Levels of support for the licensing of tobacco retailers in Australia: Findings from the National Drug Strategy Household Survey 2004-2016
Abstract Background: Assessing public opinion towards tobacco policies is important, particularly when determining the possible direction of future public health policies. The aim of this study was to describe the implementation of tobacco retailer licensing systems by state and territory governments in Australia, and use the National Drug Strategy Household Survey (NDSHS) to assess levels of public support for a retailer licensing system in each jurisdiction over time and by a range of socio-demographic and behavioural attributes. Methods: National and state/territory estimates of public support for a tobacco retailer licensing system were derived as proportions using NDSHS data from 2004 to 2016. The effect of the type of licensing system of one’s jurisdiction of residence on the likelihood of supporting such an initiative in 2016 was assessed using logistic regression while controlling for various socio-demographic characteristics. Results: Public support for a tobacco retailer licensing system ranged from a high of 67.2% (95% CI 66.5% – 67.9%) nationally in 2007 and declined to 59.5% (95% CI 58.9% – 60.2%) in 2016. In 2016, support was greatest amongst those aged 50 years and older, females, those from the least disadvantaged areas, those living in major cities, never-smokers and never-drinkers. Those living in jurisdictions with a ‘negative’ licensing system were more likely to support a licensing system than those living in jurisdictions without any system in place (OR = 1.09, 95% CI 1.01 – 1.17); however this difference disappears after adjusting for the socio-demographic and behavioural attributes of respondents (OR = 1.04, 95% CI 0.96 – 1.12). Conclusions: A clear majority of the public support a tobacco retailer licensing system, regardless of whether or not such a system is already in place in their jurisdiction of residence. While there is variation between jurisdictions in levels of support, this variation largely disappears when the different socio-demographic and behavioural attributes of the respective populations are taken into account.