BACKGROUND: The Injury Severity Score (ISS) and the New Injury Severity Score (NISS) are widely used for anatomic severity assessments, but they do not display a linear relation to mortality. The mortality rates are significantly different between pairs of the Abbreviated Injury Scale (AIS) triplets that generate the same ISS/NISS total. METHODS: The Logarithm Injury Severity Score (LISS) is defined as a change in AIS values by raising each AIS severity score (1-6) by taking the natural logarithm to a power of 5.53 multiplied by 1.7987 and then adding the three most severe injuries (i.e. highest AIS), regardless of body region. LISS values were calculated for every patient in three large independent data sets: 3,784, 4,436, and 4,018 patients treated over a six-year period at Class A tertiary comprehensive hospitals in China. The power of LISS to predict morality was then compared with previously calculated NISS values for the same patients in each of the three data sets. RESULTS: We found that LISS is more predictive of survival as well (Hangzhou: receiver operating characteristic (ROC): NISS=0.931, LISS=0.949, p=0.006; Similarly, Zhejiang and Shenyang: ROC NISS vs. LISS, p<0.05). Moreover, LISS provides a better fit throughout its entire range of predicting (Hosmer-Lemeshow statistic for Hangzhou NISS=15.76, p=0.027; LISS=13.79, p=0.055; Similarly, for Zhejiang and Shenyang). CONCLUSION: LISS should be used as the standard summary measure of human trauma.
BACKGROUND: The Injury Severity Score (ISS) and the New Injury Severity Score (NISS) are widely used for anatomic severity assessments, but they do not display a linear relation to mortality. The mortality rates are significantly different between pairs of the Abbreviated Injury Scale (AIS) triplets that generate the same ISS/NISS total. METHODS: The Logarithm Injury Severity Score (LISS) is defined as a change in AIS values by raising each AIS severity score (1-6) by taking the natural logarithm to a power of 5.53 multiplied by 1.7987 and then adding the three most severe injuries (i.e. highest AIS), regardless of body region. LISS values were calculated for every patient in three large independent data sets: 3,784, 4,436, and 4,018 patients treated over a six-year period at Class A tertiary comprehensive hospitals in China. The power of LISS to predict morality was then compared with previously calculated NISS values for the same patients in each of the three data sets. RESULTS: We found that LISS is more predictive of survival as well (Hangzhou: receiver operating characteristic (ROC): NISS=0.931, LISS=0.949, p=0.006; Similarly, Zhejiang and Shenyang: ROC NISS vs. LISS, p<0.05). Moreover, LISS provides a better fit throughout its entire range of predicting (Hosmer-Lemeshow statistic for Hangzhou NISS=15.76, p=0.027; LISS=13.79, p=0.055; Similarly, for Zhejiang and Shenyang). CONCLUSION: LISS should be used as the standard summary measure of humantrauma.