CMStatistics 2022: Start Registration
View Submission - CMStatistics
Title: Causal inference with Mendelian randomization for longitudinal traits Authors:  Yuehua Cui - Michigan State University (United States) [presenting]
Abstract: Mendelian Randomization uses genetic variants as instrument variables to determine whether an observational association between a risk factor and an outcome is consistent with a causal effect. The use of Mendelian Randomization reduces regression bias and provides a more reliable estimate of the likely underlying causal relationship between an exposure and a disease outcome. Most current Mendelian Randomization methods are focused on cross-sectional phenotypic traits. Longitudinal studies track the same sample at different time points and have a number of advantages over cross-sectional studies. It would be possible for researchers to learn more about 'cause and effect' relationships when incorporating time information. We propose a time lag model to investigate the delayed causal effects in a longitudinal study. We assume that both the current and past values of exposure contribute to the current outcome. In order to select the duration of delay included in the model, an algorithm is developed for the variable selection purpose. The point-wise testing and simultaneous testing are developed to test the existence of causal effects. The method was illustrated via simulation studies and an application to a real dataset.