TL: GREENPEACE REVIEW OF EPA'S PROPOSED MAXIMUM CONTAMINANT LEVEL FOR 2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN IN DRINKING WATER SO: Pat Costner, Greenpeace USA (GP) DT: December 13, 1990 Keywords: toxics us standards drinking water dioxins epa greenpeace reports gp chemicals chlorinated cancer diseases risks human health / ("National Primary and Secondary Drinking Water Regulations; Synthetic Organic Chemicals and Inorganic Chemicals; Proposed Rule" (55 Federal Register 143, July 25, 1990). 1. EPA's Maximum Contaminant Level Goal (MCLG) of zero for TCDD is appropriate. The proposal's goal of zero acceptable dioxin contamination in drinking water is consistent with scientific data and prudent public policy. Because TCDD is an extraordinarily potent carcinogen in laboratory animals and appears to be a potent carcinogen in humans as well (see EPA's Criteria Document for the preparation of these standards), and because these carcinogenic effects appear to operate without a threshold, the only standard that is protective of public health is zero. This goal is consistent with EPA policy for carcinogenic materials. Evidence to suggest that other adverse health effects of dioxin occur without a threshold also supports the zero standard; see for example, studies supporting a possibly linear hormonal mechanism for such adverse effects of TCDD as birth defects, developmental abnormalities, hepatic toxicity, immune dysfunction etc., in the Criteria Document, in E. Silbergeld and T. Gasiewicz, "Dioxins and the Ah Receptor," Am J. Ind Med, 16:455-474. 1989, and M. Gallo, "Rationale for A Hormone-like Mechanism of 2,3,7,8-TCDD for Use in Risk Assessment," Appendix F to USEPA A Cancer Risk-Specific Dose Estimate for 2,3,7,8-TCDD, 600/6-88/007ab, June 1988.). 2. EPA's abandonment of a goal of zero TCDD in the currently proposed regulatory standard is unacceptable. EPA has proposed a maximum contaminant level for regulatory purposes of 50 picograms TCDD per liter of water -- 50 ppq -- based on a minimum detection limit of 10 ppq, despite prior knowledge within the Agency that "a detection limit of 20 parts per quintillion in drinking water" has been established by Dr. Christoffer Rappe, University of Umea, as noted by Martha Prothro, Director, U.S. EPA Office of Water Regulations and Standards, in her letter of July 20, 1990, to John Bonine, Western Natural Resources Law Clinic, Dr. Christopher Rappe of the University of Umea in Sweden. Further, at the International Dioxin Symposium in Bayreuth, Germany, this year, Dr. Rappe presented a paper documenting the sampling and analysis of surface water and drinking water at 0.001 ppq levels. (C. Rappe, L.O. Kjeller and S.E. Kulp, "Sampling and Analysis of PCDDs and PCDFs in Surface Water and Drinking Water at 0.001 ppq Levels," Dioxin`90, Bayreuth, Germany, September 1990). In fact, as noted in Director Prothro's letter, the Agency is also aware that the Canadian government proposes requiring TCDD and TCDF concentrations in pulp and paper mill effluents to be ""non- measurable," where "non-measurable" is defined as the Limit of Quantitation (LOQ) of a reference method. For a 1 liter sample of pulp and paper effluent, Dr. Chung Chiu of Environment Canada reports a method detection limit of 2 pg/L. However, Dr. Chiu observes, "For very clean samples, such as drinking water, one could use a sample size much larger than 1 L. If one used a 10 L sample, for example, the analytes would be concentrated by a factor of 10 in the cleaned extract thus lowering the method detection limit." I.e., rather than the detection limit for TCDD of 2 pg/L proven for pulp and paper mill effluents, a detection limit for drinking water of 0.2 pg/L is readily achievable. In summary, Greenpeace suggests that it is EPA's duty to afford U.S. citizens protection from TCDD in their drinking water that is at least as stringent as Canada's protection of its rivers from TCDD discharges by pulp and paper mills. I.e., Greenpeace urges EPA to require TCDD concentrations in drinking water to be non-measurable, where non-measurable is defined as the limit of quantitation of the most sensitive proven method -- currently Dr. Rappe's method that has a detection limit for TCDD of 0.001 ppq for drinking water -- with an annual review to identify the most sensitive method, since analytical capabilities are changing so rapidly. 3. The use of a 5-fold multiplier of the method detection limit to establish the limit of quantitation is indefensible. It is a well- accepted, longstanding practice within the scientific community, including the practices and protocols of U.S. EPA, to use a 3-fold multipler in establishing the limit of quantitation. Deviation from this practice in this instance seems both arbitrary and capricious. 