Human variability can be addressed during each stage in the risk assessment of chemicals causing noncancer toxicities. Noncancer toxicities arising from oral exposure to trichloroethylene (TCE) are used in this paper as a case study for exploring strategies for identifying and incorporating information about human variability in the chemical specific hazard identification and dose-response assessment steps. Toxicity testing in laboratory rodents is the most commonly used method for hazard identification. By using animal models for sensitive populations, such as developing fetuses, testing can identify some potentially sensitive populations. A large variety of reproductive and developmental studies with TCE were reviewed. The results were mostly negative and the limited positive findings generally occurred at doses similar to those causing liver and kidney toxicity. Physiologically based pharmacokinetic modeling using Monte Carlo simulation is one method for evaluating human variability in the dose-response assessment. Three strategies for obtaining data describing this variability for TCE are discussed: (1) using in vivo human pharmacokinetic data for TCE and its metabolites, (2) studying metabolism in vitro, and (3) identifying the responsible enzymes and their variability. A review of important steps in the metabolic pathways for TCE describes known metabolic variabilities including genetic polymorphisms, enzyme induction, and disease states. A significant problem for incorporating data on pharmacokinetic variability is a lack of information on how it relates to alterations in toxicity. Response modeling is still largely limited to empirical methods due to the lack of knowledge about toxicodynamic processes. Empirical methods, such as reduction of the No-Observed-Adverse-Effect-Level or a Benchmark Dose by uncertainty factors, incorporate human variability only qualitatively by use of an uncertainty factor. As improved data and methods for biologically based dose-response assessment become available, use of quantitative information about variability will increase in the risk assessment of chemicals.