University of Oklahoma researchers have demonstrated that a smartphone-based smoking cessation app using real-time adaptive interventions can nearly double quit rates among low-income adults, according to results published in JAMA Network Open. The clinical trial compared the university's Smart-T app against the National Cancer Institute's QuitGuide app in 454 low-income smokers across the United States.
Adaptive Algorithm Delivers Personalized Interventions
Smart-T distinguishes itself from traditional smoking cessation apps through its just-in-time adaptive intervention approach. The app prompts users up to five times daily to report their urges to smoke, stress levels, mood, and environmental factors such as proximity to other smokers. An integrated algorithm processes these responses to generate personalized risk scores and delivers tailored interventions accordingly.
"Smart-T is like having a tobacco cessation counselor in your pocket," said lead author Emily Hébert, DrPH, assistant professor of family and preventive medicine at OU College of Medicine and member of the TSET Health Promotion Research Center. "While a combination of behavioral counseling and medication is the most effective way to quit smoking, in-person counseling may not be practical for everyone, especially those facing barriers like transportation or busy schedules."
Superior Engagement and Efficacy Outcomes
The randomized controlled trial enrolled 454 participants, with half assigned to Smart-T and half to QuitGuide, a static app that allows users to track cravings and provides general smoking cessation tips. All participants received nicotine replacement therapy in the form of patches, gum, or lozenges.
After six months, Smart-T users demonstrated significantly higher cessation rates compared to QuitGuide users. Beyond the primary efficacy endpoint, Smart-T users showed superior engagement metrics, used the app more frequently, reported finding it more helpful, and were more likely to request additional nicotine replacement therapy when their initial supply was depleted.
Biochemical Verification Innovation
The study implemented novel verification methods to ensure data integrity. Participants' smoking status was biochemically confirmed through breath carbon monoxide measurements using a device connected to their smartphones. Facial recognition software verified that the correct participant was providing the breath sample, representing one of the first mobile health trials to incorporate such rigorous biochemical verification methods.
Addressing Health Disparities
The research specifically targeted low-income populations, where smoking rates remain disproportionately high despite overall declines in tobacco use across the United States over the past six decades. This demographic faces unique barriers to accessing traditional smoking cessation resources, making smartphone-based interventions particularly relevant.
"Some people who smoke will not benefit from apps like Smart-T, but low-cost and always-available smartphone interventions could provide a convenient way for low-income adults to quit," Hébert noted. "We're really trying to find the best recipe for smoking cessation for everyone."
Future Research Directions
The research team plans to expand their studies to larger populations across the United States with extended follow-up periods beyond six months. The Smart-T app was developed by Michael Businelle, Ph.D., who co-leads the next research phase with Hébert through the mHealth Shared Resource and TSET Health Promotion Research Center.
The TSET Health Promotion Research Center has established itself as a global leader in mobile health technology, supporting more than 115 studies and attracting $85 million in grants since launching the Insight™ mHealth Platform in 2015.
The research was supported by the National Cancer Institute (grant No. R01CA221819), with additional funding from the Oklahoma Tobacco Settlement Endowment Trust and the Oklahoma Shared Clinical and Translational Resources through the National Institute of General Medical Sciences (grant No. U54GM104938).
