Single-vehicle (SV) crashes are of major concerns because of their high fatality rates. To understand the proneness of high injury severity for SV crashes without the influence due to striking objects, this study focuses on those SV crashes without colliding with pedestrians or fixed objects which are defined as out-of-control SV crashes. The relationship between factors including driver-vehicle/rider-vehicle, roadway and environmental characteristics is investigated using the ordered probit model. For the purpose of comparison, separate analyses for motorcycle/motor-scooter riders and motor-vehicle drivers are conducted separately.

The results showed that for both riders and drivers, variables such as age (65 and above), drink driving, error type of failing to have proper control, driving maneuvers of left and right turns as well as driving after midnight are associated with more severe injuries whereas factors such as wet, oily or sandy surfaces are related to less severe injury. Four other variables, i.e., foreign vehicle registration, probation or expired licence, high speed-limit roads, and type of median lane, have different influence on riders and drivers on injury severity. For instance, riders with foreign-registered motorcycles tend to be less prone to severe injury whereas drivers with foreign passenger vehicles are more likely to be severely injured. Additionally, factors such as road traffic type are only found to significantly influence riders but not to drivers, in that riders tend to suffer from more severe injury when involved in out-of-control SV crashes on two-way undivided roads and expressways as compared to two-way divided roads. Nationality, is found to influence drivers but not riders, in that foreign drivers are associated with less severe injury in contrast to the finding that foreign vehicles tend to aggrevate injury severity.

The results shed light on the different causes of high injury severity for riders and drivers involved in out-of-control SV crashes. From these, three situations are concluded to generally influence injury severity: risky driving behaviors, impaired judgement, and potential interaction with other vehicles. Based on the findings, targeted countermeasures may be introduced from multiple perspectives such as driver education and policy development to improve non traffic-interactive safety.