Regulatory Implementation and Indeterminate Hormesis Effects

Howard A. Latin

Professor of Law and Justice Francis Scholar

Rutgers University School of Law at Newark

123 Washington Street

Newark, NJ 07102


How do we get from here to there without losing our way? The social policy and legal literature on toxic substances control contains an avalanche of bad analyses and bad recommendations resulting from the failure to grapple with "how do we get there?" issues in a rigorous and realistic manner. Innumerable commentaries have identified some kind of desirable goaloptimally efficient regulation, equitable distribution of regulatory burdens to different classes or groups of the population, or the inclusion of a growing body of scientific knowledge in diverse toxic regulatory processesand then the commentaries discuss the theoretical benefits of attaining their idealized goal without any serious effort to examine critical implementation problems, or "how do we get there from here?"

The majority of my writing on toxic substances and pollution regulation focuses on difficult, often insurmountable, regulatory implementation constraints,1, 2 including inadequate scientific understanding, the absence of scientifically testable hypotheses, inadequate data on toxic exposures and effects, inadequate administrative funding and personnel, inadequate political support for agencies to withstand constant pressures from affected private-interest groups, inadequate bureaucratic rewards for innovation and courage, and inadequate staff time and attention in light of the overlapping multiplicity of administrative responsibilities. Could anyone offer sensible proposals for revising regulatory strategies without careful consideration of these barriers to effective regulatory implementation? I think not, and yet many commentators do submit "regulatory reform" proposals based on nothing more than the desirability of achieving some theoretical goal.

As one minor example of this unrewarding genre, consider the article by Frank Cross on the "Legal Implications of Hormesis" in this volume. Professor Cross devotes a few pages to describing the unsatisfactory efforts of regulatory agencies and courts to create regulatory standards ensuring an "ample margin of safety" or "acceptable margin of safety" for toxic exposures at relatively high dosage levels. Yet, Cross nonetheless concludes that: "Consideration of hormetic effects could distinctly enhance the benefits of public health regulation and might even provide the sort of dramatic change that could provoke a legislative breakthrough or other action to restructure environmental regulation and offer additional benefits." Cross also opines that: "The precautionary nature of conservative linear risk extrapolation, though, disappears in the presence of hormesis." And he contends that "the evidence for hormesis is sufficiently strong that government should alter its default presumption of linear dose-response models."

I disagree completely with all of these contentions, but the remarkable (though unfortunately common) aspect is that Professor Cross does not even attempt to discuss whether the implementation constraints that have prevented agencies from successfully regulating relatively high-level toxic dosages would also impede regulatory responses to the low-level dosages associated with hormetic effects. If pollution control agencies in most toxic substances contexts have been unable to identify an adequate margin of safety for higher toxic dosages, it is difficult for me to believe that those agencies would be able to identify a reasonable margin of safety for the much smaller dosages at which hormetic effects might be worth considering from a legal perspective.

Space limitations on this essay preclude a thorough debunking of the idealized conclusions offered by Professor Cross, but even a few moments of introspection should raise serious doubts about many exceptionally difficult implementation problems, no matter how enthralled readers may be with the science of hormesis.

Ambient Exposures

Thousands of tons of air pollutants, water pollutants, pesticides, and herbicides are discharged each year. Millions of people are exposed to radiation from the sun, radon contamination, dental x-ray machines, radium watch dials, and various other natural and man-made sources. Toxic substances may be released when buildings are demolished or when property where toxic wastes have been stored is disturbed in some manner. People breathe, drink, touch, and are otherwise exposed to countless small toxic exposures each year. It is therefore certain that a very large number of people have already experienced hormetic effects as a result of ambient toxic exposures. Can scientists, to say nothing of regulators, assess what the marginal hormetic effects may be for any person when a given toxic exposure is observed?

It should be obvious that different people are regularly exposed to different levels of different toxic substances in different places, and they will doubtless have different immunological responses and cell adaptations to these disparate exposures. Yet, we cannot have a separate monitoring system and a separate regulatory regime for each person. One person may derive hormetic benefits from a specific exposure while the person next door may suffer increased toxic health hazards because an identical toxic exposure puts them past the "tipping point" as a result of their unique biological and experiential characteristics. Under real-world conditions of diversity of exposures and variability of effects, the scientific finding that some toxic exposures may yield beneficial hormetic effects at low exposure levels provides no support whatever for any conclusion that toxic regulatory standards should ever be relaxed because of the chance of beneficial hormetic effects.

Even if we could identify the average exposure level for average people at which a given toxin produces desirable hormetic effects, we would seldom be willing to adjust toxic emissions regulations based purely on averages. A year ago the EPA strengthened air pollution standards because children are more sensitive to adverse health effects than the population as a whole, and the agency decided on policy grounds to increase the protection for this sympathetic sub-population. Pregnant women, asthmatics, people with pre-existing illnesses, the elderly, sub-populations with genetic propensities to contract certain diseases, and perhaps other sub-populations would also need to be considered in any legal decision to reduce the severity of toxic standards on a hormetic benefits rationale.

Another factor that might preclude adjusting toxic substances regulatory standards in pursuit of promoting hormetic benefits is the uneven geographic distributions of many toxics. Discharges from New Jersey, for example, are blown by prevailing wind patterns to some of the wealthiest areas of Connecticut, creating a distinctive cancer alley in a region where direct toxic emissions are rare. If the stringency of regulatory standards in New Jersey were reduced to promote hormetic benefits, this administrative decision may exacerbate health hazards in Paul Newman's hometown. We cannot have a separate regulatory regime for each square foot or each square mile of potentially contaminated territory, and micro-climates or micro-environments will complicate any risk assessment of toxic health hazards versus possible hormetic benefits.

