PUBLIC HEALTH ASSESSMENT
NATIONAL ELECTRIC COIL/COOPER INDUSTRIES
DAYHOIT, HARLAN COUNTY, KENTUCKY
In conducting a public health assessment, ATSDR health assessors identify and review allavailable environmental data related to contamination at the site. The on- and off-site portions ofthis section describe sampling that has been done and identify contaminants of concern. Thequality of the environmental data is discussed in the Quality Assurance and Quality Controlsubsection. Physical and other hazards not related to toxic substances, if any, are described in thePhysical and Other Hazards subsection. The following paragraphs discuss selectingcontaminants of concern and searching the Toxic Chemical Release Inventory (TRI) database.
The data referred to in the contaminants of concern tables (Appendix A) were collected byconsultants to Cooper Industries, a potentially responsible party, or by EPA. ATSDR evaluatesthe contaminants in subsequent sections of this public health assessment to determine whetherexposure to them has public health significance. ATSDR selects and discusses contaminants ofconcern using the following information:
- o concentrations of contaminants on and off the site;
o quality of field data, laboratory data, and sample design;
o comparison of on- and off-site concentrations with comparison values for noncarcinogenic andcarcinogenic endpoints; and
o community health concerns.
The listing of a contaminant in the contaminants-of-concern tables does not mean that it willcause adverse health effects if people are exposed at the specified concentrations. Rather, thelisting of a contaminant indicates which contaminants will be discussed further in this publichealth assessment. When a contaminant is considered to be of concern in one medium, thepresence or absence of that contaminant in all media will be discussed. In addition, when acontaminant is considered to be of concern on site, its potential to migrate off site also will be discussed. The potential for adverse health effects resulting from exposure to the contaminants of health concern is discussed in the Public Health Implications section of this document.
The methodology behind choosing contaminants of concern is explained further in the PublicHealth Assessment and Guidance Manual (7). The following assumptions were used in thispublic health assessment: Child: body weight = 10kg; water ingestion rate = 1 liter/day; soilingestion rate = 200 mg/day; pica soil ingestion rate = 5000 mg/day; and inhalation rate = 15meters cubed (m3)/day. Adult: body weight = 70kg; water ingestion rate = 2 liter/day; soil ingestion rate = 100 mg/day.
ATSDR uses comparison values -- contaminant concentrations in specific media consideredprotective of public health -- to select contaminants for further evaluation. ATSDR and otheragencies have developed the comparison values to provide guidelines for estimating contaminantconcentrations in media that are not likely to cause adverse health effects, given a standard daily ingestion rate and standard body weight.
The comparison values include environmental media evaluation guides (EMEGs) and cancer riskevaluation guides (CREGs). EMEGs are media-specific comparison values used to selectcontaminants of concern. CREGs are estimated contaminant concentrations expected to cause nomore than one excess cancer in a million persons exposed over a lifetime (70 years).
Maximum contaminant levels (MCLs) are limits on chemical concentrations in drinking waterthat EPA considers protective of public health (considering the availability and cost-effectivenessof water treatment technology) over a lifetime (70 years) at an exposure rate of 2 liters of waterper day (for an adult). Maximum contaminant level goals (MCLGs) are EPA drinking waterhealth goals set at levels at which no known or anticipated adverse health effects would beexperienced by exposed persons. MCLs are EPA regulations that can be enforced; MCLGs aregoals that are not enforceable. EPA's reference dose (RfD) and reference concentration (RfC) areestimates of the daily exposure to a contaminant unlikely to cause adverse health effects. Insome cases, ATSDR uses comparison values calculated from RfDs. For example, a referencemedia evaluation guide (RMEG) is a contaminant concentration in water or soil at which theexposed individual would receive a dose equal to the RfD from which it was calculated.
During its assessment of the NEC site, the preparers of this document searched the ToxicChemical Release Inventory (TRI). TRI is an on-line database, maintained by EPA, containinginformation (self-reports from chemical manufacturers and other companies throughout theUnited States) about more than 320 different substances released from facilities into theenvironment between 1987 and 1989 (8).
No releases were reported under the facility name, National Electric Coil, in the Dayhoit areabecause the company was not in business during those years. A search by zip code during thesame years showed no reported releases from other industries in the area. Site-relatedcontaminants of health concern are discussed in the following sections.
Although the property boundary of NEC ends before the Cumberland riverbank area, theriverbank adjacent to NEC is considered in this assessment to be on site because of the directdischarge of waste to the river and dumping of waste in the area.
Waste material generated by NEC was removed from the site before October 1991. Soil andwaste were transported and disposed of by Chemical Waste Management consultants undercontract to Cooper Industries (1). Buried drums were removed from the Cumberland riverbank,and hazardous materials were moved to Chemical Waste Management's facility in Emelle,Alabama. The hazardous substances were VOCs, including chlorinated solvents; PCBs,primarily Aroclors; and heavy metals. On-site sewers and drains were flushed, and contaminatedsoils were removed. The varnish tank area was capped with concrete.
Although there are sediments in the riverbank area, they are discussed in the Soil subsection because no distinction was made in the original data sets. The riverbank area, however, has been separately characterized.
Surface mixed with Subsurface
On-site soil samples were taken from four major areas: the riverbank fill area, the outfall 1 area,the equipment and drum storage area, and the southern fence line area (9,10) (Figures 2 and 3). The outfall 1 area is the previous northerly discharge point for wastes exiting a pipe from theNEC facility. Soil samples were analyzed for VOCs, PCBs, and metals. The shallowest depth ofsampling reported for the 1991 data set was 0-12 inches. Sampling depths for each area arespecified in the contaminants-of-concern tables (Appendix A); the levels provided areconcentrations in soil before removal. Contaminated soil was excavated until laboratory resultsshowed concentrations were below the EPA-established target levels. Figures 2 through 4summarize past soil contamination in the previously mentioned areas.
PCB levels exceeded comparison values in three main areas: the riverbank fill, outfall 1, andsouthern fence line areas as shown in Table 1 (10). Sampling during soil removal showed PCBconcentrations above 10 ppm from the surface to a depth of four feet in riverbank fill areas. Thehighest concentrations of PCBs were in the 0-12 inches depth interval (range: nondetectable to610 ppm) of riverbank fill. PCB concentrations along the southern fence line area were foundprimarily in the top 12 inches of soil (range: nondetectable to 58 ppm in the 1991 data set). Some surface soil sampling, 0-6 inches, was conducted in 1989 at the border between NEC andthe mobile home park. It showed PCB contamination ranging from approximately 0.29 ppm to9.1 ppm. A surface soil sample also was taken from a low lying area between the NEC site andthe mobile home park; analysis showed PCB and lead contamination (11).
TCE was detected in the four areas sampled during remedial investigations as shown in Table 1. The number of samples with concentrations above the instrument detection level also are listed inTable 1. The table shows that a majority of soil samples contained TCE; samples taken belowoutfall 1 and on the banks of the Cumberland River showed TCE surface contamination at amaximum concentration of 15,000 ppm (11). Although there is no ATSDR comparison value for TCE in soil, the health implications of exposure to TCE will be discussed in the Public Health Implications section.
Heavy metals also were found in on-site soils as shown in Table 2. The data in Table 2 areconcentrations before removal. Arsenic, beryllium, and manganese concentrations exceededcomparison values most frequently. Soil in the riverbank fill area contained lead and chromiumconcentrations above target clean-up levels. Throughout the site, beryllium levels in soilexceeded comparison values. At the south fence line, beryllium and manganese exceededcomparison values. Leaching procedures conducted on some soil samples showed that arsenic,barium, and lead were potentially mobile.
Soil gas testing was conducted in September and October of 1989 to help determine the extent ofshallow groundwater contamination, and to evaluate VOC contamination (11). Approximately30 soil gas samples were obtained on site from a depth of 3 feet. Those samples were analyzedfor vinyl chloride, trans-1,2-dichloroethene (trans-1,2-DCE), and TCE. The concentrations ofthose VOCs in soil gas are shown in Table 3. Dichloroethene (DCE) and TCE were detected insoil gas. Vinyl chloride was not detected; however, the detection limit is above the comparisonvalue. The soil gas testing was completed before hydrocarbons were discovered in the shallowgroundwater; therefore, there are no soil gas data for fuel vapors.
