You are previewing Analysis of Observational Health Care Data Using SAS.
O'Reilly logo
Analysis of Observational Health Care Data Using SAS

Book Description

This book guides researchers in performing and presenting high-quality analyses of all kinds of non-randomized studies, including analyses of observational studies, claims database analyses, assessment of registry data, survey data, pharmaco-economic data, and many more applications. The text is sufficiently detailed to provide not only general guidance, but to help the researcher through all of the standard issues that arise in such analyses. Just enough theory is included to allow the reader to understand the pros and cons of alternative approaches and when to use each method. The numerous contributors to this book illustrate, via real-world numerical examples and SAS code, appropriate implementations of alternative methods. The end result is that researchers will learn how to present high-quality and transparent analyses that will lead to fair and objective decisions from observational data.This book is part of the SAS Press program.

Table of Contents

  1. Preface
  2. Part 1 Introduction
    1. Chapter 1 Introduction to Observational Studies
      1. 1.1 Observational vs. Experimental Studies
      2. 1.2 Issues in Observational Studies
      3. 1.3 Study Design
      4. 1.4 Methods
      5. 1.5 Some Guidelines for Reporting
      6. Acknowledgments
      7. References
  3. Part 2 Cross-Sectional Selection Bias Adjustment
    1. Chapter 2 Propensity Score Stratification and Regression
      1. Abstract
      2. 2.1 Introduction
      3. 2.2 Propensity Score: Definition and Rationale
      4. 2.3 Estimation of Propensity Scores
      5. 2.4 Using Propensity Scores to Estimate Treatment Effects: Stratification and Regression Adjustment
      6. 2.5 Evaluation of Propensity Scores
      7. 2.6 Limitations and Advantages of Propensity Scores
      8. 2.7 Example
      9. 2.8 Summary
      10. Acknowledgments
      11. References
    2. Chapter 3 Propensity Score Matching for Estimating Treatment Effects
      1. Abstract
      2. 3.1 Introduction
      3. 3.2 Estimating the Propensity Score
      4. 3.3 Forming Propensity Score Matched Sets
      5. 3.4 Assessing Balance in Baseline Characteristics
      6. 3.5 Estimating the Treatment Effect
      7. 3.6 Sensitivity Analyses for Propensity Score Matching
      8. 3.7 Propensity Score Matching Compared with Other Propensity Score Methods
      9. 3.8 Case Study
      10. 3.9 Summary
      11. Acknowledgments
      12. References
    3. Chapter 4 Doubly Robust Estimation of Treatment Effects
      1. Abstract
      2. 4.1 Introduction
      3. 4.2 Implemention with the DR Macro
      4. 4.3 Sample Analysis
      5. 4.4 Summary
      6. 4.5 Conclusion
      7. References
    4. Chapter 5 Propensity Scoring with Missing Values
      1. Abstract
      2. 5.1 Introduction
      3. 5.2 Data Example
      4. 5.3 Using SAS for IPW Estimation with Missing Values
      5. 5.4 Sensitivity Analyses
      6. 5.5 Discussion
      7. References
    5. Chapter 6 Instrumental Variable Method for Addressing Selection Bias
      1. Abstract
      2. 6.1 Introduction
      3. 6.2 Overview of Instrumental Variable Method to Control for Selection Bias
      4. 6.3 Description of Case Study
      5. 6.4 Traditional Ordinary Least Squares Regression Method Applied to Case Study
      6. 6.5 Instrumental Variable Method Applied to Case Study
      7. 6.6 Using PROC QLIM to Conduct IV Analysis
      8. 6.7 Comparison to Traditional Regression Adjustment Method
      9. 6.8 Discussion
      10. 6.9 Conclusion
      11. Acknowledgments
      12. References
    6. Chapter 7 Local Control Approach Using JMP
      1. Abstract
      2. 7.1 Introduction
      3. 7.2 Some Traditional Analyses of Hypothetical Patient Registry Data
