PUBLIC HEALTH ASSESSMENT
GENERAL MOTORS (CENTRAL FOUNDRY DIVISION)
MASSENA, ST. LAWRENCE COUNTY, NEW YORK
Summary of Public Comments and Responses
General Motors/Central Foundry Division
Public Health Assessment
This summary was prepared to respond to public comments on the General Motors/Central Foundry Division draft public health assessment. The public was invited to review the draft document during the public review period which ran from March 23 to April 27, 1993. Sixteen sets of responses were received by the New York State Department of Health (NYS DOH). Some comments were consolidated or grouped together to incorporate similar concerns raised by more than one person. The toxicological data reviewed for the evaluation of human health effects included relevant papers published prior to 1995, and a major toxicological review published in August, 1995. If you have any questions about this responsiveness summary, contact the Health Liaison Program at the toll-free number 1-800-458-1158, extension 6402.
COMMENTS FROM THE PUBLIC
In the Community Health Concerns section of the public health assessment (PHA) it is stated that residents at a public meeting were in favor of complete removal of the material in the Industrial Landfill and the East Disposal Area rather than treatment. These people may have been making an emotional statement rather than one based in reason. Did they have some relevant knowledge?
The residents calling for complete removal at the public meeting are residents of the Reservation. Most reservation residents believe the land near the reservation should be restored to as close to natural conditions as possible, regardless of cost. We are not aware of any information the reservation residents have about the landfill that is not available to the general public in the repositories.
Most residents want complete removal of landfill material, is this feasible? General Motors should clean-up as much of the extended area as is physically and technically possible.
It is feasible; however, there are many factors that are considered. Those factors include: overall protection of human health and the environment, long and short term effectiveness, implementation and technical reliability, compliance with statutory requirements, community acceptance, and cost. If the wastes are sufficiently contained on-site, there is no public health reason for complete removal. Removing the wastes will only occur if they cannot be sufficiently contained on-site.
Wildlife that is consumed are in contact with PCB-contaminated soil and this will continue to occur until these soils are removed. Despite the danger of increased contamination, this process should begin.
The process of discovery, investigation, and eventual remediation are sometimes lengthy. The off-site areas where people and wildlife are found are the first areas that will be remediated.
Can the exposed, unfenced contaminated fill be capped?
Yes, but the current plan calls for these soils to be removed and contained at the plant facility within the fenced area.
Access restrictions should be implemented in contaminated areas near the Raquette River. Page 5, (B) - The unfenced area containing hazardous waste should be posted if not fenced to protect the public. Signs should be put up in particularly hazardous areas.
The contaminated soils in the unfenced area will be consolidated into the fenced area. Until that occurs, the area should be posted and this recommendation is in the health assessment. Staff from the Mohawk Environmental Services have placed warning signs in the areas near the cove where the highest levels of off-site contamination exists.
Page 9 - Surface water and sediment contamination are "near" the facility, to what extent? Page 11 (top) - the PHA states that a "large" area of the St. Lawrence River is contaminated. How large of an area is contaminated?
Background levels in the St. Lawrence River need to be determined.
We have not determined background levels. We do not believe that is necessary, because the remedial goal for the sediment removal is as low as is feasible.
Don't dredge the River! Disturbing sediments will send a plume of contaminated material downstream and will lead to new contamination.
This is one of the major points considered when the decision to dredge the river was made. Several controls will be instituted to minimize the potential for contamination to move from one part of the river to another part.
Page 9 (top) - contaminant level is "just at detection limit", what is the detection capability of equipment used?
The detection limit for the analysis of PCBs in residential water samples was 0.5 micrograms per liter or parts per billion when those samples were taken. This has been clarified in the final document.
Can the water intake for the Mohawk Reservation be moved upstream of the plant?
The first factor to consider when evaluating moving a water intake is the reason why it should be moved. There is no obvious reason for moving the intake since the highest level of PCBs detected in the water supply was almost one hundred times less than the current drinking water standard.
General Motors should pay for the testing of water samples and filters for the Mohawk Reservation. General Motors should be held responsible for all the wells that are contaminated and for the air, soil and water restoration.
All expenses associated with the investigation and remediation of a State Superfund site are eligible for cost recovery action by the State of New York. General Motors is ultimately responsible for all on- and off-site areas that are contaminated from operations at the GM facility.
We are aware that the State has conducted studies on the Native Americans concerning their exposures to pollution. Has the State conducted any studies on other people who live between the plants? They get a double dose of pollution. We live in a depressed area and many people are on welfare and unemployment and eat the fish, ducks and frogs. We know the State has studied the health effects from these exposures to the Native Americans, but has the State also considered the same effects on other people living in the area? If not, why not?
The health studies have focused on the Native Americans because they are the most potentially exposed population in the area. The results of these studies do not suggest the need to evaluate other populations. The consumption advisories issued for the area apply to all residents of the area. There is one summer camp between General Motors and Reynolds. We are not aware of any full time residences that are located in this area.
There is no mention of the previous direct contact to the Mohawk children who used to ride three-wheelers and mini-bikes over the industrial landfill before a fence was erected. This exposure may lead to health effects in the future.
Exposure to PCBs and other toxic chemicals can come from more than one Superfund site and can be carried by the air, water, through the foodchain and affected by lifestyle.
This is a true statement, and the governmental agencies involved in the investigation of the GM facility are also investigating the nearby Reynolds and Alcoa sites.
Monitoring of all biota, breastmilk, etc. should continue.
The monitoring of breastmilk has been continued beyond the original study. Fish samples are routinely collected by NYS DEC and the results of these surveys are used by NYS DOH in the preparation of advisories regarding fish consumption.
The PHA states that there are 2,000 Mohawks on the American side; this gives the impression that NYS is only concerned with NYS registered Mohawks (who may live anywhere), which I don't believe is true. This site is on the Canadian border, and those living there are concerned with the impact this site is having on their health and environment. We received comments focusing on the concern that contamination from GM can affect Canada both directly across the border and downstream. The Canadians would like their concerns considered as part of the PHA.
The PHA considers the public health effects on the population surrounding the site. The primary exposure route of concern for citizens of Canada would be ingestion of contaminated biota, particularly fish. We have reviewed the Ontario Ministry of the Environment's Guide to Eating Ontario Sportfish and found that there are advisories in effect for consumption of fish from the waters near the reservation because of organochlorine contamination.
There should be better monitoring of birth defects among the Mohawk Nation.
Birth defects are being monitored among the native American mothers who participated in the breastmilk study done by NYS DOH. Based on this survey, no statistically significant excess of birth defects were found; however, the number of women participating is relatively small. Further study of birth defects among native Americans is very difficult because there are very few births registered in New York State by Mohawk mothers. The majority of Mohawks choose to have hospital births in Canada, and other Mohawk women prefer to deliver at home.
The cumulative impacts on this area are so high people need a serious positive step to reduce some. Would education about groundwater, soils and air help? An additional recommendation is that meeting and action courses be held to educate people and to safeguard children and infants from further contamination.
Recommendations regarding education of the community and health professionals are found in the Health Activities Recommendation Panel Recommendation section.
General Motors should pay for any health costs that has a proven link to PCB contamination.
Although causation is difficult to prove, private citizens have the right to file a toxic tort case for health care costs.
There is not adequate focus on polycyclic aromatic hydrocarbons (PAHs), heavy metals and volatile organic compounds (VOCs). Also, there was no mention of the health effects of benzo(a)pyrene. While it is not present in huge quantities at the GM site, it has a significant presence in the area.
Determination of adverse health effects from potential exposure to environmental media containing PAHs and notably benzo(a)pyrene has been addressed in the revised PHA. For example, the NYS DOH has determined that chronic exposure of on-site workers to benzo(a)pyrene in soil by ingestion and dermal contact would pose a moderate increased cancer risk. There are also limited analytical data on the presence of heavy metals and VOCs in various environmental media associated with the site, and such data have been included and evaluated in the PHA.
There is only passing and indirect reference to interaction effects between contaminants. Although this is a relatively new field of study the evidence is particularly strong that organic contaminants in general and particularly PCBs interact with other contaminants usually increasing their toxicity.
With multiple contaminants the risks are evaluated first for cancer and noncancer health effects from exposure to each contaminant. If the cancer and noncancer toxicological effects associated with each of these contaminants are the same or very similar, then the individual risks are added. This assumption of additivity of effects was used to determine the health risks described in the toxicological evaluation section of the PHA for the General Motors/Central Foundry Division site. There is some degree of uncertainty associated with this. For instance, if these contaminants acted in a synergistic fashion, then the degree of risk would be greater than stated. If they acted in an antagonistic fashion, then the degree of risk would be less than stated. Although synergism or antagonism of effects from exposure to multiple contaminants is a possibility, data on toxicological mechanisms are inadequate to evaluate this further.
The Background section primarily discusses the site that is the subject of the PHA. The Public Health Implications section discusses hydrogen fluoride, the fluoride compound for which Toxic Chemical Release Inventory (TRI) data are available.
Styrene emissions should be included in the TRI data for the facility.
See Table 8. All data reported to the TRI in 1992 have been included in the final PHA, including reported styrene emissions.
General Motors should not be allowed to emit styrene from the Massena facility.
General Motors is allowed, under the current regulations, to emit a certain amount of styrene. The 1992 TRI data showing the amounts emitted are included in the final PHA.
The recommendations for what will be done with the site are unclear, will the site be cleaned up?
The site will be remediated according to the Record of Decision (ROD). This can be reviewed at the document repository. A modification to the ROD has recently been proposed by GM.
The PHA should include a map of groundwater levels to help determine flow of contaminated groundwater.
The maps are too large to be included in the PHA. If these maps were reduced they would be illegible. These maps are available for review in the repositories at various locations.
Keep all units of contamination for PCBs the same (choose either mg/kg, ppm or mcg/L).
The appropriate unit varies with the medium being discussed, but should remain consistent for a given medium throughout the document, including the tables. The unit, mg/kg, is a weight by weight measure and is used consistently for the measurement of contaminants in solids, and mcg/L is a weight by volume measure used for liquids.
Page 19, Public Health Action Taken #4 states "GM placed a cap over the Industrial Landfill, 'preventing further' percolation of rainfall through the highly contaminated soils." I suggest that 'reducing' be substituted for 'preventing further', since it is a temporary cap, not a permanent removal.
The PHA has been amended based on this comment.
A recommendation should be added concerning the need for education about the dangers during "site remediation" to those immediately downwind.
There will be a Health and Safety Plan (HASP) for remediation of the site. The HASP will include a Community Air Monitoring Plan (CAMP) which will require perimeter air monitoring during remedial activities. Public notification will occur when remediation begins.
COMMENTS FROM AREA INDUSTRIES
In general, the draft public health assessment (PHA) for GM/Massena contains many unsubstantiated statements about potential risks associated with PCBs on and off the General Motors National Priorities List Facility in Massena, New York. The authors use terms such as "high" or "unacceptable" risk throughout the document, but do not provide supporting data, assumptions, quantitative exposure assessments, references or risk calculations. Statements about potential risks should either be supported with quantitative risk assessment, or should be removed from the document.
