Population-level alcohol consumption and national homicide rates

We explored the cross-national association between population-level alcohol consumption and homicide victimization rates. The very few prior studies of this association had small homogeneous sample sizes and usually tested only for a linear effect, ignoring other commonly hypothesized explanations....

Full description

Saved in:  
Bibliographic Details
Main Author: Hockin, Sara (Author)
Contributors: Rogers, Meghan L. (Other) ; Pridemore, William Alex (Other)
Format: Electronic Article
Language:English
Published: 2018
In: European journal of criminology
Year: 2018, Volume: 15, Issue: 2, Pages: 235-252
Online Access: Volltext (Resolving-System)
Journals Online & Print:
Drawer...
Check availability: HBZ Gateway
Keywords:
Description
Summary:We explored the cross-national association between population-level alcohol consumption and homicide victimization rates. The very few prior studies of this association had small homogeneous sample sizes and usually tested only for a linear effect, ignoring other commonly hypothesized explanations. We employed a cross-sectional design, with data from 83 nations, and controlled for several possible covariates. We used exploratory data analysis, weighted least squares regression, and piecewise regression to model total, male, and female homicide victimization rates. We tested for linear effects of total per capita consumption, threshold effects, effects due to risky drinking patterns, and beverage-specific effects of per capita beer, wine, and spirits consumption. We found cross-national homicide rates were not sensitive to threshold effects and nations with riskier drinking patterns did not have higher homicide rates than nations with less risky drinking patterns. Results showed total per capita alcohol consumption was associated with total, male, and female homicide rates, though this association concealed beverage-specific effects. Per capita beer and spirits consumption was positively and significantly associated with total, male, and female homicide victimization rates, whereas our findings suggested per capita wine consumption might be negatively associated with homicide rates. The impact of alcohol consumption on cross-national homicide rates is understudied relative to other population health outcomes, and the few prior analyses did not test the four most common explanations of a possible association. Our findings provide an important contribution to better understanding this complicated relationship.
ISSN:1741-2609
DOI:10.1177/1477370817731042