Oak Ridge Reservation: Exposure Evaluation Work Group

PExposure Evaluation Work Group

May 23, 2005 - Meeting Minutes

Attendance

ORRHES Members attending:
Tony Malinauskas (Chair), Kowetha Davidson, David Johnson, Jeff Hill, James Lewis, Pete Malmquist, and Charles Washington

Public Members attending:
John Merkle and Lynne Roberson (phone)

ATSDR Staff attending:
Loretta Bush (phone), Jack Hanley (phone), Marilyn Horton (phone), Karl Markiewicz, and Bill Taylor

DOE Staff Attending:
Tim Joseph

ERG Contractors attending:
Michelle Arbogast (phone) and Liz Bertelsen (phone)

Purpose

Tony Malinauskas called the meeting to order at 5:30 p.m. The purpose of the meeting was for Dr. Karl Markiewicz to present the initial work group discussion on ATSDR's Public Health Assessment (PHA) on Polychlorinated Biphenyls (PCBs) at the Oak Ridge Reservation (ORR).

Dr. Malinauskas asked for comments on the meeting minutes from April 11. No comments were noted and the minutes were approved.

PHA on PCBs at the ORR

Presenter: Dr. Markiewicz, ATSDR

Dr. Markiewicz pointed out that this PHA only covers PCBs. He explained that Dr. Jo Freedman of ATSDR was the primary author of this PHA, but she has retired. He was asked to finalize the PHA and prepare the presentation, but Dr. Freedman had prepared many of the slides. Also, Michelle Arbogast of ERG was instrumental and helpful in assisting with this process.

A pathways evaluation slide presented a pictorial representation of the area and the exposure scenarios that will be considered. Key issues and concerns being evaluated include:

• Swallowing sediment or surface water during recreation
• Swallowing flood plain soil while gardening
• Eating fish, turtles, and geese
• Drinking potentially contaminated surface water

ATSDR is using two time periods to screen media contaminants: historic (pre-1996) and modern (1996 and later). Data sources to evaluate the historic time period include the Oak Ridge Dose Reconstruction (DR), the Oak Ridge Environmental Information System (OREIS), and the Toxicological Profile and Health Assessment Toolkit (TopHat). For the DR, ATSDR is a) considering the selection of exposure point concentrations, b) screening levels against health-based comparison values (CVs), and c) considering the suitability of using the data to make health calls. ATSDR accepted eliminating the pathways eliminated by the DR, and is considering the remaining pathways that need to be included in the screening process and in the public health implications section of the PHA. To screen contaminants for the modern time, the agency is using TopHat (state data) and ATSDR's 1998 Watts Bar Reservoir Exposure Investigation (EI).

The screening methods include:

• CVs
• Standard non-biological media: air, water, soil, and sediment
• Biological media: fish, geese, and turtles
• CVs derived for biological media for this PHA because ATSDR does not have standard CVs for these media

Pathways evaluated:

• PCBs in water
• DR eliminated drinking water pathway
• Chemical properties and environmental data indicate PCBs and water do not mix
• PCBs in sediment/soil
• Standard screening against soil CVs eliminated sediment/soil pathway
• PCBs in biota: turtles, geese, and multiple locations and species of fish
• Further evaluation

Two graphs were presented of PCBs in Watts Bar Reservoir fillets by species before 1996. Samples were from the Clinch River, Tennessee River, Watts Bar Reservoir, and Poplar Creek. The concentration of PCBs, number of samples, and median values were shown for catfish, large mouth bass, hybrid striped bass, and sunfish/bluegill. CVs for child and adult fish consumers were presented for categories based on the EI: high, moderate, and low fish consumers. About 79% of the people studied were in the low consumer group.

A slide on concentrations of PCBs in Lower Watts Bar Reservoir (LWBR) fish from 1996 and later included a breakdown of the levels by all samples, fillets, and whole fish. The median values for fillets are higher than all samples and whole fish. To make conservative assumptions and to be health protective, ATSDR uses the highest value (fillet) to calculate exposure doses.

A table for pre-1996 screening results showed that sediment, fish, and geese were the medium considered. The table showed the source/time/species, those eliminated, and those that were retained. The sediment pathway, fish in East Fork Poplar Creek (EFPC), and sunfish in Watts Bar Reservoir were eliminated. Fish by other locations, as well as moderate and high geese consumers, were retained. The post-1996 screening eliminated sediment, water, air, and turtles from further evaluation. Fish in all locations considered in the screening were retained.

As presented in the DR, ATSDR assumed people consumed 27 grams/day (g/day) and 22 pounds/year (lbs/year) of geese; geese consumption was screened out. In Dr. Markiewicz's opinion, these assumptions were high, especially considering there were closed seasons.

A thermometer graph using a logarithmic scale presented PCB effect levels and oral exposure doses. Much information is available on acute lowest-observed-adverse-effect-levels (LOAELs), an exposure occurring for 14 days or less. An intermediate LOAEL is based on exposure from 14 days to one year. Chronic LOAELS are for exposures over more than 365 days. LOAELS are based on animal studies. The graph also showed cancer effect levels (CELs), which are primarily based on studies on rats, and have mainly shown liver cancer resulting from PCB exposure.

