The effect that rollovers and tripping-event sequence surrounding rollover crashes have on the severity of driver injury was explored. Three-year crash and inventory data from Michigan (N = 35,447) and Illinois (N = 24,296) were analyzed to explore the effect of rollover, while controlling for roadway, vehicle, and driver factors. The results show that 9 percent of single-vehicle crashes are either fatal or cause incapacitating driver injuries (K + A injuries). Significantly more fatalities and incapacitating injuries occur in rollover crashes. Logistic regression models of fatal and incapacitating injuries (K + A) versus other injuries and noninjuries (B + C + 0) were estimated separately and together for the two states. The results show that the effects of independent variables are largely consistent across the two states. Driver-injury severity increases with rollovers; failure to use a seat belt; passenger cars (as opposed to pickup trucks); alcohol use; daylight; rural roads (as opposed to urban); posted speed limit; and dry pavement (as opposed to slick pavement). Restricting the injury models to rollover crashes only, it was determined that hitting point objects or longitudinal objects before rolling over resulted in a more severe injury than rolling over first. The policy implications of the findings for restraint use, crash-testing, and roadside design and hardware are discussed.
ASJC Scopus subject areas
- Civil and Structural Engineering
- Mechanical Engineering