This study examined the effect of alcohol use on the probabilities of injury, inpatient hospital stay, and emergency department visit. Data were obtained from a sample of adults (N = 1219) recruited from a Northern California county. Alcohol use measures included number of drinks, heavy drinking days, and an indicator variable for problem drinking. Models were estimated for men and women separately while controlling for confounders. Results indicate that most alcohol use measures were not significantly related to injury probability or medical care utilization. Among the exceptions, problem drinking was a significant positive predictor of any emergency department visit for both sexes. When drinkers during the past year were divided into light, moderate, and heavy drinking groups and compared to lifetime abstainers, all male drinkers had a higher probability of injury, and light and moderate female drinkers had a lower probability of an emergency department visit.