340.728.01 ADVANCED METHODS FOR DESIGN AND ANALYSIS OF COHORT STUDIES

Explores advanced methods useful for the design and analysis of cohort studies. Emphasizes methods for analyzing time-to-event data subject to staggered entries using advanced parametric and semi-parametric methods including regression tree approaches; analytical methods for incomplete observations in cohort studies; methods to measure effects of exposures on time-to-event using relative times and relative hazards; parametric survival analysis methods and taxonomy of hazard functions; coefficients of determination based on parametric models for survival data; regression methods for trajectories of biomarkers including detection of inflexion points; methods for the analysis of interventions in observational studies: confounding by indication, marginal structural models for individual effectiveness and methods for estimating population effectiveness and the determination of the optimal time to intervene. Methods are based on published papers using data from cohort studies coordinated by

Upon successfully completing this course, students will be able to: Analyze a moderately complicated real life data set from a large, long-term multicenter cohort study using advanced methods discussed in the lectures; and to write a scientific report with a “Methods” and a “Results” section of a publishable manuscript.