The basis for the chronic oral MRL is a LOAEL based on immunological effects shown in monkey studies (0.005 milligram/kilogram/day [mg/kg/day]). The intermediate oral LOAEL is based on neurobehavioral effects seen in monkey studies (0.0075 mg/kg/day). Both LOAELs include a safety factor of 300. The doses are approaching the LOAELs within one order of magnitude, which is when the agency will begin saying that a dose is too close to the margin of safety. This would be presented in the conclusions regarding fish consumption. For Canada geese, the dose is still two orders of magnitude away—at 16 times below the LOAEL.

Some of the potential non-cancerous effects include:

• Dermal—irritation, chloracne, and pigmentation of skin and nails in workers
• Neurobehavioral—insufficient in humans
• Respiratory—inconclusive in humans
• Birth weight—lower birth weight seen in animals; no information on humans
• Immunologic—inconclusive in humans
• Hepatoxicity—weak evidence in workers
• Ocular—irritation, conjunctivitis, and eye discharge in workers

Dr. Markiewicz re-iterated that the MRLs are based on neurobehavioral and immunologic effects. Dermal and ocular effects are the only types of effects seen in humans following occupational exposures. There is some dispute about these occupational studies; however, he expressed his belief that evidence is strong enough to support ocular exposures in worker-type settings. He referred to neurobehavioral studies conducted in the Great Lakes that suggest learning deficiencies and other effects in humans; however, there is a consensus that these data are insufficient. He added that people store and metabolize PCBs differently than animals, and studies indicate that animals are more sensitive to the effects of PCBs than humans.

For non-cancerous effects, based on the calculated doses:

• Doses are lower than occupational doses (looked at human epidemiological studies) and lower than levels not causing harm in humans and animals
• Doses are less than an order of magnitude away
• Lower than experimental thresholds
• It appears that animal models are more sensitive to PCBs than humans

Dr. Markiewicz discussed the potential for cancerous effects:

• Linear-no-threshold model (EPA uses)
• For every molecule of PCB, there is some level of risk
• PCBs cause cancer in rats
• 0.35–3.0 mg/kg/day in human equivalents
• Rat doses >50,000 times all exposure doses calculated
• Rat PCB metabolism different from human
• Epidemiologic studies
• 13 studies showed negative associations between PCB exposure and cancer
• Example: Dr. Renate Kimbrough's study showed that about 17% or 20% of men studied had about a 20% lower cancer incidence than the general population
• 9 studies showed "positive" associations—three without dose response (tenet of toxicology that increase in response should occur as the dose increases), 5 cohorts too small, 7 lack biological plausibility, and 5 confounded by other chemicals
• 4 inconclusive

Interim conclusions for PCBs in biota were presented:

• Fish
• All exposure doses below chronic LOAEL
• Non-cancer—limit consumption of certain species of fish from Poplar Creek, Clinch River, Tennessee River, and LWBR
• Cancer—none expected
• Geese and turtles
• Exposure doses below levels of concern

Interim overall recommendations for PCBs in biota were presented:

• Follow fish advisories
• Further reduce exposure of all biota by:
• Skin fillets and remove fat
• Prepare so fat drains away

Interim consumption recommendations were also provided:

• Sunfish—no limitation
• Largemouth bass
• Poplar Creek and LWBR (3 meals/week is ok)
• Clinch River and Tennessee River (2 meals/week is ok)
• Low consumption (3 meals/year of any species is ok)
• Child/adult avoid >2 meals/week of catfish from Poplar Creek, Clinch River, Tennessee River, and LWBR
• Child avoid >2 meals/weeks of white, hybrid striped, and striped bass from Poplar Creek, Clinch River, Tennessee River, and LWBR
• Adults avoid >2 meals/week of white, hybrid striped, and striped bass from Poplar Creek and Clinch River and >3 meals/week from Tennessee River
• Child avoid >3 meals/week of largemouth bass from Clinch River and Tennessee River

Some concerns from the Community Concerns Database were presented and discussed:

• What about PCB and mercury potential combined effects?
• An executive summary document will be prepared to address mixtures
• Were fish also tested for mercury?
• Yes
• More PCBs in the Tennessee River than the Clinch River
• Relates to source contributions
• Many implications to this question
• PCB levels are not decreasing
• Looked at levels in fish over time
• Will evaluate all media
• Other source contributions of concern
• Are blood samples a good measure of body burden?
• ATSDR believes they are a good measure of what is in a person's body
• Will try to address whether conclusions can be drawn from blood levels

Discussion

Charles Washington asked about Dr. Freedman's training. Dr. Markiewicz answered that she was a board-certified toxicologist and had worked as a toxicologist for several years.

James Lewis asked why they were not provided with prior notice about this presentation nor given advanced copies of information or briefing papers. According to Mr. Lewis, there have been past discussions about providing information so that the work group can be prepared. Dr. Markiewicz replied that the policy had changed, and they were no longer providing advance copies or handouts. Jack Hanley explained that this presentation was following the same process as used previously at this stage. This presentation would provide preliminary information and describe the general direction of the PHA.

Mr. Lewis expressed his belief that it would be beneficial to prepare the groups beforehand so that individuals could formulate questions and provide input during the discussions. His recollection was that advance information had been provided in the past. Mr. Hanley said that they will be provided with the complete public comment version of the PHA after it undergoes internal review and data validation. He noted that the purpose of tonight's presentation was to give an introduction of the material and to obtain their input. In his opinion, Mr. Lewis said, input would be improved if information had been shared prior to this meeting.

