RT Article
T1 The effectiveness of an intervention to reduce alcohol-related violence in premises licensed for the sale and on-site consumption of alcohol: a randomized controlled trial
JF Addiction
A1 Alam, M. Fasihul
A1 Williams, Anne
A1 Sivarajasingam, Vaseekaran
A1 Shovelton, Claire
A1 Shepherd, Jonathan
A1 Playle, Rebecca
A1 Murphy, Simon
A1 Moore, Simon C.
A1 Moore, Laurence
A1 Huang, Chao
A1 Hood, Kerenza
A1 Heikkinen, Marjukka
A2 Williams, Anne
A2 Sivarajasingam, Vaseekaran
A2 Shovelton, Claire
A2 Shepherd, Jonathan
A2 Playle, Rebecca
A2 Murphy, Simon
A2 Moore, Simon C.
A2 Moore, Laurence
A2 Huang, Chao
A2 Hood, Kerenza
A2 Heikkinen, Marjukka
LA English
YR 2017
UL https://krimdok.uni-tuebingen.de/Record/1866322117
AB Background and Aims: Premises licensed for the sale and consumption of alcohol can contribute to levels of assault-related injury through poor operational practices that, if addressed, could reduce violence. We tested the real-world effectiveness of an intervention designed to change premises operation, whether any intervention effect changed over time, and the effect of intervention dose. Design: A parallel randomized controlled trial with the unit of allocation and outcomes measured at the level of individual premises. Setting: All premises (public houses, nightclubs or hotels with a public bar) in Wales, UK. Participants: A randomly selected subsample (n=600) of eligible premises (that had one or more violent incidents recorded in police-recorded crime data; n=837) were randomized into control and intervention groups. Intervention and comparator: Intervention premises were audited by Environmental Health Practitioners who identified risks for violence and provided feedback by varying dose (informal, through written advice, follow-up visits) on how risks could be addressed. Control premises received usual practice. Measurements: Police data were used to derive a binary variable describing whether, on each day premises were open, one or more violent incidents were evident over a 455-day period following randomization. Findings: Due to premises being unavailable at the time of intervention delivery 208 received the intervention and 245 were subject to usual practice in an intention-to-treat analysis. The intervention was associated with an increase in police recorded violence compared to normal practice (hazard ratio=1.34, 95% confidence interval=1.20–1.51). Exploratory analyses suggested that reduced violence was associated with greater intervention dose (follow-up visits). Conclusion: An Environmental Health Practitioner-led intervention in premises licensed for the sale and on-site consumption of alcohol resulted in an increase in police recorded violence
DO 10.1111/add.13878