Analysis of Observational Health Care Data Using SAS

Chapter 10

Structural Nested Models

Daniel AlmirallCynthia J. CoffmanWill S. Yancy, Jr.Susan A. Murphy

Abstract

10.1 Introduction

10.2 Time-Varying Causal Effect Moderation

10.3 Estimation

10.4 Empirical Example: Maximum Likelihood Data Analysis Using SAS PROC NLP

10.5 Discussion

Appendix 10.A

Appendix 10.B

Appendix 10.C

References

 

Abstract

This chapter reviews Robins’ Structural Nested Mean Model (SNMM) for assessing the effect of predictors that vary over time. The SNMM is used to study the effects of time-varying predictors (or treatments) in the presence of time-varying covariates that are moderators of these effects. We describe a SAS implementation of a maximum likelihood (ML) estimator of the parameters of an SNMM using PROC NLP. The ...