Recent Brunswick Wood Preserving Studies Show Neighborhood Concern is Justified
December 2000

            In trying to downplay the significance of the latest tests, EPA has gone too far by suggesting the off-site areas are not a threat. The most recent chemical testing data for the creek area surrounding the Brunswick Wood Preserving Superfund Site shows moderately high levels of toxic chemicals. An Environmental Protection Agency (EPA) health effects study using exposure models indicates that although the levels of arsenic, petroleum chemicals, and dioxins are significant, the levels are not high enough to be acutely toxic. In other words, it would require long exposure before injury occurs. Unfortunately, the study model has several flaws that render the conclusions incomplete. The biggest problem is that the model only used exposure to Burnett Creek chemicals; it did not consider that children might also receive exposure in other areas around the site. Secondly, it does not consider whether levels are increasing or decreasing. These factors are important since the current model will change depending on the future levels. The data indicate a significant amount of off-site contamination exists in areas where children can play, and only a cleanup can make the area free from risk.         

            Received for review with reference to the Brunswick Wood Preserving Site are: Review of Off-site Soil and Sediment Data dated November 3, 2000, an October 30, 2000 Health Consultation performed by the Agency for Toxic Substances and Disease Registry (ATSDR) based on the Soil and Sediment Data, portions of the still incomplete Soil, Sediment and Groundwater Studies, and a Georgia Times Union article quoting Brian Farrier, the EPA Site Remedial Project Manager (RPM) regarding the data.

            The overall direction of the Site review and the remarks by Mr. Farrier imply that Burnett Creek and surrounding areas are not potentially hazardous to children, wildlife or from subsistence fishing activities.

    An Overview of Testing and Interpretation

            The purpose of environmental testing is to determine the extent of contamination. When contamination is found, a health-effects modeling study is used to determine if the contamination is toxic to people or the environment. Health effects are divided into two categories: cancer and non-cancer risks. Environmental cancer risk modeling is based on a probability model, in some ways not unlike “lotto” or other games of chance. Susceptibility to cancer is determined by environmental and lifestyle effects, but still requires exposure to certain cancer-causing (carcinogenic) chemicals, viruses, or energies. Each carcinogen also has a certain potency, or ability, to start cancer cells growing. That potency is different for different animal species and in all known cases is “dose-responsive.” Dose-response simply means that with more carcinogen exposure, the more likely it may cause cancer. Everyone uses the phrase “more likely” or just “likely” while not really realizing it is a mathematical probability term. In cancer modeling, a dose-response effect means that a high carcinogen dose is more likely (higher probability) to result in more cases of cancer. “Dose” is further broken down into two different effects: the concentration of the chemical and the length of exposure to the chemical.

            The interaction of all of these different factors is rolled into a probability model where all of the other factors (genetics, lifestyle, potency, exposure) are on one side of an equation and the chemical concentration is on the other side of the equation. This allows researchers to easily express the relative dangers of different chemicals in a convenient graph.

            In the lotto-type games, if someone had all 23 million possible tickets (based on a 6 ball, 53-number game) then they have a 100 hundred percent likelihood of winning. There are some human carcinogens, ionizing radiation for example, where above a certain dose the person’s likelihood of getting cancer approaches 100% as well. However, most environmental toxins contribute to the lifetime risk of cancer, rather than acting as direct risks. Again, for carcinogens those risks are expressed in terms of the “chance of getting a cancer.” For environmental clean-ups, the law places a “one-in-one-million chance” of a new cancer as acceptable, but allows up to a “one-in-ten-thousand chance.” This minimum cleanup standard can be viewed as 100 times riskier than the one-in-one-million chance, and as a policy, it is very controversial. For chemicals that are known mutagens (anything that causes mutations in DNA, usually the starting point in cancer development), or where the mechanism of action is poorly known, a one-in-ten-thousand chance may not be a safe dose. There is also considerable doubt if the one-in-ten-thousand risk level is truly safe for children, certainly for some chemicals that level is too risky. 

