AI-Enhanced App-based Intervention for Adolescent E-cigarette Cessation

This quasi-experimental study aims to develop and evaluate an AI-enhanced smartphone app designed to support adolescents aged 14-20 in quitting e-cigarettes. Given the high prevalence of e-cigarette use among youth, this app-based intervention focuses on providing personalized, real-time support for cravings and motivation to quit. The app integrates interactive features to engage users and is designed for scalability, enabling wide-reaching impact in various settings such as schools, clinics, and communities.

Study Phases and Objectives

Phase 1: Development and Usability Testing Phase 1 focuses on refining an existing beta version of the app. In this formative stage, the app's design, content, and features will be adjusted based on adolescent feedback to ensure it is user-friendly and engaging. Participants will test the app and provide insights through usability surveys and interviews, which will inform necessary changes.

Key activities in this phase include:

* Gathering feedback on usability and design.
* Modifying app features to better meet the preferences and needs of adolescent users.
* Finalizing the app to meet high usability benchmarks for deployment in the next phase.

Phase 2: Clinical Feasibility Testing In Phase 2, the app's effectiveness will be tested using a quasi-randomized design with two groups: one group of participants will begin using the app immediately, while the second group will start after a three-month delay. This approach will help determine if earlier access to the intervention leads to improved outcomes in terms of e-cigarette cessation.

The study will assess how the app impacts participants' readiness to quit, actual quitting attempts, and ongoing motivation over time. Engagement levels with the app's features, such as real-time craving support and AI-driven educational modules, will also be tracked to evaluate the intervention's overall feasibility and appeal.

App Features and Personalization

The app's core features include:

1. Goal Setting and Progress Tracking: Users set personal quitting goals, track their progress, and access daily training modules to build skills for managing cravings and quitting.
2. Real-Time Craving Management: The "urge" feature provides immediate support during cravings, using mindfulness exercises and coping strategies tailored to each user's needs.
3. AI Chatbot Support: A chatbot offers 24/7 assistance, answering questions and providing motivation based on users' quitting status and individual characteristics.

These AI-driven tools are customized according to user data and interactions within the app, ensuring the intervention feels personal and responsive to each user's progress.

Data Collection and Analysis Data will be collected on app usage, engagement with specific features, and changes in e-cigarette use over time. Analysis will include both user feedback and statistical evaluation of the app's impact on participants' quitting success. Insights from this data will contribute to the ongoing refinement of the app and inform its potential for broader use as an adolescent-focused e-cigarette cessation tool.

Anticipated Impact This study aims to create a user-friendly, scalable app that leverages AI to support adolescents in quitting e-cigarettes effectively. If successful, this digital intervention could be a valuable resource for youth cessation programs and serve as a model for similar health-related app-based interventions.

• AI-enhanced smartphone app (Behavioral): A smartphone app has been developed and is in keeping with guideline recommendations for the treatment of e-cigarette products. This app has a user-friendly Graphic User Interface (GUI) to allow users to build their own accounts and individualized contents conveniently, based on the input the users initially provide including e-cigarette use patterns, readiness to quit e-cigarette, beliefs about e-cigarette, nicotine addiction, self-efficacy, other substance use status, and parental or peer e-cigarette use status. The proposed AI model in this app will learn information from the input data, including progress toward e-cigarette cessation (e,g, changes of readiness of quitting, quit attempts), and additional data including emotional status, stress level, feedback to the previous learning modules, and then predict the result on the fly. Based on the predicted result, the app will send in-time motivational messages and mindfulness training modules.
• AI-enhanced smartphone app, but with delayed access (Behavioral): Participants in the control group will be placed on a three-month waitlist. After this period, they will receive access to the same app-based intervention as the immediate intervention group, allowing a comparison between immediate and delayed access.

Primary outcomes

• Usability [30-Day Follow-up]: The usability of the intervention will be assessed using the mHealth App Usability Questionnaire, a 21-item instrument designed for interactive mobile health apps. It measures three domains: ease of use and satisfaction, system information arrangement, and usefulness. Each item is rated on a 5-point Likert scale, yielding a total score range of 21 to 105, with higher scores indicating better usability. An open-ended question will also be included at the end of the survey to gather suggestions for app improvement.