I do not believe we can expect regulatory agencies to make these kinds of fine distinctions in a scientifically credible and logically coherent fashion even if they were capable of assessing low-exposure toxic hazards and countervailing hormetic benefits for different exposures to different toxins for different groups in different areas. And I am certain that real regulatory agencies cannot come anywhere close to the required degree of analytical precision.3

Multiple and Repetitive Exposures

Let us presume, purely for the sake of argument, that some regulatory agency can assess the comprehensive effects of low-level toxic exposures and the corresponding benefits from hormesis. If the agency sets a discharge standard at the optimal level, as Professor Cross hypothesizes, what would happen the next time the individual is exposed to more of the same toxin? Once a person has enjoyed her fair share of hormetic benefits, we cannot enclose her in a bubble to insulate her from subsequent exposures that would push her past the tipping point. Toxic exposures, cell adaptations, hormesis benefits: surely these all require a extraordinarily complex assessment of dynamically changing conditions, not a static snap-shot of exposure effects and hormetic benefits at any given instant in time. Can regulatory agencies make these complex assessments? No, their analytical capacities and fortitude under criticism have usually proven inadequate to make required determinations for the relatively easier judgments required by high-level toxic exposures. In light of the potentially severe health hazards posed by higher levels of exposures, it would not seem to me to be a wise allocation of limited resources for agencies even to attempt risk assessments of potential hormetic effects.

We have not yet come even close to the degree of complexity and uncertainty that any agency would have to confront if it wished to consider hormetic effects in a serious manner. If a given cell in a given person is stressed by a given toxic substance, it would appear impossible to assess the harmful or beneficial effects without knowing whether that cell was simultaneously or sequentially exposed to many other substances that may produce synergistic or co-carcinogenic effects at varying doses. If a given cell has not yet reached the tipping point for a given toxic, exposures to other toxic substances may exacerbate the stress with variable results. In effect, each individual low-level exposure may be below the individually calculated tipping point for that substance, but interactions between multiple exposures may push any or all toxic exposures well beyond the tipping point. We cannot have a practical regulatory regime that attempts to assess multiple interactions of multiple toxins under varied biological and external conditions.

Detectable Limits of Acceptable Exposures

We cannot put people in test tubes and measure actual cell adaptations to actual toxic exposures in an environment where exposures to many different toxics are frequent and unavoidable. The scale at which regulatory agencies must address actual discharges and human exposures may be altogether different from the scale at which scientists can measure hormetic effects and hypothesize about tipping points.

Regulatory standards are very rarely framed in terms of human exposures; they are instead written to place allowable limits on toxic discharges. The standards impose emissions limits of tons per year, pounds per day, milligrams or milliliters per cubic meter of air or water, or occasionally parts per million. These toxic discharge limits may be thousands of times larger in scale than the molecular or cell level at which potential hormesis effects must be evaluated. In effect, the smallest dosage allowed by a given regulatory program for a toxic substance may be so large that allowable exposures almost always exceed the tipping point. In this case, the existence of hormetic benefits again would provide no rationale for increasing the allowable discharge limits for a substance.

The extrapolation from discharges to dosages has frequently been controversial even for relatively large dosages. There has, for example, been a debate for decades about the relationship of lead levels in air to lead levels in blood, which was a central uncertainty in the regulatory decision about how strict the national ambient air quality standards for lead should be. Large discharges may or may not become large dosages based on a wide variety of circumstances. Small discharges may sometimes produce large dosages if exposure is frequent or significant bioaccumulation occurs. To the extent virtually all of the current toxic substances and pollution control regulations are based on discharge limits, not on dosage limits, Professor Cross' contention that increasing the scope of hormetic benefits would sometimes justify making toxic discharge regulations less stringent is speculative to say the least.


The theory of hormesis belongs in a scientist's laboratory, not in a regulatory proceeding. I do not believe hormesis will, or should, have any practical ramifications until risk-assessment science has progressed far beyond its current state.

Even if the current state of scientific understanding were sufficient to assess the dimensions of hormesis benefits for each unique individual in our society, which is far from true, we cannot have a practical regulatory system that depends on making millions of individualized risk assessments. Once we acknowledge that toxic substances control must be accomplished for very large numbers of exposed individuals in a limited number of imperfect regulatory proceedings, there is clearly no reason to conclude that attempts to attain hormetic benefits for some peopleif we could achieve that goal reliably would ever serve as an acceptable rationale for increasing allowable discharge limits and corresponding toxic risks for countless other individuals.

This conclusion does not question the validity of recent scientific findings on hormetic processes and benefits. Science and regulation are fundamentally different enterprises operating under different conditions and constraints.4 Until scientists can reliably assess high-level toxic exposures that may produce severe health effects, I see no realistic possibility that scientific risk assessments of low-level exposures with potential hormetic benefits can be translated into regulatory implementation practices that are administrable and wise. Surely, our overburdened regulatory agencies with their limited resources have better issues to focus on at this time.


1. Latin, HA. Ideal Versus Real Regulatory Efficiency: Implementation of Uniform Standards and "Fine-Tuning" Regulatory Reforms. Stanford Law Review 1985; 37: 1267-1332.

2. Latin, HA. Regulatory Failure, Administrative Incentives, and the New Clean Air Act. Environmental Law 1991; 21:1647-1720.

3. Latin, HA. Good Science, Bad Regulation, and Toxic Risk Assessment. Yale J. on Regulation 1989; 5: 89-.

4. Latin, HA. The "Significance" of Toxic Health Risks: An Essay on Legal Decisionmaking Under Uncertainty. Ecology Law Quarterly 1982; 10: 339-.