The Edmund Burke Society is pleased to provide its comments on the proposed rule, "National Primary Drinking Water Regulations: Consumer Confidence Reports," published in the Federal Register on Friday, February 13, 1998, 63 Fed. Reg. 7606.
Description and Interests of the Persons Commenting
The Edmund Burke Society is an informal group of staff members of the United States Environmental Protection Agency (EPA) who meet regularly to discuss issues relating to protection of the environment and human health. The Society has sponsored lectures and discussions, open to EPA staff and the public, featuring such eminent scientists and policymakers as Judge (now Supreme Court Associate Justice) Stephen Breyer, Zimbabwe Ambassador to the U.S. Amos Midzi, Dr. Bruce Ames, Professor Fred Singer, and the late Professor Julian Simon. As many as 50 EPA staff members have attended many of these lectures and discussions and over 150 attended the lecture given by Dr. Bruce Ames.
The Edmund Burke Society is named for the distinguished 18th century British Parliamentarian who spoke and wrote in support of American and Irish revolutions but criticized the French Revolution for its ideological extremism. Similarly, we believe strongly in protecting human health and the environment and that ideological extremism within and outside the government often clouds issues and misplaces priorities.
Members of the Edmund Burke Society share the belief that effective environmental protection uses science appropriately, allocates resources realistically and compassionately, and is built on principles of economic freedom, including free markets, decentralized priority-setting, and respect for individual property rights. The Society has 15 formal members and many other EPA staff members have at times expressed sympathy for many of the positions we support.
Society members span the full range of EPA professional disciplines including toxicology, chemistry, engineering and law. Members have worked within government and in the private sector for many years, some for over 30 years, in jobs related to environmental protection. We have members in most EPA program offices.
Members of the Edmund Burke Society, as consumers of public water, have a strong interest in ensuring clean and safe drinking water for their families and communities. We are submitting these comments as private citizens. The comments were prepared on personal time and no member of the Society stands to gain any financial interest from any activity affected by this rule. We are interested purely in protecting the public health and the environment in a rational manner and in ensuring the dissemination to consumers of truthful information about drinking water.
Regulatory Background
The Consumer Confidence Reports are described in the proposed rule as "the centerpiece of public right to know" under the Safe Drinking Water Act (SDWA). The particular provision of concern in these comments is section 1414(c)(4)(A) of the SDWA, which requires the Reports to contain brief statements in plain language regarding the health concerns that resulted in regulation of each regulated contaminant.
The proposed regulations, in Appendix B, provide specific language for each of 74 contaminants. In general, the language conveys risk information for chronic adverse health effects by stating that such effects "could" result from exposures "in excess" or "well in excess" of the Maximum Contaminant Level (MCL) "over many years." In cases where human or animal exposure to high doses have indicated that a contaminant is a possible carcinogen, the language indicates that people who drink water with the contaminant at levels above the MCL over many years "may have an increased risk of getting cancer." The Agency specifically requests comment on whether there are other ways to communicate to water system customers the degree of health risk they may face as a result of MCL violations.
Views of the Society
The members of the Edmund Burke Society believe a statement that exposure "may" increase the risk of cancer can be too easily misconstrued by the scientifically untrained general public that there is an actual risk of cancer from the levels to which customers of the water system will be exposed. We contend that such a statement does not reflect the many scientific uncertainties and assumptions underlying this conclusion. Thus, a person may through misunderstanding exaggerate the risk, particularly if the uncertainties in estimating that risk have not been explained in terms the lay public would clearly understand. Statements to customers should include at least a straightforward caution that any expressed or implied cancer risk is a hypothetical risk, unproven by available scientific evidence, rather than an actual risk determined through cause and effect studies in humans.
Scientific Information
Effectively, the position taken in almost all EPA regulations involving carcinogens, which can be found in the public rulemaking records for the contaminants regulated for carcinogenicity under the Clean Water Act, is that there could be a zero risk of cancer from the contaminant at Agency regulated levels. This is because Agency risk assessments generally indicate that the range of risks within the confidence limits expressed in the analysis include zero.
It needs to be made clear to the public, therefore, that regulation for chronic adverse effects, particularly cancer, is not based on proven dangers to humans, or supported by direct knowledge that those dangers are attributable to the action of particular contaminants in humans. These regulations, in most cases, are based on evidence from high-dose studies of laboratory animals which are then extrapolated to low-dose human exposures. Any effect these levels may have on humans, animals or the environment are not obvious, and greatly depend on the models used for high-to-low dose extrapolation, as well as assumptions that effects observed in a given animal species will also occur in humans. While analysis of the water supply clearly supports a direct correlation between bacterial contamination and adverse effects, it cannot directly determine to the same degree of certainty the actual, or implied, effects of very low concentrations -- for example, at parts per billion levels -- of a chemical in the water supply.
Because EPA cannot directly determine the actual effects of these contaminants at very low levels, it draws inferences from extrapolated animal data. While this evidence may be appropriate to impose rules to cleanup contaminants, the Agency ultimately has to acknowledge that no conclusive evidence exists of the chemical's actual effects on humans at low doses.