      4. 7.3 The Four Phases of a Local Control Analysis
      5. 7.4 Conclusion
      6. Acknowledgments
      7. Appendix: Propensity Scores and Blocking/Balancing Scores
      8. References
  4. Part 3 Longitudinal Bias Adjustment
    1. Chapter 8 A Two-Stage Longitudinal Propensity Adjustment for Analysis of Observational Data
      1. Abstract
      2. 8.1 Introduction
      3. 8.2 Longitudinal Model of Propensity for Treatment
      4. 8.3 Longitudinal Propensity-Adjusted Treatment Effectiveness Analyses
      5. 8.4 Application
      6. 8.5 Summary
      7. Acknowledgments
      8. References
    2. Chapter 9 Analysis of Longitudinal Observational Data Using Marginal Structural Models
      1. Abstract
      2. 9.1 Introduction
      3. 9.2 MSM Methodology
      4. 9.3 Example: MSM Analysis of a Simulated Schizophrenia Trial
      5. 9.4 Discussion
      6. References
    3. Chapter 10 Structural Nested Models
      1. Abstract
      2. 10.1 Introduction
      3. 10.2 Time-Varying Causal Effect Moderation
      4. 10.3 Estimation
      5. 10.4 Empirical Example: Maximum Likelihood Data Analysis Using SAS PROC NLP
      6. 10.5 Discussion
      7. Appendix 10.A
      8. Appendix 10.B
      9. Appendix 10.C
      10. References
    4. Chapter 11 Regression Models on Longitudinal Propensity Scores
      1. Abstract
      2. 11.1 Introduction
      3. 11.2 Estimation Using Regression on Longitudinal Propensity Scores
      4. 11.3 Example
      5. 11.4 Summary
      6. References
  5. Part 4 Claims Database Research
    1. Chapter 12 Good Research Practices for the Conduct of Observational Database Studies
      1. Abstract
      2. 12.1 Introduction
      3. 12.2 Checklist and Discussion
      4. Acknowledgments
      5. References
    2. Chapter 13 Dose-Response Safety Analyses Using Large Health Care Databases
      1. Abstract
      2. 13.1 Introduction
      3. 13.2 Data Structure
      4. 13.3 Treatment Model and Censoring Model Setup
      5. 13.4 Structural Model Implementation
      6. 13.5 Discussion
      7. References
  6. Part 5 Pharmacoeconomics
    1. Chapter 14 Costs and Cost-Effectiveness Analysis Using Propensity Score Bin Bootstrapping
      1. Abstract
      2. 14.1 Introduction
      3. 14.2 Propensity Score Bin Bootstrapping
      4. 14.3 Example: Schizophrenia Effectiveness Study
      5. 14.4 Discussion
      6. References
    2. Chapter 15 Incremental Net Benefit
      1. Abstract
      2. 15.1 Introduction
      3. 15.2 Cost-Effectiveness Analysis
      4. 15.3 Parameter Estimation
      5. 15.4 Example
      6. 15.5 Observational Studies
      7. 15.6 Discussion
      8. Acknowledgments
      9. References
    3. Chapter 16 Cost and Cost-Effectiveness Analysis with Censored Data
      1. Abstract
      2. 16.1 Introduction
      3. 16.2 Statistical Methods
      4. 16.3 Example
      5. 16.4 Discussion
      6. Acknowledgments
      7. References
  7. Part 6 Designing Observational Studies
    1. Chapter 17 Addressing Measurement and Sponsor Biases in Observational Research
      1. Abstract
      2. 17.1 Introduction
      3. 17.2 General Design Issues
      4. 17.3 Addressing Measurement and Sponsor Bias
      5. 17.4 Summary
      6. References
    2. Chapter 18 Sample Size Calculation for Observational Studies
      1. Abstract
      2. 18.1 Introduction
      3. 18.2 Continuous Variables
      4. 18.3 Binary Variables
      5. 18.4 Two-Sample Log-Rank Test for Survival Data
      6. 18.5 Two-Sample Longitudinal Data
      7. 18.6 Discussion
  8. Appendix: Asymptotic Distribution of Wilcoxon Rank Sum Test under <i xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xmlns:ns2="http://www.w3.org/2001/10/synthesis">H</i><sub xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" xmlns:ns2="http://www.w3.org/2001/10/synthesis">&#945;</sub>
  9. References
  10. Index