[Commentary] ALCOHOL AND VIOLENCE: A COMPLEX NEXUS OF DRINKING ENVIRONMENT AND DRINKING PATTERN

While studies of homicide victims, for example by Parker [1], Cherpitel et al. [2] and others, have demonstrated a strong association of drinking and violence, explanations of this relationship were often elusive. A part of this conundrum was whether this resulted from personal characteristics including personality and genetics of the individual victim (or perpetrator), or the drinking group, or the effects of alcohol on the victim, or the perpetrator, or both in the moment. This produced a disinhibition explanation of violence and alcohol; i.e. because drinkers are impaired by alcohol, their ability to make reasonable decisions about behavior, including conflict avoidance, is diminished. One of my favorite drawings from the ‘Dark Side’ cartoon series shows three obviously drunk dogs standing on a hill looking down into a cage in the zoo occupied by a large black panther. Two dogs are encouraging a third to enter the panther's cage with the statement, ‘Go ahead . . . you can take him!’. This, in a nutshell, presents one of the popular explanations for why drinkers become involved in violence, i.e. drunk people make stupid decisions and get into fights. Fortunately, three papers in this issue provide rich insights into the basis for explaining the association of alcohol and violence [3–5]. Hughes et al. [3] studied the relationship of drinking before and after going out into public spaces and violence in an effort to explore how personal drinking patterns and location or setting of pre-public drinking influenced alcohol-related problems in public spaces, including licensed alcohol establishments. They found that drinkers who began drinking at home or at a friend's home before going out to public drinking venues in England reported significantly higher levels of total consumption over the night out and an almost three times greater risk of being involved in a fight in such public drinking settings. The authors suggested that the relative price per drink differential between private drinking (using off-premise purchased alcohol) and in pubs and nightclubs (using on-premise purchased alcohol) may well contribute to this pattern. This ‘pre-loading’ on alcohol using low-cost sources prior to going out seemed to increase the risk of violence. They even proposed that ‘Measures to tackle drunkenness and alcohol-related violence in nightlife should expand beyond those solely targeted at nightlife environments’[3]. The second study, by Boyle et al. [4], examined self-reported drinking and personal hostility within a large cohort of individuals from the United States who were involved in the Vietnam War. They found that hostility, i.e. as measured by an abbreviated version of the Cook–Medley hostility scale (ACM), was associated with total monthly intake of alcohol as well as drinks per drinking day and heavy episodic drinking. They concluded that hostility is associated with a heavy per-occasion drinking pattern and thus, not surprisingly, elevated personal mortality. Treno et al. [5] found that individuals living in neighborhoods in California, USA with high concentrations of pubs and bars held personal values of greater acceptability of alcohol-related aggression and carried out more frequent heavy drinking at bars and pubs as well as at parties and friends' homes. While the direction of influence is unknown via this cross-sectional research, the authors concluded that both alcohol outlet densities and the personal acceptability of alcohol-related violence co-occur together. At first glance, these three studies appear to be separate aspects of the nexus of drinking with violence [3–5]. However, all three studies found an association of heavy consumption per drinking event and self-reported violence or hostility. Alcohol epidemiology has found historically that average annual drinking per person was associated with physical illness and death related to alcohol, e.g. liver cirrhosis. However, findings from these three studies suggests that heavy drinking per occasion, which naturally increases the level of blood alcohol concentration (BAC) and thus impairment, is associated strongly with violence and hostility. Such a finding focuses attention of alcohol policy and prevention to alcohol amount consumed in one setting and away from a single emphasis on average annual consumption in efforts to prevent violence. Thus, while average consumption appears to predict chronic or longer-term physical health problems, personal drinking pattern appears to predict acute or in-the-moment alcohol-related harm, especially violence, certainly at the individual level. Two studies, one conducted in England and one in California [3,5], found that public drinking venues themselves, i.e. bars and pubs, were related to heavy drinking and violence. The association appears to flow in at least two directions. First, higher per drink costs at public drinking stimulates private drinking venues where the unit cost is lower. This suggests that a heavy drinker may well arrive at bars or pubs with a high BAC and thus an increased risk of heavy public drinking and associated violence. Secondly, the neighborhoods with higher density of public drinking venues can have both higher levels of aggregate violence and a greater acceptance of alcohol-related violence, i.e. outlet density and greater acceptability of alcohol-related violence appear to coexist. The study of Vietnam War participants and the California [4,5] study found a co-occurrence of personal values (level of individual hostility) and personal norms (level of alcohol-related violence acceptability) with drinking pattern. Neither of these studies were longitudinal and thus could not suggest a causal direction. However, in these two studies we see the co-occurrence of self-reported hostility (a foundation for violence) and self-reported acceptability of alcohol-involved violence (another type of foundation for actual violence). What is to be made of all of this for alcohol epidemiology and prevention of alcohol problems? Clearly, the three separate studies present the obvious, i.e. there is no single understanding or solution to preventing violence involving alcohol [3–5]. Here we have the interactions of key personal and environmental variables. These provide greater understanding of the violence and alcohol nexus and of the co-occurrence of factors which increase the risk of alcohol-related violence associated with public drinking venues. For prevention, the results are challenging. It is not possible to focus on a single variable to prevent alcohol-related violence in public venues. In short, prevention must examine the ‘system of variables’ which increases the risk of alcohol-related violence as a means to select potentially effective strategies. For example, prevention cannot simply conduct a public education campaign to counter the association of individual hostility or acceptability of alcohol-related violence as a means to reduce public drinking violence. This approach does not address the significant influence of alcohol public drinking density and its potential to increase violence. Prevention cannot use a single focus on characteristics of one drinking venue to prevent violence, as it is very likely that these public drinking establishments as a concentrated group may stimulate more acceptability of alcohol-related violence as well as violence itself. Thus, these studies identify for prevention and epidemiology key variables which are a part of the complex of variables causing alcohol-related violence in public venues. This stimulates us to revise simple models, as described in the introductory paragraph, which are inadequate. What is missing from these three studies is a window into domestic violence associated with drinking. Two studies are specifically about drinking and violence, with an emphasis on public drinking. None of these studies makes a distinction between hostility and public versus domestic violence, which is unfortunate and is a missed opportunity. I, for one, suggest that some of the same variables which are operating for public alcohol-related violence are also operating for domestic alcohol-related violence. However, in the end, we are given more insights into the alcohol and violence nexus with more understanding of public violence than domestic violence.