Cherpitel C J
Western Consortium for Public Health, Alcohol Research Group, Berkeley, California 94709, USA.
J Stud Alcohol. 1996 May;57(3):244-8. doi: 10.15288/jsa.1996.57.244.
While alcohol has been found to be associated with fatal injury, this association may vary in different regions of the United States where drinking patterns and problems are known to vary. The purpose of this study was to analyze alcohol in injury fatalities in a "dry" area of the country (Hinds County, Miss.) compared to a "wet" area (Contra Costa County, Calif.).
Data on all unnatural causes of death during a 1-year period among those 18 years and older were abstracted from county coroner records in Hinds County, Miss. (n = 222) and Contra Costa County, Calif. (n = 304).
Across all causes of death, those in Hinds County were more likely to be alcohol-involved (57%) compared to those in Contra Costa (40%). Those in Hinds County were twice as likely as those in Contra Costa to have died from homicide (32% vs 17%), but the proportion of homicides which was alcohol positive did not differ between the two regions (51% vs 46%). Using logistic regression, gender (male) and region (South) were predictive of a positive blood alcohol concentration across all causes of fatality combined. Age (younger) and ethnicity (black) were predictive of homicide. Ethnicity (non-black), a negative blood alcohol concentration and a positive drug screen were predictive of suicide, while age (younger), ethnicity (non-black), a positive blood alcohol concentration and a negative drug screen were predictive of motor vehicle accidents.
These data suggest that regional variations in alcohol's association with fatal injury do exist, and vary by demographic characteristics and cause of death. In addition, areas of the country that have relatively low rates of per capita consumption may have higher rates of alcohol-related injury fatalities than areas with greater per capita consumption.
