The spatial effect of alcohol availability on violence: A geographically weighted regression analysis

Abstract

Studies have identified an association between proximity to outlets selling alcohol and violent crime. The strength of this association has been known to vary spatially as not all areas have the same accessibility to alcohol outlets, suggesting local context matters. Within this evidence base, however, studies have not routinely incorporated spatial variance into their models. Moreover, whilst it is acknowledged that the characteristics of the alcohol outlets (for example distinguishing between on- and off-trade premises) is important in studying the varying pathways to violent crime, studies have not often explored the role/contribution of outlet sales volume in this relationship. This study used open-source police and alcohol outlet data to explore the effect of alcohol availability and sales volume on violent crime, across small areas in England. Spatial associations were calculated using geographically weighted regression controlling for contextual variables, such as deprivation and population characteristics. Associations between alcohol outlets and violence varied spatially, for both on- and off-trade outlets and for different premise types, e.g. pubs and bars. Increased proximity to high-sales outlets was associated with an increased risk of violence in inner city contexts, where there is a concentration of night-life spaces. Through incorporating space, this study highlighted the spatially varying associations between different types of alcohol outlets and violence across small areas in England; so that licensing boards and local authorities can determine which appropriate measures are needed for targeting areas with high alcohol availability, to reduce risk of violence at the small area level.