FQPA requires determination of the combined potential risk from both dietary (food and drinking water) and non-dietary (residential) routes of exposures to pesticides. Critical to the implementation and ultimately to the impact of FQPA are the validity of the risk assessments. Fundamental to the practice of sound science in risk assessment is accuracy in representing real-world exposures and the resulting risk. Too often, simplistic and overly conservative approaches and exposure data are employed. The resulting assessments are often termed "worst-case" and the inherent inaccuracies justified as additional safety factors. A scientific, tiered approach can be employed to improve the accuracy of risk assessments and provide more realistic estimates of risk. A case study using chlorpyrifos, a widely-studied insecticide, will be presented to show the levels of refinement which can be obtained through the use of more relevant exposure data, recognition of the patterns of uses and exposures, and higher-tier probabilistic methodologies. Results show a 7- to 26-fold decrease in estimated acute dietary risk and a 30- to 80-fold decrease in estimated aggregate risk using these refinements. These decreases proved to be critical, since the change in results also change the conclusions about risk and indicate the levels are within acceptable limits.