An epidemiological approach to emergency vehicle advanced warning system development: a two-phase study

Abstract

Motor vehicle crashes involving emergency vehicle (EV; police, fire trucks, ambulances, etc.) and non-EV drivers have been a known problem that contributes to fatal and nonfatal injuries; however, characteristics associated with non-EV drivers, involved in these crashes, have not been examined adequately. This two-phase study involved: Phase 1) data analysis, using The National Highway Traffic Safety Administration's Fatality Analysis Reporting System and the National Automotive Sampling System General Estimates System to identify driver, roadway, environmental, and crash factors, and consequences for non-EV drivers involved in fatal and nonfatal crashes with in-use and in-transport EVs; and Phase 2) design and analysis of the impact of two in-vehicle driver support systems that alert non-EV drivers to approaching EVs in a simulated urban environment, based on driving performance and usability measures under distracting and non-distracting conditions. Phase 1 analysis identified potential factors associated with non-EV drivers by utilizing epidemiological methodologies and multivariate logistic regression modeling. Non-EV drivers were more often involved in nonfatal crashes with EVs when driving: distracted (vs. not distracted; OR = 1.9); with vision obstructed by external objects (vs. no obstruction; OR = 36.4 for obstruction due to buildings); at intersections of four-points or more (vs. no intersection; OR = 2.1); at night (vs. midday; OR = 2.8); and in opposite directions (vs. same directions; OR = 4.8) of the EVs. Fatal crashes were associated with driving on urban roads (vs. rural; OR = 2.2); straight through intersections (vs. same direction; OR = 3.4) of four-points or more (vs. no intersection; OR = 4.9); and at night (vs. midday; OR = 1.6) although these types of crashes were less likely to occur on dark roads (vs. daylight; OR = 0.6). Consequences included increased risk for non-EV drivers to be fatally wounded (vs. no injury; OR = 2.1) among crashes involving at least one fatality. Phase 2 consisted of eighty-five participants completing a driving simulator trial-based experiment in which they encountered EVs crossing four-way intersections. Overall, the analysis indicated improved responses and roadway safety among participants presented with the driver support systems compared to participants presented with no driver support system. Most notably, participants were at decreased risk of collisions with EVs when given a driver support system (vs. no driver support system; OR = 0.3). The presence of the driver support systems did not increase in-vehicle distractions or perceived mental workload of the driving tasks. In addition, drivers indicated a moderate level of trust and reported the systems to be somewhat useful and satisfying.The findings of this two-phase study suggest drivers have difficulties in visually detecting EVs in different environments and that the use of technology may be beneficial as an intervention to mitigate roadway crashes between non-EV and EV drivers. Future research should continue to examine these interactions to identify methods to improve roadway safety.