Testing: How much is enough?
Laboratory tests performed on animals is a troubling issue. See our sidebar for more information.
Chemical products provide many benefits to society, but they must be managed in a responsible way to minimize any adverse effects on humans and the environment. The goal of testing chemical products is to determine their inherent hazards, so we can decide what risk management actions are needed to minimize potential harm. How much testing is required to achieve this goal must be based on the educated judgment of qualified toxicologists and environmental scientists.
At The Dow Chemical Company, we have adopted a tiered testing approach that allows us to decide the appropriate studies to conduct. This product stewardship-driven approach, combined with government regulations, should provide the public with increased confidence and assure them that human health and the environment are protected.
Ground rules for testing vary according to the intended use of the product. Chemicals designed as "active" ingredients for pesticidal or pharmaceutical uses typically require a more extensive set of tests than do industrial chemicals, for several good reasons. Products that combat pests in food production, kill bacteria or speed healing in the body may be ingested or otherwise absorbed by humans. In addition, pesticides and pharmaceuticals are designed to be active in biological systems, so frequently they are inherently more hazardous. Consequently, it is important to have a full understanding of their potential impact. How much pesticide residue can remain on a food item? What side effects will a drug have, or what effects may result from improper use?
Industrial chemicals, on the other hand, are not designed to be bioactive or to be ingested or absorbed. Characteristically they are transported, used and disposed of in well-controlled situations, and testing may be used to determine what controls or limits are needed in each case to assure safe handling, use and disposal.
Regulatory and voluntary guidelines
In the United States, the Toxic Substances Control Act (TSCA) governs the testing of both "new" and "existing" industrial chemicals. New chemicals are those that are not listed on the public or confidential TSCA Inventory, while existing chemicals are those that are listed on the TSCA Inventory.
Unlike regulatory agencies in Canada, the European Union or Japan, which require a base set of test data on all new chemical emissions, the U.S. Environmental Protection Agency (EPA) relies on the use of structural activity models as surrogates for test data on many new chemicals. However, if the chemical is not appropriate to EPA models, or if the agency determines that it is within certain categories of concern, its introduction is restricted until EPA has adequate testing data to perform the required risk assessment.
EPA may require additional testing of an existing chemical under TSCA if the chemical presents an unreasonable risk of injury, or when substantial quantities are produced with the potential for extensive human or environmental exposure.
Test findings are used to regulate these chemicals under TSCA and other statutes, including the exposure levels promulgated by the Occupational Safety and Health Administration (OSHA), the disposal methods set by the Resource Conservation and Recovery Act (RCRA), and the emissions levels permitted under the Clean Air Act (CAA) and the Safe Drinking Water Act (SDWA).
The global chemical processing industry has made a voluntary commitment to characterize risk under the Responsible Care® initiative, sponsored by the Chemical Manufacturers Association (CMA) in the United States, and by national chemical associations in other countries. Under the initiative's Product Stewardship Code of Management Practices, the industry has agreed to establish a system to identify, document and implement health, safety and environmental risk-management actions appropriate to the product's potential risk. This is to be carried out by each company that manufactures or sells the product.
Why test?
Testing is necessary to determine the types of effects that can be expected, the specific levels of exposure that produce these effects, and in particular the level at which no effect is observed. By dividing this "no observable effect level" (NOEL) by an appropriate safety factor, the acceptable exposure level is set. In addition, classifying the risks of a product will help to determine the limitations of its use in specific applications. Many new chemicals never reach the marketplace because testing discovers hazards that cannot be cost-effectively managed to prevent harm.
The amount of the product present at the workplace and the likelihood of exposure during regular use also govern how much testing is needed. In some cases, a product is manufactured and used only on-site in a chemical plant, as a feedstock for another product; it is transported from one facility to another by pipeline, and there is little, if any, chance of human or environmental exposure. The only reasonably foreseeable risk in such cases is the effect of accidental spills or equipment ruptures. Testing is required to determine what effects can be anticipated from this type of unplanned, short duration exposure. Worker safety in these situations can often be maintained with the use of protective clothing and equipment specially designed for such occasions.
The growth of scientific knowledge can also govern the quantity of testing. A case in point is the recent discovery that some industrial products may disrupt the endocrine systems of humans and wildlife. This has led to recommendations for a new series of tests to determine whether some products could act as endocrine disrupters and to establish new safety levels. For this purpose, new test methods must be developed and validated. Such testing is justified as long as there is agreement on the significance of test results, and when health effects can be clearly predicted so that something can be done to minimize their potential.
Highly codified regulatory procedures for testing have been established over the past decades by EPA, the Food and Drug Administration (FDA), and the Consumer Product Safety Commission (CPSC). These procedures include a combination of non-animal (in vitro) and animal (in vivo) tests. The development of alternatives to animal laboratory tests continues to be pursued, and progress has been made in testing for skin and eye irritation by using in vitro cultures of skin or other cells.
The tiered concept
Dow's tiered testing procedure applies existing scientific knowledge to determine the hazards of a product. This procedure generates technical data about the health and environmental effects of the product in specific applications, and identifies what additional testing is warranted.
During the initial stage, called Tier 0, the product's chemical and physical properties are examined to calculate how the product is likely to behave in anticipated use scenarios. In this stage, we use the experience of many years of toxicology research as a framework for expert judgment. Can the product be closely linked to another chemical product that has already been extensively tested? A close derivative of an already tested compound, for example, or a chemical that is readily metabolized into another previously tested chemical, may not need much further testing, because its hazards can be directly inferred. Further, if a product is highly flammable or explosive, then the extensive safety procedures required for its use may so limit human exposure that full toxicity testing is unnecessary.
