Table of Contents
Regulatory Implications of Hormesis
Jeffery A. Foran, Ph.D.,
Executive Director ILSI, Risk Science Institute, Washington, DC 20036,
Tel: (202) 659-3306, Fax: (202) 659-3617
This discussion of the regulatory implications of hormesis focuses on technical and social issues that will determine whether hormesis can be incorporated in the development and implementation of risk-based regulations. Particular attention is given to the development of qualitative and quantitative regulatory limits (e.g., standards or criteria) for water pollutants, hazardous wastes, and pesticides. Hormesis is described, and is used in this article, as a dose response relationship that is stimulatory at low doses, but inhibitory at higher doses (Bailer and Oris, 1998), and as the zone of a dose/response curve, from the no observed effect level (NOEL) to 10-fold below the NOEL, where a beneficial effect exists relative to the background effect (Gaylor 1998).Standards and criteria, as well as qualitative limits on the allowable quantities of potentially toxic chemicals in the environment are typically risk-based and conservatively define the amounts of these substances that can occur in the environment without appreciable risk to ecosystems and human health. They are derived by consideration of chemical hazard or toxicity, potency as described by chemical-specific dose/response (d/r) curves, human or environmental exposure to chemicals, and uncertainty, all in the context of quantitative risk assessment. For non-carcinogens, a reference dose (RfD) or an allowable daily intake (ADI) are typically used as the basis for numeric criteria or standards derived from toxicologic information and quantitative risk assessment methods (see Gaylor 1998 for a description of RfD derivation). Hormesis has not, to date, been a component of the risk-based approach to the development of standards and criteria for environmental pollutants.
There are at least two significant road blocks to incorporating hormesis in risk-based regulations. The numerous scientific and technical questions that need to be addressed will keep a variety of research programs active for many years. Even if these research programs are successful in answering the array of technical questions, strong social objections may be raised and may ultimately prevent the use of hormesis in risk-based regulatory decision making, despite overwhelming scientific support. Both the scientific questions and the social road blocks are discussed in this article.
If regulatory decision-making activities were purely science-based, the incorporation of hormesis, should it exist and be widely and consistently demonstrated, may be a comparatively simple matter of the adjustment of standards, criteria, and other quantitative, risk-based decision points. However, several questions will need to be addressed prior to incorporating hormesis in risk-based regulations. First, can hormesis be quantified and, more important for criteria and standards development, can it be represented by a single number? The development of risk-based standards and criteria requires that risk assessment methods produce quantitative, numeric results with attendant estimates of uncertainty. Hormesis must also be estimated quantitatively if it is to be included in what might now be called risk- and benefit-based standards and criteria. Assuming that both adverse and hormetic effects can be estimated quantitatively, then we must determine how an acceptable level of an individual pollutant is to be identified, considering both toxicity and the chemical's hormetic effects. If toxicity and hormesis can be reconciled quantitatively (as suggested by Gaylor 1998), then additional, more complex questions must be addressed.
It has been widely recognized that numerous chemicals act concurrently to impose stress on ecosystems and their components (including humans), and efforts are underway to estimate risks posed by numerous co-occurring chemicals in the environment. For humans, we may assume that concurrent exposure to compounds that share a common mechanism of toxicity pose some cumulative risk, and that this cumulative risk can be quantified. Assuming each of a group of compounds that share a common mechanism of toxicity also share a common hormetic effect, can a cumulative hormetic effect be described? If so, are the individual hormetic effects additive? Are there synergistic or antagonistic hormetic effects? And ultimately, how are cumulative risks and cumulative hormetic effects to be reconciled? What do we need to know about the shapes and slopes of the dose/response curves for toxicity and hormesis to assess quantitatively the concurrent risks and benefits of these effects? Assuming that humans are constantly exposed to a background level of chemicals, radiation, and other potential stressors, is it possible that human populations are already recipients of hormetic effects, or that some humans are already potentially exposed above a cumulative hormetic zone, and perhaps approaching a zone of (cumulative) toxicity?
Standards and criteria based on risks and benefits of exposure to multiple chemicals do not account for myriad biological and physical stressors that are active in both human and ecological systems. In humans, variability in susceptibility to potentially toxic chemicals has only begun to be addressed. Human susceptibility is influenced by an array of factors such as age, stress, disease-state, and genetic composition. If quantitative risk assessment is to address and be influenced by issues of susceptibility, so must development of quantitative estimates of hormesis. Such technical issues, and many others, will accompany the consideration of hormesis in the development of risk-based standards and criteria.
Even if most of the scientific issues around hormesis can be addressed, there are profound social forces at work that shape regulatory decision-making activities. These forces may, at times, overwhelm scientific rationale for the development of standards and criteria. In fact, the social forces that affect decision-making around toxic chemicals, at virtually all levels (personal, governmental, societal, etc.), are so strong that, I argue, hormesis will not be incorporated, regardless of its scientific validity. For example, the Clean Water Act (CWA) sets a goal of zero discharge of toxic pollutants to surface waters. There has been an explicit recognition that the zero discharge goal of the CWA is not immediately attainable. In the absence of its immediate attainability, two approaches have been created to control the discharge of pollutants from point sources to surface waters - a technology-based approach and a risk-based approach. The technology-based approach requires that best available technology (BAT) economically achievable be implemented in a variety of industrial categories. Where BAT does not adequately protect the environment or human health, then risk-based standards are to be employed through the development of numeric Water Quality Criteria (WQC) that define the maximum level of toxicant in a surface water system that poses an acceptable level of risk to humans and the environment. One interpretation of the CWA is that both BAT and risk-based WQC should be continually strengthened (made more stringent) until the zero-discharge goal of the CWA is attained. Recognition of the limitations of BAT and WQC to reduce pollutant loads to zero has even led to new approaches to pollution control such as pollution prevention (source control through process changes, chemical bans, etc.), waste minimization, and other non-traditional, non-treatment based methods, which, if adopted, may achieve the zero discharge goal.
