Generating survival data for fitting marginal structural Cox models using Stata and comparing the fit

Marginal structural models (MSM) can be used to estimate the effect of a time dependent exposure in presence of time dependent confounding. Previously Fewell et al. (2004) described how to estimate this model in Stata based on a weighted pooled logistic model approximation. However, based on the current literature and some recent simulation study results, this model can be suitably fitted in other ways too and various new weighting schemes are proposed accordingly. In this presentation, first the idea behind MSM will be explained and use of various weighting schemes will be justified through simple examples and tabulations using Stata. Then the procedure of generating survival data from a Cox MSM model will be illustrated using existing Stata commands. The performance of simulated data generation and procedure of fitting MSM via Stata will be compared with other standard statistical packages such as SAS and R.