Two aquifers underlie the NEC site: an aquifer in the shallow overburden and a deep bedrockaquifer. The overburden is, on average, approximately 25 feet thick (range: approximately 0 to80 feet thick) and consists of sand, silt, gravel, and clay (5). Groundwater in the overburden aquifer flows east from the site toward the Cumberland River as shown in Figure 5.
Depth to groundwater from surface at the NEC site ranged from 9 to 23 feet in April of 1991(10). The bedrock underlying the NEC site is the Cawood Sandstone member of the HanceFormation, a tight sandstone that yields small quantities of groundwater through fractures. TheHance Formation consists of sandstone, siltstone, shale, and coal (5). Pump test data andinformation gathered from the recovery system indicate a partial connection of the bedrockaquifer with the overburden (12). Groundwater in the bedrock aquifer flows along fractures andbedding planes in a southwesterly direction as shown in Figure 6. Many of the bedrock wells are artisan.
Groundwater contamination has been assessed in both the shallow and deep aquifers. Tenmonitoring wells were initially installed on site: six in the shallow aquifer and four in the deepbedrock aquifer. Two additional on-site monitoring wells were installed during the RemedialInvestigation. Contaminant tables for overburden (shallow) and deep aquifers are shown inTables 4-7. The organic contaminants were monitored approximately once a year for severalyears, 1989 through 1991. Monitoring began again in July 1993 after the groundwater treatmentsystem was on-line. The inorganic contaminants are defined by a single sampling pre-1991 andagain in 1993.
The shallow aquifer is contaminated with organic compounds, BTEX (benzene, toluene,ethylbenzene, and xylene) compounds, and total metals. Although benzene, toluene, andethylbenzene are listed as contaminants of concern, the concentrations of xylene are not highenough for inclusion. Traces of BTEX compounds were found in the deep bedrock aquifer insamples taken in 1993. The primary organic contaminants exceeding comparison values in bothaquifers were DCE, TCE, and vinyl chloride. Chromium, lead, manganese, and nickel are abovecomparison values in both aquifers on site. Heavy metals contamination is greater in the shallow aquifer (Tables 5 and 7).
The lateral extent of bedrock aquifer contamination is shown in Figure 7. The most concentratedzone of the VOC plume is located approximately 50 to 75 feet below ground surface and near thesouthern boundary of the NEC site (12). The 1993 data, collected after the pump and treatsystem was on-line, indicates that the lateral extent of plume has lessen and the concentrationshave decreased (12). Basically, the area of groundwater contamination appears to be shrinking.
Two of the on-site bedrock monitoring wells, #10 and #11, existed during the operation of NEC. Well #10 was used for drinking water through 1989, when it was found to be contaminated withvinyl chloride (30 ppb) and cis 1,2-dichloroethene (cis-DCE) (87 ppb). Well #11, which was notbeing used, was flushed, and the contaminant levels after flushing were vinyl chloride (7.4 ppb),cis-DCE (174.4 ppb) and TCE (75.6 ppb). Because both wells were drilled into bedrock, it ispossible that they provided a vertical pathway for contaminants. A connection between aquifersmay also exist because of fractures in the bedrock covered by sandy overburden.
Contaminants in both aquifers should be remediated during the pump and treat process. Citywater lines have been extended beyond affected areas.
Outside the NEC facility
No existing data could be found on air quality near the NEC facility during its operation. Vats ofmolten lead were used during processing; fumes may have been vented directly to the air (1). The Air Quality Section of KDEP received a complaint in May of 1980 about fumes from theexhaust fans at NEC.
Air was monitored during removal operations to ensure that remedial workers and nearbyresidents were protected from airborne contaminants. Air was monitored for total particulates,lead, chromium, and PCBs. TCE and xylene were analyzed using Drager tubes, but were notdetected. Air samples taken from the breathing zone of remedial workers were screened fororganic contaminants; results showed acceptably low levels of BTEX compounds, methyl ethylketone, and acetone. The organic concentrations were below permissible exposure limits (limitsset so that concentrations are not exceeded during any 8-hour work shift of a 40-hour workweek). Twelve samples of airborne particulates trapped in air sampling filters were analyzed for PCBs. The air filters were placed along the northern, southern, and eastern property lines. No PCBs were detected.
Inside the NEC facility
Thorough ventilation is needed when solvents such as 1,2-DCE are used. According tointerviews with former employees, ventilation at NEC was inadequate (13). Because of the lackof ventilation, volatilization of TCE from the vat and other areas inside the NEC facility was an exposure route for most workers. Former employees also reported poor air quality, primarilycaused by dust, which included asbestos fibers. A trace of asbestos (amosite) was detected in bag house dust.
During Air Stripping
Air stripping of groundwater contaminated with volatile organics is part of the remedial actionfor NEC. Initially, air sampling of the exhaust from the air stripper system's vapor phase carbonunit was continuous for about 10 weeks after start up on July 30, 1993 (12). On-going airsampling is being conducted using summa canisters on at least a monthly basis. The air samplesare being analyzed for volatile organic contaminants, including vinyl chloride, trichloroethene,and 1,2-dichloroethene. The summa canister air samples for stripper exhaust (August 1993 -June 1994) contained the following reported maximum values 1125 ppb vinyl chloride, 6900 ppbtrichloroethene, and 39,000 ppb 1,2-dichloroethene. The specific volatile organics listed above were well below the US EPA emissions standards (12).
Monitoring results at the Holiday Mobile Home Park fenceline and park property are reportedunder the Ambient Air subsection for Off-site Contamination.
Contamination of the riverbank adjacent to NEC (shown in Tables 2 and 3) has been reported ason-site soil contamination. Samples taken from the Holiday Mobile Home Park are consideredoff site (south of the fence line at the southern property boundary).
Because the NEC operation was close to the Cumberland River, sediment from the river wastested for PCBs. Three upstream and seven downstream samples were analyzed. AlthoughPCBs were not detected upstream of NEC, they were detected downstream (Table 8). This pattern of PCB contamination suggests that NEC or the nearby transformer was a major source of the contamination.
A sediment sample of orange leachate, taken one-half mile from NEC in October 1989, showed amaximum barium concentration of 44.6 ppm (1). Site investigations have not included analyses of fish specimens from the Cumberland River for PCBs or metals. However, bioaccumulation of contaminants in fish is discussed under the Potential Exposure Pathways, Food Chain Pathway (Fish) subsection.
Soil gas testing was conducted in September and October 1989 to help determine the extent ofshallow groundwater contamination, and to evaluate VOC contamination (11). Approximately85 off-site soil gas samples were taken: 45 from the Holiday Mobile Home Park and 40 from theDayhoit/Fresh Meadows area. Three compounds (trans 1,2-DCE, TCE, and vinyl chloride) wereanalyzed for in off-site soil gas at a depth of 3 feet. The only compound detected was TCE; itexceeded a comparison value for inhalation in three soil gas samples obtained south of the NECsite, along the Cumberland River. TCE in soil gas was detected several hundred feet from theNEC site. The soil gas testing was completed before the discovery of hydrocarbons in theshallow groundwater; therefore, there are no data for fuel vapors. No additional soil gassampling and analysis has been completed to help define the current status of off-site soil gasmigration. Because of the high water table, however, there are no basements in the immediatevicinity, and vapors are unlikely to reach explosive levels when vented to the atmosphere. Nohealth evaluation can be made because of the lack of information on the air's volatilehydrocarbon content.
Results of samples from the border of the mobile home park and NEC that were tested for PCBsare summarized in Table 10. The results show low levels of PCBs. A maximum concentrationof 1.5 ppm aroclor-1254 was detected in surface soil near the fenceline of the NEC site and themobile home park (Table 10). The maximum concentration detected in August 1993, 0.96 ppmof aroclor-1248, was also found near the fenceline of the site (Figure 7). The maximum totalconcentration of PCBs for the 1993 data is about 1.1 ppm (Table 10).
Heavy metals were analyzed for Holiday Mobile Home Park soils in August 1993. Althoughsome metals were above ATSDR comparison values as shown in Table 11, they were withinnatural soil composition levels and background levels for the area. However, lead concentrations were elevated (around 200 ppm) at two locations compared with localbackground concentrations of approximately 14 ppm or less.
