The two‐stage clonal expansion model is a popular model for carcinogenesis data. One common form of this model is based on the approximate hazard function. In certain situations, this formulation is not appropriate, and the exact hazard should be applied. However, the difficulty
of implementing the model based on the exact hazard has deterred many from using it. This paper presents a program implementing the exact hazard model for piecewise constant dosing using SAS, a package that is readily available to most that are interested in this type of analysis. Also, an
analysis of the ED01 data is presented using this program, and comparisons are made to an earlier analysis based on the approximate hazard. By allowing for an independent background tumor mechanism, an excellent fit to the bladder tumor incidence data was obtained.
No Supplementary Data
Document Type: Research Article
Computer Sciences Corporation, National Center for Toxicological Research (HFT-20), Jefferson, Arkansas 72079-9502.
Biometry and Risk Assessment, National Center for Toxicological Research (HFT-20), Jefferson, Arkansas 72079-9502.
Publication date: 1995-04-01