Analysis of Observational Health Care Data Using SAS

Chapter 9

Analysis of Longitudinal Observational Data Using Marginal Structural Models

Douglas E. FariesZbigniew A. Kadziola

Abstract

9.1 Introduction

9.2 MSM Methodology

9.3 Example: MSM Analysis of a Simulated Schizophrenia Trial

9.4 Discussion

References

 

Abstract

Assessing treatment effectiveness in longitudinal, observational data can be complex because in observational treatment patients can change medications at any time. In addition to the need to control for selection bias at baseline due to the lack of randomization, time-varying confounders can influence treatment changes over time and, thus, affect treatment group effectiveness comparisons. One approach to producing causal treatment effect estimates—even in the presence of treatment ...