Do cumulative impact zones reduce alcohol availability in UK high streets? Assessment of a natural experiment introducing a new licensing policy

Abstract Background The Licensing Act 2003 enables English local authorities to implement Cumulative Impact Policies (CIPs). CIPs strengthen the powers of local authorities to reject licence applications for retail alcohol sales in cumulative impact zones (CIZs), where adverse effects of alcohol availability can be demonstrated. We assessed the impact of a CIP in one London borough on granting alcohol sales licences for off premise (off-licence) and on premise (on-licence) consumption and on the trading and closing times of premises. Methods We conducted a time-series analysis using Poisson regression of licensing data from April 1, 2008, to July 31, 2015, to estimate immediate and longer-term impacts of introducing CIP in 2013. Primary outcomes were number of licence applications, proportion of applications granted (success rates), and trading and closing times of licensed premises within and outside CIZs. Models were adjusted for overdispersion and underlying seasonal and secular trends, and checked for autocorrelation. In sensitivity analysis we examined the impact of CIP on outcomes in CIZs controlling for underlying trends in non-CIZs. Findings There were 409 and 278 alcohol licence applications for off-licences and on-licences, respectively. There was no significant impact of CIP introduction on number of applications submitted, or on trading and closing times of premises between areas or over time. There were statistically significant impacts immediately after CIP introduction of reduced application success rates in CIZs (−27·80%, 95% CI −42·33 to −9·60) and outside CIZs (−47·12%, −64·72 to −20·74) mostly driven by decreases in success rates for off-licence applications (−39·56% [–56·40 to −16·23] and −52·18% [–71·06 to −20·99], respectively). Subsequent years showed that application success rates for all licences increased in both areas (4·17% [0·43 to 8·05] and 13·64% [6·43 to 21·34], respectively). We found little evidence of CIP impact on primary outcomes in CIZs when controlling for non-CIZs. Interpretation Although CIPs led to short-term decreases in rates of alcohol licences granted, these reductions were not sustained. Previous research proposed that CIPs mainly affect premise type and licence conditions (eg, opening hours) rather than number of premises. Further research should explore these other potential impacts and whether non-CIZs are affected by spillover effects from CIZs. Funding National Institute for Health Research (NIHR) School