fbpx

Automated early warning decision support to detect acute decompensation in the emergency department. P2

Automated early warning decision support to detect acute decompensation in the emergency department. P2

Background

Detecting early signs of clinical decompensation, such as deviations in vital signs, is critical to reducing failure to rescue events among hospital patients. Research indicates that early warning alert scores based on clinical and physiological data predict risk of decompensation in emergency department (ED) patients. Early warning scoring systems such as the National Early Warning Score (NEWS) are commonly used on a national scale for early detection of all-cause clinical decompensation. However, there is a lack of consistent findings on the effects of using these systems to influence patient outcomes in complex settings like EDs due to a wide variety of different systems being compared and faulty implementation models. The goal of this study was to test the effectiveness of DECISIO’s decision support system, InsightIQ, to detect and intervene in clinical decompensation in the ED by evaluating reductions in hospital mortality and length of stay (LOS).

Methods

This study compared observed versus expected outcomes for mortality and LOS for a control and intervention group in an ED at a large, urban hospital. DECISIO’s decision support technology was implemented to help calculate the vital sign-based NEWS for each patient in the intervention group. The control group consisted of roughly 1650 patient admissions selected prior to implementing DECISIO’s InsightIQ technology, and the intervention group consisted of 8360 admissions selected over 9 months following the implementation of the software. The charge nurse and ED physician were notified if and when a vital sign deviation led to an overall NEWS score of 5. Upon confirming concern of clinical worsening, the team would respond with additional testing.

Results

The intervention group had a substantially reduced LOS of up to 6.5%.

The results demonstrate a clinical correlation between NEWS scores and LOS is relevant because earlier interventions improved patient outcomes

Conclusions

The results of this study indicate that implementing a near real-time decision support software system providing automated calculations and notification of early warning scores was correlated with decreasing LOS for ED patients. DECISIO’s automated decision support software system is an innovative tool in providing a vital sign-based early warning score in order to detect acute decompensation, resulting in improved outcomes for hospitalized patients in complex clinical environments such as the ED.

Abstract Title:

Automated early warning decision support to detect acute decompensation in the emergency department.P2

Authors:

Christopher Howard, Amber B Amspoker, Christopher K Morgan, Dick Kuo, Adol Esquivel, Tracey Rosen, Javad Razjouyan, Muhammad A Siddique, James P Herlihy, Aanand D Naik