Nonparametric regression for cost-effectiveness analyses with observational data -- a tutorial

2025-07-04

AI Summary

This study addresses confounding bias in cost-effectiveness analysis (CEA) using observational healthcare data. We propose a joint nonparametric causal inference framework based on Bayesian additive regression trees (BART), which simultaneously estimates heterogeneous causal effects of treatment on both health outcomes and medical costs—departing from conventional linear or unidimensional modeling approaches. By jointly modeling these dual endpoints, our method enables robust, unbiased estimation of the incremental cost-effectiveness ratio (ICER). The framework is implemented in an end-to-end, fully reproducible software package and validated empirically. Results demonstrate substantial improvements in accuracy and credibility of CEA conducted on observational data, thereby strengthening the validity of causal evidence for budget-constrained, evidence-based decision-making in healthcare.

Technology Category

• Reasoning under Uncertainty: Causality
• Machine Learning: Causal Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Healthcare decision-making often requires selecting among treatment options under budget constraints, particularly when one option is more effective but also more costly. Cost-effectiveness analysis (CEA) provides a framework for evaluating whether the health benefits of a treatment justify its additional costs. A key component of CEA is the estimation of treatment effects on both health outcomes and costs, which becomes challenging when using observational data, due to potential confounding. While advanced causal inference methods exist for use in such circumstances, their adoption in CEAs remains limited, with many studies relying on overly simplistic methods such as linear regression or propensity score matching. We believe that this is mainly due to health economists being generally unfamiliar with superior methodology. In this paper, we address this gap by introducing cost-effectiveness researchers to modern nonparametric regression models, with a particular focus on Bayesian Additive Regression Trees (BART). We provide practical guidance on how to implement BART in CEAs, including code examples, and discuss its advantages in producing more robust and credible estimates from observational data.

Problem

• Estimating treatment effects on health outcomes and costs using observational data
• Addressing confounding in cost-effectiveness analysis with nonparametric methods
• Introducing BART for robust causal inference in healthcare decision-making

Innovation

• Uses Bayesian Additive Regression Trees (BART)
• Applies nonparametric regression for CEA
• Provides practical guidance with code examples