B1280
Title: Mendelian randomization analysis with pleiotropy-robust log-linear models for binary outcomes
Authors: Jinzhu Jia - Peking University (China) [presenting]
Abstract: Mendelian randomization (MR) is a statistical technique that uses genetic variants as instrumental variables to infer causality between traits. In dealing with a binary outcome, there are two challenging barriers on the way toward a valid MR analysis: the inconsistency of the traditional ratio estimator and the existence of horizontal pleiotropy. Two novel individual data-based methods are proposed, named random-effects and fixed-effects MR-PROLLIM, respectively, to surmount both barriers. These two methods adopt risk ratio (RR) to define the causal effect of continuous or binary exposure. The random-effects MR-PROLLIM models correlate pleiotropy, account for variant selection, and allow weaker instruments. The fixed-effects MR-PROLLIM can function with only a few selected variants. The random-effects MR-PROLLIM exhibits high statistical power while yielding fewer false-positive detections than its competitors. The fixed-effects MR-PROLLIM generally performs at an intermediate level between the classical median and mode estimators. MR-PROLLIM exhibits the potential to facilitate a more rigorous and robust MR analysis for binary outcomes.
