How can we figure out how treatments work?

What is it about?

Mediation analysis helps us understand how treatments work. Mediation is a process that takes place over time. In other words, a mediator is a variable that we think a treatment will change, and then in turn will lead to a change in a health outcome. So we sometimes talk about mediation analysis as allowing us to understand more about treatment mechanisms. It helps us trace the path from treatment, to things that change early, leading to later changes and improvements. We can measure mediators and health outcomes several times over time, and then we can use methods that bring all of that data together to help us get even more information about treatment mechanisms. This paper explains how to do this, and provides some software code to get you started.

Featured Image

Photo by Josh Sorenson on Unsplash

Photo by Josh Sorenson on Unsplash

Why is it important?

Mediation is a process that occurs over time, and should be measured and modeled as such. Up to the point of publication, people had talked about models for longitudinal mediation data, but had not applied them very often. This paper, and the included software code should help make these models more accessible and widely used.