Overall, the percent of drivers involved in fatal crashes who had consumed alcohol and had blood alcohol concentration (BAC) of .08 or above prior to the crash steadily decreased from 1982 to 1997 and then leveled off (more or less). In an attempt to explain the 1982-1997 reduction and the 1997-2005 level trend, this report presents a statistical analysis of factors that influenced the historical alcohol-related driving trends from 1982 to 2005. The study is based on disaggregate logistic regression of imputed Fatality Analysis Reporting System (FARS) cases from all 50 States and the District of Columbia – to predict the probability of an alcohol-related involvement given a set of independent variables. The independent variables include alcohol-related legislation (i.e., .10 BAC, .08 BAC, Administrative License Revocation, minimum-legal-drinking-age laws), demographic factors (i.e., driver’s age and gender), per capita alcohol consumption, and external factors (i.e., day of the week, time of day, roadway function class, and posted speed limit). The independent variables explain both the decrease in alcohol-related fatal crashes (where drivers involved in fatal crashes had BAC of .08 or above) from 1982 to 1997 and its leveling off after 1997. Large portions of the reduction are explained by the effect of alcohol-laws and by the demographic trends – the aging of the population and the increased proportion of female drivers. The leveling off after 1997 does not imply that the laws are becoming less effective. On the contrary, they effectively maintain the proportion of drivers in fatal crashes who had BAC of .08 or above at the time of the crash – at the lowest level since 1982. (Author/publisher)
Samenvatting