Secondary outcomes

• Engagement - Frequency of App Use [30-Day and 3-Month Follow-up]: Automatically recorded log data from the app will track each login event by a participant.
• Engagement - Minutes of App Use [30-Day and 3-Month Follow-up]: Cumulative minutes of app usage will be captured via app log data.
• E-cigarette Use [30-day and 3-month follow up]: Self-reported number of days participants used e-cigarettes in the past 30 days. Responses range from 0 to 30 days.
• Quit attempts [30-Day and 3-Month Follow-up]: Self-reported number of quit attempts in the past 30 days, along with the longest duration of abstinence.
• Readiness to Quit [30-Day and 3-Month Follow-up]: Readiness to quit will be assessed using a modified Contemplation Ladder. The Contemplation Ladder ranges from 0 (no thought of quitting) to 10 (taking action to quit). Higher scores indicate greater readiness.
• Nicotine Dependence [30-Day and 3-Month Follow-up]: Measured using the Penn State Electronic Cigarette Dependence Index (PSECDI), a 10-item scale with a total score range of 0 to 20. Higher scores indicate higher levels of nicotine dependence.
• Beliefs About E-Cigarettes [30-Day and 3-Month Follow-up]: Assessed using a questions from the National Youth Tobacco Survey about e-cigarette harm, addictiveness, and benefits.

Sex: All, ages: 14 Years to 20 Years

Inclusion Criteria

* Adolescents who are 14 to 20 years old
* Currently use nicotine-containing e-cigarettes (those responding "Yes" to: Have you used an electronic vaping product such as PuffBar, ElfBar, Lost Mary, JUUL, Vuse, e-cigarettes, vapes, vape pens, e-cigars, e-hookahs, hookah pens, or mods at least 1 day in the last 30 days? \[CDC, 2020\])
* Interested in participating in an e-cigarette use cessation program
* Owners of an iPhone or Android smartphone who use their phone daily
* Able to read English

Exclusion Criteria

* Individuals who fall outside the age range of 14 to 20 years
* Those who have not used a nicotine-containing e-cigarette in the past 30 days
* Individuals not interested in participating in an e-cigarette cessation program
* Adolescents who do not own or regularly use an iPhone or Android smartphone
* Non-English speakers

• Hebert ET, Ra CK, Alexander AC, Helt A, Moisiuc R, Kendzor DE, Vidrine DJ, Funk-Lawler RK, Businelle MS. A Mobile Just-in-Time Adaptive Intervention for Smoking Cessation: Pilot Randomized Controlled Trial. J Med Internet Res. 2020 Mar 9;22(3):e16907. doi: 10.2196/16907. PMID 32149716
• Billingham SA, Whitehead AL, Julious SA. An audit of sample sizes for pilot and feasibility trials being undertaken in the United Kingdom registered in the United Kingdom Clinical Research Network database. BMC Med Res Methodol. 2013 Aug 20;13:104. doi: 10.1186/1471-2288-13-104. PMID 23961782
• Bell ML, Whitehead AL, Julious SA. Guidance for using pilot studies to inform the design of intervention trials with continuous outcomes. Clin Epidemiol. 2018 Jan 18;10:153-157. doi: 10.2147/CLEP.S146397. eCollection 2018. PMID 29403314
• Baskerville NB, Struik LL, Guindon GE, Norman CD, Whittaker R, Burns C, Hammond D, Dash D, Brown KS. Effect of a Mobile Phone Intervention on Quitting Smoking in a Young Adult Population of Smokers: Randomized Controlled Trial. JMIR Mhealth Uhealth. 2018 Oct 23;6(10):e10893. doi: 10.2196/10893. PMID 30355563
• Audrain-McGovern J, Rodriguez D, Pianin S, Alexander E. Initial e-cigarette flavoring and nicotine exposure and e-cigarette uptake among adolescents. Drug Alcohol Depend. 2019 Sep 1;202:149-155. doi: 10.1016/j.drugalcdep.2019.04.037. Epub 2019 Jul 19. PMID 31351341

• Study Protocol (PDF, 2024-08-11)

• Eunhee Park, PhD, RN, Principal investigator, University at Buffalo

NIH grants: R34CA287720