Machine Learning Analysis to Identify Factors Associated with Requesting Tobacco Cessation Services Among Users of an Online Self-Diagnostic Questionnaire in Mexico.

Machine Learning Analysis to Identify Factors Associated with Requesting Tobacco Cessation Services Among Users of an Online Self-Diagnostic Questionnaire in Mexico. | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Machine Learning Analysis to Identify Factors Associated with Requesting Tobacco Cessation Services Among Users of an Online Self-Diagnostic Questionnaire in Mexico. Norberto Francisco Hernández-Llanes, Ricardo Sánchez-Domínguez, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4014199/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Introduction : While traditional cessation services exist, internet-based options are gaining popularity. However, understanding online users' unique characteristics compared to traditional users is crucial. This study utilize machine learning (ML), aimed to identify these online users and their needs. Method : Through analyzing 14,182 records of adults who completed online nicotine dependence screening questionnaire, a random forest algorithm plus oversampling was used to predict request services. Results : The algorithm accurately identified 78.6% of users and rejected 68.8% of non-users. Notably, age, sex, dependence severity indicators, certain locations, and even specific occasions like World No Tobacco Day, were identified as key factors influencing service request. Discussion : These findings suggest the effectiveness of random forest algorithm in predicting potential users. Moreover, the predictor variables offer valuable insights for crafting targeted prevention and awareness campaigns, potentially leading to improved campaign effectiveness and ultimately, helping more individuals seeking cessation support. Health Economics & Outcomes Research Smoking Tobacco Use Cessation Machine Learning Random Forest Predictive model Oversampling Feature Selection Figures Figure 1 Figure 2 Figure 3 Introduction Tobacco use is a major preventable health risk factor, responsible for more than 7.7 million deaths per year and 200 million years of life lost worldwide (Reitsma et al. 2021 ). The global prevalence of tobacco use has declined in the last 20 years to approximately 22.3% in 2020 and is estimated to decline further to 20.4% in 2025 (WHO 2021 ). This decline has been proposed to be due to the coordinated and sustained implementation of cost-effective interventions envisioned in MPOWER strategies at the global level (Levy et al. 2018 ; Song et al. 2016 ), which were promoted under the Framework Convention on Tobacco Control (FCTC) of the World Health Organization (WHO FCTC 2003 ). Among these, the provision of tobacco cessation services is an effective intervention to maintain abstinence among people with tobacco dependence problems (Apollonio et al. 2016 ) or to reduce the amount of tobacco consumed (Lindson-Hawley et al. 2016 ). Cessation services encompass various treatments to help people stay tobacco-free and prevent relapse, which can range from brief counseling, behavioral interventions, and pharmacological treatment, which can be provided at the first, second, or third level of healthcare (García-Gómez et al. 2019 ). Although the beneficial health effects of smoking cessation occur at any age, early quitting significantly improves the health profile of a population by dramatically reducing the risk of death (Di Cicco et al. 2016 ). For example, quitting at the age of 50 increases life expectancy by six years and quitting before the age of 65 reduces the risk of having a heart attack by 18.2% (Reitsma et al. 2021 ). In addition, the use of cessation services reduces the costs of chronic disease at the individual, family, and societal levels, as well as the social and economic productivity (WHO 2019 ). As a result, there is a high demand for smoking cessation services among smokers. In the United States, 70% of the people who smoke want to quit and more than half of them try it every year (Babb et al. 2017 ), while in Mexico, according to the Global Adult Alcohol Use Survey (GATS), 71.3% of smokers have thought or planned to quit, and 53.6% of smokers have tried to quit in the last 12 months, although 