Introduction: Alcohol marketing significantly shapes consumption behaviours, which is particularly problematic when directed at young people. Digital platforms further amplify this influence through algorithmic targeting and immersive content. Monitoring such marketing remains challenging due to its dynamic, personalised, and often covert nature. This study aimed to develop and test SCANNER Alcohol, an AI-enabled system that uses supervised deep learning to detect alcohol brands in images and videos.
Methods: SCANNER Alcohol combines object detection and optical character recognition to identify 134 of the top alcohol brands in images and mobile screen recordings. The model was trained using 8,710 manually annotated brand logos and a human-in-the-loop refinement process. Algorithmic performance was evaluated using mean average precision (mAP), recall, and F1 score. Real-world performance was assessed using mobile screen recordings from 119 alcohol brand social media accounts and 12 hours of digital screen recordings from young people aged 8–25 years as they went about their daily digital lives. All automated detections were benchmarked against manual coding.
Results: Algorithmic testing demonstrated strong performance (mAP 0.94; recall 0.96; F1 score 0.95). In real-world testing, the model demonstrated strong performance, correctly identifying 98.9% of alcohol-branded posts in social media videos. SCANNER Alcohol achieved a low 6.7% false discovery rate, indicating high precision and low noise in the detection output.
Conclusions: SCANNER Alcohol is the first system worldwide to combine logo and text-based brand detection for automated surveillance of digital alcohol marketing. Its accuracy, scalability, and ethical design make it a valuable tool for public health monitoring and regulatory accountability. We are extending SCANNER to detect other forms of harmful digital marketing, such as for tobacco, gambling, unhealthy foods and breast-milk substitutes.

Introduction: Developing effective behavioural interventions for vaping requires a nuanced understanding of individuals’ everyday vaping practices and the contextual factors that shape patterns of product use. This study aimed to explore the daily vaping-related experiences of young Australians who vape using ecological momentary assessment.

Methods: Young Australians (14-39 year olds) who vape were invited to share their vaping-related experiences via an app-based daily diary. Data were collected over a 7-day period. Inductive reflexive thematic analysis was used for synthesis.

Results: We received 442 written entries and 11 audio recordings from 72 participants (54% men). We subsequently analysed the data provided by those who vaped daily (n=68 [95%]) given its richness. A conceptualisation of ‘a day in the life’ was created and two main themes identified: (1) vaping as part of daily routines and (2) negative and positive vaping related experiences. These experiences included disruption to daily activities and associated distress, feelings of shame and dependence, and the use of vapes for relaxation, stress relief, and self-care.

Conclusions: Among those who vape daily, vaping appears to be an integrated and often necessary part of their routine with disruptive consequences. Effectively conveying the insidious nature of vaping dependence and addiction is essential for raising awareness of associated harms and impacts, facilitating self-reflection of one’s own vaping patterns, and countering industry-driven narratives that minimise the risks of nicotine. Addressing the use of vapes as a coping mechanism for psychological distress requires the provision of alternative, evidence-based mental health supports.

Background: Commercial tobacco is the leading cause of preventable death and disease among Aboriginal and Torres Strait Islander peoples. It is responsible for half of all deaths in those aged 45 years and over. Its widespread use reflects the enduring impacts of colonisation, where commercial tobacco was systematically introduced across Australia. Using the four-stage Lopez model to describe the trajectory of the tobacco epidemic, we estimated in 2017 that Aboriginal and Torres Strait Islander peoples were in stage 3 (declining prevalence, increasing mortality). This study updates this assessment using the most recent epidemiology and policy data.

Methods: Published data on Aboriginal and Torres Strait Islander peoples’ tobacco prevalence, smoking attributable fraction (SAF), lung cancer mortality rates, community awareness of tobacco-related harms, and policy responses were extracted and assessed in line with the Lopez model.

Results: The tobacco epidemic among Aboriginal and Torres Strait Islander peoples may be transitioning to stage 4. This shift is characterised by continued declines in prevalence (54.4% in 1994 to 34.1% in 2022-23), a potential stabilisation of smoking-related mortality in those aged ≥45 years (SAF estimates stable between 2009-18 and subsequent decades; 44.8%-55.9%), high societal awareness of harms (2012-13: ~90% agree smoking and second-hand smoking are harmful; 2018-20: >60% felt smoking would or has caused sickness), and the implementation of policy and programs to address structural drivers of tobacco use (e.g. point of sale accessibility and visibility , expansion of the Tackling Indigenous Smoking program).

