Is the Current Risk Assessment Paradigm Used by U.S. EPA and Others Compatible with the Concept of Hormesis?

Kenneth A. Poirier, Ph.D. and Michael L. Dourson, Ph.D., D.A.B.T.,


4303 Hamilton Avenue

Cincinnati, Ohio 45223

Phone: 513-542-7475 x14

Fax: 513-542-7487



The use of the U.S. EPA's risk assessment paradigm (found in many of its published risk assessment guidelines) generally considers the biological effects of low level exposures in its risk assessment deliberations. However, this may not include the cluster of doses that constitute an hormetic response. Typically, the process for non-cancer endpoints determines a critical adverse effect from a dose response and identifies the critical No Observed (Adverse) Effect Level (NOAEL) and Low Observed (Adverse) Effect Level (LOAEL). The non-cancer risk assessment process assumes a threshold. In fact, low dose extrapolation occurs routinely in EPA's estimation of Reference Doses (RfDs) and Reference Concentrations (RfCs), these estimates being synonymous in practice with subthreshold doses. However, the process does not necessarily take into account the possible beneficial effects that are taking place at the low end of the dose response curve, in fact these effects may be overlooked during the construction of a dose response curve and the determination of the appropriate effect levels.

While with non-cancer toxicity risk assessments the evidence of hormesis is generally ignored, this might have more to do with the current RfD/RfC model rather than a disregard of hormesis. Since the RfD or RfC is attempting to find a dose below the toxicity threshold, a yet lower dose that enhances the organism's response to the toxicity of a chemical is generally irrelevant. Or perhaps another way to say this is that evidence of hormesis at doses lower than the RfD or RfC is entirely consistent with the definitions of the RfD and RfC. In its consideration of the biological effects of low level exposures, specific concerns of the U.S. EPA encompass low dose extrapolation of toxic response, evidence of micro- or macro-nutrient needs, and hormesis. However, the primary concern and charge of the U.S. EPA is low dose extrapolation of a toxic response.

Evidence of toxicity at doses above and below the threshold of a chemical dose exists, for example, with the micro-nutrients (i.e., essential trace elements). In these cases, EPA estimations of a RfD consider the "toxicities" on both sides of the threshold region, recognizing that in some individuals, such a region may be small or even absent. Examples of this consideration of benefit and toxicity can be found in EPA's IRIS Integrated Risk Information System (IRIS) ( Specifically the non-cancer risk assessments of the essential trace elements best illustrates this type of risk assessment (e.g., manganese, selenium and zinc).

The risk assessment for selenium is a good example of how the consideration of the total database may affect a risk assessment. The dichotomous nature of selenium was not fully realized until the 1970s with the demonstration that it was a required component of glutathione peroxidase and the subsequent establishment of the essentiality of selenium in humans. Prior to the establishment of this trace element's essentiality, selenium had been known to be a potent livestock toxicant. In more recent years, selenium toxicity in humans had become better characterized and was the basis for the development of the RfD for selenium. As the U.S. EPA deliberated the development of the selenium oral non-cancer risk assessment, the essentiality of selenium was considered as part of the risk assessment paradigm. The final risk assessment value provided about a 5-fold margin of exposure above the Recommended Dietary Allowance. It should be pointed out that the prophylatic (antitumorigenic and anticarcinogenic) properties of selenium were not considered in this analysis and in future deliberations could add another level of complexity to the analysis.

The dose response curve of this risk assessment is much like a U-shaped or inverted U-shaped relationship. The right side of the curve representing the ill effects of deficiency. The approaching trough (or crest in an inverted U-shaped curve) represents the region of adequacy and non-toxicity and finally to the left side of the curve which represents the toxicological dose response. Thus, the RfD for an essential trace element is very much like a traditional hormetic response but over a wider range of doses. The paradigm thus does include the provision for considering hormetic and hormetic-like responses, but they must be actively considered when conducting the risk assessment.

In the case of selenium, the human data was sufficient to derive the RfD without any special considerations to account for the contribution of essentiality. The RfD for zinc presented a special challenge to the risk assessors. The paradigm drove the calculated RfD below the level of zinc requirement for pregnant and lactating females. The apparent contradiction of safety and essentiality was handled by the risk assessors by a statement to the effect that in certain physiological states, excursions of zinc ingestion above the RfD are acceptable. The statement is as follows:

"This RfD for the soluble salts of zinc supplies adequate zinc to meet the requirements in adolescents and adults over a lifetime without any concurrent physiological impairment. It does not supply the Recommended Daily Allowance (RDA) to those members of the population who have greater requirements for a short, less-than-lifetime duration, for example, infants, pre-adolescent children, or, possibly, lactating women. For short-term requirements in infants, pre-adolescent children, and lactating females, refer to the RDAs (NRC, 1989)."

These RfD examples illustrate the latitude of the RfD/RfC paradigm in addressing unusual situations such as a hormetic response to a chemical. The provision for applying best scientific judgement in the derivation of a risk assessment value will allow the risk assessor to appropriately consider hormetic phenomena in traditional non-cancer risk assessments.


For cancer effects, an assumption is held that a threshold for an adverse effect does not exist with most individual chemicals. This is also based primarily on theory. Low dose extrapolation occurs routinely in EPA's estimation of cancer slope factors, risk specific dose (sometimes referred to as RSDs ­ these estimates being associated with various upper-bound, lifetime cancer risks), and unit risks which are equivalent to the RSD in the particular environmental media of interest (e.g., drinking water). Slope factors and resulting associated quantitative values (i.e., RSDs and unit risks) have been estimated for the cancer toxicity of hundreds of chemicals and are listed on EPA's IRIS.

Because of the methodology employed, a hormetic response may easily be overlooked in the risk assessment. Sometimes evidence of hormesis may be ignored or considered spurious. However, information on lower cancer incidences than control at lower doses is almost always used in the dose response model to develop the slope factor. In these situations, the hormetic response is used indirectly to affect the estimation of cancer risk although it is not explicitly stated as such. EPA's draft cancer risk assessment guidelines (EPA, 1996)1 specifically allow the incorporation of new information on modes of action into the determination of cancer risk. As more experience is gained with these guidelines, such mode of action information may include the explicit use of the hormetic response.

The evidence of an increase in non-cancer toxicity at doses lower than a specified level of cancer risk also exists, for example, with the essential trace element nutrient selenium. In these cases, EPA often chooses not to estimate a cancer slope factor. The existing cancer risk assessment guidelines allow the risk assessor to consider data in the low dose extrapolation area of the cancer dose response curve. If data were to be available to illustrate a hormetic effect for a cancer endpoint, the consideration of that data can and should be considered as part of the risk assessment.


As with many areas of science, new data force federal agencies and others to re-examine their methods for low dose extrapolations of health risk. Moreover, new models are under development that combine the estimation of health risk of both cancer and noncancer endpoints. This model development should more formally include consideration of hormesis, within the existing framework of biological effects of low level exposure.

The U.S. EPA risk assessment paradigm is compatible with the concept of hormesis. However, the risk assessor's analysis must make an active consideration of the data and the application of that data in the low dose portion of the dose response curve. The process allows for the consideration of specific and unusual responses (read hormesis), but the process must recognize the uniqueness of the data prior to initiation of risk assessment derivation.

1 U.S. EPA. 1996a. Proposed guidelines for carcinogen risk assessment. Federal Register 61(79): 17960-18011.