Alcohol and fatal accidents in the United States - A time series analysis for 1950-2002

Aims: To estimate the association between per capita alcohol consumption and fatal accidents in the United States and to compare the outcome with findings from Europe and Canada. Data and method: Yearly data on fatal accidents by gender and age were analysed in relation to per capita alcohol consumption for 1950-2002 using the Box-Jenkins technique for time series analysis. Findings: A 1-L increase in per capita consumption was on average followed by 4.4 male deaths per 100,000 inhabitants, but had no significant effect on female accident mortality. Regarding specific categories of accidents, the effect on fatal motor vehicle accidents accounted for a large part of the overall effect for men and was also significant for women. With respect to fatal falling accidents and other accidents, the only significant effects were found among young males. As concerns women, the association with per capita consumption in the US was weak in comparison with Canada and Europe. The US effect estimate for overall male accidents was however equally strong as in Northern Europe (5.2) or Canada (5.9), and stronger than that found in Central and Southern Europe (2.1 and 1.6, respectively). With respect to alcohol and fatal motor vehicle accidents, the association for men of 3.2 was stronger than in Europe and more similar to the Canadian finding (3.6). Conclusions: Per capita alcohol consumption has at least partly been an explanation for the development of male fatal accidents and particularly motor vehicle accident rates in the post-war United States. High traffic density and relatively high legal limits for drunken driving blood alcohol concentration (BAC) are suggested to explain the strong association found between alcohol and fatal motor vehicle accidents. The results also suggest that a reduction in per capita consumption would have its most preventive impact on fatal accidents among younger males. (c) 2008 Elsevier Ltd. All rights reserved.