Analysis of Observational Health Care Data Using SAS

Chapter 6

Instrumental Variable Method for Addressing Selection Bias

R. Scott LeslieHassan Ghomrawi

Abstract

6.1 Introduction

6.2 Overview of Instrumental Variable Method to Control for Selection Bias

6.3 Description of Case Study

6.4 Traditional Ordinary Least Squares Regression Method Applied to Case Study

6.5 Instrumental Variable Method Applied to Case Study

6.6 Using PROC QLIM to Conduct IV Analysis

6.7 Comparison to Traditional Regression Adjustment Method

6.8 Discussion

6.9 Conclusion

Acknowledgments

References

 

Abstract

Observational data can provide powerful answers to research questions when properly addressing selection bias due to non-randomization. The medical literature has relied mostly on conventional regression and propensity ...