Statistical approaches to causal inference in environmental epidemiology: Methodological introductions and R implementations
- VernacularTitle:环境流行病学研究中因果推断统计方法介绍及R软件实现
- Author:
Guiming ZHU
1
;
Wanying LIU
1
;
Yanchao WEN
2
;
Simin HE
1
;
Qian GAO
1
;
Tong WANG
1
Author Information
- Publication Type:Review
- Keywords: environmental epidemiology; causal inference; confounding factor; R software
- From: Journal of Environmental and Occupational Medicine 2026;43(2):253-260
- CountryChina
- Language:Chinese
- Abstract: Environmental pollution is a significant public health challenge worldwide, and investigating the causal relationship between environmental exposure and population health outcomes is a key objective of environmental epidemiology research. In recent years, the complexity of environmental exposures has increasingly come to the forefront, making it challenging for observational studies that dominate environmental epidemiology to accurately estimate causal effects. Causal inference methods are particularly advantageous in controlling for confounding factors, thus holding great potential in environmental epidemiology research. Researchers can use appropriate causal inference methods to simulate the process of randomization, providing strong support for revealing the causal relationship between environmental exposure and health outcomes. However, there is a lack of reviews on the application of causal inference methods in environmental epidemiology studies in China. Therefore, this study introduced the basic principles of common causal inference statistical methods in environmental epidemiology, summarized the applicable conditions, advantages and disadvantages of various methods, and provided R software implementation codes for these methods, aiming to offer guidance for optimizing research design and practicing causal inference statistical methods.
