METHODS: From nationwide governmental registries and databases, forensic-toxicological results were retrieved for victims (n = 200) and offenders (n = 105) of homicides committed during the years 2007-2009 and individuals killed in vehicle-related accidents (n = 1629) during the years 2006-2014. Ethanol levels in blood exceeding 0.01 g/100 ml were considered positive.
RESULTS: Using logistic regression, we found that the presence of ethanol in blood conferred a significantly increased risk of homicide offending (age-adjusted odds ratio [aOR] = 3.6, 95% confidence interval [95% CI] = 2.3-5.6) and homicide victimization (aOR = 2.1, 95% CI = 1.4-3.0). After stratification by sex, risk estimates in females were about 3-fold greater than in males for both homicide offending ([aOR = 11.0, 95% CI = 2.4-49.8] versus [aOR = 3.1, 95% CI = 1.9-4.9]) and victimization ([aOR = 5.4, 95% CI = 2.4-12.2] versus [aOR = 1.7, 95% CI = 1.1-2.8]). Sensitivity analyses yielded similar estimates.
CONCLUSIONS: The results of the present study are consistent with prior findings suggesting alcohol to be an important risk factor for homicide offending and victimization. Surprisingly, however, associations were more pronounced in females, although additional studies that control for potential confounders are warranted to facilitate speculations about causality.