The procedure used by the NYS DOH to evaluate potential health risks from exposure to contaminants of concern associated with the General Motors/Central Foundry Division site has been added to the PHA as Appendix D. Although quantitative risk calculations were completed to estimate health risks, it is not the intent of the PHA, which is not a risk assessment document, to provide detailed documentation of these calculations. Instead as discussed in Appendix D, calculated excess lifetime cancer risks and noncarcinogenic health risks were qualitatively ranked. For example, if the qualitative descriptor was "high", then the excess lifetime cancer risk from that exposure fell in the range of one per thousand to less than one per ten.
The section devoted to discussing the toxicity of PCBs (six short paragraphs on pages 13 and 14) is incomplete, is lacking complete references, and is inaccurate in several places. A NYS DOH/Agency for Toxic Substances and Disease Registry (ATSDR) document devoted to assessing health effects associated with PCB exposure needs to have an accurate and complete discussion of PCB toxicity, particularly in light of the fact that ATSDR has a lengthy toxicological profile on PCBs.
The PHA is not intended to be an exhaustive review of the toxicological literature on PCBs. Therefore, the most notable and significant toxicological properties of PCBs have been presented in a concise, but thorough, format in the PHA. In addition, the most recent ATSDR Toxicological Profile for PCBs is referenced in the PHA for those readers who may wish an extensive toxicological review of these chemicals.
The discussion was clarified in two areas. (1) The discussion of human PCB toxicity did not clearly characterize the current scientific consensus that the primary causes of Yusho poisoning were the products formed when PCBs are heated or burned, e.g., polychlorinated dibenzofurans (PCDFs), and not PCBs. (2) The discussion did not adequately address the uncertainties in evaluating if the effects observed in workers exposed occupationally and in the children of women exposed environmentally to PCBs were caused by PCBs. The text has been revised to improve clarity and avoid misinterpretation.
Data presentation in the PHA is in some areas selective, vague and confusing. The authors selectively included only pieces of the available data, and did not discuss any data which are presented in the assessment. Many isolated statements about a maximum detected sediment value or an air sample in exceedance of a NIOSH value are presented without a mean or range of values. This does not give the reader an accurate representation of conditions at the site. Data reported in the document should be discussed in terms of numbers of samples taken, ranges of values obtained, collection dates, and who collected the data. Without this information it is not possible to draw meaningful conclusions about the representativeness of the data.
The example used of the air sample that exceeded a NIOSH value is the only sample that had been collected at the time the public comment release PHA was written. Additional air data are included in this final public health assessment. Qualifiers and additions for all other data have been added to the Environmental Contamination and Other Hazards section of the final document.
Another general problem with the PHA is the tendency to attribute PCBs in the St. Lawrence and off-site areas to GM. The health assessment needs to be revised to indicate that there are other PCB sources upstream of GM which are potential contributors to the observed PCBs in surface waters, sediments, and air.
A statement regarding other potential sources of PCB contamination near the GM site has been added to the final PHA under the Off-Site Contamination section, Surface Water and Sediments subsection.
Completion of this PHA document prior to finalization of New York State's health risk assessment is ill-considered. The risk assessment final report will provide supporting rationale and calculations necessary to judge whether a health risk does in fact exist. Data from the NYS DOH breast milk study has been available since October of 1992. The results of this study demonstrated that breast milk PCB concentrations for reservation mothers were not discernibly different from those of controls (geometric mean concentrations = 0.405 ppm for reservation mothers; 0.398 ppm for controls). Compared with the national average value of 1.7 ppm reported by the NYS Health Planning Commission (1977), these values are low. Furthermore, results from the fish and wildlife sampling should be available for review. The fish sampling was completed in April of 1990. The wildlife study was completed in October of 1992.
The PHA is written and released for comment on a schedule set by the ATSDR. The public comment version that was released prior to completion of the Health Risk Assessment (HRA) would not contain information from the HRA. The final release document contains a summary of the findings of the HRA.
In the first paragraph of the Summary, NYS DOH/ATSDR state: "Although an interim cap has been placed on the landfill, contaminant migration continues from this and other on-site sources". This statement is misleading, since the RI/FS data indicates that the Industrial Landfill has a very limited impact on groundwater in this area. In GM's comments on EPA's Draft Proposed Plan (DPP) for Operable Unit 1 at this site GM quantifies the potential amount of PCBs which may be migrating from the Industrial landfill to be 1 gram/year. As comparison, tons of kilograms per year of PCBs enter the St. Lawrence River from upstream sources. Similarly, a review of the groundwater data related to other management units at the facility would indicate that groundwater at the facility is largely unaffected. The PHA should acknowledge that contaminant migration from the various management units is very limited.
Agency personnel are not in complete agreement with GM's evaluations regarding groundwater migration from the Operable Unit 1; however, this comment has been reflected in the final version of the PHA.
In the first paragraph of the Summary, two studies are mentioned, but no conclusions are discussed. The Mt. Sinai Study found that "No consistent evidence of adverse health effects reported in the literature to be associated with PCB exposure was detected in association with fish consumption in the study". The first parts of the Health Risk Assessment are complete and their conclusions are known.
The summary has been revised to include the conclusion of the studies. Additional information about the Mount Sinai Study and the breastmilk study have also been added to the Health Outcome Data section. The executive summaries of the conclusions of the HRA and breastmilk studies have been added to the PHA as Appendix C.
In actuality, many of the concerns raised in the report relate to environmental issues which are broader than the GM site. For example; the presence of PCBs in fish of the St. Lawrence River reflect numerous PCB sources, and not simply the GM site.
This statement has been amended in the final version of the PHA.
In the third paragraph of the Summary, the ATSDR Health Activities Recommendation Panel (HARP) determination is not included in the references to the report. If a recommendation was documented, it should be included in the references.
HARP does not issue a separate report. The written recommendations of the HARP group discussion are included in the PHA.
In the first paragraph of the Summary, PHA states: "An environmental education program is recommended to advise the public health professional and local medical community of the nature and possible consequences of exposure to contaminants from the GM site." In the third paragraph of the Summary, what information was obtained that determined that a lack of education about PCB health risks exists among community health professionals and the local medical community? GM has previously funded the St. Regis Mohawk Tribe (SRMT) to provide information on the issues associated with PCBs. In addition, the health professionals have participated in many of the studies mentioned in the report.
NYS DOH considers this education to be an on-going process. Efforts are necessary to continue to update local health professionals about new information as it becomes available.
In the third paragraph of the Summary, the Breast Milk Study (Chemical Contaminants in the Milk of Mohawk Women from Akwesasne) was completed in October of 1992. The results of these studies could have been addressed in this report.
In the third paragraph of the Summary, the PHA states: "A cap was placed over the Industrial Landfill, which has also been surrounded by a fence". In actuality, the entire GM facility is surrounded by security fences, and is served by monitored surveillance. Plant Security is on duty 24 hours per day.
This has been clarified in the final version of the PHA.
In the fourth paragraph of the Summary, mention of the supply of bottled water in the summary without describing the related issues discussed later in the document is inappropriate. The discussion should be removed or expanded. Mention of sampling of the public water supply without mention of the sample results in the summary of the report is also inadequate.
The summary format does not allow issues to be discussed fully in that part of the report. However, additional information was added to clarify the statement about bottled water.
In the fifth paragraph of the Summary, indications of potential air issues based on a single positive air sample result is inappropriate. No existing exposure from volatilization or particulates is known at this time.
Additional air data are presented and discussed in this final public health assessment.
In the fifth paragraph of the Summary, the NYS DOH needs to calculate exposure intakes, provide a more thorough and accurate discussion of PCB toxicity, provide quantitative risk calculations for each proposed pathway of exposure and discuss any uncertainties associated with the quantitative assessment.
As previously stated (see Response #2), the PHA provides a summary of the most notable and significant toxicological properties of PCBs and that summary has been revised to improve clarity and avoid misinterpretation.
The NYS DOH determined exposure intakes for contaminants of concern (e.g., PCBs) in various environmental media based on exposure scenarios which are now described in revised Tables 1-3 of the PHA. These exposure intakes were then used to evaluate potential health risks from exposure to these contaminants. As previously indicated (see Response #1), although quantitative risk calculations were completed to estimate health risks, it is not the intent of the PHA, which is not a risk assessment document, to provide detailed documentation of these calculations. Instead, calculated excess lifetime cancer risks and noncarcinogenic health risks were qualitatively ranked according to the procedure described in Appendix D. Finally, in regard to the issue of uncertainties associated with the quantitative assessment which the NYS DOH did for the General Motors/Central Foundry Division site, the methods and procedures that were used to estimate increased cancer risks are consistent with those used by scientists at federal agencies and many scientific advisory groups. Additional details have been added in Appendix D.
Background section, subsection A. Site Description and History - Mentioning that the surrounding land drops off sharply to the St. Lawrence River is misleading. GM has extensive surface water collection systems in place which collect site surface water for treatment prior to discharge to the St. Lawrence.
This information regarding the surface water collection and treatment system has been added to the final version of the PHA.
Background section, subsection A. Site Description and History - The PHA states "The majority of the alternatives selected by GM's consultant in the FS have been rejected by US EPA, the lead agency for the site." This sentence is misleading and indicates friction that does not exist. GM's and EPA's selected alternatives were in general agreement for the excavation of the Inactive Lagoons, the North Disposal Area, the Misc. Site Soils, the 350K gallon lagoon and the 1.5 Mgal. lagoon. GM's other alternatives were never "rejected" by EPA, rather EPA has been open to additional data from GM which may clarify certain alternatives.
This statement has been removed.
Background section, subsection B. Site Visit - Again, sampling is mentioned with no mention of results. Discussion should be complete or deleted.
This section of the PHA doesn't include environmental data. The environmental data are discussed in the Environmental Contamination and Other Hazards section.
Background section, subsection B. Site Visit - PHA states "The GM security guards are responsible for the plant facility 24 hours a day". This paragraph lead the reader to believe their responsibilities are for the main building only. This is not the case. The guards are responsible for the entire site within the boundary fence.
This information has been added to the final version of the PHA.
Background section, subsection B. Site Visit - Results of the breast milk study were discussed in May of 1992, and a Summary report was issued in October 1992. These results could have been included and discussed in this health assessment (see comment below under "D. Health Outcome Data").
The results of the HRA and the breastmilk studies have been summarized and included in the final version of the HRA in Appendix C.
Background section, subsection C. Demographics, Land, and Resource Use - The area where the site is located does not include any farms, nor any existing agriculture. The map included in Appendix A, Figure 3 of the report identifies the area as forest and abandoned farm land.
This sentence has been amended in the final version of the PHA.