Bill Taylor stated that handouts had been provided at past work group meetings. Dr. Markiewicz added that he had handed out maps and other information previously. Mr. Hanley stated that they had not provided briefing papers or copies of draft PHAs at this point, but figures or other similar materials may have been distributed. Kowetha Davidson said that they were given a copy of the presentation on the day it was made, but not given information prior to the meeting.

Dr. Taylor asked whether the y-axis for child and adults was consumption rates or doses. Dr. Markiewicz answered that these were CVs for child and adult fish consumers, which were back calculated from the chronic MRL. John Merkle noted that the CVs vary with the consumption rate, and asked whether this corresponded to the rate of elimination. Dr. Markiewicz explained that the evaluation is assuming 100% absorption without elimination. If someone ate one fish, the exposure dose would be lower than for a person eating more fish. They looked at the level that the health advisory is based on, and backed out the calculation from that. He noted that this was why there was a variation between low, moderate, and high consumers.

Pete Malmquist asked if the Clinch River was evaluated and compared to the LWBR. Dr. Markiewicz replied that four areas were broken out: Clinch River, Tennessee River, Watts Bar Reservoir, and Poplar Creek.

Charles Washington stated that some people like to eat larger fish, whereas he prefers smaller fish because of the phenomenon of smaller fish having less contamination. Dr. Markiewicz indicated that the amount—not size—of fish consumed matters. A person consuming fish once a week would be in a different category than a person consuming fish twice a year. Mr. Washington expressed his belief that smaller fish have not been in the water as long and would have lower contamination. He asked whether the tissue of larger fish was greater than the mass of smaller fish. Dr. Markiewicz said that this could be true, but he has seen data showing sunfish and older catfish with the same mercury and PCB levels even when the catfish are much larger.

Mr. Washington asked what Dr. Markiewicz had seen in this area. Dr. Markiewicz referred to the representation of the median values of concern. He stated that levels for sunfish were below CVs. Hybrid striped bass and catfish had higher median concentrations.

Mr. Merkle asked whether the units were in parts per billion (ppb). Dr. Markiewicz stated that these were in ppb. He explained that the high consumer comparison value is 6 ppb and 10 ppb in a child and an adult, respectively. For low consumers, the comparison value is 300 ppb and 700 ppb for a child and an adult, respectively. Mr. Merkle asked what the concentration was in; Dr. Markiewicz stated this was in fish tissue.

Mr. Washington asked how many fish samples were included. Dr. Markiewicz replied that there were no sunfish samples for PCBs in Lower Watts Bar. There were 26 largemouth bass, 20 hybrid striped bass, and 36 catfish samples. He expressed his belief that there were over 50,000 PCB data entries. Mr. Washington asked where the fish were collected in relation to the contaminant concentrations. Dr. Markiewicz answered that sunfish are relatively territorial and do not move more than 50–100 feet unless there is a flood, whereas other fish tend to migrate.

Mr. Washington asked where PCBs were found in relation to catfish. Dr. Malmquist replied that this would depend on when a person was fishing since various species feed at the lake at different levels and times. Tim Joseph said that catfish are bottom feeders. Dr. Markiewicz explained that fish that are higher on the predator level are more apt to have higher levels of contaminants because they tend to east smaller fish and other fish containing PCBs. Dr. Malmquist indicated that fattier fish would tend to have more contaminants. Dr. Markiewicz noted this was correct, and expressed surprise that there were no data for crappie (a fatty fish).

Mr. Washington asked about the fish in the area. Jeff Hill stated that they have a lot of crappie. Dr. Markiewicz added that there is also a lot of catfish, bass, and bluegill. According to Mr. Hill, crappie and catfish were probably the main fish that sports fishermen consumed; he said that most fisherman put bass back into the water. Mr. Washington questioned what could be compared to crappie since there were no samples. Dr. Markiewicz answered that crappie were similar to sunfish, but they had different lifestyles—sunfish tend to stay around a rock or tree, whereas crappie move around. Mr. Hill stated that sunfish typically do not eat other fish, but crappie, catfish, and bass do. Dr. Markiewicz said that sunfish also eat small invertebrates.

Mr. Washington asked if the PHA would contain information about what fishermen consume and where these fish are located in the water. Dr. Markiewicz explained that he had another slide to show where different fish species are on the food chain, but it would not provide details on where the fish are located. Dr. Davidson asked whether species that are higher on the food chain would have higher PCB levels. Mr. Hill indicated that hybrid striped bass would have higher levels; Dr. Markiewicz indicated that catfish would also, and added that bottom feeders will eat anything they can find. Mr. Hill pointed out that hybrid striped bass normally do not feed off of the bottom and that sunfish will not feed on the bottom. Dr. Markiewicz noted that catfish not only consume living things when feeding on the bottom; they also consume sludge, dirt, and other items underwater thereby increasing their exposure to PCBs.

Mr. Lewis asked about the lifespan of catfish. Dr. Markiewicz expressed his belief that sunfish live less than 10 years, but catfish can live a long time. He stated that hybrid bass grow quickly, but typically have a shorter lifespan.

