Event history analysis and inference from observational epidemiology

Abstract

Systematic inclusion of time in observational epidemiological studies may help strengthen the inference to be drawn, but new epidemiological challenges arise, such as time-dependent confounders - covariates which may change from being confounders to being intermediate variables. The focus of this presentation concerns two sets of tools: event history analysis and structural nested failure time models, both applied to a particularly intricate problem in observational epidemiology, of empirically assessing the graft-versus-leukaemia effect after bone marrow transplantation.