Author + information
- Travis J. Moss,
- Matthew T. Clark,
- James Forrest Calland,
- Kyle Enfield,
- Douglas E. Lake and
- Randall Moorman
Background: Charted vitals and labs represent intermittent samplings of a patient's dynamic, physiologic state and have been used to calculate early warning scores to identify patients at risk of clinical deterioration. We hypothesized that the addition of continuous analytics from ECG monitoring–including changes in cardiorespiratory dynamics–to intermittently sampled data improves the predictive validity of models trained to detect clinical deterioration prior to ICU transfer or unanticipated death.
Methods: We analyzed 63 years of ECG from 8,105 admissions and calculated measurements of cardiorespiratory dynamics. We excluded observations after DNR/DNI orders or transitions to comfort care. We developed models to predict deterioration within the next 24 hours using either vitals, labs, or cardiorespiratory dynamics and also evaluated models using all available data sources. We calculated the C-statistic for all models.
Results: We analyzed 395 clinical deteriorations leading to ICU transfer or unexpected death. Models using intermittent vitals and labs had C-statistics of 0.69 and 0.65 respectively. The addition of continuous cardiorespiratory dynamics improved the C-statistics of all models (Figure). An integrated model using all data sources had the best C-statistic (0.74).
Conclusions: Integration of continuous cardiorespiratory dynamics with intermittent vitals and labs improves the predictive validity of models to detect clinical deterioration in acute care patients.
Poster Hall, Hall C
Sunday, March 19, 2017, 9:45 a.m.-10:30 a.m.
Session Title: Arrhythmias and Clinical EP: Syncope and Other EP Issues
Abstract Category: 6. Arrhythmias and Clinical EP: Other
Presentation Number: 1279-092
- 2017 American College of Cardiology Foundation