• Alvin D Jeffery, Mary S Dietrich, Daniel Fabbri, Betsy Kennedy, Laurie L Novak, Joseph Coco, and Lorraine C Mion.
• Alvin D. Jeffery is a medical informatics fellow at the US Department of Veterans Affairs, Tennessee Valley Health-care System, Nashville, Tennessee, and a postdoctoral research fellow, Department of Biomedical Informatics, Vanderbilt University, Nashville, Tennessee. Mary S. Dietrich is a professor of statistics and measurement, Schools of Medicine (Biostatistics, Vanderbilt-Ingram Cancer Center, Psychiatry) and Nursing, Vanderbilt University. Daniel Fabbri is an assistant professor, Department of Biomedical Informatics, Vanderbilt University. Betsy Kennedy is a professor, School of Nursing, Vanderbilt University. Laurie L. Novak is an assistant professor and Joseph Coco is a senior application developer, Department of Biomedical Informatics, Vanderbilt University. Lorraine C. Mion is a professor, College of Nursing, The Ohio State University, Columbus, Ohio. alvinjeffery@gmail.com.
• Am. J. Crit. Care. 2018 Sep 1; 27 (5): 381-391.

BackgroundEarly warning systems lack robust evidence that they improve patients' outcomes, possibly because of their limitation of predicting binary rather than time-to-event outcomes.ObjectivesTo compare the prediction accuracy of 2 statistical modeling strategies (logistic regression and Cox proportional hazards regression) and 2 machine learning strategies (random forest and random survival forest) for in-hospital cardiopulmonary arrest.MethodsRetrospective cohort study with prediction model development from deidentified electronic health records at an urban academic medical center.ResultsThe classification models (logistic regression and random forest) had statistical recall and precision similar to or greater than those of the time-to-event models (Cox proportional hazards regression and random survival forest). However, the time-to-event models provided predictions that could potentially better indicate to clinicians whether and when a patient is likely to experience cardiopulmonary arrest.ConclusionsAs early warning scoring systems are refined, they must use the best analytical methods that both model the underlying phenomenon and provide an understandable prediction.©2018 American Association of Critical-Care Nurses.

10 keywords...

hide…