Tolerance Intervals on Bioassay Test Results to Assess Total Variability in Unbalanced Components of Variance Settings

• Activity Number: 193
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Paul Feder*
• Companies: Battelle
• Keywords: risk assessment ; bioassay ; tolerance intervals ; components of variation ; total variability ; parameter uncertainty

Abstract

Government scientific and regulatory agencies such as EPA, FDA, CDC, and NIH rely on the results of in vitro and in vivo bioassays as primary sources of data upon which to base inferences concerning safe doses or concentrations to which humans or animals can be exposed in the environment, in food, or in the workplace. Binding assays are used to distinguish among strong binding, weak binding, and non-binding chemicals

This presentation discusses and illustrates statistical methods to construct tolerance bounds on binding assay results in unbalanced components of variance settings that reflect multiple components of variation. The tolerance bound methodology is used to develop standards to screen future test results and additional test laboratories for acceptable assay performance, based on past assay results in multiple acceptably performing laboratories.

Proper estimation of the uncertainties in the bioassay parameter estimates is essential to the risk assessment basis of inferences. It is essential to assure that the estimates of bioassay parameter uncertainty reflect total variability. The total variability is composed of multiple components: Among laboratories, among ta