Cigarette smoking is a leading cause of preventable illness and death across the globe. A landmark study analyzing data from over 5,000 American smokers has revealed crucial insights into what helps people quit smoking or switch to less harmful alternatives. Led by Dr. Xiaona Liu from Smoore Research Institute, in collaboration with Professor Xuxi Zhang from Peking University School of Public Health and Dr. Ian Fearon from whatIF? Consulting, the research used sophisticated machine learning techniques to analyze over 2,000 different factors that might influence smoking behavior changes.

The study tracked smokers between 2016 and 2021, finding that while most continued smoking (69.9%), some successfully quit (18.5%), others became dual users of both cigarettes and e-cigarettes (7.2%), and a smaller group (4.4%) completely switched to e-cigarettes. “By following people over time, we could identify what really helps different groups make successful changes,” explains Dr. Liu. The research team narrowed down 396 potential factors to 27 key predictors that strongly influence smoking behavior changes.

For successful quitting, the study found several crucial factors. “People who smoked fewer cigarettes per day and waited longer after waking up before their first cigarette were more likely to quit,” notes Dr. Liu. “Environmental factors were also important – not living with other smokers and having higher education levels significantly improved quitting success.” The research revealed that having a concrete plan to quit and making previous quit attempts also increased success rates.

For those who switched to e-cigarettes, age and social media exposure emerged as surprising key factors. “Younger adults who were active on social media were more likely to switch completely to e-cigarettes,” Dr. Liu explains. “We also found that people with previous e-cigarette experience and those who perceived e-cigarettes as less harmful were more likely to make the switch.” Interestingly, those who had used prescription smoking cessation medications were more likely to switch to e-cigarettes than continue smoking.

This research have important implications for public health strategies. “Our research shows that we need tailored approaches for different groups,” emphasizes Dr. Liu. “What works for someone trying to quit completely might be different from what helps someone switch to less harmful alternatives. This understanding could help healthcare providers and policymakers develop more effective, personalized support programs.” While the study focused on U.S. adults, its comprehensive analysis and long-term tracking provide valuable insights for developing more effective smoking cessation and harm reduction strategies worldwide.

Journal Reference

Yue Cao, Xuxi Zhang, Ian M. Fearon, Jiaxuan Li, Xi Chen, Fangzhen Zheng, Jianqiang Zhang, Xinying Sun, Xiaona Liu. “Identifying Predictors of Smoking Switching Behaviours Among Adult Smokers in the United States: A Machine Learning Approach.” Cureus. DOI: https://doi.org/10.7759/cureus.69183

About the Authors

Dr. Xiaona Liu is an Epidemiologist who currently serves as the Head of Clinical and Behavioral Sciences at the SMOORE Research Institute in China. Dr. Liu previously worked as a senior physician on Infection Control at a local CDC’s office in China, and a postdoctoral researcher on Health Technology Assessment & Implementation at Erasmus University Medical Center in the Netherlands. She received a PhD in Infectious Disease Control from the Erasmus University Medical Center in 2015, a MPH from Netherlands Institute for Health Sciences in 2013, and a MSc in Social Medicine from Peking University Health Science Center in 2011. She has first- and/or corresponding- authored 22 scientific journal articles related to disease prevention and control. She has given more than 15 invited presentations at renowned international conferences.   

Dr. Xuxi Zhang is an Assistant Professor who is currently working at Department of Social Medicine and Health Education, School of Public Health, Peking University. She received a B.S. (2014) and a Master of Public Health (2017) from Peking University. She received her Ph.D. in Public Health from the Erasmus University Rotterdam in 2020. She worked as a Postdoctoral Research in National School of Development at Peking University from 2021 to March 2023. Her research interests span both healthy aging and eldercare innovation. She has published more than 30 scientific papers, and much of her work focuses on the health promotion of frailty, chronic conditions, disability, cognitive impairments and mental health among older adults.

Dr. Ian M. Fearon is a consultant specialising in the scientific basis for tobacco harm reduction. He provides scientific consulting support to tobacco and nicotine product manufacturers, and pharmaceutical companies, helping them bring novel harm-reduced products to market and to support their scientific advocacy. Dr. Fearon has authored almost 70 peer-reviewed scientific publications, which in recent years have focussed on the pharmacological and behavioural impacts of nicotine-containing products among individuals and populations.

Ms. Yue Cao is a biostatistician currently working at the SMOORE Research Institute in China, where she focuses on clinical and behavioral studies on the effects of novel tobacco products. She previously worked as a statisitician at the National Clinical Research Center for Kidney Disease, contirbuting to the development of the Chinese Renal Disease Data System. Ms. Cao received her MSc in Biostatistics from the Yale School of Public Health in 2019 and her B.S. in Mathematics and Statistics from Tongji University in Shanghai in 2017. She has a strong interest and experience in applying statistical methods and machine learning techniques to analyse public health data and its determinants.