90.2% have tried without professional support (SS et al. 2023). However, only 30% of the global population has access to smoking cessation services (WHO 2019 ), so offering online or internet-based cessation (I-BC) services has become an attractive option for healthcare providers. Approximately one third of treatment requests over the past decade have been through online services, with an increase of 43.7% during 2017 (Graham et al. 2019 ), and it has been reported that I-BC interventions may be effective in reducing smoking (G. M. J. Taylor et al. 2017 ). I-BC also has led to greater data availability, which requires more robust analytical tools such as machine learning (ML) (Yarkoni and Westfall 2017 ). ML uses algorithms to train a model that predicts the value of a variable of interest based on elements or features that act as inputs or predictors, and by considering all available information, provides greater accuracy and increases the generalization of results in healthcare (Fu et al. 2021 ; Graham et al. 2019 ). During the past decade, ML modeling have been increasingly used in tobacco research to predict smoking, nicotine dependence, or urge to use, among others (Dumortier et al. 2016 ; Fu et al. 2021 ; Thakur et al. 2022 ; Wetherill et al. 2019 ). Although there is plenty evidence on the factors associated with the demand for cessation services, including being female, early-onset tobacco use, multiple substance use, high tobacco use, not belonging to an Indigenous community, previous attempts to quit, or low perceived support in smoking quitting (Al Wahaibi et al. 2019 ; Allen et al. 2019 ; Borsari et al. 2018 ; Chin et al. 2012 )(Al Wahaibi et al., 2019 ; Allen et al., 2019 ; Borsari et al., 2018 ; Chin et al., 2012 ), the use of ML for identify predictors of request for cessation services is scarce. One study found that among adult smokers in the United States, increased use of electronic cigarettes in the past 30 days, decreased use of cigarettes in the past 30 days, smoking initiation after age 18, use of other substances, and higher body mass index were associated with increased use of cessation services using an ensemble model (random forest plus bagging) (Issabakhsh et al. 2023 )(. Another study reported that predictors of smoking cessation included age, severity of use (measured by duration of use, number of cigarettes per day, and levels of exhaled CO2), treatment use, and nicotine dependence using an artificial neural network model (Lai et al. 2021 ). This shows a promising outlook for the use of I-BC and ML for evaluation of service utilization. However, we found no evidence on the request of cessation services by users of online screening questionnaire, therefore this study aims to analyze predictors associated with the request for tobacco cessation among users of an online self-diagnosis questionnaire as first contact. Methods This is a cross-sectional, descriptive, and correlational study of records of users of an online tobacco self-assessment questionnaire in Mexico. Participants : General population who used the online tobacco self-assessment questionnaire from the Centers for Youth Integration (CIJ) (freely available at http://www.cij.gob.mx/Autodiagnostico/frmtabaco.asp ) were included. Inclusion criteria: Individuals over 18 years old. Records dated between February 8th, 2021, and June 16th, 2022. Exclusion criteria: Records with missing data. Data Collection : Upon entering the website, users answer questions about tobacco dependence symptoms and sociodemographic information, then they receive automatic feedback about their risk level of tobacco use. Users also receive several options of smoking cessation services, including voluntarily leaving contact information (such as email or phone number) or providing them with the address from the nearest treatment center. Instruments : Nicotine dependence: the Spanish version of the Fagerström Test of Nicotine Dependence (FTND) (Becoña and Vázquez 1998 ; Heatherton et al. 1991 ) was used. The psychometric properties of the questionnaire vary depending on the type of population in which it was validated, although its performance is lower in occasional smokers, users of electronic cigarettes, and smokers without smoke (Sharma et al. 2021 ). In the Mexican