Research links density of alcohol outlets to local levels of harm

Linda Hill

New geo-spatial research methods are improving our understanding of how local environments shape the level of alcohol-related problems experienced by communities. A growing body of research shows a relationship between alcohol outlet density, drinking and harm that is most marked at the neighbourhood level. The research literature reviewed below suggests that reducing the number and density of alcohol outlets can reduce heavy drinking and local alcohol-related harm, such as violent crime.

Studies in the USA in the 1980s and early 1990s showed a relationship between alcohol outlet density and alcohol consumption levels that was mutually reinforcing over time. This was followed by research in medium-sized US cities that linked alcohol outlet density to specific measures of alcoholrelated harm such as drinkdriving and violence. Findings from city-level data do vary between different parts of the USA, but analysis of neighbourhood level data consistently shows a significant link between outlet density and harm after controlling for sociodemographic and other relevant factors. These findings can inform policing and licensing decisions, as well as the wider policies of local and central government.

One of the earliest studies, of 213 California cities, showed that a 1 per cent increase in beer bars resulted in more than 1 per cent increase in public drunkenness and drink driving. Other Californian research linked high outlet density to high numbers of night-time traffic crashes and pedestrian injuries in the neighbourhoods studied, with some spill-over effects into neighbouring areas.

Research in Newark, New Jersey, showed that alcohol outlet density was the single greatest predictor of violent crime at the suburb and neighbourhood level (census tracts and census blocks). This link with harm levels holds good whether density is measured per head of population or per square kilometre. California studies linked the density of off-licensed outlets to violent assault, including violence involving young people. In an average city in Los Angeles County, each additional outlet was associated with a 0.62 per cent increase in violent offences. In New Jersey, alcohol outlets, along with bus stations and all-night businesses, appeared to become crime ‘hotspots’. The effect of high outlet density on crime was specific to the area, however, with little of spill-over into neighbouring areas that occurs with alcoholrelated driving offences.

One New Orleans study of outlet density is relevant to the debates about personal choice versus community responsibility that often come up when intervention in the alcohol market is proposed. It shows how the link between outlet density, drinking and harm works through the social norms that are created within communities. Rather than affecting individual behaviour directly, the density of alcohol outlets in a community has a structural effect on alcohol attitudes and drinking patterns across that community that then influences the attitude and drinking pattern of individuals.

In this study, urban residential census areas were randomly selected to provid three examples each for hig and low outlet density and four different levels on the socioeconomi deprivation index 24 areas in all. A random telephone survey of 2,604 adults in these areas asked abou personal drinking and attitudes to alcohol including social acceptability and perceived norms for friends, and distance from an off-licence (as a proxy for density), as well as sociodemographic information. The analysis looked at drinking and attitudinal differences between individuals and also between averages for each area. For individuals, all the variables except density were strongly linked to drinking norms; male sex and higher education were more likely to mean heavier drinking; black ethnicity and increasing age were associated with drinking less. Distance from home to the closest offlicence made no difference. However, 16 percent of the variation between individuals on perceived drinking norms and 11.5 percent of variation in selfreported drinking related to where the individual lived. When the data was analysed by neighbourhood, the average distance to nearest outlet for each neighbourhood was the only variable that linked to averages on attitudinal and drinking measures. That is, in neighbourhoods with low distances on average to the nearest off-licence, there were heavier drinking norms.

In the researchers’ view, this finding indicated that in a high outlet density area, whether you live next door to a liquor store or a kilometre away, everyone’s drinking norms are affected to some degree by the neighbourhood environment. This is consistent with social learning theory and with research from other countries that shows individual drinking behaviour is influenced by the drinking behaviour of one’s social network. For example, research links heavier drinking among 10-17 year olds and among university students to perceptions about drinking by friends that overestimate their actual consumption.