While in some cases EPA may use data of high human exposure situations to determine the possible effects on humans in the general population, the principles of extrapolation from high exposure to low actual human exposure create problems in analysis that are similar to the extrapolation from higher exposures to low exposures in a single animal species. Assessing epidemiological associations presents a further difficulty in that persons in such studies are exposed to a variety of agents besides the one under study. Ultimately there is considerable uncertainty in the amount of any adverse effects (cancer or other disease) that may occur, and it is often possible that there is no risk at all from the suspected chemical contaminants.
While the Society wishes to express strong support for close control of bacterial contaminants in water supplies, it appears that this regulation focusses to a greater degree on chemical contaminants. This could detract from public awarenes of the more likely bacterial risks. Further, the Society wishes to make clear that laboratory animal studies are a valid tool for regulating to protect public health. It is, however, very important that the public not be led to believe that the hypotheses used to regulate are determinants of actual risks. This, unfortunately, appears to us to be the message of the Consumer Confidence Reports.
A forthright and honest statement of the scientific information used by EPA to support regulation needs to acknowledge that, as a practical matter, the scientific evidence used by the Agency is uncertain to some degree and that evaluations of evidence often are based on considerations of scientific theories, projections of trends from currently available data, modeling using reasonable assumptions, and extrapolations from limited data. Science in such cases shows only a very wide range of possible risks that may be prevented through regulation.
While EPA may use this information to support regulation designed to prevent adverse effects from occurring, it is far different to leave the impression with the public that this information is based on observations that exposure to a given substance has actually resulted in cancer in humans.
Public Perception
The fact that EPA is concerned that many people may attribute greater significance to risks based on the animal data than the scientific evidence shows is aptly expressed in the rulemaking record. The preamble to the proposed rule states that the Agency is "sensitive to the concern that some water system customers may interpret the language as indicating a significantly higher level of incremental risk than would actually result from exposures at the levels that are likely to occur." (63 FR 7616) [Emphasis Added]. This concern was raised by the "expert panel" convened in June 1997. Concern that the Consumer Confidence Reports not cause unwarranted public concern or panic was among the issues raised during a series of focus groups, described in the rulemaking record, that were held by the American Water Works Association.
What is absent from the record, unfortunately, is the extent to which people believe there is scientific proof that cancer is actually caused by chemical contaminants in the drinking water. If only "some" customers believe there is a significantly higher level of risk than the science supports (as EPA seems to believe), we would have little concern. There are, apparently, no general surveys on this matter.
On the other hand, there is substantial anecdotal evidence that excessive numbers of the population at large, who are not familiar with the scientific evidence, believe that cancer is actually caused by chemical contaminants in the drinking water.
Prominent on the best seller lists for the past several years is a book titled A Civil Action by Jonathan Harr. This book describes a lawsuit brought by parents of children who contracted leukemia against two companies that were associated with causing drinking water contamination at very low levels (in the four parts per billion range) of trichloroethylene (TCE). (TCE is a regulated contaminant under the Clean Water Act and is described in Appendix B as a substance that "may" cause cancer -- Appendix B (70).) This case received substantial national publicity and people involved in it appeared on national television. The popularity of the book and the national notoriety of the case, while not in any sense a scientific survey, attests to the concerns of a very significant part of the population that drinking water contaminants do, in fact, cause cancer.
What may be most significant in the book with respect to the issue as to whether there are substantial members of the public that believe drinking water contaminants cause cancer is a public survey taken by the law firm representing one of the companies. The law firm surveyed 500 people in order to examine whether a potential jury would be willing to award money to the parents of the children who contracted leukemia. The law firm found that 80 percent would have been willing to find in favor of the parents. This, of course, does not prove that the public believes that cancer is being caused by drinking water contaminants, but it does create an impression that a very large percentage of the population would subscribe to that view.
Members of the Edmund Burke Society do not claim any expertise in the areas of public perception of risk or risk communication. We do, however, question whether EPA has adequately analyzed the effect the Consumer Confidence Reports could have on the perception of drinking water risk held by significant portions of the public.
Recommendations
The Society, accordingly, recommends that for carcinogens for which only animal studies form the basis of regulation language be included in Appendix B that states, "This contaminant has been shown to cause cancer in animals at very high doses. There have been no identified cases of cancer in people caused by this contaminant in drinking water."
If studies on humans form the basis for regulation, language should be added that states, "The risk that people may get cancer is based on studies showing that people have contracted cancer when they are exposed for many years at levels many times greater than the levels at which this contaminant is ever likely to be found in drinking water. There have been no identified cases of cancer caused by levels of this contaminant found in drinking water."
Comments on this posting?
Click here to post a public comment on the Trash Talk Bulletin Board.
Click here to send a private comment to the Junkman.
Copyright © 1998 Steven J. Milloy. All rights reserved on original material. Material copyrighted by others is used either with permission or under a claim of "fair use." Site developed and hosted by WestLake Solutions, Inc.
Material presented on this home page constitutes opinion of Steven J. Milloy.