Background on Risk Characterization
- What health effects are caused by exposure to air toxics?
- How did EPA characterize risk from the modeled 1999 exposure estimates?
- How does EPA estimate cancer risk?
- How were the cancer risk estimates affected by EPA's recently revised Guidelines for Carcinogen
Risk Assessment (EPA/630/P-03/001F) and new Supplemental Guidance for Assessing Susceptibility from
Early-Life Exposure to Carcinogens (EPA/630/R-03/003F)?
- How accurate are these risk estimates?
- Does EPA distinguish among chemicals based on the type of data that supports its carcinogenic potential?
- Can cancer risk estimates for different substances be combined?
- How does EPA estimate risks for health effects other than cancer?
- Can noncancer hazard estimates for different substances be combined?
Section 112 of the Clean Air Act identifies 188 compounds emitted from stationary, area, and mobile sources as hazardous air pollutants (also known as "air toxics"). The EPA has classified many of these substances as "known," "probable," or "possible" human carcinogens. Air toxics are associated with a wide variety of noncancer adverse health effects that include neurological, cardiovascular, liver, kidney, and respiratory effects as well as effects on the immune and reproductive systems. The seriousness of the harm can range from headaches and nausea to respiratory arrest and death. Severity varies with the amount and length of exposure, the nature of the chemical itself (e.g., how it interacts with various organs and organ systems), and the unique behaviors and sensitivities of individual people. Some chemicals pose particular hazards to people of certain ages or genetic backgrounds.
2. How did EPA characterize risk from the modeled 1999 exposure estimates?
To evaluate a chemical's potential to cause cancer and other adverse health effects, EPA examines which adverse effects a particular substance causes (in a "hazard identification"), determines the exposure to the population (in an "exposure assessment"), and evaluates the specific exposures at which these effects might occur (in a "dose-response assessment"). The evaluation is based on studies of humans, animals, and microorganisms, usually published in peer-reviewed scientific journals. In this national-scale assessment, EPA combined information from dose-response assessments with modeled exposure estimates in a "risk characterization" to describe the potential that real- world exposures to air toxics compounds might cause harm. The EPA also examined the uncertainties surrounding the characterization of risk.
3. How does EPA estimate cancer risk?
At present, EPA typically assumes a linear relationship between the level of exposure and the lifetime probability of cancer from an air toxics compound. It expresses this dose-response relationship for cancer in terms of a unit risk estimate. The unit risk estimate (URE) is an upper bound estimate of an individual's probability of contracting cancer over a lifetime of exposure to a concentration of one microgram of the pollutant per cubic meter of air. Risks from exposures to concentrations other than one microgram per cubic meter are usually calculated by multiplying the actual concentration to which someone is exposed by the URE. For example, the EPA may determine the URE of a particular air toxics compound to be one in ten thousand per microgram per cubic meter. This means that a person who inhales air containing an average of one microgram per cubic meter for 70 years would have (as an upper bound) one chance in ten thousand (or 0.01%) of contracting cancer as a result. The EPA has developed UREs for many substances, and continues to re-examine and update them as knowledge improves. More information on UREs can be found in the EPA's Integrated Risk Information System. The UREs used in this assessment, along with associated uncertainties and a summary of the EPA's risk assessment guidelines for carcinogens, are included in the Health Effects Criteria discussion.
4. How were the cancer risk estimates affected by EPA's recently revised Guidelines for Carcinogen Risk Assessment (EPA/630/P-03/001F) and new Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (EPA/630/R-03/003F)?
The revised Guidelines were released after the risk calculations for the national-scale assessment were completed. However, the effective guidance at the time we calculated the risks (an earlier draft of the revised Guidelines) was consistent in most ways with the revised document. There was one important exception to this consistency, however. The cancer guidelines now include Supplemental Guidance that makes new recommendations with regard to estimating cancer risks to children. These recommendations have not been implemented for the chemicals included in the national-scale assessment, with the exception of vinyl chloride.
The Supplemental Guidance recommends that risks to children be adjusted for carcinogenic chemicals acting through a mutagenic and linear mode of action (i.e., chemicals that cause cancer by damaging genes). Where available data for the chemical are adequate they should be used to develop age-specific potency values, as EPA has already done for vinyl chloride (and which are already reflected for that chemical in the national-scale assessment). Where available data do not support a chemical-specific evaluation of differences between adults and children, the Supplemental Guidance recommends the use of the following default adjustment factors for early- life exposures: increase the carcinogenic potency by 10-fold for children up to 2 years old, and 3-fold for children from 2 to 15 years old. These adjustments have the aggregate effect of increasing by about 60% the estimated risk for a 70-year (lifetime) constant inhalation exposure.
It is important to keep in mind that EPA recommends that the default adjustments be made only for carcinogens (1) acting through a mutagenic mode of action, (2) for which a linear dose response has been assigned, and (3) for which data to evaluate adult and juvenile differences are not available. The default adjustments are not recommended for carcinogens whose mode of action is unknown. EPA will determine as part of the IRIS assessment process which substances meet these criteria, and future national-scale assessments will reflect adjustments for those substances.
