Related Experiment Video
Updated: Dec 20, 2025
Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
Published on: May 15, 2020
Improving the ability to predict hospital mortality among adults by combining two status epilepticus outcome scoring
Yan Jiang1, Yi Yang2, Fei Feng3
1Department of Neurology, The First Affiliated Hospital, Zhejiang Chinese Medical University, Hangzhou, China.
Objective:
The goal of this study was to compare the predictive ability of the Status Epilepticus Severity Score (STESS), the Encephalitis-nonconvulsive status epilepticus (NCSE)-Diazepam resistance-Image abnormalities-Tracheal intubation (END-IT), and the combination of these two scoring tools to predict mortality among inhospital patients with status epilepticus (SE).
Methods:
A retrospective analysis was conducted of adult patients with SE who were admitted to the neurology department, the emergency department, and the intensive care unit from January 2013 to December 2017. The patients were divided into two groups: survivors and nonsurvivors. The STESS data were obtained when the patient arrived at the hospital, and the END-IT data were collected 24 h after patients were initially treated in the hospital. The ability of the scoring tools to predict death in patients with SE, alone or in combination, was evaluated.
Results:
A total of 123 patients with SE were included in the study, of which 22 died, for a mortality rate of 17.9%. The STESS and END-IT scores of nonsurvivors were both significantly higher than those of survivors (median STESS 4 vs. 2, p = 0.003; median END-IT 3 vs. 1, p < 0.001). The area under the receiver operating characteristic curve (AUC) was 0.698 for the STESS and 0.852 for the END-IT, and the cutoff values were 4 and 3, respectively. The AUC of the END-IT with the optimal cutoff value was larger than that of the STESS (p = 0.024). The sensitivity and specificity of combining the STESS and END-IT by the serial method (STESS ≥ 4∩END-IT ≥ 3) were 0.50 and 0.95, respectively, and the specificity was significantly higher than the STESS or END-IT (both p's < 0.001). The sensitivity and specificity of combining the STESS and END-IT by the parallel method (STESS ≥ 4⋃END-IT ≥ 3) were 0.91 and 0.53, respectively, and the sensitivity was higher than the STESS was (p = 0.016).
Conclusion:
Our results indicated that the combined score of the STESS and END-IT systems was a better predictor of survival of patients with SE than the scores of either the STESS system or the END-IT system alone and that combining the scores may be considered to be a new method for early identification of patients for both good and bad outcomes.
