Kenneth E Stewart1, Linda D Cowan, David M Thompson. 1. Emergency Systems Division, Oklahoma State Department of Health, 1000 NE 10th Street, Rm 1104, Oklahoma City, OK 73117-1299, United States. kenneth@health.ok.gov
Abstract
BACKGROUND: The Abbreviated Injury Scale (AIS) recently underwent a major revision from AIS 98 to AIS 05. AIS injury codes form the basis of widely used injury severity scores such as the injury severity score (ISS). ISS thresholds are often used in trauma case definitions and ISS is widely used in injury research to adjust for injury severity. This study evaluated changes from AIS 98 to AIS 05, the changes' effect on ISS distributions, and presents an application of the results. METHODS: Injury descriptions from medical records of 137 randomly selected patients in the Oklahoma Trauma Registry (OTR) were obtained. A single trained coder used AIS 98 and AIS 05 to code each injury. ISS values were calculated and grouped into 4 categories: 1-8, 9-14, 16-24, >24. Paired ISS was compared using Kappa statistics and tests of symmetry. We identified common injury diagnoses for which AIS severity changed between versions. Estimates of the proportion of patients changing ISS groups were applied to the entire OTR to assess the impact on reporting and on a model for reimbursement. RESULTS: OTR AIS 98 and manual AIS 98-based ISS values had a weighted Kappa of 0.71. OTR AIS 98 and manual AIS 05-based ISS values had a Kappa of 0.58. Manual AIS 98 and manual AIS 05 ISS had the highest Kappa of 0.81, however, though the scores differed by only 1 ISS category, there were 30 discordant pairs. The distribution of these discordant pairs was not symmetrical (Bowker's S=30; df=6; p<0.0001) with AIS 05-based ISS values consistently shifted to a lower ISS category. Reductions in AIS severity and ISS values using AIS 05 were common for extremity fractures and thorax injuries. The results suggest fewer patients would be reported to the OTR or be eligible for reimbursement. DISCUSSION: Changing from AIS 98 to AIS 05 injury coding resulted in systematic changes in AIS codes and ISS. Specific injuries and body regions were differentially affected. Trauma registries and injury researchers that use AIS based injury coding can use this information to evaluate the potential impact of changes in AIS 2005.
BACKGROUND: The Abbreviated Injury Scale (AIS) recently underwent a major revision from AIS 98 to AIS 05. AIS injury codes form the basis of widely used injury severity scores such as the injury severity score (ISS). ISS thresholds are often used in trauma case definitions and ISS is widely used in injury research to adjust for injury severity. This study evaluated changes from AIS 98 to AIS 05, the changes' effect on ISS distributions, and presents an application of the results. METHODS: Injury descriptions from medical records of 137 randomly selected patients in the Oklahoma Trauma Registry (OTR) were obtained. A single trained coder used AIS 98 and AIS 05 to code each injury. ISS values were calculated and grouped into 4 categories: 1-8, 9-14, 16-24, >24. Paired ISS was compared using Kappa statistics and tests of symmetry. We identified common injury diagnoses for which AIS severity changed between versions. Estimates of the proportion of patients changing ISS groups were applied to the entire OTR to assess the impact on reporting and on a model for reimbursement. RESULTS: OTR AIS 98 and manual AIS 98-based ISS values had a weighted Kappa of 0.71. OTR AIS 98 and manual AIS 05-based ISS values had a Kappa of 0.58. Manual AIS 98 and manual AIS 05 ISS had the highest Kappa of 0.81, however, though the scores differed by only 1 ISS category, there were 30 discordant pairs. The distribution of these discordant pairs was not symmetrical (Bowker's S=30; df=6; p<0.0001) with AIS 05-based ISS values consistently shifted to a lower ISS category. Reductions in AIS severity and ISS values using AIS 05 were common for extremity fractures and thorax injuries. The results suggest fewer patients would be reported to the OTR or be eligible for reimbursement. DISCUSSION: Changing from AIS 98 to AIS 05 injury coding resulted in systematic changes in AIS codes and ISS. Specific injuries and body regions were differentially affected. Trauma registries and injury researchers that use AIS based injury coding can use this information to evaluate the potential impact of changes in AIS 2005.
Authors: Daniel W Spaite; Chengcheng Hu; Bentley J Bobrow; Vatsal Chikani; Bruce Barnhart; Joshua B Gaither; Kurt R Denninghoff; P David Adelson; Samuel M Keim; Chad Viscusi; Terry Mullins; Amber D Rice; Duane Sherrill Journal: Ann Emerg Med Date: 2017-05-27 Impact factor: 5.721
Authors: Amy C Gunning; Menco J S Niemeyer; Mark van Heijl; Karlijn J P van Wessem; Ronald V Maier; Zsolt J Balogh; Luke P H Leenen Journal: Eur J Trauma Emerg Surg Date: 2022-08-16 Impact factor: 2.374
Authors: Rebeca Abajas-Bustillo; Francisco José Amo-Setién; César Leal-Costa; María Del Carmen Ortego-Mate; María Seguí-Gómez; María Jesús Durá-Ros; Mark R Zonfrillo Journal: PLoS One Date: 2019-05-01 Impact factor: 3.240
Authors: Zar Popal; Eva Berkeveld; Kees Jan Ponsen; Harold Goei; Frank W Bloemers; Wietse P Zuidema; Georgios F Giannakopoulos Journal: Eur J Trauma Emerg Surg Date: 2019-09-04 Impact factor: 3.693
Authors: Jan C Van Ditshuizen; Charlie A Sewalt; Cameron S Palmer; Esther M M Van Lieshout; Michiel H J Verhofstad; Dennis Den Hartog Journal: Scand J Trauma Resusc Emerg Med Date: 2021-05-27 Impact factor: 2.953