Understanding the impact of road design characteristic on teen driver's fatality.

There is little research simultaneously examining both teen drivers' characteristics and the driving environment. This study aimed to investigate how traffic engineering factors (road designs, traffic operations, and surface conditions) contribute to risks of teen driver's fatality. Understanding the role of traffic engineering factors can further improve efforts to reduce teen diving-related fatal crashes. This study used the Fatality Analysis Reporting System and General Estimate System for years 2014 to 2015. We explored numerous engineering variables: horizontal/vertical alignment, intersections types, interchange areas, traffic ways types, posted speed limit, work zone areas, types of traffic control devices and road surface conditions. Traditional variables considered in this study included sex, age, presence of passengers in relation to age group, time of day, lighting condition, season, vehicle model year, vehicle type, seatbelt use, crash type and movement prior to a crash. We ran cross-tabulations and unadjusted analyses to report unadjusted odds ratios of teen driver's fatality. A multiple logistic regression model was used to identify which variables related to road design and traffic operations remained significant after being adjusted by other variables. For traditional factors, results corroborated prior findings. All the engineering factors, except work zone area, were significant in the unadjusted analyses; four variables - horizontal alignments, posted speed limit, traffic control device type, and traffic way type - remained statistically significant in the final model. The adjusted odds ratio (AOR) of curved high-speed segment was nearly seven times (6.94; 95% CI = 3.58, 13.46) that of a straight segment with a low posted speed limit (reference group). The AOR of no traffic control was four times (4.17; 95% CI = 2.95, 5.88) that of roads with traffic controls. The AOR of a two-way undivided road was almost three times (2.75, 95% CI = 1.43, 5.29) that of a two-way median barrier road (reference group). This study suggests teen driver education programs consider addressing road design characteristics. For example, incorporating focused lessons on traffic signs and traffic control types may better prepare young people for their driving experiences. Additionally, more supervised driving time is needed on various road types, especially those that present greater risks (e.g., roads with grades/curves, intersections with no traffic signal, undivided roads, and unpaved roads).