The focus of this project is development and refinement of statistical procedures for the design and analysis of cancer screening and related studies. Problems under investigation include derivation and comparison of data analysis methods, assessment of case-control studies for screening evaluation, development of models of cancer screening, estimation of length bias in screening studies, and estimation and adjustment of lead time in cancer survival data. To assess the case- control design for screening evaluation, the MISCAN microsimulation model is being used to provide population data with and without screening. Case-control studies are then done in the screened populations, and the results are compared with the true effect to assess bias in the case-control approach. A method for estimating the effect of starting periodic screening at different ages was developed that uses data from a screened population to estimate the probability of diagnosis in an unscreened group. A noncompliance model was developed that makes possible the evaluation of periodic screening over a long duration using data from a randomized trial offering a short program of screening which only some of the target population accepts. Methods have been developed for estimating the benefit of screening unaffected by lead time bias and the average lead time by examining the differences in case survival measured both from time of entry and time of diagnosis between screened and control groups. Length bias estimators were developed using this model. Approaches for estimating the post lead time survival of screen detected cancer cases were developed. Initial work assumed that time spent in states of disease progression are independent. Subsequent research has focused on estimation under dependence of disease durations. Approaches were defined for data monitoring of cancer screening trials, including sequential methods for nonproportional hazards using maximum efficiency robust tests. Simulation procedures were developed to assess these techniques.
