Generative AI avatar videos for tobacco prevention on social media: a randomized controlled trial
Merl, N. B.; Wies, C.; Schramm, F.; Winterstein, J. T.; Chanda, T.; Brinker, T. J.
Show abstract
Short-form video platforms increasingly shape how young audiences encounter health information. Generative artificial intelligence can produce standardized avatar-based messages at scale, but randomized evidence for tobacco prevention is scarce. In this three-arm randomized online intervention study with pre-post assessment, participants aged 16 years or older were assigned to an AI avatar video emphasizing short-term smoking consequences, an AI avatar video presenting long-term cancer-related information matched to an American Cancer Society fact sheet, or the same fact sheet in written form. The primary outcome was post-intervention intention to avoid smoking and secondhand smoke exposure, adjusted for baseline intention. Among 400 randomized participants, 272 had complete data for the primary baseline-adjusted analysis. Intention increased from baseline to post-intervention in all conditions, with no statistically significant between-group differences. These findings support AI avatar videos as a scalable, social-media-compatible format for digital tobacco prevention, while not establishing superiority or equivalence.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Effectiveness of Online Training in Improving Primary Care Doctors Competency in Brief Tobacco Interventions: A Cluster Randomised Controlled Trial of WHO Modules in Delta State, Nigeria 94%
- Investigating the added value of biomarkers compared with self-reported smoking in predicting future e-cigarette use: Evidence from a longitudinal UK cohort study 93%
- The 'Icarus effect' of preventative health behaviors 92%
Similar papers in this journal
- Impact of a regional educational advertising campaign on harm perceptions of e-cigarettes, prevalence of e-cigarette use, and quit attempts among smokers 95%
- E-cigarettes With Varenicline Versus Varenicline for Smoking Cessation: A Pragmatic Randomised Controlled Trial 94%
- Associations between e-cigarette use and e-cigarette flavors with cigarette smoking quit attempts and quit success: Evidence from a US large, nationally representative 2018-2019 survey 94%
Similar papers in this journal
- Identifying Key Predictors of Smoking Cessation Success: Text-Based Feature Selection Using a Large Language Model 93%
- Varenicline Versus Cytisine for Smoking Cessation in a Primary Care Setting: A Randomized Controlled Trial 93%
- Perceptions of children and young people in England on the smokefree generation policy: a focus group study 93%
Similar papers in this journal
- Youth susceptibility to tobacco use: Is it general or specific? 92%
- Characterization of Trajectories of Physical Activity and Cigarette Smoking from Early Adolescence to Adulthood 92%
- How can we maximise the benefits of smoke-free prisons? Decision analytic model to predict potential impacts on public health 91%
Similar papers in this journal
- COVID-19, smoking, vaping and quitting: A representative population survey in England 93%
- Transitions from smoking to exclusive e-cigarette use, dual use, or stopping nicotine use in ALSPAC and their association with modifiable and sociodemographic factors 93%
- Impact of health warning labels and calorie labels on selection and purchasing of alcoholic and non-alcoholic drinks: a randomised controlled trial 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.