The Precautionary Principle: Is It a Threat to Toxicological Science?

The central figure in developing the 1970 US Clean Air Act, Senator Edmund Muskie, rather plaintively asked that the scientific information underlying this act be obtained from a one-handed scientist. He was tired of hearing scientists saying “on the one hand . . ., yet on the other hand. . . . ” The Senator was expressing the desire of lawmakers for indisputable scientific facts on which to base regulatory actions. Unfortunately, this is not the way science works, particularly in situations where there is uncertainty concerning the questions being asked, or the terms being used.

Lack of clarity in definition or consistency in use is a particular problem when considering the Precautionary Principle. My discussion of the impact of the Precautionary Principle will inevitably fall into the “on the one hand . . ., yet on the other hand . . . ” category that made Senator Muskie so unhappy. I will argue that the precautionary principle is both good public health and that it is causing public health problems; that it will both increase and decrease the importance of toxicological science in decision making; and that the European Community (EC) is both using the Precautionary Principle in some innovative ways to protect public health and the environment, and that the EC is intentionally abusing the precautionary principle by using it as a means to erect artificial trade barriers not warranted by toxicological science and risk analysis.

DEFINITIONS OF THE PRECAUTIONARY PRINCIPLE

The precautionary principle was developed primarily in Europe, beginning in the 1980s in Germany. One of the first times it received major prominence was in 1992 at the United Nations Conference on Environment and Development in Rio de Janeiro (United Nations 1992). The primary goal of this meeting was to establish an environmental agenda that would guide nations toward sustainable development. At this meeting the precautionary principle was adopted in the following form:

“Nations shall use the precautionary approach to protect the environment. Where there are threats of serious or irreversible damage, scientific uncertainty shall not be used to postpone cost-effective measures to prevent environmental degradation.”

It is informative to compare this definition with the more recent Wingspread Statement:

“When an activity raises threat upon to human health or the environment, precautionary measures should be taken even if some cause and effect relationships are not fully established scientifically” (Raffensperger and Tickner 1999).

Wingspread is notable for a number of differences from the earlier Rio declaration. The Rio declaration includes a statement that the threat of harm should be serious and irreversible, but no threat of harm is specified in Wingspread; the Rio declaration states that the action should be cost-effective, but cost-effectiveness is not part of the Wingspread statement; the whole thrust of the Wingspread statement is positive in that measures should be taken, whereas in Rio it is negative in that the lack of uncertainty should not postpone taking measures; and finally the Rio statement focuses solely on the environment whereas the Wingspread statement goes beyond the environment to the broad field of public health (Goldstein and Carruth 2004).

The problem of appropriately defining the precautionary principle is not simply academic. Like sustainable development, as a concept the precautionary principle is supportable by everyone. But the precautionary principle also has been adopted in various national and international treaties and by governmental entities in a way that provides legal teeth to what is otherwise a nebulous policy concept. The Wingspread approach is beginning to take hold in the United States where the Precautionary Principle has been approved by the San Francisco and Berkeley City Councils and is under consideration by other local authorities. Where laws and actions are at stake, definitions become very important. The European approach, which often infuriates its trading partners including the United States (Codex Alimentarius Commission 2000; Miller and Conko 2000), is to state that the precautionary principle is enshrined within European policy—but to never define it (Commission of the European Communities 2000). Rather, a policy or action is judged as to whether it will result in a precautionary approach. This has led to what Carruth and I call the Cynical American Definition:

“The Precautionary Principle is a nebulous doctrine invented by Europeans as a means to erect a trade barrier against any item produced more efficiently elsewhere” (Goldstein and Carruth 2004).