Background section, subsection C. Demographics, Land, and Resource Use - The last sentence of paragraph 2 mentions "known high concentration levels of PCBs in sediments". Page 12, paragraph 7 of the report correctly states there are some areas of low concentration PCBs, but not "high concentrations". This information is inconsistent.
The levels of PCBs in the sediments in the unnamed tributary have been found up to 3,100 mg/kg, which is considered high. The discussion on page 12, paragraph 7 of the report is in reference to surface soils, not sediments.
Background section, subsection D. Health Outcome Data - NYS DOH/ATSDR state that "The NYS DOH is completing a study of PCB levels in the breast milk of reservation mothers which was finalized in July, 1992." This sentence is confusing. If the study is finished, the results should be discussed. A 10/92 summary of this study reports that the geometric mean concentration for PCBs in breast milk was 0.405 ppm and 0.398 ppm in reservation mothers and controls, respectively. These values are not discernably different from one another, and are lower than the mean value for PCBs in breast milk reported by the NYS Health Planning Commission Report in 1977 (1.7 ppm).
Community Health Concerns section (1st paragraph) - Mention of the studies or samples analyzed by the New York State Department of Environmental Conservation (NYS DEC) should be referenced in the report. Applicable quality assurance/quality control (QA/QC) data should be reviewed and deemed acceptable before being included in the report.
The appropriate references have been made. QA/QC data were deemed acceptable for all data included in the report.
Community Health Concerns section (2nd paragraph) - The PHA states that the primary health concern at the site is PCBs in the nearby rivers and biota. The GM facility in Massena is not the sole contributor of PCBs to the St. Lawrence River. Numerous upgradient sources of PCBs in the river exist. Investigations of surface water and biota upstream of the facility show concentrations equal to or greater than those observed downgradient of the facility.
This PHA was done to evaluate potential exposures and health effects to people living and working near the General Motors facility. Other sites or sources of contamination are mentioned since they may affect the population being assessed (see response to industry comment #4).
Community Health Concerns section (2nd paragraph) - Styrene has not previously been a concern with respect to CERCLA activities at the GM facility. Raising concerns about styrene in this document without further justification is inappropriate.
The community has raised concerns about styrene. Moreover, the ATSDR guidance manual instructs health assessors to review the TRI data for determining additional sampling needs and amounts and names of contaminants that have been released by the site facility and others in the vicinity. For additional information see page 5-10 of the manual (ATSDR, 1992).
Community Health Concerns section (2nd paragraph) - Impacts of air emissions from a nearby facility should not be addressed in a health assessment for the GM facility. Any air emissions from off-site sources should be discussed in a section of the report which specifically addresses off-site sources.
See Response #26.
Community Health Concerns section (2nd paragraph) - Potential PCB issues in the Grasse River are not relevant to this site, other than as an example of other sources of PCBs to the St. Lawrence River.
The people who may be exposed to contaminants from this site may also be exposed to contaminants in the Grasse River. The PHA looks at all potential sources of exposure to the population being studied.
Environmental Contamination and Other Hazards section (1st paragraph) - the PHA states: "Several major investigations have been conducted at this site." Only two studies are referenced in the report: Mt. Sinai and GM RI/FS studies. No other studies or major investigations are mentioned. The "major investigations" should be identified and discussed.
The PHA states "Because of the massive volumes of highly contaminated materials, there are hundreds of soil samples...". Terms such as "massive volumes" and "highly contaminated" should either be explicitly defined and substantiated with data or eliminated from the text.
NYS DOH considers the Mt. Sinai study, the RI/FS study, and the HRA study to be the major investigations. The text has been clarified to address the second part of this comment.
Environmental Contamination and Other Hazards section (2nd paragraph) - The "environmental contamination data" referenced in this section and provided in Tables 1-7 do not reflect the true extent of the available data. Furthermore, the tables co-mingle data from multiple sources. This should be avoided. Data sources should be included with each table. In addition, a significant amount of necessary information is lacking from these tables. This information includes number of samples collected, frequency of detection, samplers, and sample dates. Frequency of detection should be included in determining contaminants of concern. The RI/FS provides concentration ranges and median concentrations.
Qualifiers and additions for all the data have been added to the Environmental Contamination and Other Hazards section of the final document.
Environmental Contamination and Other Hazards section, subsection A. On-Site Contamination - Air - A more complete discussion of the available data is necessary to draw meaningful conclusions about an air pathway. Referencing a maximum concentration without presenting other sampling information (e.g., mean, range, number of samples, locations collected, etc.) is misleading and inappropriate.
Additional air data are presented and discussed in this final public health assessment.
Environmental Contamination and Other Hazards section, subsection A. On-Site Contamination - Groundwater - A more complete discussion of the available data is necessary to draw meaningful conclusions about a groundwater pathway. Again, information on means, ranges of concentrations, etc. is necessary to draw meaningful conclusions about potential groundwater contamination.
Groundwater standards are violated on-site. As the discussion of the groundwater ingestion pathway indicates, these levels are not found in areas where there is a potential for ingestion to occur.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - The PHA states that "a significant amount of contamination has migrated from the GM facility to the surrounding area...". Terms such as "significant amount" are imprecise and should either be defined or eliminated from the text.
This statement has been amended in the final PHA.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - Surface Soil - The PHA states that 23 surface samples were collected from the St. Regis Indian Reservation, and that 9 samples contained PCB concentrations up to 3.3 mg/kg. Data for the remaining samples should also be discussed.
The final version of the PHA includes the median of the samples.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - Surface Soil - The PHA states that PCB concentrations as high as 400 mg/kg were detected in fill material from the GM plant site. To give an accurate idea of the extent of PCBs present in any medium, information on the number of samples taken, range of concentrations detected, the PCB Aroclor detected, and the detection limits should be presented. Referencing a minimum or maximum concentration without giving a mean and range of concentrations is misleading to the public and inappropriate for a health assessment.
The final version of the PHA includes a more complete discussion of these results. The median has been added to this section.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - Surface Soil - The fill material was not placed on the river bank, but along a roadside near the Raquette River. The material is not located at the rivers edge or within a floodplain or wetland area but rather on a flat lying area above the river bank itself.
This information has been included in the final version of the PHA.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - Surface Water and Sediments - The PHA gives PCB concentrations in St. Lawrence River surface water and sediments. GM is not the sole contributor. The PHA needs to state that sources other than GM are present.
This has been added to the final version of the PHA.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - Surface Water and Sediments - References for sample data should be included with report references (i.e., W. Stone and SRMT Sample Data).
Agreed. The appropriate references have been made.
Environmental Contamination and Other Hazards section, subsection B. Off-Site Contamination - Biota. The HRA should be included in references to the report.
This reference has been added to the final version of the PHA.
Environmental Contamination and Other Hazards section, subsection C. Quality Assurance and Quality Control. The PHA states that data collected during the RI/FS were subjected to rigorous review, but that data collected by "other parties relating to this site have not been through the same review process." What does this mean? Who are the "other parties" and have their data been subjected to any significant quality review? QA/QC data from the HRA, W. Stone and others, which did not go through rigorous data review should not be included in the report.
This statement refers to the fact that the other data were not subject to the same review process, but had adequate QA/QC. This statement has been clarified in the final document.
Pathways Analyses section. The reference to the residential water samples which were treated to remove sulfur should include a discussion of the results. What conclusions can be made here?
The discussion regarding sulfur compounds interfering with PCB analysis is found in the Quality Assurance/Quality Control section.
Pathways Analyses section. The PHA states "Because of past disposal practices and the subsequent release of PCBs, groundwater, surface and subsurface soils, river sediments, surface water, fish and wildlife, and air are contaminated." NYS DOH should define "contaminated" and discuss the available data which leads them to this conclusion.
Media are considered contaminated if the PCB levels are above background.
Pathways Analyses section. The paragraph which attempts to define "exposure pathway" and set criteria for selecting significant pathways to be addressed in subsequent sections is poorly written. The sentence which states that "An exposure pathway is categorized as a completed or potential exposure pathway if the exposure pathway cannot be eliminated." should itself be eliminated.
The explanation of exposure pathways has been revised for clarification purposes.
Pathways Analyses section, subsection A. Environmental Pathways - Surface Water and Sediment. It should again be noted in the assessment that the GM facility is not the sole contributor of PCBs to the River. The St. Lawrence River drains Lake Ontario. In addition to PCBs from the lake, other upstream sources exist. Samples of surface water collected upstream of the GM facility have been found to contain PCBs at concentrations greater than those found in downstream samples.
See Response #4 under Comments from Area Industry.
Pathways Analyses section, subsection A. Environmental Pathways - Surface Water and Sediment. The PHA states that "The PCB-contaminated sediments in the St. Lawrence - Raquette Rivers and the unnamed tributary continue to contaminate the water column through the desorption of PCBs from the sediment into the surface water." This statement is not substantiated with facts anywhere in this health assessment. The authors should either provide data which support this statement or remove it from the health assessment.
This statement has been removed from the final version of the PHA.
Pathways Analyses section, subsection A. Environmental Pathways - Surface Water and Sediment. The PHA states that "Levels decrease as distance from the plant increases, but a large area of the St. Lawrence River is contaminated." Once again, the authors are making conclusive statements without providing a factual basis. Sediments containing PCBs have been identified in the embayment of the St. Lawrence River immediately adjacent to the GM facility. While the overall embayment covers approximately 10 acres, many sediment samples from within this area and downstream areas did not contain PCBs. Furthermore, "low levels" of PCBs discussed in this section could be from sources other than GM. The PHA should acknowledge these facts.
The PHA has been amended to reflect this comment.
Pathways Analyses section, subsection A. Environmental Pathways - Consumable Biota. The PHA states that samples of fish and wildlife have been analyzed, but were not available for evaluation in the health assessment. The wildlife study was completed in October, 1992, the fish study was completed in April 1990, this health assessment should not be issued without reviewing these data (see "General Comments").
Pathways Analyses section, subsection A. Environmental Pathways - Air. The PHA states that a single air sample taken in a residential area of the St. Regis reservation had a PCB concentration in excess of the NIOSH recommended exposure limits (REL) of 1 microgram per cubic meter. Comparison of a single air sample (with no details presented in the health assessment) with a NIOSH value is inappropriate and unscientific. The PHA should present the air monitoring data.
Pathways Analyses section, subsection B. Completed Human Exposure Pathways. Under "(8)", the PHA asserts that residents are exposed to PCBs via air. This assertion is apparently supported by only one sample of a series of air samples taken (no data discussed or shown). The conclusion that a complete air pathway exists on the basis of a single barely detectable reading is unreasonable. The PHA needs to present and discuss the available air monitoring data, rather than making assertions on the basis of an isolated sample.
Additional air data are presented and discussed in this final public health assessment.
Pathways Analyses section, subsection A. Environmental Pathways - Air. Concern associated with exposure to emissions from the Reynolds facility is inappropriate for assessment of the GM site.
