Estimating Longitudinal Exposures Using Short-Term Data. L. Barraj, C. Walls, and C. Scrafford, Exponent Inc. aka Novigen Sciences, Inc.
In risk assessment, it is important to characterize both the time frame for exposure (e.g., what is the exposure duration?) and for the toxic effect (e.g., what are the time to peak effects and the time to recovery?). Exposure estimates must be compared to measures of toxic effect based on the same time frame. Consumption surveys that provide the databases used in dietary exposure assessments typically cover 1 to 3 survey days. Discussions at a the Science Advisory Panel meeting reviewing a recent Preliminary Cumulative Risk Assessment of the Organophoshate Pesticides (OPs) focused on the impact of using the same two days data over and over to estimate average exposures over multiple days (Method A), versus creating "random" multiple-day diets by pooling consumption data across individuals who share the same demographic characteristics (Method B). We used case studies using actual consumption data from a 7-day consumption survey to compare the approaches and their impact on the upper percentiles of the intake distributions. Distributions derived using Method A generally showed larger variability that those derived using Method B. The difference was most accentuated when the foods considered included infrequently consumed foods. For foods consumed on a regular basis, estimates derived using Method B generally showed less variability that those derived using Method A. The results highlight the need for studies collecting data on longitudinal activity patterns.
Go to . . .
2002 SRA Annual Meeting Table of Contents
2002 SRA Annual Meeting Author Index
Main Abstracts Menu Page
RiskWorld Home Page