Conclusion: The findings reinforce the urgency of Aboriginal and Torres Strait Islander tobacco control, including the abolition of the tobacco industry. While the Lopez model provides important insights, it was developed using Western and high-income population data. An Indigenous-led model is needed to centre sovereignty, truth-telling, and recognition of structural barriers to tobacco control, such as racism, colonisation and commercial determinants.

Aboriginal and Torres Strait Islander Governance: We recognise that public health evidence is shaped by worldviews, histories, power relations, and systems of knowledge production. As authors, we must be clear about our positionality. Team members bring Aboriginal and Torres Strait Islander lived experience and leadership (LJW, RL), Indigenous lived experience in public health (RM), and extensive experience in Aboriginal and Torres Strait Islander health research, tobacco control, and epidemiology (all authors).
The need for this study arose from ongoing conversations and priorities identified by Aboriginal and Torres Strait Islander peoples, communities and Tackling Indigenous Smoking (TIS) teams. The research question, study design, and preliminary report were presented, discussed, and approved by Thiitu Tharrmay, the Aboriginal and Torres Strait Islander governance group for Yardhura Walani. This group upholds Aboriginal and Torres Strait Islander ways of knowing, being, and doing, ensuring all research undertaken by the Centre is of service to communities. All members are Aboriginal and Torres Strait Islander people who provide strategic guidance, cultural and methodological expertise, and oversight across all stages of research.
This study was guided by the Maiam nayri Wingara Indigenous Data Sovereignty principles, which ensure that Aboriginal and Torres Strait Islander peoples lead, shape, and benefit from how data are collected, interpreted, and used. In this study, this was achieved by aligning the work with community-identified priorities, collective decision-making, and Indigenous-led interpretation of data. No new data were collected for this analysis.
We recognise that data are not just numbers, they represent people, families, histories, and futures. We are therefore committed to sharing findings in ways that honour self-determination, uphold relational accountability, and reflect Indigenous worldviews. This includes interpretation grounded in community strengths, transparency to those represented, and recognition that data carry responsibilities to communities, Country, current and future generations.

Introduction: Non-therapeutic oral nicotine products (ONPs) including pouches, lozenges, and gummies are available in a range of flavours and are marketed as discreet, convenient and cost-effective. It is not legal to market or supply non-therapeutic ONPs in Australia, however there are reports of pouches increasingly infiltrating the market through online sales, influencer promotion, and physical retail. The proliferation of ONPs reflects the tobacco industry's ongoing strategy to diversify products and appeal to new generations of users. The present study assessed the prevalence of ONP use among a nationally-representative sample of Australian adults to provide timely data that can inform public health responses to emerging commercial nicotine products.

Methods: Data were collected in February 2025 from the Social Research Centre’s probability-based online panel, Life in AustraliaTM, and were weighted. Use of (a) oral nicotine pouches, (b) chewable/dissolvable products, and (c) “other” ONPs, as well as smoking and vaping status and sociodemographic characteristics were assessed (N=9841; analytic sample, N=9802 with complete data). Data collection will be repeated in February 2026, and this more recent data—including updated ONP prevalence estimates—will be available at the time of the conference.

Results: Overall, 3.8% of Australian adults were currently using ONPs in February 2025. Compared to respondents who neither smoked nor vaped (1.6%), ONP use was higher among adults who exclusively smoked (11.2%; Prevalence Ratio (PR) (95% CI) = 7.32 (5.27, 10.17)), exclusively vaped (13.4%; PR=7.25 (4.92, 10.68)), and both smoked and vaped (22.5%; PR=12.06 (8.56, 17.00)). Use was lower among older adults (i.e., 55+ years, 1.9%; PR=0.35 (0.23, 0.52)) compared to those 18-24 years (8.3%), and higher among men/males (4.3%; PR=1.51 (1.19, 1.92)) compared to all others (3.2%).

Conclusion: Continued surveillance of ONP use among Australians is essential for monitoring patterns of use and informing public health policy and regulatory responses.

INTRODUCTION:
The oversupply of tobacco products (both legal and illicit) in Australia is a powerful commercial determinant of health. All Australian jurisdictions now require retailers to hold a licence, but none restrict the number, type, or location of outlets. The aim of this research was to measure public support for reducing tobacco licences and to model the effect of hypothetical licensing reforms in Western Australia (WA).

