Multiple contrast type tests for the evaluation of animal carcinogenicity studies

摘要

The Cochran-Armitage (CA) test is frequently used for testing the dose-response relationship in tumor incidence. This test is based on aweighted linear regression of proportions. It is well known that the CA test lacks power for nonlinear tumor outcomes. For general shape of outcomes, Hothorn and Bretz (2000) proposed a multiple contrast (MC) test. This test suggests the use of the maximum over several single contrasts, where each of them is chosen appropriately to cover a specific dose-response shape. In this work, two new test procedures are proposed and they are compared to the CA and MC tests using power.