population, although it showed low reliability, it had a high correlation with physiological measures of tobacco use and with the reported number of cigarettes consumed (Moreno-Coutiño and Villalobos-Gallegos 2017 ), hence it is considered a good indicator of the tobacco use severity. Sociodemographic data were collected: age, gender (woman, man and non-binary), member of the LGBT + or Indigenous community and state of residence. In addition, all records were automatically identified by the system, as date and time. Ethical Procedures : All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2013 (World Medical Association 2013 ) . All users of the online tobacco self-assessment questionnaire consented to provide clinical and sociodemographic data, as the questionnaire does not collect any personally identifiable information (such as name, address, email, or IP address). For those users who voluntarily provide contact information (phone, email) for cessation services request, the system deidentifies the records (removes all fields containing personal information) before generating the databases for analysis. Outcome variable : The outcome variable is the request for tobacco cessation services, which for the present study was operationalized as all those users who left their contact details on the online questionnaire to receive cessation services in the CIJ attention network. Data Analysis: The total sample was randomly divided into a train and test set (split ratio 7:3). The variables were pre-processed (recoded, centered, and standardized) before performing the analyzes, obtaining a total of 119 features to perform the analyzes. Descriptive statistics were calculated for all measured characteristics. For predictive analysis, we used the Random Forest (RF) algorithm. RF is an ensemble algorithm that averages the results of multiple decision trees (Khodadadi et al. 2023 ). It was used a 10-fold Cross Validation for internal validation and hyperparameter tunning for improving the predictive models. Due to the unbalanced nature of the group of interest (Johnson and Khoshgoftaar 2019 ), the model was trained under four oversampling conditions: upsampling, downsampling, SMOTE and ROSE. The model with the highest predictive capacity in the test set defined by the area under the ROC curve (ROC AUC), was selected (Bradley 1997 ). Performance metrics, including Sensitivity, Specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV) were obtained to assess the best model (Graham et al. 2019 ; Issabakhsh et al. 2023 ) and identify the variables that contribute the most to the model. The study followed the recommendations of the MINIMAR standard for studies using artificial intelligence in medicine (Hernandez-Boussard et al. 2020 ). Statistical analyzes were performed in R version 4.3 programming language (R Core Team, 2023 ) using the Caret library version 6.0–94 (Kuhn 2008 ) and the R Studio IDE version 2022.12.0 (Posit Team 2023 ). Results Figure 1 displays the records selected for analysis. Among all tobacco questionnaire users, 4,173 (22.7%) were excluded from the analysis because they were underage, and three records (0.01%) were excluded for incomplete information. As a result, a total of 14,182 records are available for analysis. ******************************** Insert Fig. 1 here ********************************* Table 1 shows the sociodemographic characteristics of the questionnaire users. Almost half of the users were male. Approximately 9% of the users identified as LGBT+, 3% as indigenous, and 24% request for cessation services. The average age of the users was 30 years (29.9 ± 11.8). ******************************** Insert Table I here ******************************** Table II shows the indicators of tobacco dependence, as defined by the FTND. The electronic version of the FTND had a reliability of 0.689, measured by Cronbach's alpha. Almost 80% of users smoke less than one pack per day, however, 3% smokes more than 31 cigarettes per day. Additionally, 23.2% smoke their first cigarette in the morning, approximately 30% smoke within the first hour of waking up, and 20% have difficulty quitting smoking in prohibited areas and 22% continue to smoke despite being sick. ******************************** Insert Table II here ********************************* Information on the pattern of use can be found in Supplementary Table I. The data show that most visits to the questionnaire occurred in May (77.5%). Furthermore, more than half of the users of the self-diagnosis tool (51.8%) were located in the states of Aguascalientes (4.8%), Mexico City (17.3%), State of Mexico (12.6%), Jalisco (4%), Nuevo Leon (4.8%) and Sinaloa (8.3%). Furthermore, half of the users (52.4%) accessed the questionnaire between 13:00 and 20:00 hrs. Most visits (88.2%) occurred during weekdays. No significant differences were found between the training and validation samples regarding sociodemographic and clinical variables. However, there were significant differences in the proportion of users of the state of Tlaxcala and Quintana Roo (see Supplementary Tables II to IV). Figure 2 displays the results of the predictive modeling. The area under the ROC curve ranged from 0.506 for ROSE resampling to 0.737 for downsampling. Sensitivity ranged from 0.411 to 1.000, Specificity ranged from 0.012 to 0.935, PPV from 0.234 to 0.682, and NPV from 0.840 to 1.000. ******************************** Insert Fig. 2 here ********************************* The downsampling technique was used to obtain the best predictive model (sensitivity = 0.786, specificity = 0.688, PPV = 0.445, NPV = 0.910, ROC AUC = 0.737), so it was used to evaluate the importance of predictors. Figure 3 shows the relative importance of the first 20 main characteristics, age and the month of May were the variables with the greatest relative importance for the predictive model, both having a relative importance greater than 80%. Other variables that contribute to the model's predictive capacity include those related to clinical characteristics of nicotine dependence, like tobacco consumption after wake up in the morning; sociodemographic variables such as gender, and variables related to spatial and temporal usage patterns (such as being from the states of Sinaloa, Chiapas, Mexico City, or the State of Mexico, and using the platform on the 30th and 31st, Mondays and Thursdays, and in March). These variables, although of lesser importance, still play a role in the model's accuracy. ******************************** Insert Fig. 3 here ********************************* Discussion This study aimed to analyze predictors associated with the request for smoking cessation services among users of an online tobacco self-assessment questionnaire as a first point of contact. The use of the Random Forest algorithm with downsampling technique proved to be effective in predicting the use of tobacco cessation services (ROC AUC = 0.737), correctly selecting 78.6% of users and correctly rejecting 68.8% of non-users. Random Forest is one of the most widely used ensemble algorithms (Fernández-Delgado et al., 2014 ) because it reduces the generalization error (error generated when predicting over new samples), is robust in collinearity situations, and reduces overfitting (Fang et al. 2019 ; Liu et al. 2012 ; Loupe, 2015). Another advantage of the Random Forest algorithm for estimating the importance of predictors is that they are weighted by individual random trees, so it behaves as an "average" of the individual importance of each tree (Loupe, 2015). In terms of the major predictors, our study found that age was the first significant predictor of seeking cessation services. Our findings are consistent with the literature, which suggests that older individuals, who are more likely to develop tobacco dependence, are more likely to seek formal treatment from a specialized professional compared to younger individuals (Hughes 2013 ). Research also suggests that young people may have a harder time quitting smoking due to their preference for quitting without assistance, such as nicotine replacement therapies, and instead opt for methods like "cold turkey" quitting or reducing cigarette smoking in certain situations (Dugas et al. 2016 ; Solberg et al. 2007 ). Furthermore, young adults who occasionally smoke cigarettes often label themselves as social smokers rather than smokers (Choi et al. 2010 ; Kingsbury et al. 2016 ) rather than smokers (Choi et al., 2010 ), and may perceive smoking as harmless, deny addiction, and lack motivation to quit (Levinson et al. 2007 ). We recommend including other measures in future work that assess not only physical dependence, but also consumption pattern and other characteristics related to tobacco addiction. This suggestion has been made by other authors (Paz-Ballesteros et al. 2019 ), as it is known that occasional smokers may not be detected by the FTND (Sharma et al. 2021 ). The second most important predictor was the month of May, and related to it, days 30 (5th importance level) and 31 (17th importance level) as predictors of service seeking. These dates are related to the World No Tobacco Day campaign, which will be commemorated in Mexico and worldwide on May 31, 2023 (WHO 2023 ). Evidence has shown that campaigns that promote education and health services effectively increase the use of cessation services. For example, in the United States, such campaigns generated more than 1.3 million additional calls to helplines (US Public Health Service Office of the Surgeon General and National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health 2020). Similar results were found in Korea, where the demand for treatment increased 5.5 and 3.1 times during the months in which campaigns were promoted in 2010 and 2011, respectively, compared to the average demand during other months (Park et al. 2019 ). Other predictor like gender female (15th importance level) is also a well-documented predictor in the literature, and our study is consistent with evidence that being female predicts seeking cessation services (Abdullah 2004 ; Jayakumar et al. 2020 ; Jeong et al. 2019 ; Williams et al. 2022 ). Among the clinical predictors of the FTND, item 3 "How soon after you wake up do you smoke your first cigarette?" (5th importance level), item 1 "How many cigarettes per day do you smoke?" (6th importance level) and item 4 "Which cigarette would you hate most to give up?" (8th importance level) were found to be the most important predictors of request for cessation services. These indicators have been consistently identified in the scientific literature as being associated with excessive tobacco use and nicotine dependence. Items 1 and 3 composes up the Heaviness of Smoking Index, an indicator of severity of smoking (Heatherton, et al., 1989 ), which is a predictor of quitting (Borland, et al., 2010 , Chaiton, et al., 2007 ). An issue that requires further attention in future research is the spatial pattern of self-assesstment questionnaire use, which may be closely related to the social determinants of tobacco use (Brady 2020 ). In Mexico, the most recent national household tobacco survey (Reynales Shigematsu et al. 2017 ) reported subnational differences in patterns of tobacco consumption and demand for cessation services; for example, the prevalence of tobacco consumption was higher than the national average in states such as Aguascalientes, Mexico City, and Mexico State, and our study found a higher proportion of users who expressed a desire to quit smoking in Mexico City and Mexico State, which is consistent with the aforementioned findings. 4.1 Limitations of the study : One of the main limitations of the study is that the online questionnaire did not measure many variables that the literature has related consistently to tobacco use and that could explain both the severity of use and the request of cessation services, such as socioeconomic status, schooling, type of job, or tobacco availability (Brady 2020 ; Pulakka et al. 2016 ). Other variables such as internet access and health literacy could also impact IB-C services request. Research has shown a negative correlation between Internet use and smoking severity (McCrabb et al. 2019 ), while health literacy has been found to have a positive impact on smoking cessation intentions and behaviors (Sun et al. 2023 ). The co-occurrence of psychiatric symptomatology and smoking is also related with the severity of tobacco use and the need for treatment (Szerman et al. 2022 ). For example, depressive symptoms are strongly related to frequency, severity, and level of tobacco dependence (Bakhshaie et al. 2015 ; Leyro et al. 2016 ). As a result, people experiencing these symptoms could require more care due to the cumulative effect of physical and psychological distress (Jamal et al. 2012 ). Future studies should also