Department of Statistics, The Ohio State University
Statistics and Biostatistics Colloquium Series

Government agencies such as U.S. EPA, FDA, and USDA utilize the results of toxicology studies as components of regulatory decision making concerning the safety of environmental and food or water borne chemicals to which humans may be exposed. Often these results are based on toxicity studies that were carried out at a single laboratory at a single time. Thus the variability estimates associated with these studies reflect only the intra-laboratory component of variability, often over a short duration. When developing a toxicity assay it is necessary to assure that results obtained for a particular chemical would be consistent across different laboratories or across separated times within the same laboratory.

This presentation

• Illustrates the ubiquity and the importance of inter-laboratory variation by reference to several well known data sets
• Illustrates the need to account for the influence of inter-laboratory variability in toxicologically based regulatory decision making and risk assessment
• Reviews the history and structure of a large scale national toxicological assay development program and Battelle’s role
• Illustrates Battelle’s statistical activities in supporting the assay development program and use of meta-analysis like approaches to account for inter-laboratory variability
• Identifies several statistical methodology issues arising from the program and suggests possible approaches toward dealing with them

Meet the speaker in Room 212 Cockins Hall at 4:30 p.m. Refreshments will be served.