Data on dioxin/furan contamination in soil was provided to us in a February 1993 report andaddressed by ATSDR in a May 1993 Health Consultation (15)(appendix D). Our consultationwill be summarized here. Additionally, further sampling for dioxin/furan contamination in soiltook place during August 1993. Soil sampling for dioxins/furans indicates that very lowconcentrations are present in the soil of the Holiday Mobile Home Park. The February 1993report indicated a maximum toxicity equivalence factor (TEF) of 0.005 ppb PCB. ATSDR'slevel of concern is 1 ppb. Five additional soil samples were taken in the Mobile Home Parkduring August 1993. These samples had TEFs at approximately 0.015 ppb or less.
Several surface water samples were taken from the Cumberland River, both upstream anddownstream of NEC, in March 1989 (11). Analysis did not find vinyl chloride or DCE. Also in March 1989, discharge adjacent to the Holiday Mobile Home Park was sampled for Volatile Organic Compounds (VOCs). Low levels of xylenes were found.
Air stripper discharge to Surface Water
Liquid effluent from the air stripper is being discharged to the Cumberland River. Water isdischarged after passing through the packed column air stripping tower. Cooper Industries has a Kentucky Pollutant Discharge Elimination System (KPDES) permit (13). ATSDR reviewedweekly monitoring data on this effluent collected in April through June 1994. Low levels ofTCE (nondetectable to 6.6 ppb) and 1,2 DCE (nondetectable to 18.5 ppb) were present in thedischarge. Lead (nondetectable to 8 ppb) and iron (400 to 1200 ppb) were also detected (13). The effluent was within discharge limits (monthly average and daily maximums) as set in the permit. There is no limit set for iron.
A water well survey of off-site wells, conducted in February and March 1989, showedgroundwater in the bedrock aquifer was contaminated with vinyl chloride and DCE (11)(Table12). Both of those organic contaminants are degradation products of TCE, which is fairlypersistent in subsurface waters. In September 1989, off-site wells were tested for VOCs. Additional organic contaminants were found including TCE (Table 12). Some wells have been re-tested periodically to check the extent of contamination.
Approximately 143 wells have been tested for VOCs. VOCs were not detected in most of thewells; however, approximately 13 wells (excluding the Holiday Mobile Home Park) were foundto be contaminated. A plume of groundwater contamination extends from NEC in anortheast-southwest direction (following fractures, bedding planes, and/or faults) and crossesbeneath the Cumberland River in the bedrock aquifer (Figure 8).
The 1989 soil gas sampling suggested that groundwater contamination in the overburden aquifer was limited to the NEC site and several hundred feet beyond it along the Cumberland River. Even though shallow groundwater flow is toward the river, some contamination may havemigrated toward the mobile home park. ATSDR was unable to find information about, includinglocations of, shallow off-site groundwater wells.
According to 1993 samples analyzed for inorganic contaminants, lead and zinc concentrationsare elevated in off-site bedrock wells (12). The off-site concentrations (0.445 ppm maximum forlead and 16.8 ppm maximum for zinc) are significantly higher than the on-site levels (0.127 ppmmaximum for lead and 1.07 ppm zinc in overburden, concentrations in bedrock are even lower). Chromium and nickel concentrations are elevated on-site but not off-site. Based on the data, asource other than the NEC site is suspected for the elevated off-site lead and zinc. Thiscontamination is within the VOC plume of contamination. Residents in this area have beenprovided with alternate water based on the VOC contamination.
During NEC's operations, air quality was affected by regular burning of waste materials,including contaminated sawdust. The burning took place near the fence line between NEC andthe Holiday Mobile Home Park. Although ambient air was affected, its potential effect on thepublic health cannot be evaluated because of the lack of data collection during plant operations.
Residents of the Holiday Mobile Home Park, which is south of NEC, reported that regularburning at NEC did degrade air quality. Several residents reported breathing problems duringburning episodes (1). NEC was registered as a source of air contaminants by both KentuckyDepartment of Health and by Kentucky Department of Environmental Protection's Division ofAir Pollution (1). According to wind sock information, wind directions were primarily southerly(toward the mobile home park) during March and April 1991 (10). Meteorological data obtained from an on-site station indicated predominant wind directions to be southwest to northeast, away from the Mobile Home Park. This data were collected daily over a ten month period beginning May 1993 (12).
During Air Stripping
Summa canister air sampling results are being collected during the air stripping process (12). ATSDR had results collected once a month or more from August 1993 through June 1994 (a sixto eight hour collection time for each sampling)(13). The results indicate that vinyl chloride,1,2-DCE, and TCE were not detectable at the fenceline between the mobile home park or on parkproperty. Detection limits ranged from 2.9 to 15.3 µg/m3. Predominant wind directions are tothe northeast away from the mobile home park. There is no off-site monitoring station to thenortheast. There are residences to the north approximately 2000 feet from the NEC site.
Fish tissue sampling was conducted by EPA on rock bass, golden redhorse, and channel catfishin August and November 1993. Aroclor 1260 was detected in most fish composites (Table 13). The concentrations did not exceed the USFDA action level of 2 ppm for PCBs (12). However,ATSDR recommends using the USEPA "Guidance Manual for Assessing Human Health Risksfrom Chemically Contaminated Fish and Shellfish" to evaluate the fish tissue data. Ourevaluation is provided in the Public Health Implications section, Toxicologic Evaluationsubsection, under PCBs. Mercury was detected at low levels (0.038 and 0.025 ppb) in fishapproximately 4.5 miles upstream of the NEC site near Harlan, Kentucky (location CR-REF). However, the data were deemed as potentially unreliable because spike recovery requirementswere not meet (12).
Quality assurance and quality control samples were run by the contractors to Cooper Industries. Data sources are considered reliable because of the acceptable quality of field and laboratorydata. The sample types and analyses were appropriate for known and expected contaminants atNEC. Samples with corresponding blanks that were contaminated are identified in thecontamination tables due to possible laboratory contamination rather than contamination fromthe site. Generally, if a contaminant is listed in the tables, its presence was confirmed in at least one environmental medium.
A fence completely surrounds National Electric Services. The riverbank bordering NES also has been partitioned off. It would be extremely difficult for trespassers to enter the site, therefore, physical and other on-site hazards will not be further addressed.
To determine whether nearby populations have been, are, or will be exposed to contaminantsmigrating from the NEC site, ATSDR evaluates the environmental and human componentsleading to human exposure. That pathways analysis consists of five elements: a source ofcontamination; an environmental medium (media) in which contaminants may be present or fromwhich contaminants may migrate (e.g., air, water, soil); a point of exposure (i.e., a place orsituation where humans might be exposed to the contaminated media); a route of humanexposure (i.e., how can the contaminant enter the body?); and a receptor population (i.e., howmany people, if any, are at the point of exposure?).
ATSDR classifies pathways as completed or potential. For a completed pathway to exist, thefive elements must exist, and there must be evidence that people have been, are, or could beexposed to a contaminant. A potential pathway exists when at least one of the five elements ismissing, but could exist (e.g., people may have been exposed in the past, may now be exposed, ormay be exposed in the future). A pathway is eliminated when at least one of the five elements ismissing and will never exist (e.g., there is no evidence that people ever have been or will beexposed). The completed and potential exposure pathways and estimates of the number ofindividuals exposed to the NEC site are shown in Tables 12-14 (Appendix B).
Completed exposure pathways are discussed in the following paragraphs and summarized inTables 12 and 14 (Appendix B). The presence of a completed pathway indicates that people have been exposed to contaminants in the past, are now exposed, or will be exposed. This discussion addresses on-site workers and off-site residents.
Workers were exposed to groundwater contaminants when they drank water drawn from anon-site well. The maximum number of workers at NEC during any given year of its operationwas 120-150 people. The plant began operating in 1951; groundwater contamination was notdiscovered until 48 years later, in 1989. The exact year of initial contamination is unknown. Because many workers worked more than 20 years at the plant, it is possible that they wereexposed to groundwater contaminants throughout their employment.
The potential window of exposure for residents of the Holiday Mobile Home Park(approximately 60 homes) is believed to have been about 10 years. A bedrock well providedwater to the residents of the mobile home park, which was established in 1978. Groundwatercontaminants did not have far (less than 100 to about 1300 feet) to migrate to the mobile homepark.