consider measurement of social determinants of health such as income inequality, barriers, and facilitators to access to services that could impact predictors and outcomes of both tobacco use and care seeking (Alegría et al. 2023 ; Brady 2020 ). Conclusions I-BC has become an attractive option for providing health care services to population, as it allows to reach a larger number of people at a lower cost. However, it is necessary to understand the characteristics and needs of online tool users, as they differ from traditional users. Besides, the use of web-based screening tools has increased access to mental health and substance use treatment, but it also poses new challenges for analyzing the large amounts of data generated, so the use of machine learning/artificial intelligence algorithms will become increasingly widespread. In this paper, the Random Forest algorithm with downsampling was shown to be a highly effective predictive modeling tool. It was shown to accurately identify potential users (AUC ROC = 0.737), correctly identifying 78.6% of those who requested cessation services. By using ML to identify predictors associated with request of tobacco cessation services, such as age, massive campaigns like World No Tobacco Day (May 31th), gender, or social determinants like location, it is possible to develop and implement targeted prevention campaigns and strategies that allow us to reach a larger clinical population. The results presented here could be used to guide the use of these tools in the planning and delivery of mental health services that are sensitive and adapted to the needs of the population seeking care. References Abdullah, A. S. M. (2004). Which smokers use the smoking cessation Quitline in Hong Kong, and how effective is the Quitline? Tobacco Control , 13 (4), 415–421. https://doi.org/10.1136/tc.2003.006460 Al Wahaibi, N., Al Lawati, A., Al Ruqeishy, F., Al Khatri, A., Al-Farsi, Y., Juma, T. M. A., et al. (2019). The characteristics and patterns of utilization of healthcare services among Omanis with substance use disorders attending therapy for cessation. PLOS ONE , 14 (1), e0210532. https://doi.org/10.1371/journal.pone.0210532 Alegría, M., Alvarez, K., Cheng, M., & Falgas-Bague, I. (2023). Recent Advances on Social Determinants of Mental Health: Looking Fast Forward. American Journal of Psychiatry , 180 (7), 473–482. https://doi.org/10.1176/appi.ajp.20230371 Allen, A. M., Yuan, N. P., Wertheim, B. C., Krupski, L., Bell, M. L., & Nair, U. (2019). Gender differences in utilization of services and tobacco cessation outcomes at a state quitline. Translational Behavioral Medicine , 9 (4), 663–668. https://doi.org/10.1093/tbm/iby083 Apollonio, D., Philipps, R., & Bero, L. (2016). Interventions for tobacco use cessation in people in treatment for or recovery from substance use disorders. Cochrane Database of Systematic Reviews , 2017 (1). https://doi.org/10.1002/14651858.CD010274.pub2 Babb, S., Malarcher, A., Schauer, G., Asman, K., & Jamal, A. (2017). Quitting Smoking Among Adults United States, 2000–2015. MMWR. Morbidity and Mortality Weekly Report , 65 (52), 1457–1464. https://doi.org/10.15585/mmwr.mm6552a1 Bakhshaie, J., Zvolensky, M. J., & Goodwin, R. D. (2015). Cigarette smoking and the onset and persistence of depression among adults in the United States: 1994–2005. Comprehensive Psychiatry , 60 , 142–148. https://doi.org/10.1016/j.comppsych.2014.10.012 Becoña, E., & Vázquez, F. L. (1998). The Fagerström Test for Nicotine Dependence in a Spanish Sample. Psychological Reports , 83 (3_suppl), 1455–1458. https://doi.org/10.2466/pr0.1998.83.3f.1455 Borland, R., Yong, H.-H., O’Connor, R. J., Hyland, A., & Thompson, M. E. (2010). The reliability and predictive validity of the Heaviness of Smoking Index and its two components: Findings from the International Tobacco Control Four Country study. Nicotine & Tobacco Research, 12(Supplement 1), S45–S50. https://doi.org/10.1093/ntr/ntq038 Borsari, L., Storani, S., Malagoli, C., Filippini, T., Tamelli, M., Malavolti, M., et al. (2018). Impact of Referral Sources and Waiting Times on the Failure to Quit Smoking: One-Year Follow-Up of an Italian Cohort Admitted to a Smoking Cessation Service. International Journal of Environmental Research and Public Health , 15 (6), 