Role of alcohol in unnatural deaths: a study of all deaths in Sweden.

BACKGROUND: Most previous research has concentrated on the role of alcohol in one type of unnatural death in a selected population, but the present objective was to investigate the role of alcohol in all unnatural deaths (autopsied and not autopsied).
METHODS: All cases of unnatural death from 1992 through 1996 in Sweden were analyzed (n = 23,132). Death was attributed at least in part to alcohol if the deceased was a "known alcoholic"; if the underlying or contributing cause of death was alcohol-related; if the deceased had an alcohol-related inpatient diagnosis during the 3-year period prior to death; or if the deceased tested positive for blood alcohol.
RESULTS: Just over 28% of the unnatural deaths could be associated with alcohol; the association with alcohol was more than twice as common in deaths of males (35%) as in females (16%). When only autopsied cases or only blood-tested cases were taken as the denominators, 38% and 44%, respectively, of the deaths were associated with alcohol. Alcohol involvement also was twice as common in intentional deaths (36%) as in unintentional deaths (18%). The intoxication group (78%) had the highest fraction of deaths that could be associated with alcohol, followed by the undetermined group (62%), homicide (49%), fire (41%), suicide (35%), asphyxia (29%), traffic (18%) and fall (9%) groups. In the 20- to 59-year age group, alcohol involvement was found in 51% of the males and 35% of the females (47% for males and females combined).
CONCLUSIONS: The present estimates are conservative; alcohol involvement in unnatural deaths probably is even higher, up to 44% of the total. The present estimation is an important step in policy-making to lower the number of alcohol-related deaths in Sweden.