► Proximity and density of on-premises outlets is unrelated to adolescent alcohol use.
Finally, one previously mentioned study found an association between density of on-premises outlets and drinking, but not heavy drinking (Kuntsche et al., 2008).
However, adolescents living close (within 200 m) to an off-sales outlet were more likely to drink frequently (OR 1.97, p=0.004), as were adolescents living in areas with many nearby off-premises outlets (OR 1.60, p=0.016).
Proximity and density of on-premise outlets were not associated with weekly drinking.
In contrast, proximity and outlet density of on-premise outlets (i.e. local pubs or clubs) was unrelated to adolescent alcohol use, nor was off-sales outlet density measured at the very immediate neighbourhood level (data zone).
The lack of association between the availability of on-premise drinking venues and adolescent alcohol use may be due to the greater price per unit of alcohol and stringent enforcement of age-restrictions associated with on-premise outlets.
Liquor licensing information from alcohol-control boards commonly is used to generate indicators of commercial availability namely, number of outlets/population rates and concentration of on- and off-premise outlets (Sherman et al. 1996; see also Gruenewald et al. 1997).
Geographic Information System (GIS) mapping has emerged as an innovative means of generating community indicators of outlet density (including off- and on-premise outlets) and to examine alcohol outlet density and locations in relation to alcohol-related problems, such as assaults and sale of alcohol to minors (see Gruenewald et al. 2002; Millar and Gruenewald et al. 1997).
Moreover, the presence of social companionship support in an area was moderated by alcohol outlet density, such that parents who lived in areas with higher density of on-premise outlets (i.e., bars) and a high level of social support used physical abuse more often compared with parents with high social support and low outlet density (Freisthler et al. 2014).
If privatization in Pennsylvania yielded a different set of geographic regulations, where distances varied by outlet types (e.g., off-premise outlets relative to bars) and the variety of sensitive facilities changed (e.g., other types of facilities were added to the list of protected sites), the model easily accommodates these modifications.
Self-reported social access to alcohol has also been measured in school or community surveys of youth, with participants asked to report on sources from which they obtain alcohol (i.e., commercial [on- or off-premise outlets] versus social [friends, family, etc.] sources) (see Dent et al. 2005; Harrison et al. 2000; Hearst et al. 2007; Jones-Webb et al. 1997; Wagenaar et al. 1994).
