Stitzel 1 ESTIMATED TRIAGE IMPROVEMENT AND SOCIETAL COST SAVINGS WITH NATIONWIDE IMPLEMENTATION OF AN ADVANCED AUTOMATIC CRASH NOTIFICATION ALGORITHM IN THE U.S

The objective of the study was to perform a benefits analysis of an injury-based advanced automatic crash notification (AACN) algorithm to estimate the reduction in undertriage (UT) and overtriage (OT) rates if the algorithm was implemented nationwide. In addition, the study sought to estimate the savings in societal costs using the Harm metric score. Benefits of an existing AACN algorithm were estimated using the differential between the number of motor vehicle crash (MVC) occupants correctly triaged using the AACN algorithm compared to actual triage decisions for realworld occupants. Actual triage decisions were extracted from NASS-CDS 2000-2011 to compare each occupant’s ISS (≥ 16 vs. < 16) versus the occupant’s actual triage destination (trauma center or non-trauma center). Analyses for the triage benefits included 47,361 unweighted occupants representing 9,763,984 weighted occupants with complete information on the crash mode, hospital destination, and a valid ISS. Analyses for the societal cost benefits included the subset of occupants included in the AACN algorithm resulting in 22,610 unweighted occupants. For each occupant, the Harm scores were summed across all injuries and a lifetime economic cost to society for each fatality value of $1.4 million was applied to derive the societal costs. The AACN algorithm was optimized and evaluated to minimize triage rates per American College of Surgeons (ACS) recommendations of <5% UT and <50% OT. The developed AACN algorithm resulted in <50% OT and <5% UT in side impacts and 6-16% UT in other crash modes. The weighted analysis of 12 years of NASS-CDS data revealed an UT rate of 20% and an OT rate of 54% among real-world MVC occupants across all crash modes. If the triage rates from the AACN algorithm were applied to these occupants by crash mode, the UT and OT rates would be reduced to 11% and 34%, respectively. Across a 12-year period, this would result in an UT improvement for 32,959 (44%) occupants and OT improvement for 1,947,620 (38%) occupants. For the societal cost benefit analysis, the sample sizes were small for all crash modes except frontal. Application of the AACN algorithm for the frontal cases would reduce UT rates to 7%. Application of the Harm score to the occupants who would benefit from trauma center treatment would result in over $43 million saved in societal costs. With nationwide implementation of the AACN algorithm, we estimate a potential benefit of improved triage decision-making for 165,048 occupants annually (one-twelfth of the 1.98 million occupants incorrectly triaged predicted to be triaged correctly with the AACN algorithm). Annually, this translates to more appropriate care for 2,747 seriously injured occupants and reduces unnecessary utilization of trauma center resources for 162,302 minimally injured occupants. The projected reduction in UT for the U.S. population attributable to AACN has important implications for decreasing MVC mortality and morbidity, while reduction of OT will lead to better hospital resource utilization and decreased healthcare costs.