January 20, 2000
To Whom It May Concern:
RE: Proposed rule for radon in drinking water, published in the Federal Register on November 2, 1999 (64 FR 59246), W-99-08
Physicians for Civil Defense is a nonprofit organization devoted to educating the public about threats to our population and the means of protecting against them. We devote considerable effort to risk assessment because well-intentioned but misguided regulations can not only divert needed resources from more productive uses but have severely detrimental unintended consequences.
We are opposed to the proposed rule for the following reasons:
1. The best scientific evidence shows that radon in drinking water is not a hazard and, in fact, is probably beneficial to human health.
2. Even if the EPA's risk assessment were correct, the proposed rule would impose huge costs on municipalities that far outweigh any potential benefit.
3. The proposed rule is an inappropriate use of the regulatory process to expand the agency's jurisdiction without proper Congressional authority. The proposed standards are internally inconsistent and can only lead to public confusion and unwarranted litigation.
Scientific Evidence Concerning Radon
Congress specifically directed the EPA to have the National Academy of Sciences review and validate the scientific evidence supporting the proposed limits on radon. The NAS report makes the following very clear:
For example, the NAS looked at the eight published case-control studies. The largest, performed in Sweden, involved 1,360 cases and 2,847 controls. "The lung-cancer excess was not statistically significant even for smokers or nonsmokers with over 400 Bq m-3 in the home for over 32 y."
The NAS does comment on the data set that contradicts the EPA's projections: "The ecologic study of Cohen (1995) is the most comprehensive. It encompasses about 300,000 radon measurements in 1,601 counties in the U.S. The trend of county lung cancer mortality with increasing home radon concentration is strikingly negative, even when attempts are made to adjust for smoking prevalence, and 54 socioeconomic factors....This finding contradicts the existing risk estimates at low exposure, and a sound reason for the significant negative trend should be sought."
While Cohen's data are the best in the world, and have very narrow error bars (see the figure), they are often discounted or impugned because of the "ecological fallacy." That is, the study does not identify individual exposures or confounding factors. The radon exposure is the average for the area.
In fact, it can be shown that the ecological fallacy does not apply to the calculation of a population risk, but only to the determination of the dose-effect curve, which may be affected by various confounding variables. The assumption that is needed- and that is generally made implicitly-is the equivalent-population assumption. As an example, the BEIR V report assumes that the population exposed to the Hiroshima and Nagasaki bombs is the same as the current American population. Since Cohen's test population is essentially the entire population of the United States, the use of the assumption can hardly be questioned. In contrast, to assume that the population of underground miners used in the EPA's extrapolation is equivalent to the U.S. population as a whole is a far bigger leap of inference. A very high proportion of miners smoked, and they were routinely exposed to unusually high concentrations of particulate matter, arsenic, and other toxic or carcinogenic elements found in rock.
The discrepancy between Cohen's data and the BEIR VI extrapolation is very large: about 20 standard deviations. The only way to explain it away is to assume that the largest confounding variable, smoking, has an exact negative correlation with average radon level in the county (so that the lower the radon level, the higher prevalence of smoking). There is no evidence for such a correlation, which is preposterous on its face.
With a radon concentration in water of 1000 pCi/liter, the additional contribution to indoor radon would be 0.1 pCi/liter or 3.7 Bqm-3. From the enclosed graph, it is apparent that this concentration is minute compared with the concentrations experienced by underground uranium miners. Moreover, it is within the dosage region in which a decrease in average radon levels is consistently associated with an increase in lung cancer incidence. From examining the figure, the zero-effect level is higher than 230 Bq m-3, the highest level encompassed by Cohen's data. Between 20 and 140 Bq m-3, it appears that all the actual observations have error bars extending below 1.0. Even at the highest dose levels on the graph, the best estimates of relative risk are below 2.0. Relative risks this low are generally considered questionable from an epidemiologic standpoint.
The proposed alternative maximum contaminant level (AMCL) of 4,000 pCi/liter would increase indoor radon by about 15 Bq m-3. A level 10 times that high would increase the indoor radon by only 150 Bq m-3, which is well below the zero-effect level.
The Costs of the Regulation Outweigh Any Conceivable Benefit
The EPA claims that 168 cancer deaths per year can be attributed to radon in drinking water. This is a hypothetical calculation based on the extrapolation discussed above; no observational evidence supports it. This is 0.1% of the lung cancer deaths in any given year. By the EPA's 1991 estimate, the cost of the 300 pCi/liter standard would be $180 million. This was increased to $407.6 million /year in 1997 dollars.
If the expenditure of $180 million prevented 168 premature deaths due to cancer, the cost per postponed death would be about $1 million. Actually, the benefit would be less than that, as the standard would not decrease the radon contribution from water to zero, and the EPA considers every atom to be a risk. The cost would probably also be much greater. The American Water Works Association estimated the 1991 compliance costs to be $2.5 billion, or nearly 14 times the EPA estimate. Multiplying the 1997 estimate by 14 gives a cost estimate of $5.7 billion, or $33 million per hypothetical life saved (or actually per death postponed by an unspecified length of time, with the cause of death changed from lung cancer to something else). This is an unconscionable waste of resources, which could be used far more effectively in other ways to enhance the quality, quantity, and duration of human life.
The Regulation Expands the Authority of the EPA Beyond That Authorized by Congress
The Multi-Media Mitigation (MMM) program, proposed as an alternative to the 300 pCi/L standard, expands authority beyond regulating content of water supplies to monitoring and "mitigating" indoor air quality in private homes.
If Congress intends to empower a federal agency to inspect private residences and force homeowners to reduce the concentration of a natural and ubiquitous gas in their homes, then it should pass legislation explicitly designed to do that (assuming that it could be done constitutionally). Instead, the EPA is attempting to coerce water utilities to undertake a program to meet an EPA goal in an area that has nothing to do with their responsibility to supply safe water.
The stated rationale for the proposed MCL is safety. The MMM program, however, implies that a much higher concentration in the water supply to one person's home is safe, as long as an adequate number of other persons undertake measures to reduce the concentration of radon in the air in their homes. This is like saying that the safe dose of a poison depends on the amount that is kept in the average medicine cabinet.
This position defies all reason, logic, and common sense. It is likely to inspire contempt for the regulatory process in thinking individuals, even if the vexing problem of consistent, fair enforcement could be solved (and we believe it cannot).
Conclusion
If a drinking water standard for radon is to be adopted, it should be a consistent one. The level should be set no lower than one which would increase the average concentration of indoor air to 200 Bqm-3 because the best observational data indicates that a level lower than that is associated with an increased risk of lung cancer. Citizens should not be forced to forego the benefit of a naturally occurring factor that protects against lung cancer.
Jane M. Orient, M.D., President
Physicians for Civil Defense, Re: W-99-08, p. 4