It is well recognized that alcohol increases the risk of injury. It is controversial, however, whether alcohol also has an effect on the degree and outcome of injury after controlling for the severity of impact. This review examines the evidence from experimental studies in animals regarding the potentiating effects of alcohol on trauma, and the methodologic issues that may have contributed to the contradictory findings of clinical and epidemiologic studies. Most experimental studies indicate that alcohol can adversely affect the degree and outcome of injury. In controlled laboratory conditions, alcohol is found to reduce cardiac output, to increase the susceptibility to hemorrhagic shock, and to increase the pulmonary vascular resistance after standardized experimental injuries. However, it is difficult to extrapolate these findings to humans, partly because the interactive effects of chronic and acute alcohol use on trauma are rarely considered in experimental studies. The conflicting results in studies involving trauma patients are due in part to the differences in study design, particularly selection of study populations, and in measuring and controlling for kinetic forces on the body. Studies indicating that alcohol is associated with an increased risk of serious or fatal injury are usually based on data from emergency departments or police departments. The alcohol-injury severity relationship reported in these studies is attributable to a great extent to the effects of correlates of alcohol, such as speeding and not wearing seat belts, rather than the biological effects of alcohol. Studies indicating that alcohol is not associated with the degree and outcomes of injury are mostly those involving patients who were admitted to hospitals or trauma centers. Methodologic issues concerning the alcohol-injury severity controversy, including conceptual models and future research needs, are discussed.