Multiple comparisons in carcinogenesis study with right-censored survival data

This paper considers the practical problem in animal carcinogenesis experim ents where several treatment groups are compared with a control group in a one-way layout and the observed survival data are subject to random right-c ensorship. Proposed herein are multiple testing procedures based on two-sam ple weighted logrank statistics, each comparing an individual treatment wit h the control, for determining which treatments are more effective than the control. The associated p-value of claiming a certain treatment is more ef fective than the control is also discussed. A test-based confidence set for the scale changes between each treatment and the control is then obtained. The comparative results of a Monte Carlo error rate and power study for sm all sample sizes are presented. Finally, a numerical example involving rena l carcinoma in mice demonstrates the feasibility of the proposed multiple t esting procedures and test-based confidence set. Copyright (C) 2000 John Wi ley & Sons, Ltd.