"Stopping before you start" : reducing and preventing initiation of tobacco use in the ACT

Tobacco is the leading cause of preventable death in Australia and contributes to 5.4% of disease burden in the Australian Capital Territory. Initiation of tobacco use is most likely to occur during adolescence and young adulthood (at less than 20 years). Prevention of tobacco initiation involves a combination of regulatory, educational and health promotion interventions including restrictions on the sale of tobacco products. This paper reports on the development and use of an agent-based model to explore the impact of modifying three hypothetical regulatory and health promotion interventions: 1) increasing the minimum purchasing age for tobacco products, 2) reducing retail sales of tobacco products to persons under the minimum purchasing age and 3) reducing secondary sharing of tobacco products to persons under the minimum purchasing age using health promotion messaging. The model was built using a participatory approach that engaged policy officers, health promotion officers, epidemiologists, biostatisticians and computer scientists. The structure of the model included interacting state chart representations of smoking and level of concern about tobacco use (engagement status) and a pro-smoking score, which defined the hazard rate of initiation, cessation, and relapse. The pro-smoking score was a function of several risk factors including engagement, social effect of having more or fewer smoking peers, addiction and withdrawal levels and access to tobacco products. Parameterisation of the model drew on a range of data sources with local data being prioritised where it was available. A series of scenarios comparing the impact of the interventions on smoking prevalence rates and age of initiation are reported. Of the three interventions simulated, increasing the minimum purchasing age from 18 to 21 years had the greatest impact on smoking prevalence across the population, reducing the prevalence of smoking from 8.5% (95% CI 7.8, 9.2) to 6.9% (95% CI 6.4, 7.4) five years post-intervention and 4.1% (95% CI 3.8, 4.3) 20 years post intervention (Figure 1). The interventions aimed to reduce the sale of tobacco products to minors and reduce secondary sharing produced small reductions on their own. However, when implemented in combination with increasing the minimum purchasing age, they significantly increased the impact of this intervention from ten years post-implementation, ultimately resulting in a prevalence rate of 2.8% (95% CI 2.6, 3.0) 20 years post-implementation. Given the challenges associated with ceasing tobacco use, these in silico experiments demonstrate the importance of regulatory public health interventions to delay, and therefore potentially prevent initiation.