A Changing of the Abbreviated Injury Scale that Improves Accuracy and Simplifies Scoring. Issue 3 (May 2013)

• Record Type: Journal Article
• Authors: Wang, MD; Cai, WW; Qiu, WS; Qiu, F; Lv, WS
• Is Part Of: Hong Kong journal of emergency medicine. Volume 20:Issue 3(2013)
• Page Start: 138
• Page End: 145
• Publication Date: 2013-05
• Subjects: Abbreviated injury scale -- changed injury severity score -- injury severity score -- new injury severity score -- prediction of mortality; Emergency medicine -- Periodicals; 616.02505
• DOI: 10.1177/102490791302000303
• Languages: English
• ISSNs: 2309-5407

Abstract

Objective: We present here a changing of the abbreviated injury scale (AIS). It is called the changed injury severity score (CISS), and significantly outperforms the venerable but dated the injury severity score (ISS) and the new injury severity score (NISS) as a predictor of mortality. Methods: The CISS is defined as a change of AIS values by raising each AIS severity score (1-6) by a power of 4.12 divided by 30.33 and then summing the three most severe (i.e. highest AIS) regardless of body regions. CISS values were calculated for every patient in two large independent data sets: 3455, 3900 patients treated during a five-year period at the class A grade III comprehensive hospitals in Affiliated Hospital of Hangzhou Normal University (Hangzhou) and Zhejiang Provincial People's Hospital (Zhejiang). The power of CISS to predict morality was then compared with previously calculated NISS values of the same group patients in the two hospitals. Results: We found CISS was more accurate than NISS to predict the survival. The receiver operating characteristic (ROC) of NISS and CISS in Hangzhou were 0.919 and 0.937 respectively (p=0.026), whereas for Zhejiang were 0.917 and 0.940 respectively (p=0.022). Moreover, CISS provided a better fit throughout its entire range of prediction. Hosmer-Lemeshow (H-L) statistic for NISS and CISS in Hangzhou were 24.00 (p=0.002) and 19.38 (p=0.007), whereas in Zhejiang were 22.70 (p=0.001) and 18.43 (p=0.005) respectively. Conclusions: CISS is a modified version of NISS/ISS with better statistical property and can be considered in trauma research.