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