Approximately 15 wells (two on-site wells and thirteen off-site wells, excluding the HolidayMobile Home Park) from residences and businesses in Dayhoit were found to be contaminatedwith VOCs. Those residents consumed contaminated groundwater and also used it for bathingand cooking. Several businesses such as the Kentucky Department of Transportation also hadcontaminated groundwater. The duration of exposure would depend on how long a resident oremployee used the contaminated water and the length of time that water was contaminated. Residences and businesses farther from NEC were probably contaminated at a later date andduration of exposure is anticipated to be less. For people drinking water near the source (NEC) of contamination, it is possible that duration of exposure was several decades.
Workers at NEC and some residents of Dayhoit were exposed to chlorinated solvents (TCE,DCE, tetrachloroethane, and vinyl chloride) in drinking water. ATSDR estimates thatapproximately 200 people were exposed to contaminated drinking water.
The groundwater pathway is considered to be a past completed pathway for some of the exposedpopulations because of its connection to municipal water. It is a potential current exposurepathway for a few people who still may be using contaminated wells because they have electednot to connect to municipal water. It is a future potential pathway for people who elect not to connect to municipal water and who have wells downgradient of contaminated wells.
Surface Water Pathway
Two drainage lines, outfalls 1 and 2, led from the NEC plant to the Cumberland River. Residuesfrom transformers and the degreasing pit were discharged to the river. Residents have reportedswimming in the river in the past and, to a lesser extent, in the present. Residents of HolidayMobile Home Park reported fishing near NEC, collecting metal scraps near the outfalls, andconducting baptisms in the river. Past exposure via the surface water pathway is considered to be complete because all five elements of pathway analysis exist. The routes of exposure wereincidental ingestion of water and skin contact with water. The receptor populations werefishermen wading into the river, and swimmers if they contacted surface water immediatelydowngradient of the NEC site. Because baptism is typically a ritual in which the individualparticipates only once, exposures to those individuals is not considered to be significant; thebaptizer, however, could have experienced numerous exposures similar to swimmers. Theexposure status is considered to be past, but not present or future, because contamination sources have been removed and contaminants would tend to be removed from water by volatilizing to the air or being retained by sediment.
The soil pathway is considered completed because people could have easily accessed the sitebefore the fence was installed in 1989, surface soil was contaminated by (PCBs) at the boundarybetween NEC and Holiday Mobile Home Park, and PCBs are present at low concentrations insoil of the Holiday Mobile Home Park. Ingestion or inhalation of contaminated dusts blowingfrom NEC could have added to exposures via that route. Children and adults trespassing on thesite and people living in the Holiday Mobile Home Park are considered the receptor populations. Exposure routes include ingestion, inhalation, and dermal contact. Ingestion of soil by workers is considered unlikely but possible. Soil is unlikely to have been disturbed during normal operations at NEC.
Ambient Air Pathway
Ambient air is considered a completed exposure pathway for former workers, particularly thosewho worked near the vat of TCE and those who cut asbestos from coils. Personal protectiveequipment was generally not available to the workers during plant operations (14). The workers'route of exposure was inhalation of fumes from solvents inside the plant. Outside of the plant,workers inhaled fumes from the burning of waste material and, possibly, vapors from theburn-out oven and lead pot.
Waste Material Pathway
Former employees reported that they washed their skin (face, arms, hands) in degreaser, and thatfumes from the degreaser made them weak and lethargic (14). The degreaser vat was heatedduring the processing of motors for cleaning. TCE, which readily vaporizes, is believed to be aprimary constituent of the degreaser that was used. It appears that workers were exposed bydirect skin contact and inhalation. Workers reported using dust masks periodically as protectionfrom asbestos clouds, which formed when asbestos was cut. Workers also commented on thelack of exhaust fans for ventilation. The Health Outcome Data Evaluation section also discusses workers' comments about work practices at NEC.
Food Chain Pathway - Fish
Many of the chlorinated solvents (TCE, DCE, tetrachloroethylene, vinyl chloride) identified as of health concern at the NEC site have low potential for bioaccumulation because of their high volatility and lack of persistence in the environment. PCBs, however, have highbioaccumulation potential in fish and can move up the food chain to people. Residents reportedfishing near NEC on a regular basis, and using their catch as a source of food. Although fishing has become less popular with the residents, it still takes place. Edible portions of fish tissue were sampled in 1993. Aroclor 1260 was detected in most fish composites at concentrations below 2 ppm for PCBs (12). Exposure via fish in the food chain is considered a completed pathway (past, present, and future). Although the NEC site contributed PCB contamination in vicinity of the site, there are other sources of PCBs upstream and downstream which also contribute to PCBs in fish. The power substation immediately north of NEC is a potential source in the vicinity of the site.
Potential exposure pathways are discussed here and summarized in Tables 13 and 14 (Appendix B). Exposure pathways are potential if people could have been, could now be, or could in the future be exposed to a contaminant. Some previously discussed completed exposure routes are discussed in this section as potential pathways because of differences in activities of workers and residents.
Soil Gas Pathway
Although soil gas was detected off site near Holiday Mobile Home Park, soil gas is considered a potential rather than completed exposure pathway because it is unlikely that people were exposed to contaminants in soil gas. If soil gas is vented to ambient air, the concentrations are diminished. Children could have been exposed to soil gas if they dug pits or ditches to play in. Similarly, remedial workers who did not wear protective equipment could have been exposed.
Using information known about contamination in sediment of the Cumberland River, thesediment pathway is considered potential. Although many of the VOCs would vaporize to theatmosphere, heavy metals and PCBs tend to accumulate in sediment. PCB accumulation hasbeen documented in sediment of the Cumberland River upstream and downstream of the NECsite. Although NEC is a source of the contamination, the power substation immediately north ofNEC also is a potential source. It is considered unlikely that swimmers and fishermen using the river and children in the Holiday Mobile Home Park who played on the riverbank ingestedsediments. Access to the riverbank was somewhat restricted because of its steep topography. On-site riverbank samples are discussed in the Soil Pathway section.
Food Chain Pathway - Vegetables
A few residents reported that there was a garden adjacent to the NEC site in the past and that they now have gardens at their residences. Uptake of some soil contaminants (arsenic and other heavy metals) by plants is possible. However, there are insufficient data on off-site surface soils with which to evaluate plant uptake of metals and whether they are ingested by people.
Ambient Air Pathway
The ambient air pathway is considered a past, present, and future potential pathway for residents. This is based on past operations and the current air stripping process.
Residents of Holiday Mobile Home Park may have been exposed in the past to airbornecontaminants from NEC. There is no documentation of past contamination and only one citizencomplaint on record. Two Holiday Mobile Home Park residents were treated in the past at localemergency rooms for symptoms of acute solvent intoxication, presumably from fumes comingfrom NEC. Exposures may have been more frequent and experienced by a greater number ofpeople when the property adjacent to NEC was used as a drive-in theater (before the mobilehome park was built). Possible routes of exposure were inhalation of fumes from solventsexiting the exhaust fans, fumes from the burning of waste material near the mobile home parkboundary, and, possibly, vapors from the burn-out oven and lead pot.
Air monitoring results collected during air stripping indicate no detection of VOCs in the mobile home park. However, ambient air north of the NEC site may contain VOCs. There is nomonitoring data to evaluate potential exposures of residents north of the site.
In this section, ATSDR discusses health effects that could be experienced by people exposed to site contaminants; evaluates state and local health databases; and addresses specific community health concerns.
Cancer and Noncancer Health Effects
The mere presence of a contaminant does not imply that harm will result from exposure. Todetermine whether people can get sick from exposure, ATSDR begins by estimating dailyexposure doses for each contaminant of concern by each route of exposure. We use informationabout levels of contaminants and about people's activities to estimate the exposure dose. Theestimated exposure dose is calculated by multiplying the contaminant's concentration by theintake rate of the contaminated medium, adjusted by any exposure factors (for example, forworkplace exposures we multiply by 5/7 for 5 working days a week, and by 50/52 for 50 workingweeks per year), and divided by the body weight. In this way, we can get an exposure dosemeasured in mg contaminant per kg of the person's body per day. The methodology is explainedfurther in the Public Health Assessment and Guidance Manual (7).
The estimated exposure dose is then compared to a Minimal Risk Level (MRL), which is anestimate of daily exposure to a contaminant below which non-cancer disease is unlikely to occur. To develop the MRL, ATSDR relies on information from scientific studies of the effects of exposure to contaminants on people and animals. If an exposure dose exceeds an MRL, or if no MRL has been developed, the estimated exposure dose is then compared to other health-basedguidelines such as the EPA's reference dose, or to doses which resulted in adverse health effects in people or experimental animals as described in the scientific literature. These comparisons take into account the uncertainties inherent in relying on harmful effects produced in animals to predict the possibility of effects in people, as well as differences among people.