Significant debate and discussion have ensued around the development and implementation of non-traditional approaches to pollution control. For example, several years of effort were involved in the development of new and innovative approaches to pollution control in the Great Lakes basin, as part of the Great Lakes Water Quality Agreement. These efforts were intended to lead toward the elimination of the discharge of at least some pollutants to the Great Lakes. Given the considerable human resources and time committed to this effort, as well as the strength of the philosophical and political commitment to zero discharge by some of the participants, it is difficult to imagine that some parties to the agreement would be willing to set aside the fruits of their efforts, which were pursued in part through recognition of the special need to protect the Great Lakes from both point and nonpoint sources of pollution. It is even more difficult to imagine that some of the parties would agree that a low level of chemical discharge to the Great Lakes be allowed, in fact encouraged, based on the recognition and acceptance of hormesis. In fact, the concept of desirable (hormetically influenced) levels of chemical discharge, despite its scientific merit, will likely be rejected outright as a return to the concept of acceptable pollution levels, a notion already rejected by a significant portion of the stakeholder community.
Clean-up goals for polluted sites, whether influenced by federal legislation (e.g., Superfund/CERCLA), litigation, or activism, will also be profoundly influenced by considerations of hormesis. Quantitative, risk-based approaches play an important role in the clean-up of polluted sites. Consideration of hormesis in the development of risk/benefit-based approaches to polluted site clean up may be feasible, if quantitative estimates of risk and benefit were the only issues under consideration. However, risk perception, consideration of environmental justice, and other social, non- or quasi-scientific considerations play important roles in determining to what level a site should be restored.
Regardless of their position, all parties involved in site remediation seem to agree that clean-up of most polluted sites should occur to the greatest level possible; that is to the lowest level of pollutant(s) possible. Debate arises around how a low level of pollutant or greatest level of clean-up should be defined. Economics, health and environmental risks, technical feasibility, and an array of other considerations play in this debate, but all are based on the concept that as much pollution as possible should be removed. Hormesis changes the playing field. No longer would the debate be around how much pollution should be removed, but rather, how much should be left to enhance human health or the environment; that is, how much pollution is healthy? The willingness of at least some stakeholders to enter such a debate at its conceptual level would require not only acceptance of hormesis, but rejection of pollution; that is, an acceptance that some level of chemical or other substance no longer makes the environment unfit for or harmful to living things. A portion of the societal mind set is so entrenched around the perception of harm from "pollutants," regardless of their level or concentration in the environment, that any suggestion of potential benefit from pollutants would likely be rejected outright.
Recent approaches to the regulation of pesticides provides a third example of difficulties that will be posed by consideration of hormesis in risk-based regulation. It has been argued that children are uniquely susceptible to the potentially harmful effects of pesticide exposure; thus, an additional uncertainty factor has been proposed for use in risk assessments and management of pesticides. The basis for the use of the additional uncertainty factor is the concern that children may be harmed by exposure to pesticides that are managed by the traditional use of uncertainty factors ranging from 100 to 1,000 (fold below the NOAEL). Use of an additional uncertainty factor of 10 adds a margin of safety that is assumed to protect uniquely susceptible children. The hormetic dose/response relationship displays a stimulatory range of up to 10-fold and a maximum stimulatory response of 30-60% above the control within approximately 5-fold of the NOAEL (Calabrese 1998); thus, the hormetic zone may overlap, or may even be above pesticide exposure levels defined as safe for children, as well as adults (see Gaylor 1998). At least two conclusions can be drawn from this observation. First, the hazards of low level of exposure to pesticides have been overestimated; thus, scientific and regulatory approaches to pesticide management must be revisited. Alternatively, the existence of hormesis requires no adjustment to existing approaches to pesticide management, but simply supports the safety factor approach as reasonably conservative and protective of public health. Given the contentious and very political nature of the current debate, the second outcome seems far more likely.
An array of scientific and technical questions must be addressed before hormesis can be incorporated in the development of risk-based regulations. Even where these questions are answered satisfactorily, social and philosophical road blocks will remain. Because societal concern with and management of pollutants, based on perceptions of risk, often transcend scientific rationale, I predict that the use of hormesis, should it exist, in the development of risk-based standards and criteria will not occur quickly, if at all. Society must accept that exposure to chemicals commonly perceived as toxic is not only safe, but healthy at some doses. Much of society has already accepted this view for pharmaceuticals and vitamins/trace elements. However, acceptance for other substances will be more difficult as it will require overcoming perceptions of the risk of exposure to potentially toxic chemicals at virtually any dose.
Bailer, A.J. and J.T. Oris. 1998. Incorporating hormesis in routine testing of hazards. BELLE Newsletter. 6(3):2
Calabrese, E.J. 1998. Toxicological Implications for Hormesis. BELLE Newsletter. 6(3):1.
Gaylor, D. 1998. Safety assessment with hormetic effects. BELLE Newsletter. 6(3): 6-8.