Self-reported alcohol consumption has become an important adjunct to brief intervention strategies used to reduce drinking after motor vehicle crashes (MVC). The purpose of this study was to determine the validity of self-reported alcohol consumption following an alcohol-related MVC in hospitalized nondependent drinkers. Blood alcohol concentration (BAC) was measured on admission. When the subjects were alcohol free, self-reported alcohol consumption was obtained and estimated BAC (EBAC) was calculated from self-reported data. Of 114 subjects (24 females; 90 males), only 5% (6 of 114) denied drinking prior to the crash. The correlation between BAC and EBAC was r=0.37 for females and r=0.36 for males. Males systematically under-reported but self-reports generally reflected drinking trends. Females both over-reported and under-reported drinking.