Cancer is a large group of diseases characterized by uncontained growth and spread of abnormal cells. Cancer cells multiply uncontrollably, destroying normal cells, and can spread to other parts of the body (17). A chemical capable of causing damage leading to cancer is called a carcinogen. The latency period, or amount of time between exposure and disease, may range from years to decades (18). Of the chemicals considered carcinogenic, several classes have been defined:
- Human Carcinogens -- chemicals that have been shown to cause cancer in people.
- Probable Human Carcinogens -- chemicals that have been shown in experiments to cause cancerin animals. There is inadequate evidence of their human carcinogenicity, but, over a lifetime of exposure, they are suspected of causing cancer in people.
- Possible Human Carcinogens -- chemicals that have limited experimental data indicating thatthey can cause cancer in animals. There is inadequate evidence of their human carcinogenicity,but, over a lifetime of exposure, they could cause cancer in people.
Some of the contaminants of concern at the NEC site are considered carcinogenic.
In the United States population as a whole, cancer occurs in one in three people's lifetime. It isvery difficult for scientists to determine who will get cancer, but we do know that exposure tosome contaminants can increase the chances (or risk) of getting cancer. Even if a person getscancer, scientists and physicians typically cannot know the cause of the person's cancer. Todetermine whether exposure at this site may cause cancer, a numerical increase in the risk ofcancer is estimated using the estimated exposure dose and a cancer slope factor developed by theEPA specifically for each cancer-causing chemical.
Following are discussions of health effects by contaminant with more information about thecarcinogenicity of specific chemicals.
Health Effects By Contaminant
Workers at NEC and residents and business people of Dayhoit who used contaminated wellwater may have been exposed to vinyl chloride via ingestion, inhalation, and skin contact. Themaximum level detected at the NEC site was 350 parts per billion (ppb) in a private well.
Vinyl chloride is considered a human carcinogen. Evidence of the carcinogenic potential of vinyl chloride in people is from reports of a greater than expected incidence of angiosarcoma (a kind of cancer) of the liver in workers occupationally exposed to the chemical (19).
A number of the private wells tested had elevated levels of vinyl chloride. ATSDR staff haveestimated a small increased risk of cancer for persons exposed over a lifetime to the levels ofvinyl chloride found in the contaminated wells near this site. In addition, exposure to vinylchloride at the levels found at this site is associated with adverse health effects other than cancer. ATSDR has determined a chronic oral MRL for the site using a study that reports adverse livereffects in rats exposed to vinyl chloride (19). People who drink the water from the contaminatedwells could be exposed to a dose above the MRL. Although ATSDR has found no studies ofliver effects in people after oral exposure to vinyl chloride, animal data indicate that drinking water contaminated with the levels of vinyl chloride found at the NEC site could harm the liver. Individuals with liver, renal, cardiac, or pulmonary impairments may be particularly sensitive to the effects of vinyl chloride (20).
ATSDR did not find additional studies of human or animal oral exposure to vinyl chloride atlevels similar to those found at this site. All of the reported inhalation studies were on people or animals exposed to gaseous vinyl chloride at much higher levels than those likely to be found in residences at this site.
Residents of Dayhoit who used contaminated well water may have been exposed to1,1,2,2-tetrachloroethane via ingestion, inhalation, and skin contact. The maximum leveldetected at this site was 120 ppb in a private well.
1,1,2,2-tetrachloroethane is considered a possible human carcinogen (21). Evidence of the carcinogenic potential of this chemical is from animal studies, which report a significant increase in mouse liver cancers following exposure. A human carcinogenicity study on a group of army workers exposed to tetrachloroethane fumes in a clothing processing plant reached no conclusions. The increased death rate due to various cancers was slight, and confounding factors caused the authors to conclude that the cancers may not have been due to 1,1,2,2-tetrachloroethane (21). The humans in the study, who had inhaled 1,1,2,2-tetrachloroethane, did not have an increased liver cancer rate.
Despite the absence of any indication that this chemical causes cancer in people, data fromanimal studies allow us to estimate cancer risks associated with exposure. ATSDR staff haveestimated that there is a low to moderate increase in the risk of developing cancer for personswith a lifetime of exposure to this chemical at levels such as those found in the contaminatedwells.
ATSDR staff found very few studies of noncancerous health effects following oral exposure to1,1,2,2-tetrachloroethane. All of the inhalation studies investigated occupational exposures. Studies of oral and inhalation exposures in people and animals discussed effects associated withexposure to levels much higher than those at this site. Nevertheless, people with a tendency todevelop liver disease, especially habitual drinkers of alcoholic beverages, and obese individualsshould avoid exposure to 1,1,2,2-tetrachloroethane (20).
Workers at NEC and residents of Dayhoit who used contaminated well water may have beenexposed to TCE via ingestion, inhalation, and skin contact. The maximum level detected inprivate well water was 20 ppb; the maximum level detected in on-site soil was 15,000 ppm.
Workers on site may have been exposed to TCE via inhalation and dermal contact. Althoughthere is no information about the presence of TCE in ambient on-site air, reports of previouswork practices at NEC indicate that the air would have had measurable levels of TCE. Inaddition, TCE was known to have been used in the equipment cleaning vat.
There is inconclusive evidence of the carcinogenicity of TCE. An increased incidence of tumorshas been observed in some animals experimentally exposed to TCE by ingestion or inhalation. Some laboratory studies indicate that some mice exposed to TCE by ingestion developed livercancer. Additional studies in mice suggest that inhalation exposure may result in liver and lungcancer. However, some of the TCE studies use questionable methods and have inconclusiveresults, making it difficult to conclude that TCE is an animal carcinogen. Furthermore, availableinformation is not sufficient to determine whether TCE causes cancer in people (22). Untiladequate experimental evidence is available, either from human or animal studies, it cannot beconcluded that TCE exposure is likely to cause cancer.
The maximum TCE concentrations detected in well water and in on-site soil are both estimatedto result in doses below the ATSDR MRL for intermediate exposure via ingestion. Therefore,workers exposed to soil and residents exposed to drinking water would not be expected to havenoncancerous adverse health effects. ATSDR has not developed MRLs for chronic TCEingestion.
Occupational studies indicate that workers exposed to high levels of TCE may suffer fromvarious adverse health effects; those effects would not be expected in people who receivedstrictly residential (not occupational) exposures. The following information may pertain toworkers who received occupational exposures at NEC. However, without air measurements,ATSDR cannot determine the specific health effects that may have been caused by workerexposure at this site.
One study in the literature indicates that inhalation of TCE vapor at a concentration as low as 10ppm can lead to headaches, dizziness, and sleepiness (20). At 30 ppm, TCE vapor can irritate thenose and throat. At levels between 100 and 600 ppm, workers have experienced a feeling ofdrunkenness and nausea in addition to headaches, dizziness, and sleepiness. At levels above1000 ppm, workers have had tremors, coordination problems, visual problems, and have lostconsciousness (23). Workers exposed to high, but unquantified, levels of TCE have shown additional neurologic effects: mental confusion, facial numbness, and weakness (20).
Workers also have experienced gastrointestinal effects such as nausea, vomiting, anorexia, and intolerance of fatty foods after inhalation of unmeasured levels (22). Several reports indicate that a few people have died of cardiac arrhythmias following acute, excessive occupational exposure to TCE. Dermal exposure to concentrated liquid and vapor is associated with skin irritation, burns, rashes, and dry throat (22). Repeated immersion of the hands has caused paralysis of thefingers (24). Worker exposure levels at the NEC site are not documented.
People who used contaminated well water may have been exposed to cis 1,2-dichloroethene andtotal (a mixture of cis and trans) 1,2-dichloroethene via ingestion, inhalation, and skin contact. The maximum level of cis 1,2-DCE detected in a private well at this site was 133 ppb; the maximum level of total 1,2-DCE was 39 ppb in a private well.
This chemical cannot be classified with regard to carcinogenicity. ATSDR was unable to findstudies of human or animal exposure to dichloroethene and associated cancer (25).
It has been estimated that the maximum DCE concentration detected in well water near or at the NEC site would result in a dose below the ATSDR MRL for intermediate exposure via ingestion. Therefore, noncancerous adverse health effects would not be expected. ATSDR has not developed MRLs for chronic DCE ingestion.
