to Dielectric Fluids Containing Polychlorinated Biphenyls (PCBs)." Archives of Environmental Health 1982; 37(2):69-74.
25. Fischbein A, Rizzo JN, Soloman SJ, and MS Wolff. "Oculodermatological Findings in Workers with Occupational Exposure to Polychlorinated Biphenyls (PCBs)." British Journal of Industrial Medicine 1985; 42:426-430.
26. Taylor JS. "Environmental Chloracne: Update and Overview." Annals of the New Academy of Sciences 1979; 320:295-307.
27. Taylor PR, Stelma JM, and CE Lawrence. "The Relation of Polychlorinated Biphenyls to Birth Weight and Gestational Age in the Offspring of Occupationally Exposed Mothers." American Journal of Epidemioloy 1989; 129(2):395-406.
28. Jacobson JL, Jacobson SW, and HEB Humphrey. "Effects of In Utero Exposure to Polychlorinated Biphenyls and Related Contaminants on Cognitive Functioning in Young Children." Journal of Pediatrics 1990; 116:38-45.
29. Gladen BC, Rogan WJ, Hardy P, Thullen J, Tingelstad J, and M Tully."Development after Exposure to Polychlorinated Biphenyls and Dichlorodiphenyl Dichloroethene Transplacentally and Through Human Milk." Journal of Pediatrics 1988; 113:991-995.
30. Pacific Toxicology Laboratories. Results of the Adipose Tissue Analysis for Selected Congeners of PCBs. June 1990.
31. National Health Laboratories. Results of Serum Analysis of Minden Residents for Total PCBs. 1986.
32. Kreiss K. "Studies on Populations Exposed to Polychlorinated Biphenyls." Environmental Health Perspectives 1985; 60:193-199.
33. Pacific Toxicology Laboratories. Results of Serum PCB Analysis of Minden Residents. 1990.
34. Fait A, Grossman E, Self S, Jeffries J, Pellizzari ED, and EA Emmett."Polychlorinated Biphenyl Congeners in Adipose Tissue Lipid and Serum of Past and Present Transformer Repair Workers and a Comparison Group." Fundamental and Applied Toxicology 1989; 12:42-55.
35. Emmett EA, Maroni M, Jeffreys J, Schmith J, Levin BK, and A Alvares. "Studies of Transformer Repair Workers Exposed to PCBs: II. Results of Clinical Laboratory Investigations." American Journal of Industrial Medicine 1988; 14:47-62.
36. Kreiss K, Roberts C, and HEB Humphrey. "Serial PBB Levels, PCB Levels, and Clinical Chemistries in Michigan's PBB Cohort." Archives of Environmental Health 1982; 37(3):141-147.
37. Jacobson JL, Humphrey HEB, Jacobson SW, Schantz SL, Mullin MD, and R Welch. "Determinants of Polychlorinated Biphenyls (PCBs), Polybrominated Biphenyls (PBBs), and Dichlorodiphenyl Trichloroethane (DDT) Levels in the Sera of Young Children." American Journal of Public Health 1989; 79(10):1401-1404.
38. Jensen AA. "Background Levels in Humans." In: Kimbrough RD and AA Jensen, eds. Halogenated biphenyls, terphenyls, naphthalenes, dibenzodioxins, and related products. New York: Elsevier, 1989:345-380.
39. Stehr-Green PA, Burse VW, and E Welty. "Human Exposure to Polychlorinated Biphenyls at Toxic Waste Sites: Investigations in the United States." Archives of Environmental Health 1988; 43(6):420-424.
40. CEIMIC Corporation, Data analysis sheets, Contract number 68D90024, 1990.
41. Environmental Health Research and Testing, Inc., Data analysis of 30 samples received June 18, 1990., dated June 29, 1990.
42. MDS Laboratories, Data analysis of samples received on November 6, 1990, November 30, 1990, and January 28, 1991.
REFERENCES USED BY ATSDR BUT NOT CITED
1. Swackhamer DL, Armstrong, DE. 1987. Distribution and characterization of PCBs in Lake Michigan water. J Great Lakes Res 13:24-26.
Next Section Table
of Contents
