Recommendations were made in 2008 regarding advanced automatic collision notification or AACN and the data that should be used in attempting to predict the need for trauma center care (CDC, 2008). Some have considered those recommendations and begun to produce injury predicting algorithms that can be used in part to communicate the severity of crashes to emergency medical services (EMS) and trauma personnel (Kononen et al., 2010). One possible shortcoming of many of the data sets being used and the resulting algorithms is their reliance on investigator estimated change in velocity (delta V).

Prior work has investigated the predictive ability of various occupant and crash variables as they related to occupant outcomes (Craig et al, 2009). The National Highway Traffic Safety Administration’s (NHTSA) Crash Injury Research and Engineering Network (CIREN) database provided the detailed crash and injury data as well as hospital care-based outcomes to enable that study. The current study has continued that work, but with an emphasis on studying the significance of the association between individual event data recorder (EDR) or telematics variables and patient outcomes that most justify the need for the highest level of care.

The primary aim of this study was to document the association between potential EDR or telematics variables and occupant outcomes using three frontal crash data sets. Analysis was limited to data that could be collected via telematics or voice communication and involved logistic regression analysis to document variables that were significant associated with the occupant outcomes studied. Two CIREN (non-EDR and EDR) and one National Automotive Sampling System – Crashworthiness Data System (NASS-CDS) (EDR) data sets were analyzed. The CIREN data sets were used to study the association between predictors and hospital care-based outcomes. The NASS-CDS EDR data set was used to evaluate the association between the same predictors used in CIREN data analysis and injury severity-based outcomes. Both EDR data sets were also analyzed to evaluate differences in the predictive ability of delta V obtained from an EDR versus delta V calculated as part of the crash reconstruction (using WinSMASH, e.g.).

The results of this study show that many of the recommended predictors (CDC, 2008) were significantly associated with the outcomes of interest. The study also found that EDR delta V can be a better predictor of outcomes than WinSMASH delta V. This finding may have implications for the development and application of injury predicting algorithms that could be used as part of an AACN system.