Table of Contents
If Hormesis Exists...:Implications for Risk Perception and Communication

Paul Slovic, Ph.D.

Decision Research, 1201 Oak Street, Eugene, OR 97401

Phone: (541) 485-2400, Fax: (541) 485-2403

Ironically, as society has expended great effort to make life safer and healthier, many in the public have become more, rather than less, concerned about risk (Slovic, 1987). This is particularly true for involuntary exposure to chemicals, which the public associates to a remarkable extent with danger, cancer, and death. National surveys in the United States, Canada, and France have found that about 70% of the public believe that "If a person is exposed to a chemical that can cause cancer, then that person will probably get cancer some day" (Kraus, Malmfors, and Slovic, 1992; Krewski et al., 1995). About 75% of the respondents in these surveys agreed that "If even a tiny amount of a cancer-producing substance were found in my tap water, I wouldn't drink it." More than 50% agreed that "There is no safe level of exposure to a cancer-causing chemical." This concern that any exposure to a carcinogen, no matter how small, is likely to cause cancer is linked to a desire to avoid chemicals and reduce the risks of exposure to them at any cost. About 75% of the public surveyed agreed that "I try hard to avoid contact with chemicals and chemical products in my daily life." About 62% agreed that "It can never be too expensive to reduce the risks from chemicals."

In recent years, these public attitudes have played a role in many controversies, such as dioxins in the environment, pesticide residues in food, chlorination byproducts in drinking water, and the inadvertent presence of benzene in bottled water.


There are undoubtedly many factors that contribute to public fears of chemicals. One important factor may be the assumption of low-dose linearity that is the foundation for quantitative risk assessment of carcinogens. As regulators have sought to develop more effective ways to meet public demands for a safer and healthier environment, they have come to rely heavily upon quantitative risk assessment based upon animal tests. Such tests often find evidence of cancer at high dose levels. Assuming a linear model implies that there is no level of exposure to a "high-dose carcinogen" that is without some degree of risk. Multiplying even very small probabilities of contracting cancer across large numbers of exposed individuals will likely project at least some number of deaths. This frightens people. Using upper 95% confidence bounds in the linear extrapolation makes the scenario even scarier. Thus, the many people who believe there is no safe level of exposure to a carcinogen may have learned this from hearing about the linearity assumption or seeing risk estimates projected from a linear model.

Psychological and anthropological research helps us understand the nature of the public's fear of exposure to toxic substances that are said (by scientists using a linear model) to be toxic at all levels. For example, Frazer (1959)and Mauss (1972) described a belief, widespread in many cultures, that things that have been in contact with each other may influence each other through transfer of some of their properties via an "essence". Thus, "once in contact, always in contact," even if that contact (exposure) is brief. Rozin, Millman, and Nemeroff (1986) show that this belief system, which they refer to as a "law of contagion", is common in our present culture. The implication of these notions is that even a minute amount of a toxic substance in one's food will be seen as imparting toxicity to the food; any amount of a carcinogenic substance will impart carcinogenicity, etc. The "essence of harm" that is contagious is typically referred to as contamination. Being contaminated clearly has an all-or-none quality to it­like being alive, or pregnant. When a young child drops a sucker on the floor, the brief contact with "dirt" may be seen as contaminating the candy, causing the parent to throw it away rather than washing it off and returning it to the child's mouth. This all-or-none quality irrespective of the degree of exposure is evident in the observation by Erikson (1990) that "To be exposed to radiation or other toxins. . . is to be contaminated in some deep and lasting way, to feel dirtied, tainted, corrupted" (p. 122)

A contagion or contamination model is much more likely to hold in a world in which scientists use linear extrapolation to estimate risks than in a world that recognizes beneficial effects of chemicals at low doses. We do not, for example, view ourselves as being "contaminated" by exposures to prescription drugs.

Another relevant psychological tendency is to confound perception of risk with perception of benefit. If an activity or substance conveys some benefit upon us, we are likely to perceive it as less risky (Alhakami and Slovic, 1994) and more acceptable (Starr, 1969).


If hormesis exists and is accepted by the public, risk perception and communication could be affected profoundly. As noted above, the existence of some beneficial levels of exposure to a high-dose carcinogen may at least partly offset the extreme negative perceptions and contagion effects that would occur with a substance that was "everywhere harmful".

In addition, hormesis implies the existence of a threshold below which effects are beneficial rather than harmful. Linear extrapolations of harmful effects to the point of zero exposure would not be appropriate.

Existence of a threshold means that one can describe a small exposure as equivalent to the No Observed Adverse Effect Level (NOAEL) divided by a safety factor. For example, the exposure of interest may be 100, 1,000, or 10,000 times smaller than the NOAEL.

Iain Purchase and I have done several studies in which we took the same exposure to benzene and expressed its risk in two ways. One expression was based upon a NOAEL and a margin of safety. The second was based on a linear model that produced an estimated additional lifetime risk of cancer of 1 chance in 100,000. Participants in our studies were lay persons and toxicologists. Both of these samples, differing greatly in expertise, judged the probabilistic risk of 1 in 100,000 as far greater and more serious than the equivalent risk expressed in terms of a NOAEL and a safety factor.


The existence of hormesis, if understood and accepted by the public, would have important implications for how risk is communicated and would likely lead to a reduction in concern about exposures to very small amounts of chemicals that have been found to be toxic at high doses.


Alhakami, A. S. and Slovic, P. (1994). A psychological study of the inverse relationship between perceived risk and perceived benefit. Risk Analysis, 14(6) 10851096.

Erikson, K. (1990). Toxic reckoning: Business faces a new kind of fear. Harvard Business Review, January/February, 118126.

Frazer, J. G. (1959). The New Golden Bough: A Study in Magic and Religion. New York: MacMillan (original work published in 1890).

Kraus, N., Malmfors, T., and Slovic, P. (1995). Intuitive toxicology: Expert and lay judgments of chemical risks. Risk Analysis, 12, 215232.

Krewski, D., Slovic, P., Barlett, S., Flynn, J., and Mertz, C. K. (1995). Health risk perception in Canada II: Worldviews, attitudes and opinions. Human and Ecological Risk Assessment, 1(3), 231248.

Mauss, M. (1972). A General Theory of Magic. New York: Norton (original work published in 1902).

Rozin, P., Millman, L., and Nemeroff, L. (1986). Operation of the laws of sympathetic magic in disgust and other domains, Journal of Personality and Social Psychology, 50, 703-712.

Slovic, P. (1987). Perception of risk, Science, 236, 280285.

Starr, C. (1969). Social benefit versus technological risk. Science, 165, 12321238.