A general approach to modeling environmental mixtures with multivariate outcomes

2025-04-24

AI Summary

In environmental health research, detecting weak effects of highly correlated multi-exposure mixtures remains statistically underpowered—particularly within distributed lag frameworks, where precise identification of susceptible windows and quantification of component-specific contributions are challenging. To address this, we propose a Bayesian adaptive exponential model that jointly models exposure–outcome weights and response structures for the first time. We innovatively integrate lag-spectrum clustering with exposure–response curve clustering to enable automatic discovery of unknown exposure metric structures and quantitative assessment of component importance. The method unifies distributed lag modeling with multivariate kernel machine learning. Applied to the NMMAPS dataset, it jointly models three mortality outcomes and two air pollutants (with maximum lag of 14 days), significantly improving accuracy in identifying susceptible windows and enhancing statistical power. This work establishes a novel paradigm for risk assessment of complex environmental mixtures.

Technology Category

• Machine Learning: Bayesian Learning
• Reasoning under Uncertainty: Graphical Models
• Search and Optimization: Distributed Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

An important goal of environmental health research is to assess the health risks posed by mixtures of multiple environmental exposures. In these mixtures analyses, flexible models like Bayesian kernel machine regression and multiple index models are appealing because they allow for arbitrary non-linear exposure-outcome relationships. However, this flexibility comes at the cost of low power, particularly when exposures are highly correlated and the health effects are weak, as is typical in environmental health studies. We propose an adaptive index modelling strategy that borrows strength across exposures and outcomes by exploiting similar mixture component weights and exposure-response relationships. In the special case of distributed lag models, in which exposures are measured repeatedly over time, we jointly encourage co-clustering of lag profiles and exposure-response curves to more efficiently identify critical windows of vulnerability and characterize important exposure effects. We then extend the proposed approach to the multivariate index model setting where the true index structure -- the number of indices and their composition -- is unknown, and introduce variable importance measures to quantify component contributions to mixture effects. Using time series data from the National Morbidity, Mortality and Air Pollution Study, we demonstrate the proposed methods by jointly modelling three mortality outcomes and two cumulative air pollution measurements with a maximum lag of 14 days.

Problem

• Assessing health risks from environmental exposure mixtures
• Improving power in modeling correlated exposures and weak effects
• Identifying critical exposure windows and important mixture effects

Innovation

• Adaptive index modeling for exposure-outcome relationships
• Co-clustering lag profiles and exposure-response curves
• Multivariate index model with variable importance measures