METHODS:
In December 2024, a representative sample of WA adults reported their attitudes towards reducing tobacco availability in WA. We then investigated reform scenarios that had strong community support, including banning tobacco outlets within specified proximity to schools; banning sales from or limiting sales to certain retailer types; and capping the number of tobacco outlets based on population density. Modelling was undertaken to quantify the reductions in licensed tobacco retail outlets in WA associated with each scenario.

RESULTS:
Of the 1,012 survey respondents (11% smokers), 70% agreed that tobacco should be less available or not available at all (75% non-smokers, 26% smokers). The majority (81%) supported banning tobacco sales within 500 metres of a school. Less than one third thought tobacco should be sold at convenience stores, delis, or lunch bars. Among respondents who smoked, 46% said they would probably smoke less if they had to go further out of their way to purchase cigarettes.

There were 3,079 licenced tobacco retailers in WA in July 2025, equivalent to approximately 10.3 per 10,000 people, with higher outlet density in socioeconomically disadvantaged areas. Results will be presented for each of the modelled scenarios, together with recommendations for licensing reforms.

CONCLUSION:
Licensing reforms can directly reduce the availability of legal tobacco and indirectly reduce illicit supply, by aiding enforcement. There are impactful, sustainable options that are likely to be supported by the wider community.

Background
Regulation of tobacco retail environment is an under-utilised policy measure. Previous Australian and international studies have explored retailers’ views on selling tobacco, including factors influencing decisions to stop selling tobacco and perspectives on enforced restrictions. Here we explore the experiences of retailers selling tobacco in the NT where daily smoking prevalence among people aged 14 and over is estimated at 13.3% compared to 8.3% in Australia overall.

Methods
Semi-structured interviews were conducted face to face and by phone. Anticipating that many retailers would cite time constraints as a reason for non-participation, a shorter interview of three key questions was offered as an alternative to a full interview, to maximise participation. We analysed the data thematically.

Results
Twenty-six retailers participated representing a mix of different retailer types. Many participants cared about their customers and saw themselves as providing a service. They were concerned about the impacts of tax increases on tobacco products and resulting financial stress on their customers. Many talked about customers’ freedom and choice to smoke or to not smoke. Although some participants recognised that smoking is an addiction, participants did not identify their part as suppliers in sustaining that addiction. Attitudes to potential regulatory measures varied, with some voicing strong resistance, and others willing to comply. Most participants would not volunteer to stop selling tobacco, however there was an example of a retailer choosing to not sell tobacco.

Conclusion
Tobacco retailing is a major and neglected component of tobacco control policy. Understanding the perspectives of tobacco retailers is important in order properly support them through a transition to a smoke-free society.

Intro: Digital marketing is largely data-driven, whereby automated marketing models tailor and target advertising to people who are most likely to buy a product. When it comes to the advertising of harmful and addictive products like alcohol, this means people with high-risk alcohol use are likely to be identified as a target for alcohol advertising. This study explored Australians’ experiences relating to digital alcohol marketing and assessed whether there were any differences by risk of alcohol harm.

Methods: In April 2025, we surveyed 2,037 Australian adults. Alcohol use risk level was categorised for participants who had used alcohol in the past year (n=1,799) using ASSIST-Lite. This identified participants as having either low risk alcohol use (n=945), hazardous/harmful alcohol use (n=480) or likely experiencing alcohol dependence (n=365). Survey results were analysed for differences between these risk categories.

Results: Participants likely experiencing alcohol dependence were significantly more likely to recall alcohol advertising across online platforms (23-39% vs 4-12%) and to be concerned about online alcohol advertising (52% vs 35%) than those with low-risk alcohol use. Participants likely experiencing alcohol dependence were also more likely than those with low-risk alcohol use to report that online alcohol advertising leads them to crave (47% vs 12%), purchase (71% vs 21%) and use alcohol (68% vs 18%) and to indicate that online alcohol advertising makes it more difficult for them to reduce the amount of alcohol they use (50% vs 9%). Across risk groups, there was high support for less or no alcohol advertising online (60%-73%).

Conclusions: There is a pressing need for regulation of digital alcohol advertising in Australia. People experiencing alcohol dependence should be considered as a critical population group in the development of regulation, as online alcohol advertising can be particularly harmful for this group.