Polychlorinated Biphenyls (PCBs)
PCBS are a family of synthetic chemicals that contain 209 individual compounds with varyingharmful effects. Because the health effects of PCBs are difficult to evaluate, most studies are conducted using commercially available PCBs or mixtures.
Using information from rat and mouse studies, PCBs have been classified as probable humancarcinogens (26). The animals in the studies were fed different types of PCBs; because there is insufficient information about which constituents of PCB mixtures are carcinogenic, it is assumed that PCB mixtures of any composition are potentially carcinogenic.
The MRL for PCBs was developed using information from a monkey study, which indicated thatingestion of PCBs can compromise immune system response (26). Studies of other animals show similar results. ATSDR was unable to find studies reporting immunologic effects in people.
Exposure to PCBs may have occurred and or be occurring through contact withPCB-contaminated soil in the mobile home park and through consumption of fish or ingestion ofsediment from the Cumberland River. There may have been past exposures to PCBs throughcontact with PCB-contaminated soil or wastes on-site.
Studies of worker exposure to PCBs have investigated adverse health effects caused by acombination of inhalation and dermal exposure. Occupational studies do not distinguish theindividual contribution of each route. The studies indicate that exposure to PCBs can affect the respiratory and gastrointestinal systems. Occupational exposure has been shown to adversely affect the liver. Several dermal and ocular effects, particularly chloracne (skin eruptions and cysts), skin rashes, and eye irritation, have been associated with occupational exposure to PCBs (26).Although there is no information about PCB levels in waste material and in ambient on-site air,reports of previous work practices at NEC indicate that waste materials would have containedPCBs (1). Workers on the site may have been exposed to PCBs via inhalation and dermal contact.
The adult exposure doses for the maximum PCB levels at three of the four on-site locations were above the MRL of 0.00002 mg/kg/day for PCB. The three locations and the maximum levels are the riverbank fill area (610 ppm), south fenceline (58 ppm) and outfall 1 area (49 ppm). Those exposure doses represent past exposures since the on-site contamination has been cleaned up. Exposure of children to on-site soils appears to have occurred only rarely, if at all, so exposures of children were not evaluated. An exposure dose above the appropriate MRL indicates that evaluation of the toxicological literature needs to be done to determine whether the specific exposure situation represents a health hazard.
Noncarinogenic adverse health effects are unlikely to occur based on the toxicological literature and the specific exposure situations on-site (26). The exposure doses for adults were 5.7 - 357 times lower than the lowest observed effect level of 0.005 mg/kg/day in a study of monkeys. This observed effect was a mild immunological change.
In calculating an exposure dose for adults, it is assumed that an adult daily ingests 100micrograms of soil contaminated at a specific concentration. However, it does not appear thatanyone would have been exposed on a daily basis because no one routinely worked at theselocations.
There would be an increased risk of cancer for the maximum PCB levels on-site if individualswere exposed daily for 70 years. However, this does not appear possible because the site was not contaminated for that long and daily exposure did not occur.
Soil in Mobile Home Park
Exposure doses for adults, children, and pica children were calculated for each of the 20sampling locations within the mobile home park (Figure 7). The exposure for pica childrenexceeded the MRL of 0.00002 mg/kg/day for PCB at 7 of the 20 sampling locations and forchildren at 1 of 20 locations. The adult exposure dose did not exceed the MRL at any of the 20locations. An exposure dose above the appropriate MRL indicates that evaluation of thetoxicological literature needs to be done to determine whether the specific exposure situationrepresents a health hazard.
Noncarinogenic adverse health effects are unlikely to occur based on the toxicological literature and the specific exposure situation at the mobile home park (26). All the exposure doses for pica children were 10 - 175 times lower than the lowest observed effect level in a study of monkeys. This observed effect was a mild immunological change. The exposure dose for children was 238times lower than the lowest observed effect level.
In calculating an exposure dose for a pica child, it is assumed that a child daily ingests 5 grams of soil contaminated at a specific concentration. The PCB contamination at the mobile home park appears to be focal in nature as 10 of 20 sampling locations are below the detection limit and another 3 are below 40 ppb total PCBs. The maximum PCB concentration, 1053 ppb, is right next to the fenceline and away from the mobile homes. Thus, it is unlikely any child could ingest 5 grams of soil at this concentration on a daily basis.
While PCB is considered a probable human carcinogen, there is no apparent increased risk ofcancer from exposure to PCB-contaminated soil at the mobile home park (26). This is based on 70 years of exposure to the maximum PCB level of 1053 ppb.
The adult exposure doses for consuming one fish meal a day, week, or month of channel catfishor golden redhorse exceed the chronic MRL of 0.00002 mg/kg/day for PCBs. The adultexposure dose for one meal a day or week of rock bass also exceeded the MRL. Currently, theseexposures would not be primarily site-related since PCB levels in fish upstream of the site aresimilar to those downstream of the site (Table 13).
For those consuming one fish meal a week or month, noncarinogenic adverse health effects areunlikely to occur based on a review of the toxicological literature (26). However, there is a chance of mild immunological changes for individuals eating one fish meal a day.
There would be an increased risk of cancer for anyone eating for 70 years one meal a month or more of rock bass, golden redhorse, or channel catfish caught in the Cumberland River near the site.
Children who played along the riverbank near the NEC site may have been incidentally exposedto PCBs via ingestion of river sediments. The maximum PCB level found in river sediments was0.21 ppm. Exposure doses for children exposed to sediment were calculated for a combinationof dermal contact and ingestion. The exposure doses do not exceed the MRL for PCBs, thereforehealth effects based on this scenario are considered unlikely.
Although there is no information on PCB levels in surface water of the Cumberland River nearthe NEC site, levels in sediments and prior work practices at NEC suggest that surface watercould have had measurable levels of PCBs. People who swam in the Cumberland River mayhave been incidentally exposed to PCBs via ingestion. Bathers also may have been exposed toPCBs by dermal contact.
Workers and nearby residents may have been exposed to lead fumes from the lead pot. ATSDRhas found no information on actual levels of lead in air; however, previous work practicesindicate that measurable amounts of lead may have been in the air in and around the NECbuilding. An undetermined source of lead (not believed to be related to the NEC site) wasdetected in groundwater in several private wells in Dayhoit. It is unknown whether exposures to lead in drinking water have occurred or are occurring since the extent of contamination and time of introduction of the contamination into the aquifer is also unknown.
Most lead studies discuss adverse health effects in terms of blood lead level rather than externalexposure level. In addition, occupational studies consider exposures to be multi-route(simultaneous ingestion, inhalation, and dermal exposures). Therefore, the effects of lead onhuman health will be discussed here in terms of exposure by multiple routes rather than byinhalation only.
Using recent scientific information and studies, the Centers for Disease Control recentlyre-evaluated its guidelines on acceptable blood lead levels. New data indicate that significantadverse health effects can occur in children with blood lead levels previously believed to be safe. Some health effects have been documented at blood lead levels as low as 10 micrograms lead perdeciliter blood (µg/dL) (27).
Blood lead concentrations of 10 µg/dL are associated with neurobehavioral deficits, hearingimpairments, and inhibition of hemoglobin synthesis in children (27). Blood lead levels of 10-20µg/dL in children have been shown to result in a 4- to 5-point decrease in Intelligence Quotient(I.Q.), and in electrophysiologic changes in brain activity. Blood lead concentrations greater than 33 µg/dL in children produce neurotoxic effects as well as a depression in plasma levels ofVitamin D. Neurotoxic effects of lead in children are of great concern because they may beirreversible, even after blood lead levels return to a normal range (28).
Occupational and general population studies indicate that exposure to low levels of lead inmiddle-aged men is associated with hypertension (28). Additionally, studies indicate small but significant direct associations between blood lead levels and blood pressure. Exposure to high levels of lead may cause adverse reproductive effects, particularly miscarriages and still births, in women and decreased sperm count and sperm motility in men (28).
An undetermined source of zinc (not believed to be related to the NEC site) was detected ingroundwater in several private wells in Dayhoit. It is unknown whether exposures to zinc indrinking water have occurred or are occurring since the extent of contamination and time ofintroduction of the contamination into the aquifer is also unknown.
