Estimated crash injury risk and crash characteristics for motorsport drivers

Abstract

ObjectiveMotorsport crash events are complex and driver restraint systems are unique to the motorsport environment. The National Association for Stock Car Auto Racing, Incorporated (NASCAR®) crash and medical datasets provide an opportunity to assess crash statistics and the relationship between crash characteristics and driver injury. Injury risk curves can estimate driver injury risk and can be developed using vehicle incident data recorder information as inputs. These relationships may provide guidance and insight for at-track emergency response, driver triage and treatment protocols. MethodEight race seasons of crash and medical record data (including Association for the Advancement of Automotive Medicine Abbreviated Injury Scale (AIS) scores) from the Monster Energy NASCAR Cup Series & NASCAR Xfinity Series were processed and analyzed. Multiple logistic regression modeling was used to produce injury risk curves from longitudinal and lateral resultant change in velocity, resultant peak acceleration, principal direction of force and the number of impacts per incident. Results2065 Unique IDR data files were matched with 246 cases of driver injury or sub-injury (severity below AIS 1) and 1819 no-injury cases. Multiple logistic regression modeling showed increasing resultant change in velocity, resultant peak acceleration and the number of impacts during a crash event all increase estimated driver injury risk. After accounting for the other predictors in the model, right lateral impacts were found to have a lower estimated injury risk. The model produced an Area Under the Receiver Operating Characteristics curve of 0.80. Across the eight race seasons in this study the overall average resultant change in velocity was 34.4 kph (21.4 mph) and the average resultant peak acceleration was 19.0 G for an average of 258 crashes per season. For 2011 through 2015, full time drivers experienced 134 times more crashes per mile traveled than passenger vehicles, but experienced 9.3 times fewer injuries per crash. ConclusionMultiple logistic regression was used to estimate AIS 1+ injury only and AIS 1+ with sub-injury risk for motorsport drivers using motorsport-specific crash and medical record databases. The injury risk estimate models can provide future guidance and insight for at-track emergency medical response dispatch immediately following an on-track crash. These models may also inform future driver triage protocols and influence future expenditures on motorsports safety research.