For PCBs in Watts Bar Reservoir fillets before 1996, Dr. Malinauskas confirmed that except for sunfish, eating all other fish evaluated was above or close to CVs. Dr. Markiewicz stated that this was correct for the Watts Bar Reservoir.

Mr. Washington questioned whether the sample was large enough and statistically adequate. In his opinion, Dr. Markiewicz said, there was a large enough data set to represent the various species. Dr. Joseph asked if the data were from OREIS. Dr. Markiewicz stated that this was correct. According to Dr. Joseph, this database contained a lot of fish sampling data and expressed his opinion that this was statistically good data. Dr. Markiewicz added that there was a broad range of data, and in his opinion, the data were representative of fish in these water bodies.

Dr. Davidson expressed her belief that this made sense if sunfish do not bioaccumulate, adding that bigger fish would have higher concentrations. Dr. Joseph stated that all fish bioaccumulate. Dr. Markiewicz stated that all fish do bioaccumulate, but some do so more than others. He noted ways to reduce these effects when cooking and cleaning fish. For example, a person could eat around the fatty portions of the fillets, such as not eating the "blood vein." According to Mr. Washington, many people in Tennessee do not avoid these portions.

Mr. Lewis asked where carp would fit in. Dr. Markiewicz stated that carp would probably be at the higher end of the bottom feeders, and noted that carp was fattier than sunfish or crappie.

Mr. Washington asked about area contamination sources. Dr. Joseph said that utilities are one of the largest polluters of PCBs, and explained that many years ago, PCB transformers were disposed of by dropping them into the lake. According to Dr. Joseph, these transformers are still there and leaking. Mr. Washington questioned whether any other facilities used as many transformers. At that time, Dr. Joseph said that no other facilities used more nor disposed of transformers in that manner.

Dr. Malmquist indicated that a major polluter of the lake was a man who repaired transformers on Tellico above Knoxville. The man had told the Tennessee Valley Authority (TVA) where he buried the transformers, and said that if the agency flooded the lake, the transformers would be under the water. According to Dr. Malmquist, TVA flooded the lake without digging up the transformers, which resulted in a major source of PCB contamination. Mr. Lewis expressed his belief that Eastman Kodak contributed to the PCB contamination as well.

Dr. Malinauskas asked whether any fish samples were collected upstream of the facilities. Dr. Markiewicz replied that some of the samples were from upstream locations, but that more were concentrated downstream.

Mr. Washington stated that there is contamination down to the Chattahoochee. Dr. Markiewicz stated that most waterways are contaminated to some degree.

Dr. Malinauskas questioned if ATSDR could estimate the contamination resulting from ORR operations. In his opinion, Dr. Markiewicz said, this probably could be estimated; however, it would be highly complex due to the process of sedimentation. In addition, when there are dams and locks, contaminants build up even if they are coming from hundreds of miles away.

Dr. Taylor asked whether there were geese data for post-1996. Dr. Markiewicz stated that there were not. According to Dr. Malmquist, geese were not a danger unless someone occasionally killed them as a result of closed hunting seasons. Mr. Hill pointed out that the legal seasons do not necessarily dictate who actually hunts and consumes geese.

In his opinion, Mr. Washington said, some people eat as much as the geese consumption assumptions. Dr. Markiewicz stated that this was about ½ lb per week, or a 6–8 ounce portion. He did not know anyone who ate this much geese, but indicated that it was possible.

Mr. Hill asked how geese were exposed to PCBs. Dr. Markiewicz answered that geese were exposed via eating roots, digging in sediment and dirt, and other ways. Mr. Washington questioned if geese will eat fish. Dr. Markiewicz stated that they will eat many things, including fish, if they can catch them.

Mr. Merkle suggested including time durations in the thermometer graph if it would be meaningful. Dr. Markiewicz agreed.

Dr. Joseph suggested not using log scale graphs, but putting data in terms for the public. In his opinion, he said, the public cannot understand logarithmic scales; instead, they see a straight line. Mr. Merkle expressed his belief that this would make the numbers cluster together at the low end, and Dr. Davidson agreed. Dr. Joseph indicated that showing the meaning of the number to the public was of concern. Dr. Markiewicz agreed that log scales were difficult for the public to understand. Mr. Washington agreed that this was misleading to the public. Dr. Joseph recommended using another way to show the relative range in real time and by using real numbers. Dr. Davidson expressed her belief that a linear scale would not work, but that a table could possibly be useful. Dr. Markiewicz suggested removing the log scale from the graph and putting text within brackets that said a dose is a certain amount below the CV. Dr. Joseph said that this would be a better approach.

Mr. Washington asked about organs in addition to muscle mass. Dr. Markiewicz stated that some populations, such as Asian countries, eat fish eggs (roe) that could have PCBs. Dr. Davidson asked about the liver. Dr. Markiewicz was unaware of people who consumed the liver, but noted that PCBs could be anywhere there is fat. Mr. Washington noted that people in the Far East eat the liver; Dr. Taylor added that they will use the whole fish in a stew. Dr. Markiewicz explained that his evaluation included a comparison between fillet and whole fish, and as mentioned previously, the median values were higher for fillet than all samples and whole fish. The fillet was used to calculate exposure doses, thereby assuming exposure for those with the highest dose.

