Objective: Over 200,000 in-hospital cardiac arrests occur in the United States each year and many of these events may be preventable. Current vital sign–based risk scores for ward patients have demonstrated limited accuracy, which leads to missed opportunities to identify those patients most likely to suffer cardiac arrest and inefficient resource utilization. We derived and validated a prediction model for cardiac arrest while treating ICU transfer as a competing risk using electronic health record data.
Design: A retrospective cohort study.
Setting: An academic medical center in the United States with approximately 500 inpatient beds.
Patients: Adult patients hospitalized from November 2008 until August 2011 who had documented ward vital signs.
Measurements and Main Results: Vital sign, demographic, location, and laboratory data were extracted from the electronic health record and investigated as potential predictor variables. A person-time multinomial logistic regression model was used to simultaneously predict cardiac arrest and ICU transfer. The prediction model was compared to the VitalPAC Early Warning Score using the area under the receiver operating characteristic curve and was validated using three-fold cross-validation. A total of 56,649 controls, 109 cardiac arrest patients, and 2,543 ICU transfers were included. The derived model more accurately detected cardiac arrest (area under the receiver operating characteristic curve, 0.88 vs 0.78; p < 0.001) and ICU transfer (area under the receiver operating characteristic curve, 0.77 vs 0.73; p < 0.001) than the VitalPAC Early Warning Score, and accuracy was similar with cross-validation. At a specificity of 93%, our model had a higher sensitivity than the VitalPAC Early Warning Score for cardiac arrest patients (65% vs 41%).
Conclusions: We developed and validated a prediction tool for ward patients that can simultaneously predict the risk of cardiac arrest and ICU transfer. Our model was more accurate than the VitalPAC Early Warning Score and could be implemented in the electronic health record to alert caregivers with real-time information regarding patient deterioration.
1Department of Medicine, University of Chicago, Chicago, IL.
2Department of Health Studies, University of Chicago, Chicago, IL.
3Department of Medicine, University of Pittsburgh, Pittsburgh, PA.
* See also p. 986.
Preliminary versions of these data were presented as an oral presentation at the CHEST Conference, Atlanta, GA, October 21, 2012.
Dr. Churpek is supported by a National Institutes of Health (NIH) grant (T32 HL 07605). This research was funded, in part, by an institutional Clinical and Translational Science Award grant (UL1 RR024999; PI: Dr. Julian Solway). Dr. Park consulted for and received support for travel from the National Institutes of Health Clinical & Translational Science Awards (principal investigator: Dr. Edelson received grant support Derivation and validation of a model to predict cardiac arrest on the hospital wards). Dr. Edelson is supported by a career development award from the National Heart, Lung, and Blood Institute (K23 HL097157-01), has received research support from Philips Healthcare (Andover, MA) and Laerdal Medical (Wappingers Falls, NY), and has ownership interest in Quant HC (Chicago, IL), which is developing products for risk stratification of hospitalized patients. Dr. Edelson consulted for EarlySense. Drs. Edelson, Churpek, and Gibbons have a patent pending (ARCD.P0535US.P2). The remaining authors have disclosed that they do not have any potential conflicts of interest.
Address requests for reprints to: Dana P. Edelson, MD, MS, University of Chicago Medical Center, Section of Hospital Medicine, 5841 South Maryland Avenue, MC 5000, Chicago, IL 60637. E-mail: email@example.com