The ATSDR PHA Guidance Manual instructs health assessors to review data such as that available through the Toxic Chemical Release Inventory (TRI) to identify other potential exposures that may occur to the population being studied.
Pathways Analyses section, subsection B. Completed Human Exposure Pathways. The PHA asserts that Mohawks consume fish and wildlife from "areas with high levels of contaminants". This assertion is not substantiated with verified data which document the purported consumption. Furthermore, exposure results from the breast milk study do not confirm increased exposure relative to other studies (see comment under "Page 6: Health Outcome Data").
This statement is based on observations of the staff of the Mohawk Environmental Health Services unit. The Environmental Health Services Unit personnel are Native Americans that live on the reservation and are in the best position to make such observations. This is not inconsistent with the breastmilk study results in that women of childbearing age may be less likely to eat fish from areas with high levels of contaminants. However, this section has been clarified.
Pathways Analyses section, subsection B. Completed Human Exposure Pathways. The PHA states "Direct contact with PCB-contaminated soils within the facility fence and in areas on-site not within the fenced area is occurring." Who are the receptors for this pathway? Workers? Trespassers? Worker exposures should be minimal due to proper health and safety training and precautions. Trespasser exposure should be insignificant due to fencing, monitoring and other security precautions.
Workers are the receptors for areas on-site within the fence. Trespassers are the receptors for areas on-site outside the fence.
Pathways Analyses section, subsection B. Completed Human Exposure Pathways. The area of fill soils near the Raquette River, although not located within a secured fence area is located along a road which has been blocked to reduce access to the area and is not in an area of likely activity. The soils are also not near the river bank, where fishing could possibly take place. No evidence of direct contact or ingestion of these soils exists. Any reference to such should be deleted from the report.
These measures do not preclude the potential for trespassers to come into contact with these contaminated surface soils. The referenced statement has been amended to reflect the concern raised in this comment.
Pathways Analyses section, subsection B. Completed Human Exposure Pathways. The reference to infants in the 7th paragraph is unfounded. No information exists that infants have played, or are likely to play in these areas prior to remediation.
As there are no restrictions of access to residential areas on the reservation, the potential exists for infants to play in these areas.
Pathways Analyses section, subsection C. Potential Human Exposure Pathways. If "the treatment process at the St. Regis Water Treatment Plant is efficient at removing contaminants" and samples collected to date show extremely low concentrations of contaminants, then ingestion of contaminated water should not be evaluated. Including ingestion of contaminated water as an exposure pathway because exposures may occur during potential remedial activities is not appropriate in an assessment of health under baseline conditions.
This PHA considers all exposures, including potential exposures, related to remediation.
Public Health Implications section. If the health assessment for this site has not been completed, then any statements concerning potential health risks made in this PHA are qualitative at best. Since no calculations are presented, descriptive terms such as "significant", "moderate", or "high" risks have no basis in fact and should not be made here.
As indicated in Response #1, the NYS DOH completed quantitative risk calculations for the General Motors/Central Foundry Division site. These calculations were then used to qualitatively rank excess lifetime cancer risks and noncarcinogenic health risks according to the procedure described in detail in Appendix D of the revised PHA for this site.
Public Health Implications section, subsection A. Toxicological Evaluation. This paragraph discusses the 1968 Yusho incident in Japan to support the assertion that PCBs cause numerous adverse health effects in both adults and their newborn babies. This conflicts with a previous ATSDR position that the effects observed following the Yusho incident were most likely due to the PCDF contaminants which were present in the oil. In their Toxicological Profile on PCBs (February, 1992 draft), the NYS DOH/ATSDR state: "Although there is a historical linkage between Yusho and PCBs, and some regulatory documents ascribe health effects from these incidents to PCBs, effects from the incidents are not reviewed in this profile because the effects cannot be attributed specifically to the Kanechlors." NYS DOH/ATSDR state further: "There appears to be general agreement that the PCDF contaminants, particularly the more potent isomers, contributed significantly to the health effects observed in the Yusho and Yu Cheng patients." To be consistent with ATSDR's official position on the Yusho incident, the authors of this health assessment should delete all references to the Yusho incident as an illustration of adverse health effects caused by human exposure to PCBs.
The current scientific consensus is that the effects observed after the Yusho and Yu-Cheng incidences were primarily caused by the breakdown products (PCDFs) of heated PCBs. However, recent work by active investigators of the Yu-Cheng incident suggest that it is premature to totally exclude the possibility that some PCBs may have contributed to some of the effects (Chen et al., 1992; Yu et al., 1991). Moreover, ATSDR (1995) states that the Yusho and Yu-Cheng incidents are indirectly relevant to PCB toxicity because they show the sensitivity of humans to dioxin-like compounds such as PCDFs and some individual PCB congeners. The discussion of Yusho and Yu-Cheng has been retained in the text, but revised to reflect these developments.
Public Health Implications section, subsection A. Toxicological Evaluation. NYS DOH/ATSDR state: "A study conducted by the New York State Health Department of women exposed to PCBs in their workplace found evidence of a link between this exposure to PCBs and lower birthweight children (Taylor, 1984)." This statement misrepresents the findings of Taylor et al.. Taylor found that the small difference in birth weight between low and high exposed groups was due to a difference in gestational age. When adjusted for gestational age, the difference in birth weight between the two groups is negligible (3.330 + 0.490 vs. 3.272 + 0.476 kg). Furthermore, the fact that average gestational age for babies born to both "high exposure" (274.3 + 20.1 days) and "low exposure" mothers (280.9 + 14.2 days) is greater than that of a full-term infant (270 days, give or take 2 weeks) is meaningful. The observed gestational ages suggest that, if anything, babies born to mothers in the "high exposure" group were less overdue on the average than babies born to the "low exposure" group. This is hardly an adverse health effect. Furthermore, the design of the Taylor study did not account for other risk factors such as tobacco use, underlying medical conditions, maternal height and previous history of low birth weight which could influence gestational age and birth weight. Taylor et al. (1984) concludes: "The small number of observations, lack of information on important influencing factors, binary exposure measure, and uncertainty in assigning biological significance to birthweight and gestational age differences of this magnitude mandate that such a conclusion be considered tentative."
In itself, Taylor et al. (1984) only suggests that occupational exposure to PCB may reduce birthweight. However, Taylor et al. (1989), which also should have been cited and was included in the references, showed that after adjustment for variables known to effect birthweight, there was a significant relationship between increased estimated PCB serum level and decreased birthweight and gestational age, and that the decrease in birthweight was at least partially related to a shortened gestational age. Reference to Taylor et al. (1989) has been added to the text.
Public Health Implications section, subsection A. Toxicological Evaluation. The authors state: "Recent evidence suggests that some behavioral anomalies and low birthweights occurred in children whose mothers had a relatively high exposure to PCBs from consumption of contaminated fish over a period of several years (Fein, Jacobson)." This statement is misleading. There are many flaws in the studies of Fein and Jacobson which limit their usefulness in assessing hypothetical PCB toxicity. A major flaw in these studies is that they fail to account for the potential effects of DDT, mercury and other chemicals which are present in Lake Michigan fish. They also fail to quantify PCB exposure among the "exposed" and "nonexposed" groups, and fail to take into account alcohol use, tobacco use and nutritional status of the women participating in the study. All of these considerations could impact perceived differences in behavioral or physiological indicators of human development. NYS DOH/ATSDR should either discuss the limitations of these studies or eliminate them from the text.
These studies, like many epidemiological studies, examined the correlations between PCB exposure and effects and can not prove that PCBs were responsible for the observed effects. This is a limitation of all epidemiological studies. The work by Jacobson, Fein, Rogan and Gladen were recently discussed at a major EPA-sponsored meeting of experts on the developmental neurotoxic effects associated with PCB exposure. The studies' flaws, including those cited above, were discussed and evaluated. The summary conclusion of the workgroup responsible for evaluating the human studies was:
In terms of qualitative risk assessment, it was concluded that the studies
were of sound quality and both were well-designed and well executed.
The fact that two independent, well-designed studies showed that
low-level exposure to PCBs was associated with psychomotor and
cognitive deficits in humans suggests that prenatal PCB exposure is
of public health concern. However, the epidemiologic evidence
cannot stand alone, and must be appraised in terms of its consistency
with the results of animal studies and with the results of future
Animal experiments have clearly shown that PCBs cause developmental and neurological effects in rodents and rhesus monkeys and that prenatal (before birth) PCB exposure produces lower birthweights and neurological effects in monkeys (ATSDR, 1995; Golub et al., 1991). These effects are similar to those observed in the children studied by Jacobson, Rogan and colleagues. Thus, the human and animal data are consistent. The text has been revised to improve clarity and avoid misunderstanding of the studies.
Public Health Implications section, subsection A. Toxicological Evaluation. The authors assert that Rogan and Gladen found behavioral anomalies among children born to PCB-exposed mothers, but fail to discuss any details. This is misleading. In the initial study conducted by Rogan et al. in 1986, the only detectable difference between exposed and nonexposed groups was the observation of hypotonicity and hyperreflexia among the exposed group. In this study, Rogan et al. concluded that these differences may not be important if they did not persist. A follow-up of the children assessed in the 1986 study did in fact show that no important differences between the two groups were apparent. There were no differences between exposed groups in McCarthy scores of development at 3, 4 or 5 years of age. Furthermore, there was no statistically significant relationship between PCB exposure and grades obtained at school. Once again, NYS DOH/ATSDR should present more details of these studies or eliminate them from discussion.
Gladen et al. (1988, 1991) also found a significant correlation between increased PCB exposure and decreased scores on the psychomotor development index of the Bayley test of child development when the children were tested at 6 months, 12 months, and 24 months. A similar, but nonsignificant correlation was found at 18 months. Thus, neurological effects lasted for about two years after prenatal exposure ceased. This persistence beyond the neonatal period raises concerns about the human developmental risks of low-level perinatal PCB exposure. The text has been revised to improve clarity and avoid misunderstanding of the studies.
Public Health Implications section, subsection A. Toxicological Evaluation. NYS DOH/ATSDR assert that PCBs are known to cause a variety of health effects in laboratory animals "at exposures generally less than one order of magnitude greater than exposure from consumption of PCB-contaminated fish from the St. Lawrence River and tributaries near the site." This assertion is riddled with problems, and is not supported by any rationale or calculations presented in the health assessment. First, regulatory agencies such as the United States Environmental Protection Agency have refrained from deriving an RfD (noncarcinogenic Reference Dose) for PCBs on the basis of these effects due to the great uncertainty associated with which effects are occurring, the biological importance of some of the effects, and the doses at which these effects might actually be occurring. What values are the authors of this document choosing as the basis for their comparison? Second, the health assessment fails to present any information on fish consumption. What is the justification for assuming a 0.5 lb per week consumption rate? What PCB concentrations in fish are used to calculate human exposure? What PCB intake was calculated? Third, one cannot simply compare human exposure directly with animal exposure data to reach the conclusion that adverse health effects in humans are likely to occur. NYS DOH/ATSDR cannot make the assertation that human exposure is greater than exposures which may cause adverse effects in animals without presenting all of the data, assumptions and calculations which lead them to this conclusion. NYS DOH/ATSDR should either present this information or eliminate the statement from the report.
