|
Activity Number:
|
35
- Epidemiological Models for Genetic Data, Biomarkers, and Rare Outcomes
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 7, 2022 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #322732
|
|
|
Title:
|
Multivariable Functional Mendelian Randomization Incorporating Longitudinal Data
|
|
Author(s):
|
Hanfei Xu* and Ching-Ti Liu
|
|
Companies:
|
Boston University and Boston University
|
|
Keywords:
|
mendelian randomization;
multivariable;
longitudinal data;
functional data analysis
|
|
Abstract:
|
The relationship between skeleton and obesity is still under debate, and this work is motivated by studying their causal relationship with multiple correlated longitudinal obesity indices. Mendelian randomization (MR) is a popular approach to investigate causal relationships between health indicators and complex traits. However, it lacks a way of incorporating multiple time-varying exposures into MR analysis. We propose models using functional data analysis approach to handle multiple time-varying exposures under multivariable MR framework. We also introduce the concept of mean functional exposure, yielding interpretable causal effect estimates. Our simulation study shows that the proposed models perform better than methods only utilizing single measurement, in terms of power and bias of the effect size. We apply our models on Framingham Heart Study data to study the respective direct causal effects of body mass index and waist-hip ratio on bone health related measures. Our method advances the research of causal inference by making better use of longitudinal data from multiple exposures, thus can provide insights into the relationship between exposures and the outcome of interest.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2022 program
|
