Several recent reports have described abdominal injuries occurring as a result of seat belt use, raising concerns about seat belts as an agent of injury in motor vehicle crashes. The purpose of this study was to characterize the distribution of abdominal injuries after motor vehicle crashes in belted and unbelted patients admitted to trauma centers. The mortality was higher in unbelted than belted patients (7% vs. 3.2%, respectively, p < 0.0001). Unbelted patients also had significantly more frequent and more severe head injuries (50.0% vs. 32.9%, respectively, p < 0.001). The incidence of abdominal injury was equal in both unbelted and belted patients (13.9%), but the spectrum of orgas injuried was different in the two groups. Gastrointestinal tract injuries (stomach, small bowel, colon and rectum) were significantly more freouent in belted vs. unbelted patients (3.4% vs. 1.8%, respectively, p = 0.001). The frequency of liver and spleen injuries was the same in both groups. This study demonstrates that in patients admitted to trauma centers after motor vehicle crashes, belted and unbelted patients have an equal incidence of abdominal injury, but belted and unbelted patients have a different spectrum of injuries. Hollow viscus injuries are more common in belted crash victims. Seat belt use was associated with significantly fewer head injuries and deaths. Physicians evaluating trauma victims after motor vehicle crashes should be aware of the fact that the types of abdominal injuries may vary substantially depending on seat belt use.