A meal size of one-half pound (or 227 grams) was used to derive fish consumption rates for each adult meal. This meal size is used in determining NYS DOH fish consumption advisories and corresponds to the 90th percentile fish meal size for U.S. population fish consumers, as found in a U.S. Department of Agriculture dietary survey (Pao et al., 1982). Based on data from a 1979-1980 Mohawk dietary survey and assuming a meal size of 227 grams, the average fish consumption rate (total fish, all individuals) for Mohawk Indian adults was about 32 grams per day or approximately one-half pound per week. Based on a mean PCB concentration of 2.4 ppm in fish (see Table 6) and a fish consumption rate of one-half pound per week (or about 32 grams per day) for adults weighing 70 kilograms, PCB intakes of greater than 1 x 10-3 mg/kg/day were calculated.
To estimate the risks of noncarcinogenic effects, PCB intakes were compared to an ATSDR-derived minimal risk level 2 x 10-5 mg/kg/day for the PCB commercial mixture, Aroclor 1254 (ATSDR, 1995). This toxicity value is the same as the US EPA's recently derived oral reference dose for exposure to this Aroclor (US EPA, 1995a). An oral reference dose or minimal risk level is an estimate (with an uncertainty spanning perhaps an order of magnitude or greater) of a daily exposure level for the human population that is likely to be without any appreciable risk of adverse health effects. If exposure to a contaminant exceeds the oral reference dose or minimal risk level, there may be concern for potential noncarcinogenic effects because the margin of protection is less than that afforded by the reference dose or minimal risk level. As a rule, the greater the decrease in the margin of protection, the greater the level of concern. In general, when the estimated intake of a contaminant is greater than about 10 times its reference dose or minimal risk level, the risk is characterized as high (a discussion of procedures for evaluating potential health risks has been added to the revised PHA as Appendix D). The estimated potential intake for PCBs from fish consumption at the General Motors/Central Foundry Division site that could pose a high risk of noncarcinogenic health effects is about 60 times greater than its minimal risk level and only about 4 times less (that is, less than one order of magnitude) than a dose that caused adverse effects in animals.
Public Health Implications section, subsection A. Toxicological Evaluation. NYS DOH/ATSDR assert that people who regularly consume one-half pound of fish per week from the St. Lawrence River and tributaries near the site have a "high increased risk of developing cancer over their lifetimes." NYS DOH/ATSDR cannot make such a statement without supporting evidence. This statement should either be substantiated with exposure data, reference toxicological data and risk calculations, or should be eliminated from the assessment. Furthermore, suggesting that any PCB exposure leads to a "high increased risk of developing cancer" is inconsistent with ATSDR's position (and the position of the scientific community as a whole) that there is insufficient evidence to conclude that PCBs cause cancer in humans.
PCBs cause cancer in laboratory animals exposed to much higher doses than the levels in fish from the St. Lawrence River and tributaries. Chemicals that cause cancer in laboratory animals exposed to high levels over their lifetimes may also increase the risk of cancer in humans who are exposed to lower levels over long periods of time. Although it is not known whether PCBs cause cancer in humans, the results from animal studies are sufficient to conclude that these chemicals are potential human carcinogens. For this analysis, the increased cancer risk was estimated using information on exposure levels of PCBs in fish (see Response #60) and interpreting them using the cancer potency estimate of 7.7 (mg/kg/day)-1 derived for PCBs by the US EPA (1995c). This increased cancer risk was estimated to be eight per thousand for persons eating 32 grams per day of fish contaminated with 2.4 ppm of PCBs. An estimated increased cancer risk of eight per thousand is considered high (see Appendix D) and led to the conclusion that increased cancer risk from ingestion of PCB-contaminated fish is high.
Public Health Implications section, subsection A. Toxicological Evaluation. Conclusions from the HRA are made, but throughout the report the HRA is discussed as unreleased. The last sentence is very speculative.
The HRA entitled: Health Risk Assessment for the Akwesasne Mohawk Population from Exposure to Chemical Contaminants in Fish and Wildlife from the St. Lawrence River Drainage on Lands of the Mohawk Nation at Akwesasne and Near the General Motors Corporation Central Foundry Division Facility at Massena, New York has been released by the NYS DOH. This HRA presents calculated noncarcinogenic and carcinogenic health risks for wildlife consumption that indicate that these risks are high, particularly for duck consumption. This conclusion, therefore, supports the statement in this PHA that "...the data suggest that the risks of noncarcinogenic and carcinogenic effects could be high" (in regard to ingestion of PCB-contaminated wildlife).
Public Health Implications section, subsection A. Toxicological Evaluation. NYS DOH/ATSDR state: "In addition, PCBs produce a variety of noncarcinogenic toxicities (previously discussed) at exposures several orders of magnitude greater than exposure from off-site ambient air. Although risks of noncarcinogenic health effects from these exposures aren't completely understood, the existing data suggest that they would be low." This statement should be removed from the health assessment for several reasons. First, one cannot comment on the level of exposure without calculating exposure from reliable and representative air monitoring data. Such data and calculations are not presented anywhere in the health assessment. Second, the animal studies which address noncarcinogenic health effects are not sufficient to determine what effects might actually occur, and the doses at which they might occur. Without adequate animal dose-response data, it is impossible to discern a threshold for noncarcinogenic toxicity in animals. This is why regulatory agencies such as EPA have not been able to decide upon a noncarcinogenic reference dose for PCBs other than Aroclor 1016. Third, one cannot directly compare human exposure with animal exposure to reach conclusions about potential adverse health effects. Drawing conclusions about human risks based on the uncertain animal PCB exposure database is even more unscientific. Fourth, there is no evidence to support "chronic exposures".
- Quantitative risk calculations to determine the level of risk based on very limited air monitoring data were made according to the procedures discussed in Appendix D, which has been added to the PHA. Additional language has been added to clarify these statements.
- Animal experiments have clearly shown that PCBs can damage many organs and organ systems, including the skin, liver, blood and the nervous, reproductive and immune systems (ATSDR, 1993). Overall, these data are sufficient to establish dose-response relationships between PCB exposures in animals and toxic effects, including the identification of thresholds below which no effects were observed. Although there are some gaps in the data on PCB toxicity, these gaps are not large enough to preclude use of the data to evaluate the potential human risks of PCB exposure. Risk assessment methods include factors to compensate for such gaps, and such factors have been used by the U.S. EPA (1995a,b) and ATSDR (1995) to establish reference doses or minimal risk levels for evaluating the risks of PCB exposure. The ATSDR-derived minimal risk level of 2 x 10-5 mg/kg/day (ATSDR, 1995) for Aroclor 1254 was used to determine the noncarcinogenic risk from exposure to PCBs in ambient air. This risk was determined to be "minimal", according to the procedures discussed in Appendix D and not "low" as previously indicated. The text of the PHA has been revised accordingly.
- A basic fundamental principle of risk assessment is that effects of chemicals observed in animals, if properly qualified, are applicable to humans. This premise also underlies all of medicine and pharmacology. Many medicines, for example, are first tested on animals to see what are the beneficial effects and the toxic, side-effects. The results of animal studies are not, usually, directly applied to humans because of the biological differences between animals and humans. However, scientists have developed successful ways of extrapolating the animal results to humans and have used those methods to develop doses that can be safely tested on humans. This approach has lead to the development of many effective human medicines.
- NYS DOH has revised the toxicological evaluation section of this public health assessment. The available information indicates that, except during remedial activities, ambient air levels in residential areas are likely to be less than 20 ng/m3. If chronic exposure to 20 ng/m3 of PCBs occurred, it would pose a low increased cancer risk over a lifetime of exposure.
In risk assessment, data on humans are often absent or limited, and animal data must be used to evaluate the potential health effects of chemical exposures. Risk assessment methods also use factors to compensate for interspecies difference between humans and animals. These factors can be conservative (more likely to overestimate human risk than to underestimate it) when, as is the case for most chemicals, we know little about the relative differences between humans and animals in their response to chemical exposures.
In particular, the available information supports the use of animal data to evaluate the potential human health risks associated with PCB exposures. Animal studies have shown that the most sensitive responses to PCBs are developmental and neurological effects in rodents and rhesus monkeys, and that prenatal (before birth) PCB exposure causes lower birthweights and neurological effects in monkeys (ATSDR, 1995; Golub et al., 1991). These effects are similar to those observed in epidemiology studies of environmentally exposed women and their children (see Response #59). This consistency supports using the results of animal experiments, properly qualified, to evaluate the noncarcinogenic risks of human PCB exposures.
Public Health Implications section, subsection A. Toxicological Evaluation. NYS DOH/ATSDR assert that workers who might be chronically exposed to soils and sediments on-site would have a "high increased cancer risk". This statement should be removed for the following reasons: 1) Chronic worker exposure is nonexistent due to training and health and safety precautions; 2) one cannot comment on the level of risk (high, low or otherwise) without a quantitative assessment of exposure and risk calculations; these are not provided anywhere in the health assessment; 3) ATSDR as well as other agencies and the scientific community have concluded that there is insufficient evidence to conclude that PCBs cause cancer in humans; 4) there is too much uncertainty in the animal database to make meaningful conclusions about the capacity for PCBs to cause adverse noncarcinogenic health effects in humans.
In regard to the first issue, the NYS DOH cannot totally dismiss the possibility that some workers, irrespective of how much training they received in safety precautions, may have become exposed to on-site soils and sediments. Accordingly, increased cancer risks to workers on-site were determined by the NYS DOH (see Appendix D for procedures). Due to capping of the on-site soil in 1988, such exposures are now unlikely. The text of the PHA has been revised to indicate that the potential for further exposure by workers is now significantly reduced. In regard to the remaining three issues of Comment #65, the second, third and fourth issues have been addressed in Response #1, #61 and #63, respectively.
Public Health Implications section, subsection A. Toxicological Evaluation. NYS DOH/ATSDR assert that "high risks" are associated with potential exposure to PCBs in the Raquette River fill area. This statement should be removed or modified for reasons 2 through 4 in the previous comment.
Furthermore, the road to this area of concern is blocked. The area is in a remote location and distant from the SRMT reservation property. The likelihood of anyone playing in the area is low due to its location and the existence of dense brush throughout the area. Creation of a fence may likely draw interest to an area unnoticed before. No footpaths exist in this area.