Mr. Washington asked what people in East Tennessee were consuming. Mr. Hill said that this would depend on the term "whole fish." When he cooks fish, he takes the meat off of the bone along with the skin, whereas some people take the skin off. For whole fish, people clean the fish; remove the organs, head, tail, and fins; and cook it with the skin. Dr. Markiewicz explained that the whole fish samples were taken after a whole fish was placed into a blender and homogenized.

While on the Oak Ridge Health Agreement Steering Panel (ORHASP), Dr. Joseph conducted a study on a commercial fishery that operated in the past (about 15–20 years before the study) on Watts Bar. The fishery made fish patties out of shad, using everything except the guts. The study evaluated radionuclides and considered the maximum concentrations detected. The fishery sold to two populations: one in Chicago and one in New York. Thus, people do consume whole fish.

Mr. Lewis asked whether snakes consume fish. Dr. Markiewicz replied that water moccasins eat fish and noted that all snake species are good to evaluate since they have a lot of pesticides, PCBs, and heavy metals; therefore, they are good indicators of the contamination in an area.

Mr. Hill asked about a person who consumes a lot of catfish and geese. Dr. Markiewicz replied that the conclusions will indicate the groups that should limit their consumption in certain areas. Mr. Hill questioned where the individuals who lived on the lake and mostly lived off the land would be found on the chart. Dr. Markiewicz stated that these people would probably be high on the chart, adding that 108 g/day was the high end used for consumption (based on the EI). He explained that the U.S. Environmental Protection Agency (EPA) uses 170 g/day for subsistence-based populations. These levels—108 and 170 g/day—are still within the same order of magnitude, but below and less than an order of magnitude from the LOAEL. This evaluation made many conservative assumptions, such as assuming 100% absorption. Also, Dr. Markiewicz explained the debate regarding wet weight studies that evaluate uncooked fish. Because fish lose some weight when cooked, the level of PCBs might appear higher because the sample will be smaller; however, some PCBs may be lost as a result of cooking. According to Dr. Markiewicz, EPA indicates that using wet weight samples incorporates these considerations. Mr. Hill expressed his belief that people who eat a lot of catfish and geese would look at this information and not find themselves represented. He questioned where the high fish consumer and high geese consumer were evaluated.

Mr. Lewis asked whether dredged soil was sold and used as topsoil. To his knowledge, Dr. Malmquist said, Watts Bar Reservoir was never dredged. Dr. Markiewicz had seen this used before and would look into this further. Mr. Lewis recalled seeing this used for gardening. Mr. Hill noted that there was a proposal to dredge Watt Bar Reservoir in the past 10 years, but it was not approved. He has bought river bottom dirt before, but not from the Tennessee River.

Mr. Washington asked about the uptake of PCBs by plants. Dr. Markiewicz explained that most plants are water-based, whereas PCBs are lipophilic. Mr. Washington said that transformers were discarded into lakes, ponds, and rivers. Dr. Markiewicz said that PCBs mainly bind to sediment or organic matter; therefore, the uptake in plants would be minimal. Some intake could occur if the plant is in close contact and has a waxy cuticle; however, the PCBs would not enter the vascular parts of the plant. People could, however, eat some if they did not wash their vegetables.

Mr. Lewis asked whether the sediments always stay at the bottom of the river or if there is a difference at certain times depending on the river flow. He also questioned whether there was a higher concentration when transformers were disposed of in the river as opposed to now. Dr. Markiewicz noted that there was also migration of PCBs into sediments. When evaluating aquatic scenarios, the top layer of sediment is the critical area, called the "living zone." If there is a higher concentration in the living zone, then there will be a higher concentration in the species within the zone. During times of discharge, there would have been higher levels in this zone.

Mr. Lewis asked whether correlations were made between the massive floods that have occurred and when the fish samples were collected. Dr. Markiewicz looked at the trend over time. He found that levels are staying relatively constant, which differs from other sites he has seen where levels decrease over time. He was not sure if this was a result of floods and distribution of sediments, and noted that he was unsure how the barrier locks and dams affected the sediment distribution. In the past, Mr. Washington was involved with collecting unofficial sediment samples in the morning and late afternoon following the floods. The primary interest was in PCBs, and according to him, the levels were fairly high.

During Dr. Markiewicz's discussion on the potential for cancerous effects, Dr. Malinauskas asked what ORR doses were being referenced. Dr. Markiewicz answered that these were all of the doses calculated. Because fish were the highest, he was primarily talking about these. He added that the margin of safety would actually be greater if he had calculated doses for water, soil, and sediment. Mr. Merkle questioned the accuracy of using the term "ORR doses." Dr. Markiewicz agreed and suggested replacing this with "exposure doses." After discussion, Dr. Taylor noted that exposure doses would be appropriate.

Mr. Lewis asked if a summary of the epidemiologic studies would be provided in the PHA. Dr. Markiewicz was still preparing this part of the document, but could include this. Dr. Malinauskas suggested only citing the sources. Mr. Merkle recommended including the definitions of cohort and confounding. Mr. Lewis expressed his belief that an example of one of the epidemiologic studies should be included. Dr. Markiewicz was most familiar with Dr. Kimbrough's study of about 7,000 male and female General Electric capacitor workers. The workers were divided into two groups: people who worked with PCBs and people who worked at the facilities. The workers were followed in a longitudinal study. According to Dr. Markiewicz, no association was found between PCB exposure and cancer; in fact, the study actually found a lower incidence rate of cancer in males compared to the general population.

