• Kelly C Vranas, Jeffrey K Jopling, Timothy E Sweeney, Meghan C Ramsey, Arnold S Milstein, Christopher G Slatore, Gabriel J Escobar, and Vincent X Liu.
• 1Department of Medicine, Clinical Excellence Research Center, Stanford University, Stanford, CA. 2Division of Pulmonary and Critical Care, Department of Medicine, Oregon Health and Science University, Portland, OR. 3Department of Surgery, Stanford University, Stanford, CA. 4Biomedical Informatics Research, Stanford University, Stanford, CA. 5Division of Pulmonary and Critical Care, Department of Medicine, Stanford University, Stanford, CA. 6Health Services Research and Development, Portland VA Medical Center, Portland, OR. 7Division of Research, Kaiser Permanente, Oakland, CA.
• Crit. Care Med. 2017 Oct 1; 45 (10): 1607-1615.

ObjectivesIdentifying subgroups of ICU patients with similar clinical needs and trajectories may provide a framework for more efficient ICU care through the design of care platforms tailored around patients' shared needs. However, objective methods for identifying these ICU patient subgroups are lacking. We used a machine learning approach to empirically identify ICU patient subgroups through clustering analysis and evaluate whether these groups might represent appropriate targets for care redesign efforts.DesignWe performed clustering analysis using data from patients' hospital stays to retrospectively identify patient subgroups from a large, heterogeneous ICU population.SettingKaiser Permanente Northern California, a healthcare delivery system serving 3.9 million members.PatientsICU patients 18 years old or older with an ICU admission between January 1, 2012, and December 31, 2012, at one of 21 Kaiser Permanente Northern California hospitals.InterventionsNone.Measurements And Main ResultsWe used clustering analysis to identify putative clusters among 5,000 patients randomly selected from 24,884 ICU patients. To assess cluster validity, we evaluated the distribution and frequency of patient characteristics and the need for invasive therapies. We then applied a classifier built from the sample cohort to the remaining 19,884 patients to compare the derivation and validation clusters. Clustering analysis successfully identified six clinically recognizable subgroups that differed significantly in all baseline characteristics and clinical trajectories, despite sharing common diagnoses. In the validation cohort, the proportion of patients assigned to each cluster was similar and demonstrated significant differences across clusters for all variables.ConclusionsA machine learning approach revealed important differences between empirically derived subgroups of ICU patients that are not typically revealed by admitting diagnosis or severity of illness alone. Similar data-driven approaches may provide a framework for future organizational innovations in ICU care tailored around patients' shared needs.

20 keywords...

hide…