Zinc is a naturally occurring element and can be found in our air, water and food. Eating a small amount of zinc is important in maintaining good health; too much zinc, however, can causeillness (29). Studies have been conducted in people who were exposed to zinc for several week at levels similar to those to which people drinking contaminated water are estimated to be exposed. Some people had changes in their blood (decreased serum HDL), some experienced abdominal cramping and vomiting, and some had damage to their immune systems (30).
Other Heavy Metals
ATSDR staff estimated the ingestion doses of arsenic, manganese, beryllium, and chromium that workers might receive by incidental ingestion of contaminated soil on the site. None of the metals were estimated to have been ingested at a level of public health concern. ATSDR staff also estimated the ingestion doses of arsenic, manganese, beryllium, and chromium that children might receive through incidental ingestion of contaminated soil and sediment at the riverbank. Pica children, or children who eat an unusually large amount of soil (up to 5 grams, or about a teaspoon a day), and who play in contaminated soils and sediments several times a week, may be exposed to arsenic, manganese, beryllium, and chromium at levels of public health concern. However, unless children eat unusual amounts of soil and play often in contaminated soil, none of the metals were estimated to be ingested at levels of concern. Ingestion of heavy metals in edible plants is not considered a significant route of exposure.
Although ATSDR has no data that report actual levels of asbestos, workers have reported visibleclouds of asbestos in the air in the NEC building (more information about workers and asbestosis discussed in the Health Outcome Data Evaluation section). Asbestos fibers were probablydispersed into the breathing zone, resulting in inhalation exposures by NEC workers. Forasbestos fibers at least 5 micrometers in length, the National Institute for Occupational Safetyand Health has set a recommended exposure limit of 0.1 fibers per cubic meter to protect workerhealth (31). Numerous studies of workers indicate that the respiratory system is adversely affected following inhalation of asbestos. Such exposure is associated with asbestosis, a disease characterized by fibrosis of the lung, shortness of breath, and frequently by rales and cough (32).
Additionally, asbestos is a known human carcinogen; inhalation of asbestos is associated with lung and gastrointestinal cancers (32). Smokers exposed to asbestos dusts from inhalation are at a higher risk of developing lung cancer than are nonsmokers with similar exposures (20). ATSDR has insufficient information to evaluate adverse health effects associated with workers' asbestos exposure at this site.
ATSDR staff have reviewed the mortality statistics in WONDER, the CDC on-line cancer(neoplasms, all organ sites) mortality rate database, over a five-year period for the United States, the state of Kentucky, Harlan County, and adjoining Bell County. Age-adjusted, gender-specific rates were compared. Age-adjusted means that the rate is adjusted to take into account differences in age distribution between areas. For example, a county with relatively more young people (people less likely to die) than another would have a lower crude death rate even if more people were dying at a younger age; the large number of young people would keep the rate low. An age-adjusted rate would prevent the larger number of young people from masking an unusually high death rate.
Because county-level data contain information about the whole county, elevated death rates for acommunity as small as that near the NEC site might not affect county rates as a whole. Countydata may provide information about trends in a community, but only a community-specificanalysis can truly answer whether cancer rates near the site are elevated. Becausecommunity-specific information is not available at this time, county-level cancer data will beexamined here as a background data source only. As a result, if an elevated death rate is seen atthe county level, it cannot be related to the site. Likewise, if no elevated death rate is seen, thatdoes not mean that the site has had no adverse effect on the death rate in the community.
The cancer data may be summarized as follows: Harlan County men had the highest cancerdeath rate of the four groups of men (U.S., Kentucky, and the two nearby counties); Bell Countywomen had the highest cancer death rate for women; Harlan County had the next highest rate. Throughout the United States, men consistently have higher cancer death rates than women. Table 15 lists the rates.
* Neoplasms (tumors) for all organ sites.
** Rates are per 100,000; data are for years 1983-1988.
ATSDR staff also examined crude death rates associated with liver cancer for the four areasbecause vinyl chloride was in a completed pathway and is associated with liver angiosarcoma,one type of liver cancer. Because liver cancer is relatively rare, death rates associated with livercancer would be expected to be relatively low. In the case of relatively rare cancers andassociated low death rates, even one extra death from that type of cancer can significantly elevatea death rate. In the case of more common cancers and associated higher death rates, one moredeath makes no difference in the rate. Therefore, an elevated community rate of liver cancercould affect the county rate. Harlan County men and women had death rates from liver cancerthat were lower than Bell County, and both genders in both counties had rates much lower thanKentucky and the United States. Liver cancer, which is associated with vinyl chloride exposure,apparently is not common in Harlan County. On the other hand, because of the small numbersevaluated (one death in 415,708 people-years [number of people times the number of yearsconsidered] for Harlan County and one death in 342,242 people-years for Bell County), thoserates are considered unstable. There have been only two deaths associated with liver cancer inHarlan County over a 10-year period, and it is difficult to draw any conclusion about differences between the counties. Table 16 shows the data on liver cancer deaths.
* Rates are per 100,000; data are for years 1979-1988
Vital Statistics Report
ATSDR staff reviewed data contained in the State of Kentucky's 1990 Vital Statistics Report and noticed nothing remarkable. Health statistic trends such as cancer incidence rates are extremely unstable when examined for a one-year period. Because readily available information contained at the Vital Statistics Bureau describes cancer mortality rates at the county level, analyzing several years of vital statistics information would yield similar results to the WONDER data discussed above. The Vital Statistics Bureau is able to provide vital statistics at the zip code level. This information may be useful in conducting a disease and symptom prevalence study.
Kentucky State Cancer Registry
ATSDR staff received the 1991 and 1992 Kentucky Cancer Registry reports. However, theinformation contained in these reports covers cancer incidence for only two years, and thepopulation was analyzed down to the county level. Since health statistic trends such as cancerincidence rates are extremely unstable when examined for a two-year period, and since anytrends occurring in the community around the site would be masked in an analysis of the entirecounty, no meaningful analysis can be conducted at this time.
Medical Records Review
On May 14, 1992, an ATSDR staff member reviewed the medical evaluations performed by theUniversity of Cincinnati Medical Center (UCMC) on Dayhoit, Kentucky, residents and formerNEC workers. ATSDR conducted the review to help determine occupational and environmentalexposures to site contaminants and to identify health impact. This section summarizes themedical records findings, including the opinions of the examining physician at UCMC (33).
Thirty-five people were seen in the UCMC Center for Occupational Health, and the medicalrecords of one deceased resident were reviewed. Twenty-three men and 13 women wereevaluated. The group of 36 included 24 Dayhoit residents and 14 former NEC employees (2 ofthe employees also were residents of Dayhoit). Eight of the residents have lived or now live inthe Holiday Mobile Home Park. Eighteen people, including former NEC workers and Dayhoitresidents (some living in the mobile home park), were tested for lead and PCBs. None of thelaboratory evaluations revealed elevations of either lead or PCBs. The levels found wereconsidered normal (<10 µg/dl) for an unexposed rural population. One person was tested only for lead and one person only for PCBs. Those tests also were normal.
The former employees of NEC worked with TCE and other chlorinated solvents, PCBs, asbestos,paints, varnishes, epoxy resins, silica, and lead. Personal protective equipment was generally not available to the workers. Employees reported that NEC management did not provide materialsafety data sheets (information now required by the Occupational Safety and HealthAdministration) for any of the substances used at the Dayhoit facility. Medical records indicated the following working conditions (33): Most employees used TCE from open containers at theirwork stations; many regularly washed their hands and faces with TCE. TCE also was kept in anopen tank (approximately 6 feet by 8 feet at the surface and 9 feet deep). Employees could notshower at the facility after work; therefore, they took their contaminated uniforms home. Workers ate and smoked at work stations. Some workers warmed food in the ovens that bakedthe solvent-coated machinery. Asbestos was routinely cut with a band saw in an open area of the plant, resulting in clouds of dense dust.
According to the examining physician at UCMC, former workers were exposed to high levels ofcontaminants, particularly TCE and PCBs, because of the lack of good industrial hygiene andsafety practices. The physician drew that conclusion after taking exposure histories, examiningthe workers, and reviewing their medical records. Most of the former workers reported frequentsymptoms of solvent intoxication, including lightheadedness, dizziness, headache, nausea,fatigue, and a staggering gait. Two workers were hospitalized for acute solvent intoxication. Another worker reported having solvent intoxication at the time of an automobile accident. Twoformer workers have toxic encephalopathy (a disease of the brain) as a result of their pastexposures at NEC.
