Knowledge enriched EHR features predict hospital readmission well
Knowledge-Enriched Structured EHR Features for 30-Day Hospital Readmission Prediction on MIMIC-IV
Machine Learning
Summary
Predicting if a patient will return to the hospital within 30 days is important for better care. The authors found that using structured patient health records combined with medical knowledge about diseases, procedures, drugs, and lab results can predict readmissions almost as well as methods that read doctors’ notes. Their method is easier to interpret and faster to run. This means hospitals could use existing data in a smarter way to identify patients at risk of returning soon.
What this means in practice
- •For hospital data teams: Build readmission risk models using structured EHR data enriched with medical knowledge for efficient and interpretable patient risk scoring.
- •For health software developers: Create clinical decision support tools that predict readmission risk without requiring computationally heavy processing of clinical notes.
Tested on one dataset.
Authors
Mohamad Najafi, Hongyun Fu, Mathias Brochhausen, Jian Wu, Yaohang Li
Abstract
Recent approaches to 30-day hospital readmission prediction rely on pre-trained language models applied to discharge summaries. Although these methods achieve strong performance, they depend on the availability of clinical notes, incur substantial computational costs, and yield representations that lack interpretability. We propose a knowledge-enriched feature representation that augments structured Electronic Health Record (EHR) data with four medical knowledge sources: disease ontology mapping, procedure classification, drug ingredient vocabulary, and organ system laboratory aggregation, without using clinical notes. Each feature dimension corresponds to a named clinical concept, yielding a sparse and interpretable patient representation. The approach is evaluated with six classifiers on a MIMIC-IV v2.2 cohort. Under 20-fold cross-validation, the best configuration achieves an AUROC of 0.743. This performance is comparable to that of previously reported methods on this dataset, including both those using only structured data and those incorporating clinical notes, while requiring considerably less computational cost. Interpretability analysis shows that demographics, organ system labs, drug ingredient features, and first-level ontology disease categories drive prediction, while deeper hierarchy levels contribute negligibly. These findings indicate that knowledge-enriched structured features offer a competitive and efficient alternative to embeddings from clinical notes for 30-day readmission prediction.