Dr. Malinauskas recommended not providing an example. Mr. Lewis was not suggesting that all of the studies be provided, but indicated that a summary similar to what Dr. Markiewicz just provided would be helpful. Dr. Davidson agreed with Dr. Malinauskas, and noted that a lot of work would be required to summarize all of these studies. She suggested only giving an overview of the studies that were reviewed. Drs. Davidson and Malinauskas agreed that if one study was selected, then all of them would need to be included. Dr. Malmquist expressed his belief that 90% of the public would not read the document. He stated that people who are interested will look up the study themselves.

Mr. Lewis stated that there is bias and information. Dr. Malinauskas replied that all 26 studies were information. Dr. Markiewicz noted that these could be provided in an appendix. Dr. Taylor explained that the purpose is to identify health effects that they are discussing. The PHA could either refer to the studies that went into the CVs or refer the reader to the toxicological profile. The PHA should summarize some of the more important studies, he said, and pointed out that there are various ways to convey this information.

In his opinion, Mr. Lewis said, they had to provide the public with something that they could grasp onto and understand. Dr. Markiewicz suggested using language to indicate that a dose is "1,000 times greater than doses calculated," instead of using the thermometer graph, and possibly also including a table showing the estimated doses compared to those in the studies. He agreed that they needed to have more transparent messages. Mr. Lewis stated that a significant amount of detail was not required, but that some description would be helpful.

In his opinion, Mr. Lewis said, they are assuming that people are interested in the data and the calculations. However, he stated that they needed to be mindful of people who have had exposures and are concerned about health effects when conveying this information to the public. Dr. Markiewicz noted that the data needed to be in perspective, such as saying that the doses are "12,000 times below doses shown to cause health effects in animals or people." Dr. Davidson suggested summarizing the data set instead of the individual studies.

Dr. Malinauskas expressed his suspicion that all 13 negative associations did not have objections. Mr. Merkle expressed his belief that they probably could not pass the null hypothesis.

Mr. Lewis asked what ATSDR concluded from its EI. Dr. Markiewicz stated that there were several conclusions, one of which was that the level of PCBs in serum was more related to age than to fish consumption. It did not look at health outcome data; it investigated serum levels and how much fish people consumed. The study answered the question it was designed to answer.

Mr. Hill asked what ATSDR had expected to see and if it was seen. In his opinion, Mr. Lewis said, ATSDR had stopped short with the investigation, and as a result, concerns still remained. Dr. Davidson said that the investigation was not stopped—the hypothesis had been answered. Dr. Taylor explained that the purpose of the investigation was to look for exposures, with the surrogate for exposures being blood level of PCBs. Blood levels were measured from the study population, and the levels were found to be acceptable. Mr. Hill asked whether the blood levels were lower than those expected to cause health effects. Dr. Taylor said that this was correct except for five individuals. Dr. Markiewicz noted that uncertainty remains about the meaning of the levels because no scientific consensus exists about them. Dr. Taylor added that there was a benchmark used to evaluate the PCB blood levels.

Mr. Hanley explained that the U.S. Department of Energy's (DOE) remedial investigation (RI) for Watts Bar had used EPA risk assessment modeling. The modeling indicated that only people exposed to PCBs through consuming large amounts of certain fish species could potentially have health effects. To validate these findings and assess PCB blood levels, ATSDR drew blood from 116 people and compared their levels to those in the general population. The EI concluded that PCB and mercury blood levels were very similar to levels found in the general population.

Mr. Washington asked whether the general population referred to the U.S. or East Tennessee. Mr. Hanley said that the levels were compared to levels of PCBs found in people in different parts of the country in the early 1980s. Since the 1980s, levels of PCBs in the general population have decreased. In this PHA, ATSDR will compare the levels for high and moderate consumers of Watts Bar fish to the most recent U.S. levels in the National Health and Nutrition Examination Survey (NHANES). Dr. Markiewicz stated that only comparisons will be made—conclusions from blood levels will not be drawn. Mr. Hanley concurred, and added that they will not be drawing health conclusions (from PCBs levels in blood), but showing how the levels relate to the general population.

Mr. Hill asked when the blood levels were taken. Mr. Hanley said that they were taken in the fall of 1997. Mr. Hill questioned whether these were originally compared to levels from the 1980s. Mr. Hanley said that this was correct, but that new data were available to enable comparisons to 2000 data. Mr. Hill asked if they already knew what the comparisons were. Dr. Markiewicz answered that they know the national levels in the NHANES data, but they are still working on comparing the two. He said that there are various ways to calculate blood levels, and that he was waiting for answers from the agency on how to best evaluate the numbers in a scientific way.

Dr. Malinauskas asked whether the levels are higher or lower than the 2000 data. Of the people measured in the EI, Dr. Markiewicz said that some people will be above the national average, but questioned what this would mean. He indicated that comparisons could be made, but he expressed his belief that science was not strong enough to explain what the levels in blood mean.

