The myth of the vanishing zero.

Analytical chemists continue to refine their ability to measure minute quantities of potentially carcinogenic environmental contaminants. As a result, the "zero" or "no detect" limit is constantly being pushed downward. Today's zero becomes tomorrow's quantifiable concentration. In effect, analytical detection and quantification levels are becoming vanishingly small and, some believe, absurdly low.

For example, it is currently possible to reliably measure such chemicals as dioxin in the parts per quadrillion (ppq) level and parent pesticides and their metabolites at the parts per trillion (ppt) level. Would a more accurate human cancer risk estimate be produced for a probable human carcinogen, such as dioxin, at a concentration two to three orders of magnitude greater than its level of quantitation than at the quantification level itself? This article will explore this question and determine whether such a belief is a scientific myth.

The myth

We have dubbed the belief that risk professionals can more accurately quantitate or communicate possible carcinogenic health risks associated with "high level" versus "low level" exposures to potential carcinogens as "the myth of the vanishing zero."

Risk professionals routinely perpetuate this myth. For example, a distinguished risk communication expert commented, during a recent lecture, that scientists' ability to accurately explain human health effects associated with exposure to carcinogenic chemicals decreases as laboratory quantification levels reach down into the parts per trillion (ppt) and parts per quadrillion (ppq) ranges. The lecturer was asked, "... if a potentially carcinogenic chemical--now measurable at the ppq or ppt level--were only measurable at the parts per billion (ppb) or parts per million (ppm) level of a decade or more ago, would your ability to provide the public with accurate health risk estimates significantly improve?" He answered "no" and acknowledged that carcinogen risk estimates provided at the ppb or ppm level would probably be no more accurate than at the ppq or ppt level.

From the speaker's response, it appears that when conducting risk assessments for potential human carcinogens, it may be equally valid to estimate human health risks from exposures at the ppq or ppt levels as at the ppb or ppm levels. Why? Addressing this question requires a brief look at the subjective components of the risk assessment process.

Risk assessment calculations estimate human morbidity or mortality on the basis of numerous policy, value and scientific judgments. During the risk assessment process judgments may be made on the:

* experimental design, including the species and strain of test animals, form of the chemical and its delivery vehicle, length of experiment and experimental parameters to be measured;

* data collection techniques and statistical designs to evaluate experimental results;

* carcinogenicity status using the U.S. EPA's weight-of-evidence approach (such rankings determine whether EPA will quantify the risks using a carcinogen approach);

* model(s) to be used to quantitate possible human health risks; and

* acceptable/unacceptable risk levels (e.g., de minimis or one-in-a-million lifetime cancer risk).

This list clearly illustrates that professional judgment, the melding of science, experience and policy, is used by risk assessors to generate input data for use in mathematical models for calculating carcinogenic risks. Use of professional judgment is a "science based" activity, and, as such, it is not a fully objective exercise.

Impetus for quantitating risks

Numerical estimates of risks posed by exposure to potential human carcinogens will continue to be produced so long as risk managers need to regulate a substance or determine when to issue health advisories. Current quantitative risk assessment techniques provide quantitative answers to what appear to be qualitative questions.

While there may be no one correct answer when estimating human health risks from exposure to potentially carcinogenic environmental contaminants, there is usually only one correct regulatory answer. The risk assessment process can yield several risk level concentrations, each equally valid since they rely heavily on professional judgment. The desire for uniformity in scientific approach and the need for regulatory consistency and enforcement (not the search for one and only one correct answer) has, until recently, driven the governmental regulatory system to produce a single appropriate risk estimate. There is currently limited movement towards requiring presentation of several risk estimates, based on different scenarios, in risk assessment reports. However, it is unclear how presenting a range of risks will provide risk managers with a more accurate or clear picture of what are the "true risks" from a site, facility or activity. Given the array of human judgments that shape quantitative risk assessments of potential human carcinogens, human health risks described by a given mathematical approach at the ppm level can be just as valid as those described at the ppq level. This is especially true when estimating human cancer risks by extrapolating from high dose animal experiments to low level human exposures for which no animal or human toxicity data are available. This is especially true when using the assumption that the relationship between carcinogen dose and cancer induction are linear in nature.

Perpetuating the myth

Why is the myth of the vanishing zero being constantly perpetuated? Several possibilities come to mind. Risk assessment is a rapidly evolving art and science that encompasses numerous disciplines. Professionals in one sub-discipline must rely on experts in other segments of the profession to explain their issues. Risk professionals, not fully acquainted with risk modeling, toxicology and analytical chemistry may readily accept the myth as truth when spoken often enough by persons they believe are experts in such matters. In essence, the myth becomes accepted truth when repeated often enough by authority figures.

