Enabling Counterfactual Survival Analysis with Balanced Representations

Abstract

Balanced representation learning methods have been applied successfully to counterfactual inference from observational data. However, approaches that account for survival outcomes are relatively limited. Survival data are frequently encountered across diverse medical applications, i.e., drug development, risk profiling, and clinical trials, and such data are also relevant in fields like manufacturing (e.g., for equipment monitoring). When the outcome of interest is a time-to-event, special precautions for handling censored events need to be taken, as ignoring censored outcomes may lead to biased estimates. We propose a theoretically grounded unified framework for counterfactual inference applicable to survival outcomes. Further, we formulate a nonparametric hazard ratio metric for evaluating average and individualized treatment effects. Experimental results on real-world and semi-synthetic datasets, the latter of which we introduce, demonstrate that the proposed approach significantly outperforms competitive alternatives in both survival-outcome prediction and treatment-effect estimation.

arXiv commentAccepted at ACM Conference on Health, Inference, and Learning (ACM CHIL 2021). Code at https://github.com/paidamoyo/counterfactual_survival_analysis

References in corpus (11)

• Random survival forests
• DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network
• Variational Inference with Normalizing Flows
• Estimating heterogeneous survival treatment effect in observational data using machine learning
• Deep Survival Analysis
• Balancing Covariates via Propensity Score Weighting
• Adversarial Time-to-Event Modeling
• Countdown Regression: Sharp and Calibrated Survival Predictions
• Estimating heterogeneous treatment effects with right-censored data via causal survival forests
• Survival Function Matching for Calibrated Time-to-Event Predictions
• Variational Learning of Individual Survival Distributions

Cited by in corpus (2)

• Counterfactual Phenotyping with Censored Time-to-Events
• SurvITE: Learning Heterogeneous Treatment Effects from Time-to-Event Data