This project addresses statistical problems generated from collaboration with scientists in other program areas and general statistical problems of current interest encountered in biomedical research. This project is a continuing activity of the Branch. In FY98, Branch statisticians have contributed to the following areas of statistical research: screening methodology for diseases in general populations; measurement of structural uncertainties in morbidity data arising from the choice of disease definition; investigation of selection bias in epidemiological surveys; application of time series to longitudinal imaging of the developing brain in heathly individuals and patients with neurological diseases; and development of statistical methods for the analysis of sensitivity, specificity, and predictive value of diagnostic tests; development and application of nonparametric regressions for the analysis of neuroimaging data; and the application of recursive partitioning for the evaluation of risk factors in epidemiological studies. Other active areas of statistical research are: the development of nonlinear response models when response is nondecreasing and is a time-dependent function of two compartments; statistical modeling of longitudinal data to identify bench mark events and their time of occurrence in the clinical course of disease; adapting analysis methods for the estimating the level of disease risk for individual factors in the presence of many correlated potential risk factors; and the formulation of analysis methods for survival data with time- dependent covariates, the presence of competing risks, and informative censoring.
