Evaluation of the efficacy of Search out Severity Score and National Early Warning Score for urgent Endotracheal intubations or unanticipated intensive care unit (ICU) admission for an elderly patient in general medical wards

Main Article Content

Sunee Suwanpasu
Werapattra Praparpak
Sukanya Poonsap


Purpose: To investigate the efficacy of Search out Severity Score (SOS) and National Early Warning Score (NEWS) score in predicting urgent endotracheal intubations or unanticipated intensive care unit admission

Design: Prospective analytical study

Method: The sample was patient’s ages 60 years and over admitted to general medical wards at King Chulalongkorn Memorial Hospital between November 1, 2019, to July 31, 2020.  Purposive sampling was selected based on criteria. A sample of 85 was obtained. Data were collected using a questionnaire about personal information and observation form composed of respiratory rate, heart rate, systolic blood pressure, body temperature, urine output, and conscious level for the calculation of NEWS and SOS. The data were analyzed for sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC)

Finding: The NEWS and SOS systems at 4 hours before the event were the most effective, with the cut-off values ​​of NEWS> 7 and SOS> 4 giving the best accuracy (sensitivity 71.4%, 64.3% specificity 98.2%, 98.2%. Respectively). The area under the ROC curve [AUC]) was 0.90 (95% CI 0.81-0.99) and 0.86 (95% CI 0.75-0.97), respectively. The prediction efficiency decreased over time before the event. However, at 8, 12, 24 hours before the incident, it was still accurate at a good-fair level (AUC 0.82, 0.82, 0.74, and 0.77, 0.79, 0.72, respectively).

Conclusion: The NEWS system provides excellent accuracy, while the SOS system gives a good accuracy level. Therefore, the NEWS system should be used for the early detection of abnormalities and as a guideline for the rapid response to changes in the vulnerable elderly. It keeps patients out of the crisis and is safe and able to predict worsening patients.


Download data is not yet available.

Article Details

Research articles