"Marginal Structural Models and Inverse-probability-of-treatment Weighted Estimators"

Dr. James M. Robins

Harvard School of Public Health

In observational studies, with exposures or treatments that vary over time, standard approaches for adjustment of confounding are biased when there exist time-dependent confounders that are also intermediate variables. This talk introduces marginal structural models (MSMs), a new class of causal models that allow for improved adjustment of confounding in those situations. The parameters of a MSM can be consistently estimated using a new class of estimators: the inverse-probability-of-treatment weighted estimators. The relationship of MSMs to propensity score methods is considered. I also discuss how to conduct a sensitiivity analysis using MSMs that alows for confounding by unmeasured confounders.

Papers of Robins and colleagues that cover the topic of the seminar; these and lots of others are available in PDF on his bibliography web page (different from the page linked above):

• "Association, causation, and marginal structural models," Robins, J.M. (1999)
• "Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men," Hernán, M., Brumback, B., and Robins, J.M. (1999)
• "Marginal structural models," Robins, J.M. (1998)
• "Marginal structural models and causal inference in epidemiology," Robins, J.M., Hernán, M. and, Brumback, B. (1998)

Return to Biostatistics Working Group