Elastic Integrative Analysis of Randomized Trial and Real-World Data for\n Treatment Heterogeneity Estimation

2020/05/21 by Shu Yang, Yang, Shu, Gao, Chenyin +4 · 10 citations
Mathematics · #Advanced Causal Inference Techniques #FOS: Computer and information sciences #Methodology (stat.ME) #Statistical Methods and Inference #Statistical Methods in Clinical Trials

paper · pdf · doi:10.48550/arxiv.2005.10579

openalex publication_date 2020/05/21 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

We propose a test-based elastic integrative analysis of the randomized trial\nand real-world data to estimate treatment effect heterogeneity with a vector of\nknown effect modifiers. When the real-world data are not subject to bias, our\napproach combines the trial and real-world data for efficient estimation.\nUtilizing the trial design, we construct a test to decide whether or not to use\nreal-world data. We characterize the asymptotic distribution of the test-based\nestimator under local alternatives. We provide a data-adaptive procedure to\nselect the test threshold that promises the smallest mean square error and an\nelastic confidence interval with a good finite-sample coverage property.\n

Citations

• Estimating causal effects of treatments in randomized and nonrandomized studies.

Cited by

• When Does Trial-Real-World Data Fusion Improve Precision? Model Auditing and Selection-Aware Inference for Adaptive-TMLE
• Testing Effect Homogeneity and Confounding in High-Dimensional Experimental and Observational Studies
• Integrating RCTs, RWD, AI/ML and Statistics: Next-Generation Evidence Synthesis
• Heterogeneity-Aware Federated Causal Inference Leveraging Effect-Measure Transportability
• Estimate Time-Varying Exposure Effects via Ensemble Learning-based Marginal Structural Model with Application to Adolescent Cognitive Development Study
• Rate doubly robust estimation for weighted average treatment effects
• When Is Causal Inference Possible? A Statistical Test for Unmeasured Confounding
• The Identification Power of Combining Experimental and Observational Data for Distributional Treatment Effect Parameters
• Minimax Rates and Adaptivity in Combining Experimental and Observational Data
• Adaptive Data-Borrowing for Improving Treatment Effect Estimation using External Controls

Related

• Comparison of Methods that Combine Multiple Randomized Trials to Estimate Heterogeneous Treatment Effects
• Statistical Inference for Heterogeneous Treatment Effects Discovered by Generic Machine Learning in Randomized Experiments
• Statistical Inference for Heterogeneous Treatment Effect with Right-censored Data from Synthesizing Randomized Clinical Trials and Real-world Data
• A Variance-Based Test for Heterogeneous Treatment Effects
• Concise Summarization of Heterogeneous Treatment Effect Using Total Variation Regularized Regression