Testing the impact of local alcohol licencing policies on reported crime rates in England.

BACKGROUND: Excessive alcohol use contributes to public nuisance, antisocial behaviour, and domestic, interpersonal and sexual violence. We test whether licencing policies aimed at restricting its spatial and/or temporal availability, including cumulative impact zones, are associated with reductions in alcohol-related crime. METHODS: Reported crimes at English lower tier local authority (LTLA) level were used to calculate the rates of reported crimes including alcohol-attributable rates of sexual offences and violence against a person, and public order offences. Financial fraud was included as a control crime not directly associated with alcohol abuse. Each area was classified as to its cumulative licensing policy intensity for 2009-2015 and categorised as 'passive', low, medium or high. Crime rates adjusted for area deprivation, outlet density, alcohol-related hospital admissions and population size at baseline were analysed using hierarchical (log-rate) growth modelling. RESULTS: 284 of 326 LTLAs could be linked and had complete data. From 2009 to 2013 alcohol-related violent and sexual crimes and public order offences rates declined faster in areas with more 'intense' policies (about 1.2, 0.10 and 1.7 per 1000 people compared with 0.6, 0.01 and 1.0 per 1000 people in 'passive' areas, respectively). Post-2013, the recorded rates increased again. No trends were observed for financial fraud. CONCLUSIONS: Local areas in England with more intense alcohol licensing policies had a stronger decline in rates of violent crimes, sexual crimes and public order offences in the period up to 2013 of the order of 4-6% greater compared with areas where these policies were not in place, but not thereafter.

Real-time modelling of influenza outbreaks--a linear regression analysis.

Seasonal outbreaks of influenza exert a considerable burden on health services, and are notorious for their variability from year to year. Making use of historical data from the Scottish sentinelle surveillance since 1972, a potential candidate model has been derived based on simple linear regression. It was applied with a measure of success in the 1999-2000 winter season.