Marginal Structural Models

Marginal structural models (MSMs) are a statistical technique used to estimate the causal effect of treatments on an outcome of interest. MSMs are used to improve upon the limitations of conventional regression models, such as those that use traditional adjustment methods to account for confounding. For example, MSMs can be used when trying to quantify the effect of a treatment over time, often in the context of randomized clinical trials or longitudinal studies. Compared to traditional regression models, MSMs are better able to control for underlying measures, such as the effects of age, sex, or other prognostic factors, and to produce more accurate estimates of the causal effect of a treatment. MSMs are therefore essential for the reliable and valid estimation of causal effects.