Online Program Home
My Program

Abstract Details

Activity Number: 175
Type: Contributed
Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #321339
Title: A Latent Variable Model with Scaled Nonlinear Effects for Multiple Outcomes
Author(s): Zhenzhen Zhang* and Brisa N. Sanchez
Companies: University of Michigan and University of Michigan
Keywords: latent variable ; non-linear ; multiple outcomes ; linear mixed model ; lead exposure ; BASC-II

Multiple outcomes can strengthen epidemiological studies on environmental exposure effects by using information from correlated sources to reduce estimation error and increase power of detection. Besides taking into account the correlation among multiple outcomes, some models in the literature have also adopted the idea of a scaled exposure effect to examine the overall impact of environmental exposure on outcomes that can be similarly affected. In this paper, we extend that idea and also incorporate the estimation and testing of non-linear exposure effects on multiple outcomes in a latent variable model. We apply our model to the study of lead exposure effect on children's behavioral development in a Mexican cohort, where the negative influence of lead on multiple psychometric scales exhibits a ceiling effect for high-level exposure.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association