            Toxicity is considered by a different model. For nearly all-environmental chemicals, there is very good animal toxicity data. Generally, toxic effects in animals, such as liver or kidney failure, can be “extrapolated” to humans. Extrapolation means that the dose for a toxic injury in animals and humans can be compared based on body-weights. This is in contrast to cancer, where there is no easy comparison of animal cancers with human cancers. For toxicity, a “Hazard Quotient” (HQ, or hazard index, HI) is developed again using the chemical concentration as the benchmark. Chemicals are usually considered hazardous of the HQ exceeds a value of 1.

            If the cancer risk is high, say one-in-one-thousand, and/or the hazard quotient is high, say over 3, the response from EPA may be to order an immediate cleanup of the area. These levels are indicative of acute toxicity or short-term cancer risk. Unfortunately, the decision system is bureaucratic and inefficient. EPA collects or orders samples collected. The analysis is performed and the data sent to the ATSDR (Agency for Toxic Substances and Disease Registry). ATSDR has no say in how the study is conducted; they only respond to the data sent to them. Their remarks are then sent back to the EPA for comments on their comments. At the Brunswick Wood Preserving site several studies have been done with the on-site areas, different studies done with some off-site areas, and now this study only looks at the creek areas. In fact, one person could be exposed in several areas to varying concentrations of the same chemical, but these studies would not consider that total effect.

        In contrast to Region 4 EPA’s conclusion that the off-site contamination poses no risk to neighborhood children, the facts plainly tell a different story:

_ Given the long latency period for cancer development, children exposed to sediments containing carcinogenic polycyclic aromatic hydrocarbons are enduring too much risk.

_ The data shows arsenic levels up to 3.8 parts-per-million. At nearly 8 times the Cancer Risk Evaluation Guide, this is sufficiently high to justify a fast track cleanup up the affected areas.

_ The ATSDR Health Consultation actually found that a significant number of sediment samples from residential areas exceed chronic guidelines for dioxin. This should be a cause of concern at the EPA and in the community. Unfortunately, ATSDR and Region 4 EPA dismissed the importance of these studies by saying “safety margins” built into the guidelines will protect children. This is unscientific reasoning.

_ EPA should have requested a comprehensive assessment by ATSDR rather than the limited assessment as was done. Region 4 is well aware that bias can result from a limited assessment.

_ The most striking fact about the data is that only 10 samples were taken but nearly all contained toxins from the Brunswick Wood Preserving Site. These sediments are heavily affected and the potential for toxic injury exists. Certainly a more comprehensive testing and analysis is needed in this area.

            Probability is a complex science, but look at it this way: if you never buy a lotto ticket, you never have a chance of losing; likewise, if you are never exposed to a certain carcinogen, you can never get cancer from it. Region 4 EPA clearly does not understand this very basic law of probability and is assigning “safe” levels without a scientific understanding of probability laws.

            Worse yet, EPA is second-guessing the scientific advisory panels which set the safe levels by saying that there are “safety margins” that Region 4 can exceed at will. This is similar to a pilot ignoring National Transportation Safety Board rules on safe aircraft operation by assuming the NTSB set the values arbitrarily high. Clearly, EPA has exceeded their mandate of authority in producing conclusions such as stated in these documents. 

            The result of considering each small study independent of the whole is that Region 4 EPA does not properly consider all of the potential exposure. They have subsequently justified long delays in cleanups by stating that no immediate risk occurs, when the facts speak differently. The real reason for the long delays in finding the pollutants and rendering them harmless is not known. Perhaps Region 4 does not want to spend “their” money cleaning up the Brunswick Wood Preserving Site. Regardless of the reason, once someone is exposed to a mutagen or carcinogen that exposure cannot be undone. The earlier in life that exposure occurs the more likely that harm will accrue. EPA’s assessment of the potential harm at Brunswick Wood Preserving Site is threatening to area children.

                Written by R. Kevin Pegg, Ph.D.; edited by Dr. Mary S. Saunders. Copies of the newsletter are available from the GEC, at the Glynn County library, or at www.NucleicAssays.com/tags  on the Internet.

"This project has been funded wholly or partly by the U.S. Environmental Protection Agency under Assistance Agreement Number 1984482-98-0 to The Glynn Environmental Coalition, Inc. The contents of this document do not necessarily reflect the views and policies of the U.S. Environmental Protection agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use."