Causal Explainability of Machine Learning in Heart Failure Prediction from Electronic Health Records

TL;DR

This paper introduces a new computational framework for causal discovery in mixed-type clinical data, enhancing understanding of variable importance in heart failure prediction beyond correlation.

Contribution

It proposes a novel method for causal structure discovery with mixed data types and demonstrates its effectiveness in identifying causal variables in heart failure prediction.

Findings

Nonlinear causal models outperform linear ones in clinical data.

Causal strength correlates with feature importance in nonlinear classifiers.

Correlated features are often causal but rarely identified as effects.

Abstract

The importance of clinical variables in the prognosis of the disease is explained using statistical correlation or machine learning (ML). However, the predictive importance of these variables may not represent their causal relationships with diseases. This paper uses clinical variables from a heart failure (HF) patient cohort to investigate the causal explainability of important variables obtained in statistical and ML contexts. Due to inherent regression modeling, popular causal discovery methods strictly assume that the cause and effect variables are numerical and continuous. This paper proposes a new computational framework to enable causal structure discovery (CSD) and score the causal strength of mixed-type (categorical, numerical, binary) clinical variables for binary disease outcomes. In HF classification, we investigate the association between the importance rank order of three feature types: correlated features, features important for ML predictions, and causal features. Our results demonstrate that CSD modeling for nonlinear causal relationships is more meaningful than its linear counterparts. Feature importance obtained from nonlinear classifiers (e.g., gradient-boosting trees) strongly correlates with the causal strength of variables without differentiating cause and effect variables. Correlated variables can be causal for HF, but they are rarely identified as effect variables. These results can be used to add the causal explanation of variables important for ML-based prediction modeling.