Wqsreg: a Stata command for weighted quantile sum regression

Area of Science:

• Environmental Epidemiology
• Biostatistics
• Statistical Genetics

Background:

• Weighted Quantile Sum (WQS) regression is a key method for analyzing complex mixtures in environmental epidemiology.
• Its application has been limited due to a lack of accessible software outside the R environment.
• Addressing this gap is crucial for advancing research on multidimensional exposures.

Purpose of the Study:

• Introduce `wqsreg`, the first Stata command for Weighted Quantile Sum (WQS) regression.
• Provide a user-friendly implementation of WQS regression for continuous, binary, and count outcomes.
• Facilitate the analysis of complex mixtures in epidemiological studies.

Main Methods:

• Developed `wqsreg`, a Stata command for WQS regression.
• Implemented features such as bootstrap, training/validation splitting, and repeated holdout procedures.
• Demonstrated the command's utility with exposome data, analyzing 38 exposures and a continuous outcome.

Main Results:

• `wqsreg` offers a flexible and user-friendly approach to WQS regression in Stata.
• The command successfully quantified associations between multiple environmental exposures and a health outcome.
• Graphical displays of individual weights provide insights into predictor contributions.

Conclusions:

• The `wqsreg` command expands access to WQS regression for Stata users in environmental epidemiology.
• This tool supports the investigation of complex mixtures and multidimensional exposures.
• Promotes the use of advanced statistical methods for analyzing correlated predictors in health research.