Examination of independent contributions of driver, crash, vehicle and geometric characteristics to driver fatalities

Abstract

In 2016, the National Highway Traffic Safety Administration (NHTSA) had recorded a total of 37,461 road-related mortalities within the United States alone. To alleviate the effects of road fatalities concerning economy and health, numerous factors must be scrutinized for their distinct contribution to such events. In Bédard et al. (2002), the authors examined the independent contribution of driver, vehicle, and crash factors to driver fatalities. Findings established that older drivers, higher travelling speed, seatbelts, vehicle wheelbase, model year and blood alcohol content have a direct or indirect effect on driver fatalities. This research study extends the previous work from Bédard et al. (2002) by applying 17 years of extra crash data. Additional variables not found in the original Bédard et al. (2002) study, such as geometric aspects, crash characteristics, atmospheric, light, and road surface conditions are also considered. This research also examines the impact of geometric factors on mortality trends for collisions with fixed objects. Crash data were obtained from the Fatality Analysis Reporting System (FARS). To assess the relationship between the various factors and driver fatalities, univariate and multivariable logistic regression are used to compute odds ratios (adjusted and unadjusted). Also, Poisson regression was carried out to determine the expected total number of driver fatality counts based on roadway geometric factors. Results from univariate and multivariate regressions indicate that increased driver ages, increased vehicle speeds, increased vehicle ages, females, left-side crashes, and BAC ≥ 0.15 mg/L were associated with higher odds of driver fatalities. Conversely, drivers wearing shoulder and lap belts were found to have lower odds of fatalities. Poisson regression showed that curved alignment, dry weather, dark conditions, minor arterial roads, male drivers, non-junctions, weekends, passenger cars (sedans), and 20 -29 and 80+ years age categories to have a significant impact on driver fatality counts. Furthermore, based on research results some countermeasures are recommended such as, providing driver simulator, and on-road driving evaluation training programs, emergency medical services (EMS) without any delays, and keeping fixed objects away from the roadway and others.