biologic specimens. In either case, exposure serves as a surrogate for dose. Exposure assessments based on measurements of environmental contaminants attempt to quantify the amount of the contaminant that contacts a body barrier over a defined period. Exposure can occur through inhalation, skin contact, and ingestion. Exposure also can be assessed by measuring the compounds of interest—or their metabolites—in human tissues. Such biologic markers of exposure integrate absorption from all routes. The evaluation of those markers can be complex, because most are not stable for long periods. Knowledge of pharmacokinetics is essential to the linkage of measurements at the time of sampling with past exposures. Similarly, the assessment of markers that could be markers of effect—such as DNA adducts—shows promise, but does not necessarily provide accurate measurements of past exposure; that is, there is little evidence that currently measured DNA adducts are related to occupational or environmental exposures experienced years before.

Because quantitative assessments based on environmental or biologic samples are not always available for epidemiologic studies, investigators rely on a mixture of qualitative and quantitative information to derive estimates. There are a few basic approaches to exposure assessment for epidemiology (Armstrong et al., 1994; Checkoway et al., 1989). The simplest compares the members of a presumably exposed group with the general population or with a non-exposed group. That approach offers simplicity and ease of interpretation. If, however, only a small fraction of the group is exposed to the agent, the increased risk posed by exposure might not be detectable when the risk of the entire group is assessed.

A more refined method assigns each study subject to an exposure category, typically high, medium, low, or no exposure. Disease risk for each group is calculated separately and compared with a reference or non-exposed group. That method can identify the presence or absence of a dose–response trend. In some cases, more-detailed information is available for use in quantitative exposure estimates, which are sometimes called exposure metrics. They integrate quantitative estimates of exposure intensity (such as air concentration or extent of skin contact) with exposure duration to produce an estimate of cumulative exposure. Ideally, these refined estimates reduce errors associated with misclassification and thereby increase the power of statistical analysis to identify true associations between exposure and disease.

The temporal relationship between exposure and disease is complex and often difficult to define in epidemiologic investigations. Many diseases do not appear immediately following exposure. In the case of cancer, for example, the disease may not appear for many years after the exposure. The time between an exposure and the occurrence of disease is often referred to as a latency period (IOM, 2004). Exposures can be brief (sometimes referred to as acute exposures) or protracted (sometimes referred to as chronic exposures). At one extreme the exposure can be the result of a single insult, as in an accidental poisoning. At the other, an individual exposed to a chemical that is stored in the body may continue

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement