Although it is well established that alcohol is associated with increased probability of traumatic injury, the assumption has been that the relationship is attributable to alcohol's impairment of judgment and psychomotor performance. Detailed analysis of data from over 1,000,000 crashes indicates that, when injury-related variables such as driver belt usage, vehicle deformation, mode of crash, speed, driver age, and vehicle weight are controlled, the drinking driver is still more likely to suffer serious injury or death compared to the nondrinking driver. These findings do not support the commonly held notion that alcohol is protective against injury but rather suggest that alcohol increases the vulnerability to injury in any given crash.