A high density of alcohol outlets increases the visual presence of alcohol in a neighbourhood, and may have similar effects on attitudes and norms as high exposure to television, print and billboard advertising for alcohol. The presence of alcohol outlets, including signage and empty bottles or intoxicated patrons in the street, may also have a ‘broken window’ effect, suggesting that ‘no one cares’ in this neighbourhood. Concentrations of bars and bottle stores can act as ‘attractors’ of socially disinhibited people and help create the conditions for nonnormative activities such as drug use and prostitution. This, as well as the effects of increased drinking in the neighbourhood, may contribute to the link between outlet density and violence.

The type of outlet linked to increases in harm varies between the US studies. Differences appear to be related to local patterns of drinking and risk and reflect licensing system differences between US states. In New Orleans it was the density of off-licensed premises that made a difference, with researchers pointing out that the data for on-licensed premises would include many serving the tourist industry rather than the local population. No such distinction could be made in the New Jersey study, where licensed premises usually have both on and off sales. In Californian studies, the density of licensed restaurants affected alcohol-related crash rates, as alcohol was twice as likely to be consumed in restaurants than in bars , but bar density was linked to higher assault rates. Research on different types of licensed premises in Perth, Western Australia, did not consider density but linked intoxication levels, drink-driving, alcohol related crashes and violence to late-closing hotels (pubs), particularly those with high volume alcohol sales.

High local crime rates are better explained by alcohol availability and outlet density than by the socio-demographic characteristics of the neighbourhood – although those characteristics may contribute to the level of community organisation and control over what happens in a neighbourhood. In the USA, whites and those with higher incomes are most likely to drink frequently and heavily. Yet in Chicago, areas with high African American population had six times as many offlicensed outlets as areas with mainly white populations. The concentration of alcohol outlets, particularly off-licences, was a major predictor of homicide levels, after controlling for other neighbourhood factors. That is, higher ratios of blacks or young males in a neighbourhood did not explain a higher crime rate. The California and Los Angeles studies also showed that alcohol availability and outlet density in communities were stronger predictors of the level of violence than race or ethnicity. Researchers go so far as to suggest that the overrepresentation of African-Americans in US violence statistics may result from neighbourhood-level risk factors rather than from personal or cultural characteristics. A review of crime in the USA attributed a decline of homicide to falling alcohol consumption, and presented evidence that tighter regulation of alcohol availability could help reduce violence.

These studies link density to harm statistics, which are most readily available for drinkingdriving and violent crime. The Chicago study, which reports community action in lower income Chicago neighbourhoods, describes other negative effects of high liquor store density:

Typical complaints include the serving of minors, littering, loitering, harassment and intimidation of pedestrians and customers, public urination, drug dealing, prostitution, assault and even murder. These specific grievances represent more general quality-of-life, public health and safety problems that feed into the economic and social deterioration of an area. For example, safety issues, like drug dealing and stealing, raise local business operating expenses and create a poor business environment.

Increased competition between many bars and bottle stores in an area may encourage cost-cutting and irresponsible management. This study showed that customer visits to liquor stores or bars often did not involve other retail shopping or consumer activity. High numbers of alcohol outlets may be displacing a retail mix that would benefit other businesses.

Many councils want to develop vibrant entertainment areas that attract tourists as well as local residents. Restaurants, bars and bottle stores provide jobs and other economic benefits, and increasing numbers of licensed premises may be thought desirable for the community. Concentrating alcohol-related problems in one area – but away from residential zones – may even seem to be a good idea from a policing perspective.

This review of research shows that concentrations of alcohol outlets can in fact increase the local problems to be dealt with. Local governments may wish to give careful consideration to the location and density of bars and bottle stores and to the role that these kinds of business play in the retail mix of their city centres.

References

Gruenewald, P.J., L. Remer and R.Lipton (2002) Evaluating the alcohol environment: Community geography and alcohol problems. Alcohol Research and Health 26(1): 42-48.