Dr. Markiewicz would not be discussing blood levels tonight, but the information would be presented in the PHA. He has had conversations with chief scientists in Atlanta, but at this time he could not provide the information with certainty. Mr. Hill asked if they would not see this information until the document is released for public comment. Dr. Markiewicz said that he could have another presentation beforehand. Dr. Malinauskas questioned how having this information in advance would make a difference. Mr. Hill stated that this made a difference to him because the national population might have decreased from 1980–1997, but this area may not have. Dr. Markiewicz explained that the NHANES data did not target high fish consumers as the EI did; instead, the study includes various compounds seen in blood. He cautioned that they needed to be sure to not give people the wrong impression that everyone in this area has higher than national average levels because this is not what the information means.

Mr. Merkle stated that specific town data were needed to make comparisons. Dr. Malmquist asked if they would need to resample the EI population or take another sample if comparing the levels to more recent data. Dr. Markiewicz expressed his belief that people would still be following the same exposure patterns, and thus the data would be representative. He suggested including a table in the PHA of the findings from the EI and the NHANES. Dr. Davidson expressed concern that the NHANES measures exactly what ATSDR is reporting. She noted that most people believe that PCBs are only one chemical, when they are not. Dr. Markiewicz stated that there are 209 different PCBs, and agreed that the data have to be close to compare.

In his opinion, Mr. Lewis said, a chain reaction is sent off in a community when blood samples are collected and then the agency comes back and says that the investigation had no meaning. Dr. Taylor responded to Mr. Lewis, saying that the investigation was meaningful and useful. In fact, Dr. Taylor said, the investigation provided useful information about consumption data, which was what the agency was looking for. At the time of the investigation, the study provided valuable, site-specific information about blood levels compared to the general population. He noted, however, that interpreting these data regarding health effects is difficult and something that the agency is still grappling with. Dr. Taylor said that they had to look at this in perspective by determining whether the consumption rates are still the same as 3–6 years ago (probably so), and if the PCB levels in fish are the same (ATSDR is looking at data for this).

Dr. Joseph added that the EI also responded directly to public request. Mr. Lewis stated that he did not mean to imply that the investigation was pointless. He indicated that the objective was to provide enough information to make a reasonable judgment, and in his opinion, nothing was done in this area. Dr. Taylor said that this study had to be correlated to more health information to provide this type of information.

Dr. Davidson stated her belief that you could not say that the EI did not tell them anything. She said that the study determined blood levels in the community and whether the levels were higher than levels throughout the country. She added that everyone in the U.S. has PCBs in their bodies, and in her opinion, it is extremely useful to know that the levels are not higher than other parts of the country. Mr. Hill, however, said that he had not heard anything to indicate that this area has lower than national levels. In his opinion, it was significant that the national levels are decreasing, while the levels in this area are staying the same. Dr. Davidson said that they are still in the same place as the national average if the levels are the same as the 2000 levels.

Mr. Washington asked what would happen if the PCB levels have increased. Mr. Merkle said that possibly the levels in this area were low from the beginning if the levels are staying about the same, but the national average was dropping. Mr. Hill said that this did not reflect what they have been hearing tonight. Dr. Markiewicz explained that the PCB levels in fish do not appear to be decreasing in this area. Regarding comments Mr. Lewis made about having samples taken, Dr. Markiewicz agreed that the public expects answers following something of this nature. He said that ATSDR needed to do an adequate job of explaining that this was an EI that answered specific questions, and convey that there is a lot of uncertainty involved. In her opinion, Dr. Davidson said, foremost people want to know if they are or are not normal.

Mr. Lewis expressed concern that they continued to miss the point about reaching the public. He said that he is still trying to figure out their objectives.

Dr. Malmquist asked about the national levels of PCBs dropping, while the fish levels are not decreasing in this area. He questioned if this meant that the levels of PCBs were not decreasing in this area, and if so, why not. He also asked when the national samples were collected compared to their samples in 1997. Dr. Markiewicz answered that blood levels in this area were only collected in one snapshot. He said that they are comparing fish or sediment levels over time—not only to 1 year. He added that the national samples were first collected in about 1991.

Dr. Malinauskas asked if they were only focusing on people who claim that they ate fish in the EI. Dr. Taylor said that they looked at people who ate fish and turtles. According to Mr. Merkle, the investigation found that 116 out of 550 people ate such little fish that they could not be studied. Dr. Malinauskas stated that the national average was not people consuming fish. Dr. Markiewicz said that this was correct, and noted that discussions on this topic had to capture that the national average does not focus on fish consumers while the EI did.

Mr. Hill asked if there was a national average on fish—not blood. He questioned whether they could look at national averages of PCBs in fish from 1980–current. Drs. Joseph and Markiewicz were unsure that these data were available, but Dr. Markiewicz asked about the Food and Drug Administration (FDA) market basket surveys. Dr. Taylor indicated that the FDA has conducted these for decades, but was unsure if these looked at PCBs. Mr. Hill requested that someone look to see if these are available. Dr. Joseph asked what this would tell them. Mr. Hill replied that this would tell them whether PCBs in the nation have decreased.

Dr. Malmquist questioned eating less fish from the LWBR when 80% of it comes from the Tennessee River. Since the Clinch River only produces 20% of the water volume, he asked whether this meant that the Clinch River had much higher PCB concentrations than the Tennessee River. He noted that the recommendations indicated that it was alright to consume fish from the Tennessee River, but not the LWBR, which in his opinion did not make sense when most of the flow comes from the Tennessee River. Dr. Markiewicz said that they could get an idea of relative levels by looking at the different species from the various areas. He understood Dr. Malmquist's comments about flow because generally the higher the flow, the higher the degree of sedimentation. He said that PCBs will bind to sediment or organic matter; whatever contribution is greatest for PCB-bound molecules is probably where the highest levels will be.

