Smoking rates among adults in the United States have fallen precipitously since the first Great American Smokeout was organized in San Francisco in 1977, when over a third of Americans smoked cigarettes. Today, that number is less than 12.5%. While those changes are dramatic, and worthy of celebration - called “one of the most significant public health successes in modern U.S. history” by Surgeon General Jerome M. Adams in his 2020 report on smoking cessation - it is important to acknowledge that cigarette smoking remains the leading cause of preventable disease, disability, and death in the United States, accounting for 1,300 deaths every day. And while cigarette smoking has decreased broadly across the country, tobacco use among some minors has actually increased over the last two decades. Clearly, there is still work to be done, and researchers at the UTHealth Houston School of Public Health are developing new and intriguing techniques, utilizing mobile health technology, to help smokers quit.
mHealth, or mobile health, is a term commonly used to reference the use of mobile communication devices - mobile phones, tablets, and wearable smart-devices, for health services. With the ubiquity of the technology and devices across demographics, mHealth has emerged as an important tool for reaching hard-to-reach and underserved populations, and improving access to public health information and tools.
Irene Tami-Maury, DMD, DrPH, MSc, assistant professor of Epidemiology, Human Genetics & Environmental Sciences at the School of Public Health, leads the Health Equity Research Group (HERG) at the school, who have adapted and pilot tested a text-message-based smoking cessation intervention for sexual and gender minority (SGM) individuals – lesbian, gay, bisexual, transgender, and queer (LGBTQ+) people. SmokefreeSGM has been tailored to the needs and experiences of LGBTQ+ smokers. SmokefreeGSM is based on the SmokefreeTXT intervention developed by the National Cancer Institute (NCI), but is tailored specifically for the LGBTQ+ community using input from tobacco specialists, current and former LGBTQ+ smokers, and leading experts in behavioral health, epidemiology, and biostatistics. The goal of the NCI-funded program is to use a randomized control trial to determine the extent to which text-based interventions like SmokefreeTXT and SmokefreeSGM can help LGBTQ+ people who want to quit cigarettes.
Participants are screened and assessed upon entry to the project, before being randomly assigned to either SmokefreeTXT (the control group), or SmokefreeSGM. Neither participants nor researchers are told which group an individual is assigned to. Overall, participants spend eight months in the program. Follow-up assessments are conducted at one, three, and six-month intervals through smoking status self-report and nicotine saliva tests.
SmokefreeSGM participants receive complimentary nicotine patches to help wean them off of cigarettes, and they receive daily text messages two weeks before, and up to four weeks after they quit smoking – motivational messaging encouraging them to stay smoke-free, informational messages reminding them of the important reasons why they quit, strategies for dealing with cravings, triggers, and stressful situations, among others. They also receive prompts and queries designed to track their mental and emotional well-being. Importantly, on-demand help is available to participants who need help staying smoke-free.
At this time, SmokefreeSGM is in its final phase (randomized control trial) recruiting potential study participants who self-identify as LGBTQ+ adult individuals and are currently smoking cigarettes. Interested individuals will participate in a two-part virtual screening process, and if enrolled, are given access to the text-messaging program, receive a supply of nicotine patches, and participate in the four virtual assessments and a final virtual interview. Each of these assessments (including the individual interview) last less than six minutes, for which study participants are compensated.
A separate study, funded by SWOG/The Hope Foundation and conducted by Tami-Maury and HERG, uses a mobile phone application to enhance the knowledge and skills of healthcare professionals providing smoking cessation services to their patients. The Decision-T app aims to help providers make quicker and better decisions about cessation treatment approaches for their patients, and is based on the 5A’s framework: Ask, Advise, Assess, Assist, and Arrange. Providers ask patients about their smoking behavior, advise them to quit smoking, assess their willingness to quit, assist them with finding available resources, and arrange follow-up meetings to evaluate patient progress. Decision-T guides healthcare professionals in estimating their patients’ smoking behavior and provides a personalized quitting plan for the patient to use. The research team has completed the beta testing phase of the app. The feedback from the small sample of healthcare providers testing Decision-T has been outstanding in terms of its usability, functionality, and acceptability. These preliminary findings are encouraging for wider implementation of the Decision-T app, which planned by HERG.
Emily T. Hébert, DrPH, assistant professor in the Department of Health Promotion and Behavioral Sciences is also developing interventions that utilize mobile technology to aid smoking cessation. Hébert was awarded an NIH Career Development Award in 2019 for a study using machine learning to develop just-in-time adaptive interventions (JITAI) for smoking cessation. Machine learning is a robust data analytic strategy that can produce highly accurate predictive models from large and dynamic datasets in real-time. The objective is to use supervised machine learning to develop an automated algorithm to quantify smoking lapse risk at the individual level, and deliver a personalized intervention in real-time.
Phone sensors, wearable technology, and real-time data collection methods make it possible to collect a wealth of personalized environmental and physiological data such as location, heart rate, and mood. Using environmental and situational cues provided by the technology, such as cravings or proximity to other smokers, the technology can predict lapses and interrupt them with timely, and tailored support, delivered via mobile devices, when it is most needed. If successful, this project and the methodology it implements could be a profound demonstration of the power of machine learning, and how it can be adapted for other health behaviors and concerns, including diet, physical activity, or other substance use disorders. The study will begin enrollment in 2023.
“Most smokers want to quit,” says Hébert, “but many don’t use evidence-based treatments like behavioral counseling or nicotine replacement therapy. Since 85% of adults in the U.S. own a smartphone, our hope is to use mHealth interventions to help increase access to effective treatments like these, and help more people quit.”
While this project will be among the first to use machine learning methods to predict risk of individual smoking lapse in real time, Hébert has participated in other related research using mobile technology and machine learning. In a study published in 2020, Hébert and her colleagues found that dynamic smartphone apps that tailor intervention content in real-time may increase user engagement and exposure to treatment-related materials, and may be capable of providing similar outcomes to traditional, in-person counseling. A second study published that year used an algorithm to develop a just-in-time adaptive intervention for individuals in a clinic-based smoking cessation program. The study demonstrated the utility of data-driven approaches in estimating smoking lapses and developing JITA.