Between 2000 and 2003, the National Highway Traffic Safety Administration (NHTSA) of the United States Department of Transportation (USDOT) funded demonstration projects designed to reduce impaired driving through wellpublicized and frequent enforcement in seven States:​ Georgia, Louisiana, Pennsylvania, Tennessee, Texas, Indiana, and Michigan. Significant reductions in fatal crashes in the intervention States relative to surrounding States were obtained in Georgia, Tennessee, Indiana and Michigan when an interrupted time-series analysis of Fatality Analysis Reporting System (FARS) data was used comparing the ratio of drinking to non-drinking drivers in fatal crashes. Significant reductions in a second measure, alcohol-related fatalities per 100 million vehicle miles traveled (VMT), were also obtained in Indiana and Michigan. The other three States showed only marginal, non-significant changes relative to their comparison jurisdictions or States. As compared to surrounding States, fatal crash reductions in Georgia, Tennessee, Indiana, and Michigan ranged from 11 to 20 percent. In these four States, the programs were estimated to have saved lives ranging from 25 in Indiana to 43 in Tennessee to 57 in Michigan to 60 in Georgia. Some common features of the programs that experienced significant reductions included the use of paid media to publicize the enforcement (in three States), using a statewide model rather than selected portions of the State (all four States), and the use of highly visible and frequent sobriety checkpoints (in three States). In summary, it appears that a variety of media and enforcement procedures that supplement ongoing statewide efforts can yield meaningful crash reduction effects among alcohol impaired drivers.