AI Tools Could Help Identify the Best Ways to Help Young People Quit Vaping
Key Takeaways:
- Researchers at the University at Buffalo used machine learning and explainable AI tools to identify which vaping cessation strategies may work best for different individuals.
- The study found that starting to vape before age 18 – especially before age 15 – was one of the strongest predictors of continued nicotine use.
- Researchers believe AI-driven approaches could help universities and public health teams move from generic stop-vaping programmes to more personalised interventions.
Understanding why young people struggle to quit vaping
Young adults between the ages of 18 and 24 are now among the heaviest users of e-cigarettes in the United States, with 38.4% of young people reporting habitual vaping. Rates of e-cigarette use are particularly high in Western New York, where vaping prevalence exceeds that seen in New York City.
Although awareness of the potential health risks associated with vaping has increased, many people still find it difficult to stop using e-cigarettes. Researchers say this challenge can be even greater for younger individuals, whose brains may be more susceptible to nicotine dependence.
These concerns prompted cancer researchers at the University at Buffalo (UB) to investigate why some young people continue vaping while others successfully quit. Their goal was not only to better understand vaping behaviour, but also to identify which cessation strategies may be most effective for different individuals.
The team conducted an online survey involving 119 people who vape, approximately three quarters of whom were aged between 21 and 26 years old. Their findings were published in PLOS Digital Health.
The senior corresponding author of the study was Supriya D. Mahajan, Ph.D., associate professor of medicine in the Jacobs School of Medicine and Biomedical Sciences at UB.
Researchers explore better ways to support vaping cessation
The study was driven by the researchers’ experiences treating people with nicotine dependence in clinical settings.
“As cancer researchers in the divisions of Hematology/Oncology and Allergy, Immunology, and Rheumatology at UB, we see the direct clinical consequences of nicotine dependence in our patients,” says Satheeshkumar Poolakkad Sankaran, DDS, first author of the study and research scientist in the Division of Hematology/Oncology in the Department of Medicine.
“We wanted to understand not only who is vaping but also who is successfully quitting—and then translate those insights into better cessation support for our cancer patients and into broader social determinants of health research.”
To investigate this, the researchers applied artificial intelligence techniques, including machine learning, to determine why some stop-vaping strategies appear to work for certain people but not for others.
The researchers noted that these findings could potentially extend beyond vaping cessation and inform wider public health approaches.
Using AI to predict who may successfully quit
The research team tested five different computer models designed to predict which individuals were most likely to successfully stop vaping.
According to Poolakkad Sankaran, some of the most effective models were also among the simplest.
“The simplest and most reliable ones were like a smart checklist that automatically figured out which life factors mattered most in deciding whether or not to stop vaping,” says Poolakkad Sankaran.
The researchers also explored more advanced forms of “explainable AI” – systems designed to help humans understand how AI reaches its conclusions.
Explainable AI offers insight into individual barriers
One explainable AI model used in the study was called Accumulated Local Effects (ALE). According to the researchers, this tool helps visualise how specific factors influence vaping cessation outcomes across larger groups of people.
“For example, the model called Accumulated Local Effects (ALE) shows how each factor—for example, being under age 21—changes the odds of quitting across the whole group, almost like a graph of ‘what-if’ scenarios,” says Poolakkad Sankaran.
The team also used another explainable AI approach known as Local Interpretable Model-Agnostic Explanations (LIME), which focuses on individuals rather than groups.
“Another model, Local Interpretable Model-Agnostic Explanations (LIME), zooms in on individual people,” says Poolakkad Sankaran. “It can look at one specific vaper and say, ‘For this person, social triggers are the biggest barrier—here’s exactly how much they lower their chance of success.’”
Researchers believe these tools could eventually help clinicians and counsellors provide more personalised support rather than relying on standardised approaches for everyone.
Earlier vaping initiation linked to greater difficulty quitting
One of the clearest findings from the study was the strong relationship between early vaping initiation and continued nicotine use later in life.
The researchers found that individuals who began vaping before the age of 18 – particularly before age 15 – were significantly more likely to continue vaping and struggle with cessation.
“Starting before age 18, and especially before 15, was one of the strongest predictors of continued use,” says Poolakkad Sankaran.
“This tells us that prevention must begin early, before the brain’s reward system becomes wired to nicotine. For kids who already started young, the message is hopeful but urgent: The sooner they get help, the better their chances.”
The researchers suggested that vaping cessation programmes aimed at younger individuals should be specifically tailored to this age group.
“These should be age-tailored strategies: short, frequent digital nudges; peer support; and trigger-management tools, because their developing brains make these people especially vulnerable but also especially responsive to timely intervention,” he explains.
Universities could play a key role in vaping prevention
Poolakkad Sankaran believes universities could become important centres for implementing AI-driven vaping cessation support.
“UB is uniquely positioned to translate these exploratory machine learning/explainable AI results into real-world programs that reduce nicotine addiction, lower long-term health care costs and address health disparities affecting Buffalo’s young population,” he says.
The researchers suggest that university health services and local public health departments could use these findings to build personalised text-message campaigns targeting the most common triggers associated with vaping in local communities.
They also propose that campus applications or quit-support services could identify students at higher risk – including younger individuals, frequent users and people vulnerable to social vaping triggers – and provide immediate tailored support.
AI tools could be integrated into existing stop-vaping programmes
The research team now plans to integrate their predictive models into digital vaping cessation tools, including expanded versions of existing programmes such as “This is Quitting.”
The aim is to help counsellors better understand why a particular student may be struggling and which intervention strategies are most likely to succeed.
“Because the study was done locally with Western New York participants, the findings already reflect the realities our students and young adults face,” says Poolakkad Sankaran.
The researchers say the work highlights how predictive analytics and machine learning could help public health professionals move beyond one-size-fits-all approaches.
He adds that the research demonstrates how machine learning and predictive analytics can help public health teams move from one-size-fits-all programs to precision interventions by identifying who is most at risk and what will actually help them before they drop out of treatment.
“Our study pushes the field forward by showing that explainable AI (XAI) can make these powerful tools transparent and trustworthy for clinicians and policymakers,” he says.
“Instead of a black-box prediction, we deliver actionable, human-understandable explanations that can be directly built into digital health apps and community programs.”
Source: Medical Xpress
Read More