Home | Help Center

Endless possibilities in academia

Causal Inference in Environmental Epidemiology: Application of Targeted Maximum Likelihood Estimation to Assess the Effect of Air Pollution Reduction Policies on Cardiovascular Mortality

Authors: Mohammad Tarique¹, Lei Yang², Chao Luo³*


Affiliations:

¹ University of Missouri, Columbia 65211, USA

² Xi'an Jiaotong University, Xi'an 710049, China

³ The Affiliated Huaian No.1 People's Hospital of Nanjing Medical University, Huaian 223300, China


*Corresponding Author:Chao Luo, Email: hayylch@njmu.edu.cn

Abstract

This study employs targeted maximum likelihood estimation to rigorously estimate the causal effect of comprehensive air pollution reduction policies on cardiovascular mortality across 156 urban areas from 2005 to 2022. By analyzing longitudinal mortality data, air quality monitoring, and policy implementation records, we found that sustained PM2.5 reductions of 10 μg/m³ achieved through multi-sectoral policies were associated with an 11.2% decrease in cardiovascular mortality. The TMLE approach, which efficiently adjusts for time-varying confounding and model misspecification, provides robust causal evidence supporting the cardiovascular benefits of air quality regulations. Policy effects were modified by baseline pollution levels, population age structure, and concurrent public health interventions.

Keywords: Causal Inference; Environmental Epidemiology; Air Pollution; Cardiovascular Mortality; Targeted Maximum Likelihood Estimation; Policy Evaluation; Time-Varying Confounding

Cite
[Copy]