An Approach to the Design and Analysis of Screening Studies in Toxicology

Publisher Website
Google Scholar
Abstract
An increasingly important aspect of toxicology is the use of screening tests for detecting the presence or absence of a single end point of effect, such as mutagenicity or neurobehavioral effects. Such screens have a common set of operating characteristics that are not widely apreciated and that make traditional approaches to statistical analysis insensitive and inefficient in comparison to other available methods. All too often, control and historical data are not used to strengthen the analysis process. The characteristics of screens are presented and reviewed, along with overviews of 26 sets of data from functional observational battery screen (neurobehavioral) studies. Two alternative approaches to statistical analysis of screening tests (a control chart approach and a graphic-exploratory data analysis approach) are presented, along with a review of traditionally used contingency table, rank sum, and ANOVA methods. The performances of these methods in analyzing the test case screen datasets are compared in terms of power, sensitivity, and efficiency. Both alternative approaches are shown to be superior to traditional approaches in performance toward meeting the objectives of screens.