The most reliable information about alcohol involvement comes from fatal crashes. In 2007, the National Highway Traffic Safety Administration (NHTSA) estimated that 32 percent of all traffic deaths occurred in crashes in which at least one driver had a BAC at or above 0.08 percent and that some alcohol was present in 37 percent of drivers involved in all traffic fatalities. 4 Thirty-five percent of pedestrians killed in crashes also had BACs at or above 0.08 percent. Such statistics do not mean that a third or more of all fatal crashes are caused solely by alcohol, because alcohol may be only one of several factors that contribute to a crash involving drinking drivers. A 2007 Institute study estimated that 13,452 of the approximately 15,000 alcohol-related crash deaths in 2005 were directly attributable to alcohol. These lives could have been saved if all drivers had been restricted to no blood alcohol. About 8,900 lives could have been saved by reducing BACs to less than 0.08 percent, and 11,100 lives could have been saved by reducing drivers' BACs to less than 0.05 percent.5 It has also been estimated that 15 percent of all crashes occurring between 4pm and 2am would be avoided if no driver had a BAC of 0.05 or above.6 Alcohol involvement is much lower in crashes involving nonfatal injuries and lower still in crashes that do not involve injuries at all. A study conducted during the 1960s estimated that 9 percent of drivers in injury crashes and 5 percent of drivers in non-injury crashes in Grand Rapids, Michigan, had BACs at or above 0.10 percent.7 A 1977 study found that 12 percent of drivers in injury crashes in Huntsville, Alabama, and San Diego, California, had BACs at or above 0.10 percent, compared with 1 percent of a sample of drivers not involved in crashes.8 According to a research review conducted during this same period, studies of fatally injured drivers found that 40-55 percent had BACs at or above 0.10 percent.9
