Stable discovery of interpretable subgroups via calibration in causal studies

2020/08/23 by Raaz Dwivedi, Tan Yan, Dwivedi, Raaz +11 · 3 citations
Mathematics · Pharmacology, Toxicology and Pharmaceutics · #Advanced Causal Inference Techniques #Applications (stat.AP) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Methodology (stat.ME) #Pharmacogenetics and Drug Metabolism #Statistical Methods in Clinical Trials

paper · pdf · doi:10.48550/arxiv.2008.10109

openalex publication_date 2020/08/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01

Abstract

Building on Yu and Kumbier's PCS framework and for randomized experiments, we introduce a novel methodology for Stable Discovery of Interpretable Subgroups via Calibration (StaDISC), with large heterogeneous treatment effects. StaDISC was developed during our re-analysis of the 1999-2000 VIGOR study, an 8076 patient randomized controlled trial (RCT), that compared the risk of adverse events from a then newly approved drug, Rofecoxib (Vioxx), to that from an older drug Naproxen. Vioxx was found to, on average and in comparison to Naproxen, reduce the risk of gastrointestinal (GI) events but increase the risk of thrombotic cardiovascular (CVT) events. Applying StaDISC, we fit 18 popular conditional average treatment effect (CATE) estimators for both outcomes and use calibration to demonstrate their poor global performance. However, they are locally well-calibrated and stable, enabling the identification of patient groups with larger than (estimated) average treatment effects. In fact, StaDISC discovers three clinically interpretable subgroups each for the GI outcome (totaling 29.4% of the study size) and the CVT outcome (totaling 11.0%). Complementary analyses of the found subgroups using the 2001-2004 APPROVe study, a separate independently conducted RCT with 2587 patients, provides further supporting evidence for the promise of StaDISC.

Citations

• On the Application of Probability Theory to Agricultural Experiments. Essay on Principles. Section 9
• A new framework to enhance the interpretation of external validation studies of clinical prediction models
• A New Vector Partition of the Probability Score
• A New Initiative on Precision Medicine
• Mining frequent patterns without candidate generation
• VERIFICATION OF FORECASTS EXPRESSED IN TERMS OF PROBABILITY
• Estimating causal effects of treatments in randomized and nonrandomized studies.
• Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment
• Veridical data science
• CausalML: Python Package for Causal Machine Learning
• Recent Developments in the Econometrics of Program Evaluation

Cited by

• Cross-World Assumption and Refining Prediction Intervals for Individual Treatment Effects
• CLEAR: Calibrated Learning for Epistemic and Aleatoric Risk
• PCS Workflow for Veridical Data Science in the Age of AI

Related

• Stable Discovery of Interpretable Subgroups via Calibration in Causal Studies
• Sharp Inference on Selected Subgroups in Observational Studies
• Propensity Score Analysis with Guaranteed Subgroup Balance
• Improving subgroup analysis using methods to extend inferences to specific target populations
• An Algorithm for Identifying Interpretable Subgroups With Elevated Treatment Effects