Causal Machine Learning Is Not a Panacea: A Roadmap for Observational Causal Inference in Health

TL;DR

This paper discusses the potential and limitations of causal machine learning in observational health data, emphasizing the need for careful validation and responsible application to ensure valid clinical insights.

Contribution

It provides a roadmap and template for applying causal ML responsibly in observational health data, highlighting the importance of validating assumptions and modeling choices.

Findings

Causal ML has limitations that are under-recognized across disciplines.

Proper validation of causal assumptions is crucial for reliable results.

Causal ML can generate useful hypotheses if applied rigorously.

Abstract

Objective: The growing availability of large-scale observational clinical datasets and challenges in conducting randomized controlled trials have spurred enthusiasm in using causal machine learning (ML) for causal inference in observational data. We present a roadmap for applying causal ML to observational data. Materials and methods: We outline the importance of assessing validity assumptions within available data and applying causal ML responsibly for clinical experts using causal ML and ML practitioners with limited clinical expertise. Observations: Despite advances in causal ML, its limitations remain largely under-appreciated across disciplines. This gap in shared knowledge may impact the validity of findings. Discussion: Causal assumptions must be satisfied and modeling choices justified. Otherwise, these approaches risk producing biased or misleading results, with consequences for clinical research and patient care. Conclusion: Causal ML can be a powerful tool for generating causal hypotheses. We provide a template to strengthen the rigor and interpretability of causal analyses.