In his opinion, Mr. Lewis said, people were interested in where PCBs were located. He recommended giving an indication of the TVA monitor locations and showing where the concentrations were higher in the various water bodies. He said that this would provide people with information on what is in the various areas, and suggested using a general overview map.

Dr. Malinauskas asked about the PCB sources. Dr. Joseph said they were ubiquitous, but that there were differences among them. Dr. Malinauskas asked if there were sources above the ORR. Dr. Joseph noted that there were sources above the reservation, as well as in Norris. Dr. Malinauskas asked if the levels decrease; Dr. Joseph said that the levels always lessen, but that they are present in Norris. Mr. Hanley did not recall the PCB detection levels in Norris, but knew that they were in the database. He indicated that they might be some of the background samples. Dr. Malinauskas asked if there were fish measurements in Norris. Mr. Hanley said that there might be background levels in fish in Norris or near Norris. Dr. Markiewicz recalled that the information was limited. Ms. Arbogast said that they could research the information.

Mr. Hill asked if there were fish consumption warnings for PCBs in lakes located upstream. Mr. Hanley said that there are warnings in Loudon and along the Tennessee River, but not the Clinch River. Mr. Hill asked about the Emory River; Mr. Hanley did not recall any warnings. Mr. Hanley noted that the warnings are still in effect for Loudon and along the Tennessee River down through Watts Bar.

Mr. Washington asked Dr. Malmquist to reiterate his comments about contamination related to the Tennessee River and Clinch River. Dr. Malmquist said that on the Tennessee River below Kingston, 80% of the flow comes out of Knoxville and 20% comes from the Clinch and Emory Rivers. Mr. Washington asked whether the Clinch River had more contamination. Dr. Malmquist explained that if the recommendations say not to eat many fish from the LWBR, but that you can from the Tennessee River, then this indicates that the Clinch River is highly contaminated. Mr. Hill added that it would have to be contaminated for 20% to contaminate 80%; Dr. Malmquist said this was correct. Dr. Markiewicz noted that the LWBR could also be a collection point. Mr. Hanley stated that the DR Staff tried to determine where most of the PCBs originated, and concluded that most came down the Tennessee River.

Dr. Davidson asked if PCBs are sensitive to heat. Dr. Markiewicz stated that some volatilization could occur causing chemical changes. Dr. Davidson explained that most people who fry their fish reuse the oil. Since PCBs are fat soluble, she questioned whether PCBs could leach into the oil, thereby increasing the concentration of PCBs in fat that people are reusing. Dr. Markiewicz said that this was true, and noted that this was a concern as most high consumers probably fry their fish. Dr. Davidson suggested recommending that people change their oil when cooking.

Dr. Joseph asked how ATSDR would respond to a person who asks what a certain blood level means for his or her body. Dr. Markiewicz noted that this would be difficult to answer because tremendous uncertainty is associated with this issue.

Mr. Hill proposed an action item: Determine the PCB levels in fish within this area over the specific time period of interest; assess whether these levels have stabilized, decreased, or increased; and compare these levels as close as possible to national levels in fish. Answer the following: 1) Have local levels stayed the same (over the longest time period that can be stated)? and 2) How do these levels compare to the national or regional levels over the specific time period?

Dr. Joseph expressed his belief that it might be possible to compare levels regionally, but not nationally. Mr. Merkle said that ATSDR needed to use numbers that can support statements about comparisons between local and national values. Dr. Markiewicz knew that ATSDR had data for PCBs in fish for Paducah, Kentucky. Dr. Joseph said that the state and EPA also had data. Dr. Davidson noted that they might find some similar areas.

According to Mr. Hill, he had been told that transformers in this area were always located downstream of the dam so if a spill occurred or draining was needed, a valve could be opened and the PCBs would drain into the lake downstream of the dam where there is a good flow. This practice discontinued 30–40 years ago, and in his opinion, the contaminants should be settling out or moving downstream by now. Dr. Markiewicz added that this was downstream of LWBR.

Mr. Lewis asked when actions were taken to stop PCB use. Mr. Hanley recalled that this was incorporated into the PHA, but if not, he would ensure that it was included. Dr. Taylor referred to the toxicological profile, indicating that PCB production occurred from about 1930 until it ceased in 1977. Mr. Hill said PCBs were still used until the early 1980s; Dr. Markiewicz stated that PCB transformers are actually still used today.

Dr. Joseph reiterated his concern about using a linear scale instead of a logarithmic scale to put information into perspective for the public. He presented an example of a non-logarithmic graph containing text boxes.

Additional Business

Dr. Malinauskas reminded the group that comments on the White Oak Creek PHA were due to him by June 6. After discussions on the next meeting date, Dr. Taylor suggested planning this after tonight's meeting. Dr. Malinauskas adjourned the meeting at 7:40 p.m.

Action Item

Determine the PCB levels in fish within this area over the specific time period of interest; assess whether these levels have stabilized, decreased, or increased; and compare these levels as close as possible to national levels in fish. Answer the following: 1) Have local levels stayed the same (over the longest time period that can be stated)? and 2) How do these levels compare to the national or regional levels over the specific time period?