Statistical evaluation of mortality in long-term carcinogenicity bioassays using a Williams-type procedure

► We propose a method for analysis of mortality in long-term carcinogenicity bioassays. ► Williams-type hazard ratios or survival time ratios are used as effect size. ► We develop simultaneous confidence intervals for the hazard ratios. ► The intervals can be interpreted in terms of significance and biological relevance. ► An extension of the approach allows the pooled analysis over sex and strains.