Lowenfels A B, Wynn P S
Department of Surgery, New York Medical College, Valhalla 10595.
Ann Epidemiol. 1992 May;2(3):249-56. doi: 10.1016/1047-2797(92)90057-w.
We used data for the years 1965 to 1987 from 19 countries to study the relationship between per capita alcohol consumption and vehicular fatalities. Cross-sectional data for the most recent years show a strong relationship between per capita alcohol consumption and vehicular deaths (r = .83, P < .001). When we examined time trends, we noted an increase in per capita alcohol consumption and vehicular deaths for the years 1965 to 1973, a dissociation of the two variables in response to the oil crisis during the 1970s, and a recent gratifying reduction in both variables for the period 1980 to 1987. On average, for the years 1980 to 1987, a 1% reduction in per capital alcohol consumption was associated with a 1% reduction in vehicular deaths (95% confidence interval, .9 to 1.1). For many countries alcohol consumption has already peaked and these findings provide quantitative support for the beneficial effects of continued efforts to control overall per capita consumption of alcohol.
我们使用了19个国家1965年至1987年的数据来研究人均酒精消费量与交通事故死亡率之间的关系。最近几年的横断面数据显示，人均酒精消费量与交通事故死亡人数之间存在密切关系（r = 0.83，P < 0.001）。当我们研究时间趋势时，我们注意到1965年至1973年人均酒精消费量和交通事故死亡人数有所增加，20世纪70年代的石油危机使这两个变量出现了分离，而最近在1980年至1987年期间这两个变量都出现了令人欣慰的下降。平均而言，在1980年至1987年期间，人均酒精消费量每减少1%，交通事故死亡人数就减少1%（95%置信区间为0.9至1.1）。对于许多国家来说，酒精消费量已经达到峰值，这些发现为持续努力控制人均酒精总消费量的有益效果提供了定量支持。