4. EPA's proposed approval of unprecedentedly high cancer risks is indefensible. According to the text of the proposal, the MCL proposed -- 50 ppq -- allows cancer risks to the public of 2.5 per 10,000, based on EPA's 1985 cancer potency estimate and assuming intake of 2 liters of water daily by adults weighing 70 kilograms. To our knowledge this is the highest cancer risk that EPA has ever attempted to legitimize in a regulation. One per million remains EPA's official de minimis standard, though we are aware of several instances in which EPA has proposed "acceptable" risks of one per 100,000, and one proposal in which one per 10,000 was deemed acceptable (benzene emissions from ovens and furnaces associated with the steel industry). This proposal allows calculated risks 250 times higher than the de minimis standard, 25 times higher than the occasional exception, and 2.5 times higher than the highest risks the Agency has ever allowed. The 2.5 per 10,000 lifetime risk posed by the proposed dioxin standard is equivalent to approximately 892 deaths each year in the U.S. population of 250 million (assuming a 70 year lifetime). 892 deaths per year can hardly be deemed negligible. This discussion should not be construed to suggest that we consider scientifically or ethically acceptable either the quantitative risk assessment process or the risk management policy which institutionalizes as "negligible" risks which are involuntary and preventable. 5. The unprecedented approval of exposures which exceed reference doses is unacceptable. The criteria document calculates an RfD of 35 pg/L TCDD based on studies (Allen 1979 //Joe, give full cite?//) which found adverse reproductive effects among rhesus monkeys exposed to dioxin at dietary doses as low as 50 parts per trillion (ppt), which is equivalent to 1.5 ng/kg/day. The proposed standard of 50 ppq thus legitimizes exposures that are more than 40 percent greater than the dose calculated as protective against adverse reproductive effects. We are aware of no instance in which the Agency has set exposure standards higher than calculated reference doses. Such a standard, which effectively authorizes exposures which cause reproductive dysfunction, represents an abandonment of EPA's duty to protect the public. 6. Failure to include data showing adverse reproductive effects at lower doses is unacceptable. Since Allen's 1979 rhesus monkey studies, follow-up research has shown adverse reproductive and developmental effects at dietary doses of 25 ppt and even 5 ppt, which are equivalent to 0.75 anf 0.15 ng/kg/day, respectively. (see R. Bowman et al, "Behavioral Effects in Monkeys Exposed to 2,3,7,8-TCDD Transmitted Maternally During Gestation and For Four Months of Nursing," Chemosphere, 18:235, 1989). The Criteria Document does not make use of this new data. The emergence of new evidence of reproductive effects at dietary doses 10 times less than the data used to calculate the RfD is certainly significant. The RfD (and thus the MCL based upon it) may thus be ten times higher than an RfD that is protective of public health. If so, the MCL proposed would exceed an RfD based upon the 1989 study by as much as 14 times. 7. Failure to consider other health effects. Other adverse effects of TCDD -- i.e., neurological, endocrine, immunological -- appear to occur at extraordinarily low doses, although there is little information to quantify a does/response relationship. The lack of data should not be construed to mean that exposures are safe. Without information to show that the RfD calculated will not result in such effects, the MCL cannot be considered protective for these health effects. 8. Failure to consider non-average subpopulations. As noted above, the cancer risks and reference dose for threshold effects are calculated based on a 70 kg individual consuming 2 L of water each day. The proposal text does not offer an estimate based on other parameters, but we note that the ratio of daily water intake to body weight for some individuals may be far higher than the .028 L/kg. Infants fed water-based nursing formulas, children, athletes, and those living in hot, dry climates are leading possibilities. For example, the Criteria Document //full cite, Joe?// estimates this ratio for children at 0.1 L/kg. Cancer risks for such individuals will be much higher than 2.5 per 10,000. The quantity of TCDD ingested by such individuals will, of course, result in doses which exceed the reproductive RfD by a greater amount, as well. Schantz, S.L., D. A. Barsotti, and J.R. Allen, " toxicological effects produced in non-human primates chronically exposed to 50 parts per trillion 2378-tetrachloro-p-dioxin (TCDD), Toxicol. & Appl. Pharm. Vol. 48, p. A180. ENDS[Greenbase Inventory October 27, 1991 ] =======[#]=======