Building Prediction Models for 30-Day Readmissions Among ICU Patients Using Both Structured and Unstructured Data in Electronic Health Records

Abstract

ICU readmissions are associated with poor outcomes for patients and poor performance of hospitals. Patients who are re-admitted have an increased risk of in-hospital deaths; hospitals with a higher readmission rate have reduced profitability, due to an increase in cost and reduced payments from Medicare and Medicaid programs. Predicting a patient's likelihood of being readmitted to the ICU can help reduce early discharges, the risk of in-hospital deaths, and help increase profitability. In this study, we built and evaluated multiple machine learning models to predict 30-day readmission rates of ICU patients in the MIMIC-III database. We used both the structured data including demographics, laboratory tests, comorbidities, and unstructured discharge summaries as the predictors and evaluated different combinations of features. The best performing model in this study Logistic Regression achieved an AUROC of 75.7%. This study shows the potential of leveraging machine learning and deep learning for predicting ICU readmissions.

MiPy: A Framework for Benchmarking Machine Learning Prediction of Unplanned Hospital and ICU Readmission in the MIMIC-IV Database

“…ICU Readmission [38][39][40][41] Early Prediction (At ICU admission or first hours after admission) Binary Classification [42,43] At ICU Discharge Binary Classification [22,23,42,[44][45][46][47][48][49][50][51][52][53] Multi-class Classification [45,54] Unsupervised [55] Hospital Readmission [14,41,56] Early Prediction (At hospital admission or first hours after admission Binary Classification [24,57,58] At Hospital Discharge Binary Classification [24,57,[59][60][61][62][63][64][65][66][67][68][69][70][71][72] Multi-class Classification [73,74] Regression [73] To start, the base tables ("Admissions", "ICU_stays", and "Patients") must be joined...…”