Abstract
This article presents a review of some important considerations for scientists and regulatory decision makers who are charged with the task of drawing a conclusion from an existing and growing body of bioassay data to evaluate the toxicity of a compound. Two general approaches are outlined, based on classic statistical methods and on Bayesian statistical methods. Their advantages and disadvantages are discussed in relation to the overall decision-making process. The article concludes by presenting an application of the Bayesian approach to the body of bioassay data available at the time of the regulatory decision concerning the link between saccharin and bladder cancer.