Four other former employees have an illness similar to toxic encephalopathy, but diagnosis oftheir conditions cannot be made with medical certainty; the four have other plausibleexplanations for their disturbed nervous system function or neurologic deficits. For example, oneman had a history of severe head trauma in addition to the chronic solvent intoxication; therefore,it is difficult to determine what caused his current neurophysiologic problems. Other formerNEC workers reported symptoms of possible peripheral neuropathy (a disease of the nervesoutside the brain and spinal cord).
Numerous former NEC employees reported chronic skin ailments, particularly chronic dryness,itching, and rashes. Some of the conditions were noted when those people were examined in theUCMC clinic. Because such skin problems are also prevalent in unexposed populations, it isdifficult to determine if the problems were caused by exposure to TCE and other solvents used at NEC.
Some former NEC employees were evaluated for disease associated with asbestos exposure;pulmonary function tests (PFTs) and chest radiographs were performed. Chronic obstructiveairway disease changes were noted on some of the PFTs and x-rays, but there was no evidence ofasbestos-related disease. However, asbestosis and asbestos-related cancer may develop decadesafter exposure.
One former worker gave a history of severe solvent intoxication at work and metal fume fever. He now has tremors, paresthesias (abnormal sensations) in his extremities, autoimmune type nephritis, and vasculitis (autoimmune type kidney and blood vessel disorder). It is difficult to determine if the toxic exposures caused the vasculitis because that condition can occur spontaneously. The vasculitis also could cause several of the other medical problems that this man has, including neuropathy and nephropathy (nerve and kidney disorders). The history of one former worker suggests reactive airway disease. Further testing would be required to make that diagnosis.
Several of the Dayhoit residents provided to UCMC an environmental history that described agasoline odor or greasy feeling associated with the well water in the area. They also reportedfrequent solvent odors emanating from NEC, and noted soot and fumes caused by frequentburning of waste materials.
Two Holiday Mobile Home Park residents were treated in local emergency rooms for symptomsof acute solvent intoxication. Chronic skin problems associated with the use of well water alsowere described by patients. Some abnormal skin conditions were noted when several Dayhoitresidents were evaluated in the Center for Occupational Health. Because many of thedermatologic problems experienced by the Dayhoit residents also are seen in unexposedpopulations, it is difficult to determine their cause. Many residents described water-relatedproblems from showering, such as dizziness, headaches, and coughing; the physician attributedthose symptoms to the volatilization of solvents from shower water.
Some residents reported asthma, which seemed to improve when their well water was replacedwith water supplied by the City of Harlan. Another resident seemed to have a greater severityand frequency of migraine headaches after exposure to contaminated well water. The reviewingphysician indicated that those conditions probably were not caused by the contaminated water,but could have been exacerbated by the exposure.
1. Is cancer a problem in our community?
Cancer is a group of diseases characterized by rapid, uncontrolled growth of tissue that invadesadjacent structures and can spread to distant areas of the body. Many cancerous tumors cancause death if not adequately treated. Although many malignant diseases are grouped together ascancer, they are actually different conditions with distinct causes and outcomes. For example,colon cancer is associated with a low-fiber, high-fat diet, and basal cell carcinoma is associatedwith sun exposure. Colon cancer can spread throughout the body, eventually causing death;basal cell carcinoma may never spread to distant body sites or cause death. Members of thecommunity reported concern about a wide range of cancers to ATSDR staff during the publicavailability session for the NEC site; the types of cancer are listed in Appendix C.
There is not enough information to determine whether or not there is a pattern of cancer in theDayhoit community. In order to determine whether the community has an elevated cancer rate,public health investigations are needed. The Health Activities Recommendation Panel (HARP)has determined that a disease and symptom prevalence study is indicated for the NEC site(HARP recommendations are listed in the Recommendations section of this public healthassessment). Without such a study, ATSDR cannot state whether there is an elevated cancer ratein the Dayhoit community.
2. We are suffering from nerves, rashes, blisters, nausea, diarrhea, and/or persistent colds. Are those problems site related?
Several citizens are concerned about the prevalence of neurologic, psychiatric, skin, and immunedisorders in the community. Although many of the problems noted by Dayhoit residents couldoccur in unexposed populations, those ailments also may be associated with solvent exposure. Some conditions (e.g., Alzheimer's disease) have no known cause. A careful clinical andlaboratory evaluation of potentially exposed persons will help to determine if neurologic,psychiatric, or skin problems are causally related to environmental toxins. Sometimes, whenclinical and laboratory evaluations cannot determine the source of a condition, an epidemiologicanalysis of a community is needed to determine if there is an excess of a particular disease. ATSDR has recommended further evaluation of health-related concerns reported by the Dayhoitcommunity. ATSDR's health follow-up activities are determined by the Health ActivitiesRecommendation Panel (HARP) and are listed in the Recommendations section of this publichealth assessment.
Many of the digestive, cardiovascular, and ear, nose, and throat disorders reported by residentsduring the public availability session (Appendix C) are seen in most American communities andare not related to environmental toxin exposure. Some of the illnesses could be related toexposure to environmental toxins, but that would be very unlikely. Diarrhea was a particularproblem mentioned by Dayhoit residents. Although it does not appear that contaminants fromthe NEC site would cause that condition, diarrhea may be caused by drinking water that has hadinadequate sewage treatment. Biologic organisms in the sewage (e.g., Giardia and otherpathogenic bacteria) can cause diarrhea. There is no regional sewer system in the Dayhoit area.
3. Can contaminants be stored in women's bodies and later be passed on to their children?
The community reported a concern about levels of PCBs and lead. Many individuals wereconcerned that those substances could cause adverse health consequences in unborn children. Both PCBs and lead can cause developmental problems in children born to mothers who havehigh PCB or lead levels during pregnancy. PCBs and lead can be transmitted to infants in breastmilk if the nursing mother has elevated levels in her body. PCBs and lead can be measured inblood samples. A group of 18 former NEC employees and Dayhoit residents, including women,were tested for PCBs and lead at the University of Cincinnati Medical School. Sample resultsdid not show elevated levels of either PCBs or lead. Despite the test results, it is not knownwhether some women of childbearing age living in Dayhoit have increased levels of PCBs orlead. Biologic monitoring of those women would be necessary to determine their levels of PCBsand lead.
4. Is the water from our wells safe to drink, now and in the future?
If the wells are within the groundwater VOC plume of contamination (Figure 7) or near it, theyshould be periodically tested until the groundwater has been remediated (Recommendation 1). Water contaminated with chlorinated solvents should not be used for drinking. It is possible thatwells within the plume are not contaminated because of fracture zones and varying well depths. If a well is within the plume of contamination and has not been tested, we recommend that it betested. If a sweet odor or sharp, unpleasant, ether-like odor or taste is coming from the water, thewell should be tested even if it is not within the plume. Chlorinated solvents have noticeableodors in water at the following concentrations: vinyl chloride, more than 3 ppm; TCE, 25-50ppm. Wells are not likely to be contaminated from NEC sources if they are outside of thecontamination plume and water drawn from them has no odor.
Additionally, water in the area is contaminated with lead and zinc from an unknown source. Until this contamination has been characterized (see Recommendation 2), ATSDR can not saythat groundwater is or will be safe to drink.
5. Will air strippers used in remediation affect our health?
ATSDR recommends monitoring of air and water effluent discharged from the air stripper(Recommendation 3). The Agency recommends continuing with air monitoring at the commonboundary of the NEC site and the Holiday Mobile Home Park and off-site in the Holiday MobileHome Park. Additionally, we recommend monitoring downwind (north/northeast) of the NECsite and modeling of air emissions. Modeling of air emissions under worst case scenarioconditions (maximum concentrations of contaminants and meteorological conditions such asinversions) is needed to determine possible health implications. Air sampling should beconducted to confirm the model.
The water effluent concentrations are within the permitted release to surface water for Aprilthrough June 1994 and do not indicate that health effects will occur. Specific questions about theair-stripping operation and monitoring should be directed to EPA.
6. Is it safe to eat produce from our gardens?
Surface soil in the Holiday Mobile Home Park has been tested for PCBs, dioxin/furans, andheavy metals. Low levels of these contaminants have been found but not at levels of healthconcern. The testing indicates that the soil does not contain hazardous contaminants that couldbe taken up by plants and consumed at toxic levels by humans.