Statistical Methods in Toxicology

This book contains selected papers from a workshop on modern statistical methods in toxicology held during the EUROTOX '90 conference in Leipzig. The papers deal with the biostatistical evaluation of the commonly used toxicological assays, i.e. mutagenicity, long-term carcinogenicity, embryotoxicity and chronic toxicity assays. The biological background is considered in detail, and most of the related statistical approaches described. In five overview papers, the present state of the art of the related topics is given, while in several contributed papers new approaches are discussed. The most important features are: - A new view on the per-litter analysis problem in em- bryotoxicity assays. - A highly sophisticated treatment of the so-called muta-tox problem in mutagenicity assays. - A detailed discussion of the multiplicity problem based on the closed testing procedure. This volume provides readers with an overview of modern biostatistical methods for several toxicological assays and is in part intended for direct, practical use.

Contenu

Modern statistical methods in toxicology - an overview.- Biostatistical analysis of mutagenicity assays.- Statistical analysis of the Ames assay.- Statistical analysis of the micronucleus test with the Fisher-Pitman permutation test.- A statistical analysis of HGPRT-test data by means of computer simulations.- Measuring intra-assay agreement for the Ames salmonella assay.- Tabular overview of statistical methods proposed for the analysis of Ames Salmonella assay data.- Biostatistical analysis of long-term carcinogenicity studies.- Analyses of longterm carcinogenicity studies.- Mathematical models for the initiating and promotive action of carcinogens.- Statistical problems in estimating a threshold in a dose-response model.- Biostatistical analysis of embryo-toxicity studies.- Per-litter analyses for studies in developmental toxicity.- Biometrical analysis of strain differences and litter effects, a case study.- Statistical evaluation of offspring parameters in embryotoxicity studies.- General principles in biostatistical analysis of toxicological studies.- General statistical principles in testing of toxicological studies.- Determination of endpoints in acute and chronic toxicity tests by using generalized linear models.- The need of software validation in routine toxicology.- An application of mixed model ANOVA for the assessment of motor activity in rat behavioural toxocity studies.- Sponsors.