Superfund Research Program

August 2025

Paper of the Month

Researchers at the Northeastern University Superfund Research Program Center recommended new guidelines for which statistical methods are best suited for different chemical mixture studies. The team created a new software package, called CompMix, which helps other scientists identify the best statistical methods to use to answer their specific research question.

Chemical mixture analyses often have different goals. For example, some seek to characterize the overall effect of a mixture, but others seek to determine which exposure(s) within the mixture drive those effects. To determine which statistical methods were best for which types of questions, scientists used CompMix to simulate 11 different common methods for analyzing chemical mixtures.

The findings suggest that there is no one-size-fits-all approach. Researchers identified the best methods to identify mixtures and their interactions, create a summary score for environmental mix, and more. CompMix is available as an R package to encourage seamless adoption of new method guidelines.

To learn more, see the NIEHS Environmental Factor Newsletter.