Uwe Siebert

Real World Health Care Data Analysis


Скачать книгу

6.5.2 Computation of Different Matching Methods

       6.5.3 1:1 Nearest Neighbor Matching

       6.5.4 1:1 Optimal Matching with Additional Exact Matching

       6.5.5 1:1 Mahalanobis Distance Matching with Caliper

       6.5.6 Variable Ratio Matching

       6.5.7 Full Matching

       6.6 Discussion Topics: Analysis on Matched Samples, Variance Estimation of the Causal Treatment Effect, and Incomplete Matching

       6.7 Summary

       References

       Chapter 7: Stratification for Estimating Causal Treatment Effects

       7.1 Introduction

       7.2 Propensity Score Stratification

       7.2.1 Forming Propensity Score Strata

       7.2.2 Estimation of Treatment Effects

       7.3 Local Control

       7.3.1 Choice of Clustering Method and Optimal Number of Clusters

       7.3.2 Confirming that the Estimated Local Effect-Size Distribution Is Not Ignorable

       7.4 Stratified Analysis of the PCI15K Data

       7.4.1 Propensity Score Stratified Analysis

       7.4.2 Local Control Analysis

       7.5 Summary

       References

       Chapter 8: Inverse Weighting and Balancing Algorithms for Estimating Causal Treatment Effects

       8.1 Introduction

       8.2 Inverse Probability of Treatment Weighting

       8.3 Overlap Weighting

       8.4 Balancing Algorithms

       8.5 Example of Weighting Analyses Using the REFLECTIONS Data

       8.5.1 IPTW Analysis Using PROC CAUSALTRT

       8.4.2 Overlap Weighted Analysis using PROC GENMOD

       8.4.3 Entropy Balancing Analysis

       8.5 Summary

       References

       Chapter 9: Putting It All Together: Model Averaging

       9.1 Introduction

       9.2 Model Averaging for Comparative Effectiveness

       9.2.1 Selection of Individual Methods

       9.2.2 Computing Model Averaging Weights

       9.2.3 The Model Averaging Estimator and Inferences

       9.3 Frequentist Model Averaging Example Using the Simulated REFLECTIONS Data

       9.3.1 Setup: Selection of Analytical Methods

       9.3.2 SAS Code

       9.3.3 Analysis Results

       9.4 Summary

       References

       Chapter 10: Generalized Propensity Score Analyses (> 2 Treatments)

       10.1 Introduction

       10.2 The Generalized Propensity Score

       10.2.1 Definition, Notation, and Assumptions

       10.2.2 Estimating the Generalized Propensity Score

       10.3 Feasibility and Balance Assessment Using the Generalized Propensity Score

       10.3.1 Extensions of Feasibility and Trimming

       10.3.2 Balance Assessment

       10.4 Estimating Treatment Effects Using the Generalized Propensity Score

       10.4.1 GPS Matching

       10.4.2 Inverse Probability Weighting

       10.4.3 Vector Matching

       10.5 SAS Programs for Multi-Cohort Analyses

       10.6 Three Treatment Group Analyses Using the Simulated REFLECTIONS Data

       10.6.1 Data Overview and Trimming

       10.6.2 The Generalized Propensity Score and Population Trimming

       10.6.3 Balance Assessment

       10.6.4 Generalized Propensity Score Matching Analysis

       10.6.5 Inverse Probability Weighting Analysis

       10.6.6 Vector Matching Analysis

       10.7 Summary

       References

       Chapter 11: Marginal Structural Models with Inverse Probability Weighting

       11.1 Introduction

       11.2 Marginal Structural Models with Inverse Probability of Treatment Weighting

       11.3 Example: MSM Analysis of the Simulated REFLECTIONS Data

       11.3.1 Study Description

       11.3.2 Data Overview

       11.3.3 Causal Graph

       11.3.4 Computation of Weights

       11.3.5 Analysis of Causal Treatment Effects Using a Marginal Structural Model

       11.4 Summary

       References

       Chapter 12: A Target Trial Approach with Dynamic Treatment Regimes and Replicates Analyses

       12.1 Introduction

       12.2 Dynamic Treatment Regimes and Target Trial Emulation

       12.2.1 Dynamic Treatment Regimes

       12.2.2 Target Trial Emulation

       12.3 Example: Target Trial Approach Applied to the Simulated REFLECTIONS Data

       12.3.1 Study Question

       12.3.2 Study Description and Data Overview

       12.3.3 Target Trial Study Protocol

       12.3.4 Generating New Data

       12.3.5 Creating Weights

       12.3.6 Base-Case Analysis

       12.3.7 Selecting the Optimal Strategy

       12.3.8 Sensitivity Analyses

       12.4 Summary

       References

       Chapter 13: Evaluating the Impact of Unmeasured Confounding in Observational Research