The effect of liquor licensing restrictions on assault: a quasi-experimental study in Sydney, Australia

Aims: To determine whether restrictions on the availability of alcohol in two inner-city entertainment areas in Sydney, Australia (1) reduced the incidence of assault in those areas, (2) increased the incidence of assault in nearby areas (where the restrictions did not apply), (3) resulted in a net reduction in overall levels of assault (4) and/or whether the reductions in assault were most pronounced during the daily time-periods when liquor trading restrictions were in operation. Design: Structural time–series modelling was used to estimate and compare trends in assault in areas/times affected by the new restrictions on alcohol availability with trends in assault in areas unaffected by the new restrictions. Setting: Sydney, Australia. Participants/measurements: The primary outcome measure was the police-recorded monthly incidence of non-domestic assault in the 78 months between January 2009 and June 2015. Findings: Following the reforms, we found reductions in assaults of 45% [beta = −0.599, 95% confidence interval (CI) = −1.107, –0.091] and 22% (beta =0.260, 95% CI = −0.397, –0.123), respectively, in the Kings Cross and Sydney CBD Entertainment Precincts. In the Kings Cross Entertainment Precinct, reductions in assault were observed in all three daily time-periods. In the Sydney CBD Entertainment Precinct reductions in assault were observed only in the second and third daily time-periods. Assaults did not increase in entertainment areas adjacent to or within easy reach of the target areas. Conclusion: Restrictions on the availability of alcohol appear to reduce the incidence of assault