The ecology of domestic violence: the role of alcohol outlet density Michael Livingston School of Population Health, University of Melbourne, Melbourne, Australia and AER Centre for Alcohol Policy Research Turning Point Alcohol and Drug Centre, Fitzroy, Australia Abstract. Studies have consistently found positive associations between the density of alcohol outlets and levels of vio- lence in areas. Few studies have examined whether this relationship holds for domestic violence. This study assesses whether alcohol outlet density is related to domestic violence and whether this relationship is due to alcohol availabil- ity or to co-occurring economic disadvantage and social disorganisation. Cross-sectional data on family incidents, liquor outlets and socio-demographic characteristics were obtained for 217 postcodes in Melbourne, Australia. These data were used to construct models assessing the association between alcohol outlet density and domestic violence, both with and without controlling for socio-demographic factors. Models were tested for spatial autocorrelation, and spa- tial-error models were developed to control for its influence. Outlet density was significantly associated with rates of domestic violence, even controlling for socio-demographic factors. The density of hotels (pubs) was positively associ- ated with domestic violence rates and the density of restaurants and bars was negatively associated with domestic vio- lence. Socio-economic disadvantage was also associated with domestic violence rates. The density of packaged liquor outlets was not associated with rates of domestic violence. The results present a mixed picture, and further study is required to develop a clearer understanding of the links between alcohol availability and domestic violence. Keywords: alcohol availability, outlet density, violence, Australia. Introduction A substantial number of studies have linked alco- hol availability to a range of alcohol-related prob- lems (Gruenewald, 2007; Livingston et al., 2007). In particular, there has been an almost uniform finding from cross-sectional studies that the density of alco- hol outlets in an area is positively associated with rates of violence in that area. This finding has been replicated in a wide variety of U.S. settings (e.g. Scribner et al., 1999; Stevenson et al., 1999; Gorman et al., 2001; Lipton and Gruenewald, 2002; Nielsen et al., 2005; Reid et al., 2005; Gruenewald et al., 2006; Gyimah-Brempong and Racine, 2006) and in handful of Australian studies (Stevenson et al., 1999; Chikritzhs et al., 2007; Livingston, 2008a). An increasing number of longi- tudinal analyses have also found significant associa- tions between alcohol outlet density and violence over time, providing stronger evidence that the rela- tionship is causal (Norström, 2000; Gruenewald and Remer, 2006; Livingston, 2008b; Yu et al., 2008). These studies have focussed on assault rates, using either hospital or police data to estimate the rates of violence occurring in geographical areas. Thus, the violence generally examined in the current literature is largely public violence, often taking place in and around licensed premises. It is worth noting that Freisthler and colleagues have undertak- en a series of studies that have demonstrated statis- tical associations between alcohol outlet density and child abuse and maltreatment, both cross-sectional- ly (Freisthler, 2004; Freisthler et al., 2004, 2005, Corresponding author: Michael Livingston AER Centre for Alcohol Policy Research Turning Point Alcohol and Drug Centre Gertrude Street, 54-62, Fitzroy 3065, Australia Tel. +61 3 8413 8407; Fax +61 3 9416 3420 E-mail: michaell@turningpoint.org.au Geospatial Health 5(1), 2010, pp. 139-149