NYS DOH's justification for determining noncarcinogenic and carcinogenic risks to PCBs in various environmental media which have been affected by the General Motors/Central Foundry Division site has already been presented (see Response #s 2, 15, 61 and 63). In regard to exposure to PCBs in the Raquette River fill area, the text of the PHA has been revised to indicate that potential exposure to PCBs in these soils could pose a moderate risk of carcinogenic effects. This estimated cancer risk was based on only one reported level of PCBs (i.e., the highest level detected) and therefore may not be representative of the risk from exposure to these soils. Additional language has been added to clarify this in the final document. Finally, the NYS DOH cannot dismiss the possibility that trespassing may occur in this area even though it is apparently not readily accessible.
Public Health Implications section, subsection A. Toxicological Evaluation - Organic Compounds. The health assessment states: "The health risks of exposure to PCBs have already been discussed (see #1 above)." The discussion of health effects to which the authors refer is limited, inaccurate and incomplete.
The health assessment also states: "Chronic exposure to the levels of PCBs found in on-site groundwater would pose a very high increased cancer risk over a lifetime of exposure." This statement, and others like it (e.g., end of 5th paragraph on p. 16 and 1st partial paragraph on p. 17) which assert risks associated with on-site groundwater use, should be removed from the health assessment. There is no use of groundwater on-site, nor is there any plausible chance of groundwater use in the future. One cannot have risks without a complete pathway. Moreover, PCB levels were found in groundwater in only two areas (the Industrial Lagoon area, at monitoring wells 24 A/B, and 14 A/B, and the Industrial Landfill at Monitoring Well 16 A/B) and PCBs do not migrate significantly through groundwater. The likelihood of a drinking water well being installed at either of these locations is very small.
Although there is no apparent use of on-site groundwater, the NYS DOH cannot rule out the possibility that there is a potential for ingestion, dermal and inhalation exposure to contaminants from contaminated on-site groundwater by plume migration. Therefore, the potential exists for adverse health risks.
Public Health Implications section, subsection B. Health Outcome Data Evaluation. NYS DOH/ATSDR state that Selikoff and Hammond (1986) found a direct relationship between blood levels of PCBs and mercury, and the amount of fish eaten by Akwesasne residents, then make a statement about a correlation between mercury levels and fish consumption. What does this have to do with potential adverse health effects associated with human exposure to PCB? The study failed to demonstrate statistically meaningful correlation between serum PCB concentration and fish consumption. No adverse health effects were correlated with serum or fat PCB concentrations. This section should either be expanded to give a more complete discussion of the study or should be eliminated from the health assessment.
The NYS DOH neither stated nor implied in the PHA that the Mt. Sinai study demonstrated a statistically significant correlation between serum PCB concentration and fish consumption. Instead, the NYS DOH stated that the Mt. Sinai study concluded that there was a direct relationship between blood level of PCB and the amount of fish eaten by the Akwesasne residents. This statement was based on the study's investigators noting in their report (Selikoff et al., 1986) "a trend of increasing average serum PCB concentration with increasing fish consumption is evident." The NYS DOH also neither stated nor implied that this study showed that adverse health effects were correlated with serum or fat PCB concentrations.
The purpose of the Health Outcome Data Evaluation section of the PHA is to report and evaluate the results of any epidemiologic analysis of the residents living in the vicinity of the site. The Mt. Sinai study reported that fish consumption was significantly associated with blood and hair mercury levels. Most mercury accumulated in fish is methylmercury. This form is one of the most toxic forms of mercury and is readily absorbed in humans. These findings were then included in the Health Outcome Data Evaluation section of this PHA.
Public Health Implications section, subsection C. Community Health Concerns Evaluation. Statements about the "once lucrative commercial fishery and guided fishing trip businesses" should be eliminated. This is irrelevant to the health assessment document which does not include an evaluation of potential economic impacts.
Public Health Implications section. Data are not provided to corroborate that historically commercial fishery and guided fishing businesses were lucrative enterprises.
This is a concern to the population being assessed; however, the words "once lucrative" have been removed.
Conclusions section. Under "(1)" NYS DOH/ATSDR state that "Based on the information reviewed, this site is a public health hazard because of past and continued human exposures to hazardous substances at levels which may adversely affect their health." Once again, NYS DOH/ATSDR has not provided the references, data discussion, assumptions or calculations to support this conclusion.
This conclusion category is based on criteria created by ATSDR. This category is used for sites that pose a public health hazard as the result of long-term exposures to hazardous substances (ATSDR, 1992, p. 8-3). Also see response to comment 1 (area industries).
Conclusions section. NYS DOH/ATSDR state "...the Health Risk Assessment being conducted by GM...". The HRA is not being conducted by GM, but is partially funded by GM.
This statement has been changed in the final PHA.
Public Health Action Plan section, Public Health Actions Planned. Individuals or groups referenced as the "local medical community" and/or "public health professionals" should be identified. GM believes that much of the local health and medical "community", specifically the SRMT Health Services Staff has been involved in various studies related to this site, and are knowledgeable of the potential health effects associated with PCBs.
We consider the education of the local health community to be an on-going process. As with most areas of medicine, new information and results of studies continue to become available, and NYS DOH wants to ensure that local medical community is aware of all information currently available. These statements have been clarified in the final version.
Reference section. No references have been included for any site specific studies by NYS DEC or NYS DOH but their data is used throughout the report.
Mt. Sinai Study reference does not include all pages relevant to the GM site which should be referenced (pp. 127-154).
The appropriate references have been included in the final version of the PHA.
Figure 2 - The GM Foundry Outfall is incorrectly referenced. The outfall is no longer a permitted discharge for GM and has been sealed since 1988. The outfall serves only for drainage associated with runoff from State Route 37.
The map also inaccurately illustrates the GM property line. The true property line has no bends. Also, the plant perimeter fence extends along the northern side of the plant property, completely enclosing the active portion of the site.
This figure has been changed in the final PHA.
- Recommendations section. Again, what evidence exists that there is a lack of understanding among the local health community? GM has previously provided funding to the SRMT, and the local health community is involved in many of the studies related to this site.
- Recommendations section. Under (4) NYS DOH/ATSDR is suggesting that site and non-site-related emissions be evaluated. GM believes that the air emissions monitoring associated with Recommendation 2 (i.e., during site remedial activities) will adequately address site related emissions. Non-site related emissions should not be addressed in this NYS DOH/ATSDR document.
- Under (5) NYS DOH/ATSDR indicates that the TRI data for the site does not include styrene emissions. This topic should be addressed separately, and is not considered appropriate within the context of this Report.
- See response to #48 (comments from Area Industries).
- See response to #49 (comments from Area Industries).
- Communication with the public regarding health advisories is considered to be an ongoing process.
Tables 1, 1A and 2 - The "comparison values" listed in these tables are not defined anywhere in the document, but are obviously used by NYS DOH/ATSDR to reach conclusions about risks. These values are meaningless because the reader has no idea where these numbers come from or how they were derived. These values need to be defined and discussed in the text of the health assessment.
A PHA comparison value is a contaminant level in water, soil or air that is not likely to cause health effects given the potential for exposure. Therefore, comparison values are one of the factors used to evaluate if the contaminants at a site are likely to pose a health threat. Each environmental medium (air, soil, water) has its own comparison value for a contaminant. If a contaminant concentration is above its comparison value, the contaminant is evaluated further to determine if exposure is of public health significance. A general discussion of comparison values and the selection of contaminants for further evaluation has been provided at the beginning of the section on "Environmental Contamination and Other Hazards" and in Appendix D of this PHA for the General Motors/Central Foundry Division site. Exposure scenarios to determine comparison values for contaminants associated with each environmental medium are described in the footnotes of revised Tables 1-3 of the PHA.
Table 3 - Data from this table includes multiple sources; all data sources should be identified. Data from different sources, subjected to different levels of data review and quality should not be combined. Data sources should be included in references.
All data sources cited are included in the references and added to the table. Data from different sources were combined to describe off-site contamination in one table.
Table 4 - The data source is not identified or included in references.
This has been referenced in the final version of the PHA.
Background section. The PHA states that the last site visit on the GM property was December 6, 1988. The site should have been visited more recently than 4.5 years prior to the preparation of the PHA.
The document has been updated to indicate the most recent site visits.
Background section. Provide additional data to support the assertion that the Grasse River is used for swimming.
These statements are based on observations made by both personnel from the Mohawk Environmental Health Services Unit and the Massena District Office of the NYS DOH.
The section on Community Health Concerns does not adequately differentiate between the St. Lawrence, Grasse, and Raquette Rivers with regard to the site.
The community is concerned about contamination in all of the rivers.
Environmental Contamination and Other Hazards section. What is the definition of "near surface soil"?
The soil samples were collected at a depth of 0-2 feet. NYS DOH/ATSDR considers 0-3" or 0-6" to be surface soil, 0-2' would be considered "near surface". This information was added to the PHA.
Environmental Contamination and Other Hazards section. What is the significance of 2 mg/kg of PCBs in sediment?
The 2 mg/kg of PCB level is the proposed clean-up level for sediments. This information was added to the PHA.
Pathways Analyses section. The information suggesting exposure to PCBs via the air pathway is weak and could be unduly alarming to nearby residents.
The air pathway is not fully defined because of the lack of adequate data. We do not consider this to be alarming to the nearby residents because it is listed as a community health concern that needs to be addressed by an air study.
Pathways Analyses section. The statement that "a large area of the St. Lawrence River is contaminated" is subjective and potentially misleading.
This statement was amended.
Pathways Analyses section. The one positive detection of PCB contamination in well water used for drinking water was never confirmed. Therefore, it is inappropriate to refer to "significant positive results."
This sentence has been amended in the final PHA.
Public Health Implications section. Epidemiological data provided on exposure to PCBs may be unnecessarily alarming to local residents as it is anecdotal in presentation and does not relate to the nature or degree of potential exposure at the site.
There are three types of epidemiological studies discussed in the PHA: studies of (1) Yusho victims, (2) workers and (3) women in the general population. Studies of the Yusho victims have only indirect relevance to potential human exposures at the site, and the text has been revised (see Response 2 under summary of comments received from area industry). However, results of the other two types of studies are directly relevant because the exposures were via air or food, which are potential sources of exposure at the General Motors/Central Foundry Division site. Although the degree of potential exposure at the site has not been fully evaluated, the assumption that the results of the epidemiological studies of workers and women in the general population are applicable to the local population is reasonable.
Public Health Implications section. Data presented on wildlife is from a small sample size, and therefore, cannot be considered representative of wildlife in the vicinity of the site. While the document acknowledges that the database is limited, it should make clear that 1 to 3 animals are not an adequate sample size to represent potential PCB levels in wildlife. Data should be provided to substantiate that wildlife is a large portion of the Mohawk Indian diet.
This comment is really two comments:
- The PHA includes contamination data on a total of 16 mallard and black ducks, the species most frequently consumed by the Mohawks. This sample size is adequate to characterize contaminant levels in ducks consumed by the Mohawks.
