The focus of this project is development and refinement of statistical procedures for the design and analysis of cancer screening and related studies. Statistical problems under investigation include development and comparison of data analysis methods, assessment of case-control studies for screening evaluation, and development of models of cancer screening. Each of these problem areas is common to screening and prevention trials in which the Division participates, but the methods for screening studies must address the special lead time and length biases inherent in screening programs. Properties of case-control studies in the context of screening evaluation are being considered. Screening effect as estimated from a case-control study within a randomized trial is compared with the trial result as a standard, initially using data from the HIP study. Alternative definitions of cases, controls and exposure will be assessed. An age-dependent cancer screening model that assumes no birth cohort effect and the invitation of a random sample of individuals by age to screening allows one to design a trial to estimate survival for screening patterns different from those in the trial, and to infer screening effect when all individuals undergo some screening. A computer simulation model of colorectal cancer screening is also under development. Research into short term endpoints which might be valid proxies for mortality in screening evaluation focuses on the rate of advanced stage disease. Analysis of data from the HIP study suggests the rate of Stage III and IV disease closely mimics mortality.
