• Hideo Tohira, Ian Jacobs, David Mountain, Nick Gibson, Allen Yeo, Masato Ueno, and Hiroaki Watanabe.
• School of Primary, Aboriginal and Rural Health Care, University of Western Australia, Western Australia, Australia. soba@tororo.net
• J Trauma. 2011 Dec 1;71(6):1829-34.

BackgroundThe Abbreviated Injury Scale 2008 (AIS 2008) is the most recent injury coding system. A mapping table from a previous AIS 98 to AIS 2008 is available. However, AIS 98 codes that are unmappable to AIS 2008 codes exist in this table. Furthermore, some AIS 98 codes can be mapped to multiple candidate AIS 2008 codes with different severities. We aimed to modify the original table to adjust the severities and to validate these changes.MethodsWe modified the original table by adding links from unmappable AIS 98 codes to AIS 2008 codes. We applied the original table and our modified table to AIS 98 codes for major trauma patients. We also assigned candidate codes with different severities the weighted averages of their severities as an adjusted severity. The proportion of cases whose injury severity scores (ISSs) were computable were compared. We also compared the agreement of the ISS and New ISS (NISS) between manually determined AIS 2008 codes (MAN) and mapped codes by using our table (MAP) with unadjusted or adjusted severities.ResultsAll and 72.3% of cases had their ISSs computed by our modified table and the original table, respectively. The agreement between MAN and MAP with respect to the ISS and NISS was substantial (intraclass correlation coefficient = 0.939 for ISS and 0.943 for NISS). Using adjusted severities, the agreements of the ISS and NISS improved to 0.953 (p = 0.11) and 0.963 (p = 0.007), respectively.ConclusionOur modified mapping table seems to allow more ISSs to be computed than the original table. Severity scores exhibited substantial agreement between MAN and MAP. The use of adjusted severities improved these agreements further.

17 keywords...

hide…