There are many different actions that can be taken under the heading of the precautionary principle, or the precautionary approach as it is often described. It is useful to consider such actions under the public health and prevention nomenclature of primary and secondary prevention. Simply put, primary prevention is an approach that focuses on there never being a problem in the first place; e.g., an individual does not start smoking or, even better, there are no cigarettes being produced. Secondary prevention depends upon the early detection of a problem, such as detecting high blood pressure before someone has a stroke or is otherwise symptomatic. Much of what is done as a result of environmental risk assessment and risk management is secondary prevention (Goldstein 2004), particularly when it is related to the standard approaches to dealing with known toxic chemicals. Many risk assessors would argue that the precautionary principle is already part of risk assessment (Starr 2003). Among the potentially precautionary standard risk assessment practices are 10-fold “safety” factors; the 95% upper confidence limit; and the use of exposure models containing prudent default assumptions. The primary preventive approaches of the precautionary principle go well beyond more stringent risk assessment. Its components include taking preventive action in the face of uncertainty; shifting the burden of proof to the proponents of an activity such as industry of government; exploring a wide range of alternatives; and increasing public involvement in decision making (Nussbaum et al. 2004; Raffensperger and Tickner 1999; Tickner 2002).

THE PRECAUTIONARY PRINCIPLE AND WORLD TRADE

The implications of how to define the precautionary principle have nowhere been more evident than for some of the trade issues to which it is being applied. In order to keep a level playing field, various world trade agreements depend upon harmonizing risk assessments so that a nation cannot just arbitrarily decide that another nation’s product is unhealthy. Unfortunately, there are a variety of examples in which it appears that the European community has used the precautionary principle as a means to exclude items from elsewhere in the world (Goldstein and Carruth 2004; Majone 2002). Before describing these as inappropriate uses of toxicological science, let me first emphasize that we in the United States have much to learn from the EC as it wrestles with core issues of democracy that impact on the control of environmental hazards. The current debates about the governance relationships among citizens, states, and the multistate European confederation are unparalleled in the US since the Federalist Papers; and are of profound global importance.

As I do not want current broader EU-US issues to cloud the discussion, I will start with an issue that does not involve the United States. An example of action based upon the precautionary principle that fits under the heading of secondary prevention is the adoption by the EC of a very stringent standard for aflatoxin. This was done in essence by adding an additional safety factor for the genotoxicity of aflatoxins. For all nondairy products designated for human consumption, the EC standard is one-fifth that of the United States and for milk it is one-tenth of that of the United States. This has resulted in the exclusion of an estimated $700 million yearly of trade from Sub-Saharan Africa to the advantage of European growers. These Sub-Saharan nations are among the poorest in the world, with relatively fixed trade routes following old colonial patterns. The difference in risk is less than one cancer per year in Europe. The Joint FAO/WHO Expert Committee on Food Additives (JECFA), the expert body of the Food and Agricultural Organization and the World Health Organization, at the request of the Codex Alimentarius Commission, reviewed the difference between the U.S. and European standard and found that it was of no health significance (International Program on Chemical Safety 2001). There is also concern that the EC standard is halting the usual harvesting of Brazil nuts in the Amazon area with the result that the trees are being cut down rather than providing resources for indigenous groups (Newing and Harrop 2000).

An example of a trade barrier related to primary prevention aspects of the precautionary principle is that of the EU’s ban on beef from cattle that have previously been treated with estrogenic growth agents (EC Measures Concerning Meat and Meat Products (Hormones) 1998; EC Measures Concerning Meat and Meat Products 1997). Canada and the US brought this issue to the World Trade Organization (WTO) who decided in their favor. The EU argued that despite the lack of recognition of risk by JECFA, or even its own scientific bodies, the precautionary principle permitted them to ban beef from hormone-treated animals as a risk could not be ruled out, and that a formal risk assessment was not needed. Among the arguments advanced by Canada and the US were that the EC was inconsistent in not having limits on natural estrogens, and was inconsistent in permitting the use in swine of antibiotics whose residues in pork were carcinogenic in laboratory animals. The WTO ruling basically said that the precautionary principle was not yet generally accepted as a legal doctrine capable of supplanting risk assessment. Much of the subsequent EU activity in justifying the precautionary principle can be seen as a means of reversing the WTO opinion (Rogers 2000; European Environmental Agency 2001). The use of the precautionary principle as a trade barrier will likely be central to the forthcoming WTO hearings on the EU’s ban of genetically modified foods.

