Commentary on Kuhns et al. (2011): Drinks and death – alcohol and the homicide victim


The carefully conducted meta-analysis presented by Kuhn and colleagues is an important contribution to our understanding of alcohol and violence [1]. The significant harms associated with alcohol intoxication and dependence have been well documented. A great deal of such attention, however, has focused on the negative health effects of the drug, including major organ pathology, cognitive impairment and psychopathology, and the role of alcohol in motor vehicle accidents. Alcohol and violence has also received attention. Indeed, the importance of this field of research is seen in the strong correlation between national alcohol sales and homicide rates [2]. Overall, however, the focus of such work has been on the risk of committing violent acts [3]. This is clearly of major importance. It is, however, as Kuhn and colleagues demonstrate, only half the problem [1]. Victimology is also crucial.

As one who has conducted research in this field [4,5], I applaud the authors for their work in this vital area. The authors report that, across a wide range of countries and cultures, there is remarkable consistency in the proportion of homicide victims who test positive for alcohol. Consistently, nearly half of victims were alcohol-positive at the time of death, and average alcohol concentrations are high in such cases. Indeed, one of the major findings of research is that the toxicology of offenders and victims are close to identical. Moreover, it is well known that offenders are more likely to be victims of assault. These are, essentially, the same pool of people. A typical scenario is of two friends drinking for many hours, an intoxicated argument breaking out, followed by a brief fight and a death, often from a single punch.

Kuhn and colleagues make excellent recommendations for studies reporting on the toxicology of homicide. As they note, we often see studies in which even basic demographic comparisons, such as age and gender, are not performed. We also do not typically see toxicology stratified by type of assault (e.g. gunshot, blunt force injury). Urine concentrations, which may indicate higher recent alcohol concentrations than seen in blood samples, are sometimes reported, sometimes not. Indeed, as the authors note to their frustration, even the units of measurement in which alcohol concentrations are reported varies considerably, and are not immediately comparable. For us to understand fully how alcohol relates to violence, whether victims or offenders, these are basic and crucial data. I would add that the toxicology and location of death (e.g. hotel, street, home) are also rarely reported, despite there being major public health implications that may arise from such data. I can but share the authors’ frustrations. Of course, autopsies and toxicology analyses are conducted for medico-legal reasons, and not for research purposes. Even so, standard reporting would have benefits for both clinical forensic pathologists and for researchers in being able to conceptualize an individual case or a case series.

While this study dramatically illustrates the carnage associated with alcohol, we must also look beyond this drug. At least a quarter of homicide victims have illicit substances detected in their blood [4–6]. Indeed, the prevalence of drugs such as psychostimulants and opioids are many orders of magnitude higher than would be expected from population statistics. Of course, we also commonly see multiple drug intoxication in victims, particularly illicit drug users. The same consistency in reporting suggested by Kuhn and colleagues for alcohol should also apply to the illicit drug users. The risk of violent death is extremely high among dependent illicit drug users, and it we are to understand the dynamics of drugs and violent death, comparable data are necessary.

In summary, the epidemiological data suggest that we may, potentially, reduce homicide rates by reducing rates of alcohol consumption and binge use. The work by Kuhn and colleagues is an excellent indicator of the extent of the problem, and of how we may improve our means to study this major public health problem.

Declaration of interests