Second, the myth makes intuitive sense. The less you have of a chemical, the less of a toxic effect you would expect: in fact, the less you have of a chemical, the less you are able to clearly explain its expected toxic action in humans. One of the basic tenets of toxicology is that the dose makes the poison. Stated another way, toxicity is a matter of dose. If a substance is inherently highly toxic, it will be so at what some would term "low levels." It is clear, then, that the dose of a substance that causes an effect (effect = dose x toxicity) in an organism is related to its inherent toxicity, not to a designation of "minute quantities" that implies negligible risk.

As a third possibility, perhaps perpetuation of the myth by some risk professionals and acceptance by the public could serve as a way to decrease public anxiety caused by the finding of any level of a potentially carcinogenic substance in an environmental sample.

Finally, perpetuating the myth might lessen the common public perception that any level of environmental degradation by potential human carcinogens equals an unacceptable human health risk. In fact, it appears that some segments of the public do use any finding of a potentially toxic substance (e.g., analytical reporting limits) as their trigger for personal health concern. When this is the case, presentation of measured chemical levels and associated health risk estimates does little to reduce their anxiety and level of concern.

Based on the concept of such an analytical trigger, environmental professionals can expect increased public anxiety associated with detection of environmental contaminants and demands for control or remediation at concentrations that might be considered well below levels of public health or environmental concern.

Anxiety experienced by some risk professionals who provide the public with information on chemical exposures and potential human health risks is not a function of the so-called risk accuracy issues posed by the "vanishing zero." Rather, it is probably more closely associated with the emotional surge and sometime political pressure exerted by chemically exposed and at-risk individuals or communities with which these professionals must deal. Exposed persons and communities are becoming less likely to have their anxiety quelled by simplistic explanations of relative risk that use the quasi-scientific argument that exposure to low concentrations of a substance translates into negligible risk.

Summary and conclusions

This article has introduced, examined and unmasked the "myth of the vanishing zero." The true purpose of perpetuating the myth may reside in the desire of some persons to establish a "perceptual risk threshold." This threshold would not be based on science, policy or law, but rather on the induced perception that persons exposed to "vanishingly small" quantities of a potentially carcinogenic substance need have no concerns about the health consequences. Put another way, if the concentration of a chemical is sufficiently low, then there is nothing to worry about. In the age of toxic torts, inducing such perceptions in exposed individuals or others could have significant legal and economic consequences.

Professional and public acceptance of the myth can result in unwarranted discounting of valid public concern about human health risks that can occur from exposure to potentially carcinogenic substances in the environment. It can also undermine the validity and usefulness of human health risk assessment, a crucial tool for public health protection. In fact, rather than hampering scientists, improving the detection and quantitation levels for environmental contaminants enables them to better understand the processes by which contaminants move through environmental compartments; in most cases allowing detection well before contaminants reach a level of regulatory or human health concern.

Improved analytical reporting limits can provide risk assessors with better data on which to formulate risk estimates. For example, if a substance can only be quantitated in water at 50 ppb and its level of health concern is 1 ppb, findings of no chemical contamination does not mean that the substance is not there. It could be in the water sample above levels of health concern but be unmeasurable by the analytical technique employed. If analytical methods were minimally capable of measuring chemical residues at their levels of health concern, risk assessors could provide the public estimates of risk with higher levels of confidence.

More accurate predictions of possible human health impacts from exposure to environmental contaminants will become feasible only as toxicological data bases expand and risk assessment models that use such data are refined -- rather than depending on the analytical capability to detect contaminants. Significant improvement in these two areas is unlikely in the near future and regulatory scientists will continue to rely on existing risk assessment procedures to address environmental contamination problems. It is important, therefore, for scientists concerned with comprehending and dealing with environmental contamination to recognize the fallacy of the "vanishing zero" argument: we are no more, or no less, able to assess or explain the actual human health impacts of many carcinogenic environmental contaminants at the ppq level than at the ppm level.

Discomfort with continually improving analytical capabilities to detect and quantify minute levels of potentially carcinogenic environmental contaminants results, instead, from burgeoning public concern generated by findings of what appears to be any level of these substances. In many instances, findings of any reportable concentration of such materials becomes manifested as a cycle of citizen political action, extensive media coverage, time consuming and costly litigation, imposition of fines or judgments against the polluting party, and finally the promulgation of new or tighter environmental pollution control laws along with their ancillary regulatory requirements and costs.

As a profession it is time we addressed the substantive questions posed by the "vanishing zero" myth rather than diverting limited resources to address red herrings. The credibility of risk professionals and public support for rational risk management decisions based on prudent risk assessment findings hang in the balance.

David A. Belluck, Ph.D., Risk Writers, Ltd., 3108 46th Ave. South, Minneapolis, MN 55406.