We've made some changes to EPA.gov. If the information you are looking for is not here, you may be able to find it on the EPA Web Archive or the January 19, 2017 Web Snapshot.

About Using Predictive Models and Tools To Assess Chemicals under TSCA

On this page:

On other pages:

The Toxic Substances Control Act (TSCA) provides for the regulation of industrial chemicals by EPA.  Under TSCA, EPA is charged with assessing, and if necessary, regulating all phases of the life cycle of industrial chemicals including manufacturing, processing, use, and disposal.

If the EPA determines that a chemical may present an unreasonable risk to humans or the environment, the Agency can request additional information be submitted or may choose to take regulatory action to control the risks.

EPA’s methodology for risk assessment of chemicals under TSCA integrates predictive methods into the hazard and exposure analysis, and has been used for over 25 years, reflecting several specific regulatory requirements that define the framework under which the EPA must operate.

Section 5 of TSCA specifies the premanufacture reporting requirements for new chemicals.  The only information that must be provided in the notifications are:

  • Chemical identity information, including molecular weight, trade names, and byproducts and impurities
  • Production, import and use information, including the amount for each use
  • Human exposure and envirnmental release information
  • Any existing test data in the possession of the notifier

Read section  5 of TSCA.

As a consequence, the empirical hazard and exposure data available to EPA for industrial chemicals is often limited.  Since TSCA places the burden of proof on EPA to demonstrate risk and/or exposure for chemicals, the Agency may need to predict over 150 attributes for a chemical during an assessment.

EPA has computerized many of these predictive models and methods used to evaluate chemicals where data are lacking. These computer models assess a particular aspect of a chemical's possible impact on humans or the environment. The Agency may make predictions concerning chemical identity, physical/chemical properties, environmental transport and partitioning, environmental fate, environmental toxicity, engineering releases to the environment, and environmental concentrations.

EPA uses a variety of methods to make predictions which include:

  • Structure activity relationships (SAR)
  • Nearest analog analysis
  • Chemical class analogy
  • Mechanisms of toxicity
  • Mass balance models and measured data
  • Models based on measured laboratory data
  • Chemical industry data
  • Professional judgement

The results of an exposure assessment are combined with a hazard (toxicity) assessment to predict the potential of an industrial chemical to cause risk to humans or the environment. If a potential for risk may exist EPA will examine these new chemical notices in greater detail and regulate if necessary.

EPA’s Sustainable Futures Program offers training

EPA's Sustainable Futures program provides training to stakeholders on the use and interpretation of the models and tools and provides regulatory incentives for chemical developers to screen new chemicals for potential risks early in the development process.

EPA's Sustainable Futures website contains:

  • Presentations
  • Information on how EPA offers regulatory incentives to graduates of the program 
  • Training material
  • Information on interactive workshops
  • Documents on how to apply the predictive models described in this site and how to interpret the results

Visit EPA’s Sustainable Futures Program

Use of EPA models and methods to create safer chemicals

EPA's Safer Choice has developed several programs that apply EPA's predictive models and methods to help stakeholders evaluate human health and environmental attributes of chemicals in products to support more informed decisions on chemical management. Its website information to industry and consumers on safer chemicals and chemical products.

Visit the Safer Choice EPA website.