Modeling cancer incidence in A-bomb survivors: a perspective

Abstract

Several manuscripts have been published which implement regression methods or stochastic processes methodology to model the incidence of a variety of cancers obtained via epidemiologic studies of radiation exposure. Cancer incidence data obtained from the Life Span Study (LSS) cohort have been of particular interest due to the extensive amount of information available from numerous years of follow-up of the Japanese A-bomb survivors. Conventional methods of exploring goodness-of-fit within each model class have indicated that both types adequately fit the data, but the model predicted curves for excess relative risk (ERR) and excess absolute risk (EAR) can vary dramatically. A simulation study is carried out to compare and contrast an excess relative risk regression model to a stochastic two-stage cancer model to determine the extent of the model predicted risk differences and why they occur.