5. How accurate are these risk estimates?
- First, many of the air toxics compounds in this assessment were classified as probable carcinogens, which means that data were not sufficient to prove these substances definitely cause cancer in humans. It is possible that some are not human carcinogens at environmentally relevant doses, and that the true cancer risk associated with these air toxics is zero.
- Second, all UREs used in this assessment were based on linear extrapolation from high to low exposures. To the extent that true dose-response relationships for some air toxics compounds are less than linear, this assumption may result in overestimates of cancer risk.
- Third, UREs for most of these substances were developed from animal data using conservative methods to extrapolate between species. Actual human responses may differ from the predicted ones.
- Fourth, most UREs used in this assessment (benzene is a notable exception to this) were based on statistical procedures that give the upper bound on the URE, which will likely result in an overstatement of risk.
On balance, we believe that true cancer risk from the assessed air toxics compounds would probably be less than that calculated in this assessment, although the possibility remains that it could be greater.
The EPA uses a system called the weight-of-evidence for carcinogenicity for characterizing the extent to which available data support the hypothesis that a compound causes cancer in humans. Under the EPA's 1986 risk assessment guidelines, the weight-of-evidence was described by Groups (A through E). Group A contains "known" carcinogens, or compounds for which evidence is sufficient to demonstrate a causal relationship between exposure and cancer incidence in humans. Group B contains "probable" carcinogens, for which evidence of cancer in humans is suggestive (Group B1) or evidence of cancer in animals is conclusive (Group B2). Group C contains "possible" carcinogens, for which animal evidence is suggestive but not conclusive. Group D contains agents for which no evidence exists (so it cannot be said whether the compound is or is not a carcinogen). Group E contains compounds for which adequate negative evidence exists (so it can be said that the compound is not a carcinogen). The approach outlined in the EPA's proposed guidelines for carcinogen risk assessment (1999) considers all scientific information in determining whether and under what conditions an agent may cause cancer in humans, and provides a narrative approach to characterizing carcinogenicity (rather than using categories or groups).
Because risk estimates are probabilities, cancer risks associated with different substances can be added together as long as the substances cause cancer by (1) similar mechanisms, or (2) completely independent mechanisms. Addition of cancer risk estimates is inappropriate only where substances interact in ways that either enhance or inhibit each other's carcinogenic potency. Had it been available, information on non-additive interactions would have been considered (as recommended in the EPA's 1986 Guidelines for the Health Risk Assessment of Chemical Mixtures, 52 FR 34014-34025). Because no such information was identified, the EPA used the guidelines' default assumption that cancer risks from different air toxics compounds may be added.
The EPA typically expresses dose-response relationships for effects other than cancer in terms of the inhalation reference concentration (RfC). The RfC is a concentration of the compound in air thought to be without adverse effects even if a person is exposed continuously. In other words, exposures below the RfC will probably not cause adverse noncancer health effects.
To express noncancer hazards the EPA uses the RfC as part of a calculation called the hazard quotient (HQ), which is the ratio between the concentration to which a person is exposed and the RfC. A value of the HQ less than one indicates that the exposure is lower than the RfC and that no adverse health effects would be expected. A value of the HQ greater than one indicates that the exposure is higher than the RfC. However, because many RfCs incorporate protective assumptions in the face of uncertainty, an HQ greater than one does not necessarily suggest a likelihood of adverse effects. Furthermore, the HQ cannot be translated to a probability that adverse effects will occur and is not likely to be proportional to risk. An HQ greater than one can best be described as indicating that a potential exists for adverse health effects.
The EPA has developed RfCs for many substances, and continues to re-examine and update them as knowledge improves. More information on RfCs can be found in the EPA's Integrated Risk Information System. The RfCs (and equivalent values) used in this assessment, along with associated uncertainties and a summary of the EPA's risk assessment guidelines for effects other than cancer, are included on the Health Effects Criteria page.
Because different pollutants may cause similar adverse health effects, it is often appropriate to combine hazard quotients associated with different substances. The EPA has drafted revisions to the national guidelines on mixtures that support combining the effects of different substances in specific and limited ways. Ideally, hazard quotients should be combined for pollutants that cause adverse effects by the same toxic mechanism. However, because detailed information on toxic mechanisms was not available for most of the substances in this assessment, the EPA used a simpler and more conservative method also outlined in these draft guidelines. The combined noncancer hazards associated with respiratory irritation caused by 41 pollutants and neurological effects associated with 20 pollutants were calculated using the hazard index (HI), defined as the sum of hazard quotients for individual air toxics compounds that affect the same organ or organ system. The HI is only an approximation of the combined effect because some of the substances may affect the target organs in different (i.e., non-additive) ways. As with the HQ, a value of the HI below 1.0 will likely not result in adverse effects over a lifetime of exposure. However, a value of the HI greater than 1.0 does not necessarily suggest a likelihood of adverse effects. Furthermore, the HI cannot be translated to a probability that adverse effects will occur and is not likely to be proportional to risk. An HI greater than one can be best described as indicating that a potential may exist for adverse effects to respiratory or nervous system.