- The second part of the comment implies that large amounts of a contaminated food must be eaten to produce a dose large enough to be a potential health concern. However, if a food is sufficiently contaminated, eating relatively small amounts may significantly increase health risks. The Health Risk Assessment established Mohawk consumption rates of waterfowl and other game, based on a dietary survey. Based on these consumption rates and contamination data, the Health Risk Assessment concluded that duck consumption is a significant source of PCB exposure for the 12% of the Mohawk population who consume ducks.
Public Health Implications section. The Mt. Sinai study is quoted with regard to the relationships of mercury to blood levels, but no data are provided for correlations for PCBs and blood levels. This data should be provided in the document, if it exists.
The Mt. Sinai study includes data on blood PCB levels determined for 47 adult residents of St. Regis Mohawk Indian Reservation. These data show that the average blood PCB levels in these subjects are comparable to the levels found in individuals without occupational or other identified PCB exposures. This information was added to the PHA.
Presentation of only maximum contaminant concentrations for soils and groundwater in summary data tables and text of the PHA is misleading because maximum sample levels can substantially exceed average or typical values.
We agree that ranges and median would provide a more complete picture. A qualifier has been added to the data presentation comments in the Environmental Contamination and other Hazards section.
The discussions of potential exposure of residents on the St. Regis Mohawk Indian Reservation to air emissions from the Reynolds plant do not include sufficient supporting documentation to allow adequate public comment. Further, discussion of the Reynolds Plant does not belong in this assessment of the GM facility.
We agree there is not sufficient supporting documentation because of data limitations. The PHA discusses potential public impacts on the population affected by the GM site. It discusses other potential environmental impacts on the population such as air emissions from other nearby industrial facilities. We believe Reynolds air emissions are in this category.
The identification of groundwater as a past potential human exposure pathway is not supported.
This pathway is identified as a potential pathway because we can't conclusively prove that the one set of positive results are wrong (see the explanation of potential exposure pathway under the Pathways Analyses section).
The discussion of potential risks associated with inorganic chemicals in groundwater on pp. 16 to 17 of the PHA should include an analysis of background levels.
The levels of inorganic chemicals in groundwater on-site are above the levels in the upgradient wells, which are considered background.
The discussion of the air pathway on page 12 of the PHA should note that the data suggesting the presence of site-related contaminants in air are very limited.
The information in the discussion of air sampling has been qualified.
The carcinogenic potential of PCBs should be evaluated on an Aroclor-specific basis. Not all PCBs exhibit the same toxic potential. Only PCB mixtures (Aroclors) with 60 percent chlorine have been shown to be carcinogenic in laboratory animals. Less chlorinated PCB mixtures are either not carcinogenic or are substantially less potent than the more highly chlorinated mixtures. It is important to identify which of the Aroclors are present at the site. Excess lifetime cancer risks associated with Aroclors having less than 60 percent chlorine, if any, are less than those predicted using the EPA estimate of carcinogenic potency, which is based on Aroclor 1260.
Most of the human exposure to PCBs comes from food, especially fish, meat and dairy products. Data on the total PCB concentration in the fish and wildlife (water fowl) collected near the GM Foundry show that most of the PCBs are molecules containing 4 chlorines or more (Skinner, 1992; Sloan and Jock, 1990). These molecules also make up most of the molecules found in the commercial PCB mixtures Aroclors 1248, 1254 and 1260, which contain 48%, 54% and 60% chlorine by weight, respectively (ATSDR, 1995). Monitoring data also show that most of the PCBs in human breastmilk are molecules with 4 or more chlorines (Fitzgerald et al., 1992). Data on the carcinogenic effects of the individual PCB molecules found in animals and humans are not available and cannot be used to directly evaluate the carcinogenic risks associated with human exposure to PCBs in fish and wildlife. However, there are data on the carcinogenic effects in animals of PCB mixtures containing the PCB molecules found in fish, wildlife and humans. These data have been consistently used by government agencies, including the US EPA, ATSDR and US FDA to evaluate the carcinogenic risks associated with PCB exposures. Estimates based on this approach may not be as accurate as those based on the toxicities of the individual molecules because the composition of the environmental mixtures of PCBs differ from the composition of occupational mixtures or mixtures tested on laboratory animals. This uncertainty cannot be avoided given the limited data on the toxicities of individual PCB molecules.
Studies on rats show that PCB mixtures containing 60% chlorine (Aroclor 1260 and Clophen A60) are clearly carcinogenic in rats (ATSDR, 1995; US EPA, 1995c). Studies on PCB mixtures containing 54% chlorine (Aroclor 1254) are suggestive of carcinogenic effects in both the liver and stomach of rats (NCI, 1978; Morgan et al., 1981; Ward, 1985). A PCB mixture containing 30% chlorine (Clophen A30) caused a carcinogenic effect in the liver of rats, although the effect was smaller than that observed in rats exposed to Clophen A60 (Schaeffer et al., 1984). Other PCB mixtures, including Aroclors 1232, 1016, 1242 and 1248 have not been adequately tested for their carcinogenic potency in rats. Thus, some general evidence suggests that highly chlorinated PCB mixtures are more potent inducers of liver tumors in rats than mixtures containing less chlorine. However, additional studies on all the Aroclors, preferably with the same strain of rats, are needed to confirm the relationship for rats and to determine the size of the potency differences.
Faced with these uncertainties, most public health agencies evaluate the carcinogenic potential of PCBs on a Aroclor-specific basis, usually using data on Aroclor 1260. When exposure is associated with eating fish and wildlife, which contain the PCB molecules that are relatively abundant in highly chlorinated mixtures (Aroclors 1248, 1254 and 1260), evaluating human risk based on Aroclor 1260 is reasonable. Aroclor 1260 is the most potent of the tested Aroclors and we don't know which PCB molecules in Aroclor 1260 are responsible for its potency and whether these molecules accumulate in humans environmentally exposed to PCBs. Many of the same PCB molecules are found in all three Aroclors (Hong et al., 1993).
Conclusions regarding the magnitude of risk posed by PCBs in the discussion of public health implications (pp. 13 to 17) are not sufficiently documented to allow public comment.
Page 15, section 6 - sample collected and analyzed by Reynolds Metals Company demonstrate no exceedances of residual fluoride particulate on vegetation.
So noted, although NYS DOH does not have these data.
James et al. (1993) reviewed the Yusho and other incidents, including occupational exposures to PCBs, and concluded that the only adverse effects from PCB are dermal.
James et al. (1993) summarized the major findings associated with occupational exposures to PCB. They examined the noncarcinogenic effects on the many organ or organ systems by reviewing clinical studies (which examine, for example, test results on liver, kidney and lung function) and mortality studies (which examine the causes of death) of PCB-workers. They concluded that the evidence suggests that the only adverse health effects attributable to occupational PCB exposures are dermal. There remains the possibility that other effects are associated with occupational PCB exposure (e.g., Lawton et al., 1985). In addition, conclusions based on the mortality data may be premature because many of the workers are still alive, and usually, conclusions based on small number of deaths are less reliable than conclusions based on larger number of deaths. Moreover, the review did not cover reproductive/developmental or carcinogenic effects of PCB exposures. The text has been revised to improve clarity and avoid misinterpretation of the occupational studies.
Public Health Implication section 6. Actual sampling results do not show exceedance of NYS DEC Ambient Air Guidelines.
So noted. This section discusses potential exposures based on some limited data. The evaluation is based on the toxic effects of PCBs, not on a comparison to ambient air guidelines.
Page 17, section C - There is no evidence to support the need for further air studies for Reynolds Metals Company.
There are little data available for air quality on the reservation especially for air contaminants other than PCBs. The community remains concerned about air emissions from nearby industry.
Air emissions from RMC: RMC is currently in compliance with all applicable air regulations - NYS DEC & EPA.
So noted, however, there is not mention in the document to the contrary.
Page 9, C - QA/QC: Data not subjected to a rigorous review process should not be included for the purposes of this study.
All data included in the report underwent adequate QA/QC.
Page 11, Air: RMC data does not support DEC's contention that any vegetation samples have exceeded standards for residual fluoride.
This statement was removed from the final health assessment.
Recent analysis performed by the states of Michigan and Connecticut found no significant medical problems in studies of individuals consuming fish with PCBs.
We could not identify or find the exact reports to review, given the limited information provided.
On page eleven of the draft PHA, it is noted that although samples have been collected and analyzed, not all of the data are available for evaluation. How can the risk assessment be finalized, or even commented on, if the data are not available?
This statement in the health assessment was incorrect and has been removed.
PCB's are a group of molecules of which very little is known in terms of risk to the environment or human health. PCB's are not all alike. The PCB molecules in question here are designated with 4-digit numbers, of which the last two digits indicate chlorine content. 1260 PCB is a highly chlorinated biphenyl with 60% chlorine. This particular molecule (one of 209 PCB molecules) is a suspected carcinogen. Studies of its toxicity span the spectrum of effects ranging from nil to mild to serious. Negative effects in these studies are only observed after long exposure to maximum tolerated doses. Other PCB molecules, such as 1248 and 1232, may not be of any concern at all in terms of environmental health.
1260 PCB (which is Aroclor 1260) is not a single molecule but a mixture of 80 to 100 different individual PCB molecules. The toxic effects of Aroclor 1260 have been observed in many types of studies, not just after long exposure to maximum tolerated doses (ATSDR, 1995). The toxic effects of Aroclor 1260, as those of all chemicals, depend on the dose and the length of exposure. PCB 1248 and 1232 are not single molecules, but mixtures of different PCB molecules. Aroclors 1248 and 1232 have caused toxic effects in animals (ATSDR, 1995). The human health risks associated with environmental exposures to these mixtures would depend on exposure, and can be cause for concern if the exposure levels are high.
It is stated that "Whether or not PCB causes cancer in human is not known" but then a risk assessment based on "on results of animal studies" was done.
PCBs have caused cancer in laboratory animals. Although it is not known whether PCBs cause cancer in humans, agencies and other scientists (US EPA, FDA, ATSDR, IARC) assume that chemicals that cause cancer in animals have the potential to cause cancer in humans, unless there is evidence to the contrary. There is no such contrary evidence for PCBs; consequently, PCBs are considered potential human carcinogens and the results of animal studies were used to evaluate the human carcinogenic risks of PCB exposures. Additional discussion of the risk assessment methods for evaluating carcinogenic risk can be found in Appendix D.
The use of animal studies to estimate excess human cancer risk is under fire, and has been criticized by consultants and in articles appearing in Science magazine and the New York Times. These criticisms focus on whether the results at high dose animal experiments are relevant to evaluating the human risk at much lower environmental doses.
This is a controversial issue. However, animal studies currently are the best information we have for evaluating the human carcinogenic potential of most chemicals and the use of high-dose studies maximizes the likelihood that an animal study will detect a carcinogenic effect, if the chemical is actually a carcinogen. Thus, chemicals that cause cancer in animals are assumed, unless there is evidence to the contrary, to have the potential to cause cancer in humans. This is a conservative assumption to protect the public health. Additional discussion of the risk assessment methods for evaluating carcinogenic risk can be found in Appendix D.
