Predicting Discharge Destination of Critically Ill Patients Using Machine Learning

Decision making about discharge destination for critically ill patients is a highly subjective and multidisciplinary process, heavily reliant on the ICU care team, patients and their caregivers’ preferences, resource demand, staffing, and bed capacity. Timely identification of discharge disposition can be useful in care planning, and as a surrogate for functional status outcomes following critical illness. Although prior research has proposed methods to predict discharge destination in a critical care setting, they are limited in scope and in the generalizability of their findings. We proposed and implemented different machine learning architectures to determine the efficacy of the Acute Physiology and Chronic Health Evaluation (APACHE) IV score as well as the patient characteristics that comprise it to predict the discharge destination for critically ill patients within 24 hours of ICU admission. We conducted a retrospective study of ICU admissions within the eICU Collaborative Research Database (eICU-CRD) populated with de-identified clinical data from adult patients admitted to an ICU between 2014 and 2015. Machine learning models were developed to predict four discharge categories: death, home, nursing facility, and rehabilitation. These models were trained and tested on 115,248 unique ICU admissions. To mitigate class imbalance, we used synthetic minority over-sampling techniques. Hierarchical and ensemble classifiers were used to further study the impact of imbalanced testing set on the performance of our predictive models. Amongst all of the tested models, XGBoost provided the best discrimination performance with an area under the receiver operating characteristic curve of 90% (recall: 71%, F1: 70%). Our findings indicate that the variables used in the APACHE IV model for estimating patient severity of illness are better predictors of hospital discharge destination than the APACHE IV score alone. Incorporating these models into clinical decision support systems may assist patients, caregivers, and the ICU team to begin disposition planning as early as possible during the hospitalization.

[1]  François Chollet,et al.  Keras: The Python Deep Learning library , 2018 .

[2]  Daniel Fabbri,et al.  Predicting Discharge Dates From the NICU Using Progress Note Data , 2015, Pediatrics.

[3]  José Salvador Sánchez,et al.  Index of Balanced Accuracy: A Performance Measure for Skewed Class Distributions , 2009, IbPRIA.

[4]  Joydeep Ghosh,et al.  Cluster Ensembles --- A Knowledge Reuse Framework for Combining Multiple Partitions , 2002, J. Mach. Learn. Res..

[5]  Vikas Agarwal,et al.  A model to aid in the prediction of discharge location for stroke rehabilitation patients. , 2003, Archives of physical medicine and rehabilitation.

[6]  P. Riehmann,et al.  Interactive Sankey diagrams , 2005, IEEE Symposium on Information Visualization, 2005. INFOVIS 2005..

[7]  Alex A. Freitas,et al.  A survey of hierarchical classification across different application domains , 2010, Data Mining and Knowledge Discovery.

[8]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[9]  Gustavo E. A. P. A. Batista,et al.  A study of the behavior of several methods for balancing machine learning training data , 2004, SKDD.

[10]  E. Draper,et al.  Improving intensive care unit discharge decisions: Supplementing physician judgment with predictions of next day risk for life support , 1994, Critical care medicine.

[11]  B. Cuthbertson,et al.  Predicting death and readmission after intensive care discharge. , 2008, British journal of anaesthesia.

[12]  C. Camargo,et al.  Admission vitamin D status is associated with discharge destination in critically ill surgical patients , 2015, Annals of Intensive Care.

[13]  B. Wells,et al.  Predicting patient discharge disposition after total joint arthroplasty in the United States. , 2010, The Journal of arthroplasty.

[14]  Stef van Buuren,et al.  MICE: Multivariate Imputation by Chained Equations in R , 2011 .

[15]  Saxon Ridley Critical care. , 2003, Anaesthesia.

[16]  Michael J. Breslow,et al.  Readmissions and Death after ICU Discharge: Development and Validation of Two Predictive Models , 2012, PloS one.

[17]  Joydeep Ghosh,et al.  C 3E: A Framework for Combining Ensembles of Classifiers and Clusterers , 2011, MCS.

[18]  T. Williams,et al.  Delayed discharges from an adult intensive care unit. , 2004, Australian health review : a publication of the Australian Hospital Association.

[19]  J. Krall,et al.  Predicting discharge destination of stroke patients using a mathematical model based on six items from the Functional Independence Measure. , 1996, Archives of physical medicine and rehabilitation.

[20]  Yoshua Bengio,et al.  Random Search for Hyper-Parameter Optimization , 2012, J. Mach. Learn. Res..

[21]  I. Tomek,et al.  Two Modifications of CNN , 1976 .

[22]  Federico Esteban,et al.  Pursuing Optimal Prediction of Discharge Time in ICUs with Machine Learning Methods , 2019, AIME.

[23]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[24]  A. Adithya Parallel Coordinates , 2015 .

[25]  M. Kattan,et al.  Predicting discharge to a long-term acute care hospital after admission to an intensive care unit. , 2014, American journal of critical care : an official publication, American Association of Critical-Care Nurses.

[26]  Alistair E. W. Johnson,et al.  The eICU Collaborative Research Database, a freely available multi-center database for critical care research , 2018, Scientific Data.

[27]  Suman V. Ravuri,et al.  A Clinically Applicable Approach to Continuous Prediction of Future Acute Kidney Injury , 2019, Nature.

[28]  Nitesh V. Chawla,et al.  SMOTE: Synthetic Minority Over-sampling Technique , 2002, J. Artif. Intell. Res..

[29]  Kay See Tan,et al.  Trends in Critical Care Beds and Use Among Population Groups and Medicare and Medicaid Beneficiaries in the United States: 2000–2010 , 2016, Critical care medicine.

[30]  David P. Dobkin,et al.  The quickhull algorithm for convex hulls , 1996, TOMS.

[31]  Frank E. Harrell,et al.  Prediction models need appropriate internal, internal-external, and external validation. , 2016, Journal of clinical epidemiology.

[32]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[33]  David Sontag,et al.  Temporal Convolutional Neural Networks for Diagnosis from Lab Tests , 2015, ArXiv.

[34]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[35]  Reid C Thompson,et al.  Using a Guided Machine Learning Ensemble Model to Predict Discharge Disposition following Meningioma Resection , 2017, Journal of Neurological Surgery Part B: Skull Base.

[36]  Anis Sharafoddini,et al.  A New Insight Into Missing Data in Intensive Care Unit Patient Profiles: Observational Study , 2018, JMIR medical informatics.

[37]  Pierre Geurts,et al.  Extremely randomized trees , 2006, Machine Learning.