The use of a primary preventive precautionary approach also frequently occurs in US law. An interesting example demonstrating a switch to primary precaution is the hazardous air pollutant (HAP) provisions of the 1990 Clean Air Act amendments (Goldstein and Carruth 2004). Prior to 1990, to regulate a compound as a HAP, the Environmental Protection Agency (EPA) had to go through an involved and time-consuming process. It began by listing a compound as one reasonably anticipated to cause adverse health effects (and is not regulated by setting an ambient standard). After suitable hearings, the EPA would then use risk-based and other considerations to decide which sources of the listed compound to regulate. The burden of proof was on the government to show the potential for adverse consequences. Relatively few compounds had been regulated under this approach, primarily those believed to be human carcinogens. in 1990, faced with impatience at this slow approach, and the recognition that common air contaminants responsible for many tons of emissions were uncontrolled, Congress dramatically changed the HAP regulatory process (Clean Air Act Amendments of 1990). Congress listed over 180 compounds to be regulated for which the only rule-making procedure was the onerous one of attempting to remove them from the list by demonstrating safety, i.e., the burden of proof was switched to the industry. Further, Congress specified that emission sources should be required to use the maximum available control technology (MACT), reasoning that all sources should do as well as the best (defined operationally as the upper 12th percentile). Risk considerations are taken into account only secondarily. If, after MACT, there is still a one in a million risk to the maximally exposed individual (MEI), additional steps are to be taken. These steps are currently under consideration and likely will go through court challenges in the next few years. The use of the MEI as the target in a public health bill is problematic—it does not matter whether the source is in the middle of a populous major metropolitan area or upwind of the Mojave Desert (Goldstein 1989).

If a similar switch in the burden of proof, and away from risk based approaches, was under consideration today, the debate would be phrased in terms of the precautionary principle. As it is now 15 years from the passage of the 1990 Clean Air Act amendments, the change in the HAP provisions allows an opportunity to consider the effectiveness of the precautionary principle in action. Carruth and I have pointed out potential public health shortfalls to the precautionary approach taken in the HAP amendments (Goldstein and Carruth 2003b). From the vantage point of a toxicologist interested in understanding the mechanism of chemical effects, the most important is that once a compound is listed, why would there be any interest in studying it further? In fact, there does appear to be a decline in EPA’s research interest in HAPs in recent years. Another issue is that the compounds on the HAP list have varying degrees of toxicity, e.g., benzene and toluene. But if they are both on the list and subject to the same controls, what is the incentive to use the least toxic agent? Also, nonlisted compounds are likely to be in the large universe of compounds for which there is less toxicological information and even more likelihood for unwanted surprise. Yet the HAP amendments may well push toward use of nonlisted compounds as it allows the industry source to avoid regulation. There is also a concern that the MACT standards, which have taken years and much debate to establish, will inhibit investment in even better technology—for example, air pollution control technology that might result from an advance in materials science. All of these concerns can be counteracted by provisions enacted into the 1990 HAP amendments—but whether they will be is uncertain. Accordingly, despite the 1990 Clean Air Act HAP amendments likely leading to a decrease in tonnage of air pollutants emitted, it is still questionable whether there will be any significant lessening of toxicity directly caused by HAPs.