The evidence on PCBs as an actual cause of cancer in humans is lacking even in highly exposed PCB workers.
The International Agency for Research on Cancer (IARC, 1987) has evaluated the evidence on the carcinogenicity of PCBs in humans and has classified the degree of evidence as limited. This means that a positive association has been observed between exposure to PCBs and human cancer for which a causal interpretation is considered by the IARC experts to be credible, but not proven, since chance, bias or confounding factors could not be ruled out with reasonable confidence. The U.S. EPA (1995c) has classified the human evidence on carcinogenicity as inadequate. This means that there are not enough data to make a decision on whether PCBs cause cancer in humans. These assessments differ, at least, in part because the available studies do not clearly show that PCBs are human carcinogens. Neither do the data show that they are not human carcinogens. The issue may be resolved when scientists get more data on the cause of death among workers occupationally exposed to PCBs. In the meantime, the NYS DOH assumes that PCBs are potential human carcinogens given that they are carcinogenic in animals (see two previous responses).
The James et al. (1993) article also stated that the two rice-oil poisoning incidents in Japan and Taiwan are now generally seen by investigators as the result of the ingestion of PCDFs, but it is stated that "it is not clear which of the contaminants (PCBs or other chemicals) caused the effects". The authors should not rely on the Yusho incident for information on PCBs.
Regarding the Fein, Jacobson studies and the Rogan, Gladen studies referenced regarding low birthweight, it is stated that "In both studies, the possibility that the women may have been exposed to other toxic chemical was not fully evaluated." According to the commenter, "This could quite clearly mean that PCBs are not the culprit."
If no adverse effects have been found in workers who have been in close and direct contact with concentrated PCBs, how does the DOH rationalize basing a risk assessment on a chemical in quantities of parts per million in water and in the soil in an area that is characterized as low density, residential farm lands?
Human effects reported after occupational exposures to PCBs include skin, eye and respiratory tract irritation and less frequently, effects on the liver and the nervous and digestive systems (ATSDR, 1995). The conclusion that PCBs caused these effects is only tentative because the workers were also exposed to other chemicals, including the contaminants of PCBs. However, the observed effects are similar to those induced in animals exposed to PCBs. In addition, the limited epidemiological data on the mortality of PCB workers are inadequate to conclude that there are no noncarcinogenic effects associated with occupational exposure to PCBs.
Epidemiology studies of environmentally exposed women and their children show a relationship between increased exposure to PCBs and reproductive/developmental effects, specifically lower birthweight and neurobehavioral effects (ATSDR, 1995; US EPA, 1993). These studies, like many epidemiological studies, observed an association between PCB exposure and effects, but cannot prove that PCBs were the sole factor responsible for the observed effects. This is a limitation of all epidemiological studies. However, the observed effects are similar to those induced in animals exposed to PCBs (ATSDR, 1995; Golub et al., 1991; US EPA, 1993).
Collectively, these data indicate that there may be human health hazards associated with the very low PCB levels typical of environmental exposures.
Government has made major leaps of logic in declaring that what they believe in as a probable cancer causing agent such as 1260 should be considered a known cancer causing agent.
The NYS DOH has not suggested that Aroclor 1260 should be considered a known human cancer causing agent. Given the clear evidence that Aroclor 1260 causes cancer in rats and the absence of good data that it doesn't cause cancer in humans, it is biologically plausible and prudent to regard PCBs as if they are potential human carcinogens that present a carcinogenic risk to humans (IARC, 1987, US, EPA, 1986). Also see above Response #109 and #110.
The EPA and NYS DOH consider 1248, 1232 and all other PCBs to be in the same risk classes of the 1260 PCBs, even though there is no scientific evidence that these molecules with less chlorine content than the 1260s are any risk at all to human health or the environment. In comparison, an editorial in "Science" stated that "From the standpoint of health effects, there is no justification to base regulation of all PCBs on tests with Aroclor 1260."
One use of public health assessments is to evaluate the potential health effects associated with environmental contamination. Generally, the levels of contaminants in environmental samples are not high enough to dramatically and adversely affect human health, but are sufficiently high (that is, higher than levels in uncontaminated areas) to raise concerns among the local population and within government agencies over the potential long-term health effects of continued exposures. Consequently, agencies such as the NYS DOH and ATSDR are requested to do public health assessments.
The potential for harmful health effects under any exposure situation depends on the toxicity of the chemical and the amount of a person's exposure, which includes both the level of exposure and the length of exposure. Animal studies have identified the type of toxic effects PCBs can cause in animals and some of the same effects have been reported in humans occupationally and environmentally exposed to PCBs, although the causal role of other chemicals remain undetermined (ATSDR, 1995). These studies are also provided enough information to calculate the strength (potency) of PCBs to cause these effects. Measuring PCB levels in fish, wildlife, soil, water and air have identified the potential sources of PCBs for individuals living near the GM Foundry. These data can be used to calculate the PCB exposures of these individuals. When combined, the potency and exposure data provide estimates of the potential risk associated with various exposure scenarios, such as eating contaminated fish or working or playing in contaminated soil. If these risks are low enough, then the environmental exposures are not considered a significant public health concern. If the risks are high; however, then the environmental exposures are considered a public health concern and remedial action to reduce those risks may be recommended. This basic approach is used by many federal and state agencies, including the NYS DOH, ATSDR and the US EPA, to evaluate human risk associated with environmental contamination.
REFERENCES TO RESPONSE TO COMMENTS ON GM MASSENA PHA
Agency for Toxic Substances and Disease Registry. 1991. Toxicological Profile for Selected PCBs (Aroclor-1260, -1254, -1248, -1232, -1221 and -1016). Update Draft. U.S. Department of Health and Human Services. Atlanta, Georgia: U.S. Public Health Service.
Agency for Toxic Substances and Disease Registry. 1992. PHA Guidance Manual. U.S. Department of Health and Human Services. Atlanta, Georgia: U.S. Public Health Service.
Agency for Toxic Substances and Disease Registry. 1995. Toxicological Profile for Polychlorinated Biphenyls. Draft for Public Comment (Update). U.S. Department of Health and Human Services. Atlanta, Georgia: U.S. Public Health Service.
Chen, Y.C.J., Y.L. Guo, C.C. Hsu and W.J. Rogan. 1992. Cognitive development of Yu-Cheng ("Oil Disease") children prenatally exposed to heat-degraded PCBs. JAMA. 268: 3213-3218.
Gladen, B.C., W.J. Rogan, P. Hardy, J. Thullen, J. Tingelstad and M. Tully. 1988. Development after exposure to polychlorinated biphenyls and dichlorodiphenyl dichloroethene transplacentally and through human milk. J. Pediatr. 113: 991-995.
Gladen, B.C. and W.J. Rogan. 1991. Effects of perinatal polychlorinated biphenyls and dichlorodiphenyl dichloroethene on later development. J. Pediatr. 119: 58-63.
Golub, M.S., J.M. Donald and J.A. Reyes. 1991. Reproductive toxicity of commercial PCB mixtures: LOAELs and NOAELs from animal studies. Environ. Health Perspectives. 94: 245-253.
Hong, C.S., B. Bush, J. Hiao and H. Qiao. 1993. Toxic potential of non-ortho and mono-ortho coplanar polychlorinated biphenyls in Aroclors, seals and humans. Arch. Environ. Contam. Toxicol. 22: 118-123.
International Agency for Research on Cancer. 1987. Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs Volumes 1 to 42. Supplement 7. Geneva: World Health Organization.
James, R.C., H. Busch, C.T. Tamburro, S.M. Roberts, J.D. Schell and R.D. Harbison. 1993. Polychlorinated biphenyl exposure and human disease. J. Occup. Health. 35: 136-148.
Lawton, R.W., M.R. Ross, J. Feingold and J.F. Brown, Jr. 1985. Effects of PCB exposure on biochemical and hematological findings in capacitor workers. Environ. Health Perspect. 60: 165-184.
Morgan, R.W., J.M. Ward and P.E. Hartman. 1981. Aroclor 1254-induced intestinal metaplasia and adenocarcinoma in glandula stomach of F344 rats. Cancer Res. 41: 5052-5059.
National Cancer Institute. 1978. Bioassay of Aroclor 1254 for Possible Carcinogenicity. Technical Report Series No. 38. Bethesda, MD: U.S. Department of Health, Education and Welfare, U.S. Public Health Service.
Pao, E.M., K.H. Fleming, P.M. Guenther and S.J. Mickle. 1982. Foods Commonly Eaten by Individuals: Amount Per Day and Per Eating Occasion. Home Economics Research Report No. 44. U.S. Department of Agriculture, Human Nutrition Information Service, Hyattsville, MD.
Schaeffer, E., H. Greim and W. Goessner. 1984. Pathology of chronic polychlorinated biphenyl (PCB) feeding in rats. Toxicol. Appl. Pharmacol. 75: 278-288.
Selikoff, I.J. and E.C. Hammond. 1986. Environmental Contaminants and the Health of the People of the St. Regis Reserve. Volume II. Environmental Sciences Laboratory, Mount Sinai School of Medicine of the City University of New York, New York, NY.
Taylor, P.R., C.E. Lawrence, H.-L. Hwang and A.S. Paulson. 1984. Polychlorinated biphenyls: Influence on birthweight and gestation. Am. J. Public Health. 74: 1153-1154.
Taylor, P.R., J. Stelma and C.E. Lawrence. 1989. The relation of polychlorinated biphenyls to birthweight and gestational age in the offspring of occupationally exposed mothers. Am. J. Epidemiol. 129: 395-406.
U.S. EPA. 1986. Guidelines for Carcinogen Risk Assessment. Fed. Register. 51: 33992-34003.
U.S. EPA. 1993. Workshop Report on Developmental Neurotoxic Effects Associated with Exposure to PCBs. EPA/630/R-92/004. Washington, DC: Risk Assessment Forum.
U.S. EPA. 1995a. Aroclor 1254. On-Line. Integrated Risk Information System. Cincinnati, OH: Office of Research and Development, Environmental Criteria and Assessment Office.
U.S. EPA. 1995b. Aroclor 1016. On-Line. Integrated Risk Information System. Cincinnati, OH: Office of Research and Development, Environmental Criteria and Assessment Office.
U.S. EPA. 1995c. PCBs. On-Line. Integrated Risk Information System. Cincinnati, OH: Office of Research and Development, Environmental Criteria and Assessment Office.
Ward, J. 1985. Proliferative lesions of the glandular stomach and liver in F344 rats fed diets containing Aroclor 1254. Environ. Health Perspect. 60: 89-95.
Yu, M.L., C.C. Hsu, B.C. Gladen and W.J. Rogan. 1991. In utero PCB/PCDF exposure: Relation of developmental delay to dysmorphology and dose. Neurotoxicol. Teratol. 13: 195-202.