Search

Introduction Cigarette smoking continues to pose a threat to public health. Identifying individual risk factors for smoking initiation is essential to further mitigate this epidemic. To the best of our knowledge, no study today has used machine learning (ML) techniques to automatically uncover informative predictors of smoking onset among adults using the Population Assessment of Tobacco and Health (PATH) study. Aims and Methods In this work, we employed random forest paired with Recursive Feature Elimination to identify relevant PATH variables that predict smoking initiation among adults who have never smoked at baseline between two consecutive PATH waves. We included all potentially informative baseline variables in wave 1 (wave 4) to predict past 30-day smoking status in wave 2 (wave 5). Using the first and most recent pairs of PATH waves was found sufficient to identify the key risk factors of smoking initiation and test their robustness over time. The eXtreme Gradient Boosting method was employed to test the quality of these selected variables. Results As a result, classification models suggested about 60 informative PATH variables among many candidate variables in each baseline wave. With these selected predictors, the resulting models have a high discriminatory power with the area under the specificity-sensitivity curves of around 80%. We examined the chosen variables and discovered important features. Across the considered waves, two factors, (1) BMI, and (2) dental and oral health status, robustly appeared as important predictors of smoking initiation, besides other well-established predictors. Conclusions Our work demonstrates that ML methods are useful to predict smoking initiation with high accuracy, identifying novel smoking initiation predictors, and to enhance our understanding of tobacco use behaviors. Implications Understanding individual risk factors for smoking initiation is essential to prevent smoking initiation. With this methodology, a set of the most informative predictors of smoking onset in the PATH data were identified. Besides reconfirming well-known risk factors, the findings suggested additional predictors of smoking initiation that have been overlooked in previous work. More studies that focus on the newly discovered factors (BMI and dental and oral health status,) are needed to confirm their predictive power against the onset of smoking as well as determine the underlying mechanisms. [ABSTRACT FROM AUTHOR]

Background We model the potential impact of relaxing current nicotine vaping product (NVP) restrictions on public health in Australia. Aims and Methods A Restricted NVP Scenario was first developed to project current smoking and vaping rates, where a U.S. smoking model was calibrated to recent Australian trends. To model less restrictive NVP policies, a Permissive NVP Scenario applied rates of switching from smoking to vaping, initiation into NVP and cigarette use, and cessation from smoking and vaping based on U.S. trends. The model measures vaping risk relative to the excess mortality rate of smoking. The public health impacts are measured as the difference between smoking- and vaping-attributable deaths (SVADs) and life years lost (LYLs) in the Restricted and Permissive NVP Scenarios. Sensitivity analysis is conducted regarding the NVP excess risk and other factors. Results Assuming an NVP excess risk of 5% that of smoking, 104.2 thousand SVADs (7.7% reduction) and 2.05 million LYLs (17.3% reduction) are averted during 2017–2080 in the Permissive NVP Scenario compared to the Restricted NVP Scenario. Assuming 40% NVP excess risk, 70 thousand SVADs and 1.2 million LYLs are averted. The impact is sensitive to the rate at which smokers switch to NVPs and quit smoking, and relatively insensitive to the smoking initiation and NVP initiation and cessation rates. Conclusions The model suggests the potential for public health gains to be achieved by relaxing NVP access regulations. However, the model would benefit from better information regarding the impact of NVPs on smoking under a relaxation of current restrictions. Implications Australia has implemented a strong array of cigarette-oriented policies, but has restricted access to NVPs. The Smoking and Vaping Model offers a framework for modeling hypothetical policy scenarios. The Australian model shows the potential for public health gains by maintaining cigarette-oriented policies while relaxing the current restrictive NVP policy. Modeling results under a permissive NVP policy are particularly sensitive to the estimated rates of smoking cessation and switching to vaping, which are not well established and will likely depend on past and future cigarette-oriented policies and the specific NVP policies implemented in Australia. [ABSTRACT FROM AUTHOR]

Introduction: The U.S. Food and Drug Administration (FDA) has proposed lowering the nicotine content of cigarettes to a minimally addictive level to increase smoking cessation and reduce initiation. This study has two aims: (1) to determine when cigarette manufacturers had the technical capability to reduce cigarette nicotine content and (2) to estimate the lost public health benefits of implementing a standard in 1965, 1975, or 1985.Methods: To determine the technical capability of cigarette companies, we reviewed public patents and internal cigarette company business records using the Truth Tobacco Industry Documents. To evaluate the impact of a very low nicotine content cigarette (VLNC) standard on smoking attributable deaths (SADs) and life-years lost (LYLs), we applied a validated (CISNET) model that uses past smoking data, along with estimates of the potential impact of VLNCs derived from expert elicitation.Results: Cigarette manufacturers recognized that cigarettes were deadly and addictive before 1964. Manufacturers have had the technical capability to lower cigarette nicotine content for decades. Our model projected that a standard implemented in 1965 could have averted 21 million SADs (54% reduction) and 272 million LYLs (64% reduction) from 1965 to 2064, a standard implemented in 1975 could have averted 18.9 million SADs and 245.4 million LYLs from 1975 to 2074, and a standard implemented in 1985 could have averted 16.3 million SADs and 211.5 million LYLs from 1985 to 2084.Conclusions: Millions of premature deaths could have been averted if companies had only sold VLNCs decades ago. FDA should act immediately to implement a VLNC standard.Implications: Prior research has shown that a mandated reduction in the nicotine content of cigarettes could reduce the prevalence of smoking and improve public health. Here we report that cigarette manufacturers have had the ability to voluntarily implement such a standard for decades. We use a well-validated model to demonstrate that millions of smoking attributable deaths and life-years lost would have been averted if the industry had implemented such a standard. [ABSTRACT FROM AUTHOR]