Using Machine Learning to Predict ICU Transfer in Hospitalized COVID-19 Patients

Fu-Yuan Cheng; Himanshu Joshi; Pranai Tandon; Robert Freeman; David L Reich; Madhu Mazumdar; Roopa Kohli-Seth; Matthew Levin; Prem Timsina; Arash Kia

doi:10.3390/jcm9061668

Using Machine Learning to Predict ICU Transfer in Hospitalized COVID-19 Patients

J Clin Med. 2020 Jun 1;9(6):1668. doi: 10.3390/jcm9061668.

Authors

Fu-Yuan Cheng¹, Himanshu Joshi^{1

2}, Pranai Tandon³, Robert Freeman^{1

4}, David L Reich^{4

5}, Madhu Mazumdar^{1

2}, Roopa Kohli-Seth⁶, Matthew Levin^{5

7}, Prem Timsina¹, Arash Kia¹

Affiliations

¹ Institute for Healthcare Delivery Science, Icahn School of Medicine at Mount Sinai, 1425 Madison Avenue, New York, NY 10029, USA.
² Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, 1425 Madison Avenue, New York, NY 10029, USA.
³ Respiratory Institute, Icahn School of Medicine at Mount Sinai, 10 E 102nd St, New York, NY 10029, USA.
⁴ Hospital Administration, The Mount Sinai Hospital, 1 Gustave L. Levy Place, New York, NY 10029, USA.
⁵ Department of Anesthesiology, Perioperative and Pain Medicine, 1 Gustave L. Levy Place, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA.
⁶ Institute for Critical Care Medicine, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA.
⁷ Department of Genetics and Genomic Sciences, 1 Gustave L. Levy Place, Icahn School of Medicine at Mount Sinai, New York, NY 10029, USA.

Abstract

Objectives: Approximately 20-30% of patients with COVID-19 require hospitalization, and 5-12% may require critical care in an intensive care unit (ICU). A rapid surge in cases of severe COVID-19 will lead to a corresponding surge in demand for ICU care. Because of constraints on resources, frontline healthcare workers may be unable to provide the frequent monitoring and assessment required for all patients at high risk of clinical deterioration. We developed a machine learning-based risk prioritization tool that predicts ICU transfer within 24 h, seeking to facilitate efficient use of care providers' efforts and help hospitals plan their flow of operations.

Methods: A retrospective cohort was comprised of non-ICU COVID-19 admissions at a large acute care health system between 26 February and 18 April 2020. Time series data, including vital signs, nursing assessments, laboratory data, and electrocardiograms, were used as input variables for training a random forest (RF) model. The cohort was randomly split (70:30) into training and test sets. The RF model was trained using 10-fold cross-validation on the training set, and its predictive performance on the test set was then evaluated.

Results: The cohort consisted of 1987 unique patients diagnosed with COVID-19 and admitted to non-ICU units of the hospital. The median time to ICU transfer was 2.45 days from the time of admission. Compared to actual admissions, the tool had 72.8% (95% CI: 63.2-81.1%) sensitivity, 76.3% (95% CI: 74.7-77.9%) specificity, 76.2% (95% CI: 74.6-77.7%) accuracy, and 79.9% (95% CI: 75.2-84.6%) area under the receiver operating characteristics curve.

Conclusions: A ML-based prediction model can be used as a screening tool to identify patients at risk of imminent ICU transfer within 24 h. This tool could improve the management of hospital resources and patient-throughput planning, thus delivering more effective care to patients hospitalized with COVID-19.

Keywords: COVID-19; critical care; intensive care units; random forest; supervised machine learning.