Evaluating the effects of the Licensing Act 2003 on the characteristics of drinking occasions in England and Wales: a theory of change-guided evaluation of a natural experiment

Abstract

Background and Aims: The Licensing Act 2003 deregulated trading hours in England and Wales. Previous evaluations have focused upon consumption and harm outcomes, finding mixed results. Several evaluations speculated on the reasons for their results, noting the role of changes in the characteristics of drinking occasions. This study aimed to test proposed mechanisms of effect for the Licensing Act 2003 by evaluating changes in characteristics of drinking occasions. Design, Setting and Participants. Interrupted monthly time–series analysis of effects in England and Wales versus a Scottish control series, using 2001–08 data collected via 7-day drinking occasions diaries by the market research company Kantar (n = 89 192 adults aged 18+). Measurements: Outcomes were start- and end-time of each reported occasion; variation in finish time; prevalence of pre-loading, post-loading and late-night drinking; and alcohol consumption (in units). Findings: After the introduction of the Act, occasions shifted later at night in England and Wales [finish time +11.4 minutes; 95% confidence interval (CI) = 3.6–19.2]. More occasions involved pre-loading in England and Wales relative to Scotland (0.02% increase; 95% CI = 0.01–0.03). There was no evidence of changes in variation in finish time, post-loading, late-night drinking or alcohol consumption. Conclusions: The Licensing Act 2003 in England and Wales appears to have had only limited effects on the characteristics of drinking occasions. This may help to explain its lack of substantial impacts on alcohol harms.