REGISTRATION EVALUATION AND AUTHORIZATION OF CHEMICALS (REACH)

Currently under consideration by the EU is a new far-reaching act that is firmly based upon the precautionary principle (Registration, Evaluation, and Authorisation of Chemicals 2005). It is a work in progress, with amendments being acted upon that may change the details. REACH does have provisions for prioritization of chemicals based upon exposure, or based upon hazard—but not based upon risk. In many ways, REACH is the antithesis of the US Toxic Substances Control Act through its requirement of a substantial amount of safety data for all chemicals, new or existing in commerce. Through this requirement it presumably will counteract the major problems we have had in the US with existing chemicals, such as the counterproductive move toward using MTBE and other oxygenated fuels without adequate toxicological testing (Goldstein and Erdal 2000). Also, through its insistence on the testing of all chemicals, REACH presumably will avoid the potential problem posed by unlisted chemicals in the HAP provisions. Like the 1990 US Clean Air Act HAP amendments, REACH is justified by its proponents in part on the basis of frustration with the slowness of existing regulatory approaches. This expectation of a speedier regulatory response has not been borne out for the US HAP amendments—15 years after the passage of the act much remains unfinished, including the inevitable court cases.

Alan Boobis has pointed out that there are far fewer toxicologists in Europe than in the US, and has questioned whether there are sufficient toxicologists to accomplish the task (Boobis 2005). Accordingly, REACH may greatly increase the demand for toxicologists; although in my view, much of these funds would be better spent on developing the scientific basis for newer equivalents of the Ames test that would be effective in primary prevention. A similar criticism can be made of the current approach to testing high production volume chemicals (US Environmental Protection Agency 2005).

TOXICOLOGY AS PRIMARY PREVENTION

Toxicology should be given more recognition as a primary precautionary approach. An excellent example is the Ames test, which was developed based upon scientific advances in understanding basic bacterial genetics; the role of mutagenesis in carcinogenesis; and the importance of liver metabolism in activating indirect carcinogens. To develop new consumer and industrial products, the chemical industry routinely uses the Ames test, and similar tests, to weed out potentially harmful chemicals from their development processes. The focusing of new chemical development and marketing on less harmful products is an excellent example of a primary preventive precautionary approach achieved through advances in toxicological sciences (Goldstein and Carruth 2003a; Goldstein 2004).

Unfortunately, the contributions of toxicology to primary precautionary approaches through safety assessment are not well understood by many who advocate the precautionary principle. Some are distrustful of scientists and are opposed to what they see as a technocracy that is controlling their lives. This is not surprising, in view of the frequency with which industry or government has used scientific uncertainty as a basis to stall needed actions (Rampton and Stauber 2002; Rosner and Markowitz 2002). The justifiable concern is that without a precautionary approach, industry will operate under the CATNIP principle—Cheapest Available Technology Not Involving Prosecution.

COMMUNICATION WITH THOSE WHO ADVOCATE THE PRECAUTIONARY PRINCIPLE

Potentially problematic is the lack of communication between those active in the field of toxicology or environmental risk assessment and those most strongly advocating the precautionary principle Of note is that of the 34 signers of the Wingspread Statement on the Precautionary Principle described above, none appear to have been members of the Society of Toxicology or of the Society for Risk Analysis (Goldstein 2004). This lack of communication among different disciplines concerned with environmental protection may well limit understanding of the role of toxicological science in preventing disease. As this could be of grave consequence to our field, our reaching out to those who are advocates of the precautionary principle is particularly important. Let me emphasize that many, and perhaps most, active supporters of the precautionary principle are well aware of the role of science in preventing disease, and that there is a growing literature on the subject of science and the precautionary principle (Kriebel et al. 2001; Foster, Vecchia, and Repacholi 2000; Goldstein 1999; Cranor 2003).

CONCLUSION

Among the major external threats to the field of toxicology are animal rights activists who would rather put humans and animal pets at risk rather than subject any laboratory animal to study; so called ethicists who seem opposed to any controlled experimental exposures of human subjects to chemical agents, even at levels at which we are exposed in our homes and in gasoline stations; the whole field of toxic torts; and, I would add, the precautionary principle. Each of these threats has their potentially valuable side in promoting the use of toxicology or developing new approaches—such as in vitro techniques that can replace laboratory animals. Similarly, although the precautionary principle can be a threat to the field of toxicology, it also challenges us to think carefully about the goals of toxicology and of the importance of broadly communicating about what we do and why we do it.