Tier 1: If an immediate link cannot be made with another previously studied chemical, a series of simple tests are indicated to characterize the product further. These tests may include environmental degradation tests, the further study of the structure of the molecule either in a computer simulation or through a laboratory biodegradation test, and short-term animal tests to identify potential health effects.
Tier 2: If the Tier 1 tests indicate the potential for adverse effects, more extensive follow-up testing may be warranted. If, for example, the short-term animal tests reveal possible organ system toxicity, then additional animal tests are conducted to examine this potential.
Tier 3: In this phase, future uses for the product - such as contact with foods or personal care products, either as an ingredient or a packaging material - are examined. If the product is a monomer which may be used in plastic for food packaging, for example, tests will be run on the finished polymer to determine the availability of non-polymerized leachate and its effect, if any, on health or the environment. If the final use is in a consumer product that may give off vapors or contact the user's skin, a series of tests will be needed to determine inhalation effects or dermal toxicity.
Tier 4: If the end product under consideration is to go into widespread commercial use with a possibility for long-term human exposure - for example, a gasoline additive or a dry cleaning solvent - then long-term studies may be appropriate or necessary to determine potential chronic effects. This can involve lifetime chronic toxicity studies or multigenerational animal studies. In addition, if widespread contact with the environment is likely, the effects on wildlife should be determined, calling for such tests as fish reproduction studies.
Naturally, no one set of rules can determine when and how to advance from one tier to the next. Educated judgments by experts must be made in each case, based on the nature of the product, the expected end use and the volume to be produced.
Where do we go from here?
Although the introduction of new industrial chemicals has been regulated in the United States and other countries for several years now, many chemicals that were already in use when these regulations were enacted were "grandfathered" and have been subject to separate programs to fill information gaps. Many of these chemicals are produced in large volumes - so-called high production volume (HPV) chemicals - at facilities throughout the world. Recognizing this situation and the need to coordinate testing across the industry and various governments, the Organization of Economic Cooperation and Development (OECD) devised a voluntary testing program to accelerate the collection of a standard base set of toxicology and environmental fate data on these HPV chemicals.
This program, while producing some success, has been criticized recently as being too slow. In response, EPA, CMA and the Environmental Defense Fund (EDF) agreed on a voluntary testing program, funded by the industry, designed to fill toxicology and environmental fate data gaps on 2,800 HPV chemicals by the year 2004. This represents at least a tenfold acceleration in the pace of the current OECD program. The International Council of Chemical Associations (ICCA) has recently agreed to accelerate the pace of testing for another 1,000 high-priority chemicals. These two voluntary programs, taken together, compose a very credible response to the criticism that insufficient hazard information exists on the most common chemical products in use.
Not all of the data gaps that may be identified for certain HPV chemicals will require additional testing. The principles described in Dow's tiered approach can be applied to decide when existing information about certain HPV chemicals is adequate or when testing is truly required.
There is some talk today about "reverse onus." Some environmental non-governmental organizations want product manufacturers to prove their products are safe before being permitted to manufacture or sell them, rather than regulatory agencies having to prove the products cause harm before they can restrict or ban them. Although many responsible chemical manufacturers might agree with this in principle, the technical challenges are likely to revolve around how much testing is required to prove safety. The amount of testing required is likely to differ from product to product, depending on the chemical and physical properties, availability of data on similar products, anticipated uses and reasonably foreseeable exposures. Testing frameworks based on volume of product manufactured or sold, like those in effect in Canada, Japan and Europe, if harmonized globally, could provide a workable solution.
Ethical concerns prompt search for alternatives to animal research
Over the years, a number of questions have arisen as to the ethics and necessity of using live animals in laboratory tests. Manufacturers of chemical products have the responsibility to provide adequate hazard information to guide the safe handling and use of their products, but at the same time have a concern for the well-being of laboratory animals for both humane and scientific reasons. The following is quoted from Animal research: A necessary safeguard to human health (1992), which highlights Dow's position on animal research:
The use of laboratory animals in research is an emotionally charged issue that puts society in a moral dilemma where the sanctity of life is concerned. We all want the continuing benefits of medicine and biotechnology. But in good conscience, researchers cannot go directly from unconfirmed chalkboard theory to testing in humans. To do so would be to take grave risks with the health and safety of the world's human population.
Animal testing has played a key role in virtually every major medical breakthrough this century and has helped ensure that the food we eat, the medicine we use and the many products we depend on each day meet the highest standards of safety. Dow is bound by federal testing regulations and a commitment to customer safety to use animals judiciously in product safety research. We are also required by state, federal and our own internal requirements to provide a humane environment for the animals we use.
Dow maintains a long-term commitment to reduce, refine and, as much as is practical, replace the use of animals in its laboratories. Long before animal rights became a national issue, Dow followed a self-imposed program to use only a minimal number of animals. This program has worked. Despite the fact that the number of Dow products has increased over the past decade, the number of animals used in our product testing has dropped sharply. Dow uses alternative, non-animal tests whenever possible. Some alternative methods utilize computer models, cultured cells, bacteria or insects as replacements for mammals.
Dow is active on a global basis within its own laboratories to conserve laboratory animals whenever possible. Dow management and scientists support efforts to use research animals as wisely, humanely and as sparingly as possible until advances in science provide testing alternatives that are acceptable to the scientific and regulatory communities.
This article originally appeared in the 05/01/1999 issue of Environmental Protection.