Risky alcohol use and violence against women: cause or consequence?

The aim of the current study was to determine (a) whether there is a relationship between the quantity of alcohol consumed by a woman when drinking alcohol and the risk of physical violence and (b) whether experiencing physical violence increases the quantity of alcohol consumed by a woman.

The research questions were addressed by fitting a series of random and fixed effect logistic regression models to data drawn from 22 waves of the HILDA survey, a nationally representative longitudinal survey of Australian households. The dataset for the current study involves 143,517 survey responses from 15,934 women. We examine (1) the cross-sectional relationship between alcohol use and physical violence among women (2) the relationship between alcohol use (or violence) at wave t and physical violence (or alcohol use) at wave t + 1 and (3) the within-subject relationship between alcohol use and physical violence (and vice versa).

The risk of is nine per cent for a woman who consumes 1-2 standard drinks at a sitting, 15 per cent for a woman who consumes 7 to 8 standard drinks at a sitting and 20 per cent for a woman who normally consumes 13 or more standard drinks at a sitting. Controlling for other factors, a woman who reports having been assaulted in the previous 12 months consumes an average of 13 per cent more alcohol at a sitting than a woman who has not been assaulted in the previous 12 months.

Prevention resources are always limited and are therefore best targeted at those who are most at risk, regardless of whether the correlate plays a causal role or is simply a signal of heightened risk. Women reporting they have been assaulted to health, welfare or criminal justice authorities should be assessed to determine whether they also have concurrent drug and alcohol problems.