Reynolds, R.I., H.D. Holder and P.J.Gruenewald (1997) Community prevention and alcohol retail access. Addiction 92: S261-S269.

Watts and Rabow cited in Alaniz, ML. (1998) Alcohol availability and targeted advertising in racial/ethnic minority communities. Alcohol Health and Research World. 22(4): 286-289.

Gruenewald, P.J., A.B. Millar and P. Roeper (1996) Access to alcohol. Alcohol Health and Research World 20(4): 244-251.

Gorman, D.M., P.W. Speer, P.J.Gruenewald and E.W. Labouvie (2001) Spatial dynamics of alcohol availability, neighborhood structure and violent crime. Journal of Studies on Alcohol, 62(5): 628-637.

Scribner, R., D. Cohen, S. Kaplanand S.H. Allen (1999) Alcohol availability and homicide in New Orleans: Conceptual considerations for small area analysis of the effect of alcohol outlet density. Journal of Studies on Alcohol 60(3): 310-317.

Scribner, R.A., D.P. MacKinnon andJ.H. Dwyer (1995) The risk of assaultive violence and alcohol availability in Los Angeles County. American Journal of Public Health. 85(3): 335-340; Gruenewald, Millarand Roeper (1996); Alaniz (1998). Gorman et al. (2001).

Scribner, R., D. Cohen and W. Fisher (2000) Evidence of a structural effect for alcohol outlet density: A multilevel analysis. Alcoholism: Clinical and Experimental Research. 24(2): 188- 195.

Scribner, Cohen and Fisher (2000).Skog, O-J. (1985) The collectivity of drinking culture: A theory of the distribution of alcohol consumption. British Journal of Addiction 80: 83-99; Paton-Simpson, G. (2001) Socially obligatory drinking: A sociological analysis of norms governing minimum drinking levels. Contemporary Drug Problems 28: 133-177.

Wyllie (1997) Love the ads, love thebeer: Young people’s responses to televised alcohol advertising. Doctoral thesis, Department of Community Health, University of Auckland; Kipri,K. 2002. Tertiary student harzardous drinking: Epidemiology and development of a brief intervention trial. PhD theses. University of Otago.

Babor, T. et al. (eds) (2003) Alcohol: No Ordinary Commodity. Research and Public Health. Oxford University Press/World Health Organization, Geneva. August. Chapter 10; Alaniz (1998).

Gorman et al. (2001).

Parker, R. Nash and R.S. Cartmill (1998) Alcohol and homicide in the US: Or one reason why US rates of violence may be going down. Journal of Criminal Law and Criminology. 88(4): 1369-1398.

Schribner, Cohen, Kaplan and Allen (1999).

Gruenewald, Millar and Roeper (1996) Lipton, R. and P. Gruenewald (2001) The spatial dynamics of violence and alcohol outlets. Journal of Studies in Alcohol 63: 187-195.

Chikritzhs, T. and T. Stockwell (2002) The impact of later trading hours for Australian public houses (hotels) on levels of violence. Journal of Studies on Alcohol, 63(5): 591-600; Stockwell, T. (1997) Liquor outlets and prevention policy: The need for light in dark corners. Addiction 92(8): 925- 930; Stockwell, T., P. Somerford and E. Lang (1992) The relationship between licence type and alcohol related problems attributed to licensed premises in Perth, Western Australia. Journal of Studies on Alcohol 53: 495-498.

Maxwell, A. and D. Immergluck (1997) Liquorlining: Liquor store concentration and community development in lower-income Cook County neighborhoods. Woodstock Institute, Chicago.

Schribner, Cohen, Kaplan and Allen (1999); Alaniz (1998).

Parker and Cartmill (1998).

Maxwell and Immergluck (1997)

Note: The work on which these articles are based was funded by the New Zealand Ministry of Health as part of a project on Planning for the Sale of Alcohol. A research and policy review and an issues paper from this project are available on www.ndp.govt.nz under Alcohol.