Inverse Probability Weighting-Based Mediation Analysis for Microbiome Data

• Activity Number: 231 - SPEED: SPAAC SESSION I
• Type: Topic-Contributed
• Date/Time: Wednesday, August 11, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Biometrics Section
• Author(s): Yuexia Zhang* and Jian Wang and Jiayi Shen and Jessica Galloway-Pena and Samuel Shelburne and Linbo Wang and Jianhua Hu
• Companies: University of Toronto and University of Texas MD Anderson Cancer Center and University of Southern California and Texas A&M University and University of Texas MD Anderson Cancer Center and University of Toronto and Columbia University
• Keywords: causal inference; confounder; high-dimensional mediators; interventional indirect effect; microbiome study

Abstract

Mediation analysis is an important tool to study casual associations in biomedicine and other scientific areas, and has recently gained attention in microbiome studies. With a microbiome study of acute myeloid leukemia (AML) patients, we investigate whether the effect of induction chemotherapy intensity levels on the infection status is mediated by the microbial taxa abundance. The unique characteristics of the microbial mediators--high-dimensionality, zero-inflation and dependence--calls for new methodological developments in mediation analysis. The presence of an exposure-induced mediator-outcome confounder, antibiotics usage, further requires a delicate treatment in the analysis. To address these unique challenges brought by our motivating microbiome study, we propose a novel nonparametric identification formula for the interventional indirect effect (IIE), a measure recently developed for studying mediation effects. We develop the corresponding estimation algorithm, and test the presence of mediation effects via the bias-corrected and accelerated bootstrap. Extensive simulation studies and real data analysis show that the proposed method has good finite-sample performance.