1234. https://doi.org/10.3390/ijerph15061234 Bradley, A. P. (1997). The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognition , 30 (7), 1145–1159. https://doi.org/10.1016/S0031-3203(96)00142-2 Brady, K. T. (2020). Social Determinants of Health and Smoking Cessation: A Challenge. American Journal of Psychiatry , 177 (11), 1029–1030. https://doi.org/10.1176/appi.ajp.2020.20091374 Chaiton, M. O., Cohen, J. E., McDonald, P. W., & Bondy, S. J. (2007). The Heaviness of Smoking Index as a predictor of smoking cessation in Canada. Addictive Behaviors, 32(5), 1031–1042. https://doi.org/10.1016/j.addbeh.2006.07.008 Chin, D. L., Hong, O., Gillen, M., Bates, M. N., & Okechukwu, C. A. (2012). Occupational Factors and Smoking Cessation among Unionized Building Trades Workers. Workplace Health & Safety , 60 (10), 445–452. https://doi.org/10.1177/216507991206001006 Choi, Y., Choi, S. M., & Rifon, N. (2010). “I Smoke but I Am Not a Smoker”: Phantom Smokers and the Discrepancy Between Self-identity and Behavior. Journal of American College Health , 59 (2), 117–125. https://doi.org/10.1080/07448481.2010.483704 Di Cicco, M. E., Ragazzo, V., & Jacinto, T. (2016). Mortality in relation to smoking: the British Doctors Study. Breathe , 12 (3), 275–276. https://doi.org/10.1183/20734735.013416 Dong, J., Feng, T., Thapa-Chhetry, B., Cho, B. G., Shum, T., Inwald, D. P., et al. (2021). Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care. Critical Care , 25 (1), 288. https://doi.org/10.1186/s13054-021-03724-0 Dugas, E. N., Wellman, R. J., Kermack, A., Tremblay, M., & OʼLoughlin, J. (2016). Reasons Young Smokers Do Not Use NRT Even When it is Available Free-of-Charge: An Exploratory Study: The Canadian Journal of Addiction , 7 (2), 14–21. https://doi.org/10.1097/02024458-201606000-00003 Dumortier, A., Beckjord, E., Shiffman, S., & Sejdić, E. (2016). Classifying smoking urges via machine learning. Computer Methods and Programs in Biomedicine , 137 , 203–213. https://doi.org/10.1016/j.cmpb.2016.09.016 Fang, G., Annis, I. E., Elston-Lafata, J., & Cykert, S. (2019). Applying machine learning to predict real-world individual treatment effects: insights from a virtual patient cohort. Journal of the American Medical Informatics Association , 26 (10), 977–988. https://doi.org/10.1093/jamia/ocz036 Fernández-Delgado, M., Cernadas, E., Barro, S., & Amorin, D. (2014). Do we Need Hundreds of Classifiers to Solve Real World Classification Problems? Journal of Machine Learning Research, 15, 3133–3181. Fu, R., Kundu, A., Mitsakakis, N., Elton-Marshall, T., Wang, W., Hill, S., et al. (2021). Machine learning applications in tobacco research: a scoping review. Tobacco Control , tobaccocontrol-2020-056438. https://doi.org/10.1136/tobaccocontrol-2020-056438 García-Gómez, L., Hernández-Pérez, A., Noé-Díaz, V., Riesco-Miranda, J. A., & Jiménez-Ruiz, C. (2019). Smoking Cessation Treatments: Current Psychological and Pharmacological Options. Revista de investigación Clínica , 71 (1), 1410. https://doi.org/10.24875/RIC.18002629 Graham, S., Depp, C., Lee, E. E., Nebeker, C., Tu, X., Kim, H.-C., & Jeste, D. V. (2019). Artificial Intelligence for Mental Health and Mental Illnesses: an Overview. Current Psychiatry Reports , 21 (11), 116. https://doi.org/10.1007/s11920-019-1094-0 Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., & Fagerstrom, K.-O. (1991). The Fagerstrom Test for Nicotine Dependence: a revision of the Fagerstrom Tolerance Questionnaire. Addiction , 86 (9), 1119–1127. https://doi.org/10.1111/j.1360-0443.1991.tb01879.x Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., Rickert, W., & Robinson, J. (1989). Measuring the heaviness of smoking: using self-reported time to the first cigarette of the day and number of cigarettes smoked per day. British Journal of Addiction, 84(7), 791–799. https://doi.org/10.1111/j.1360-0443.1989.tb03059.x Hernandez-Boussard, T., Bozkurt, S., Ioannidis, J. P. A., & Shah, N. H. (2020). MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care. Journal of the American Medical Informatics Association , 27 (12), 2011–2015. https://doi.org/10.1093/jamia/ocaa088 Hughes, J. R. (2013). An updated algorithm for choosing among smoking cessation treatments. Journal of Substance Abuse Treatment , 45 (2), 215–221. https://doi.org/10.1016/j.jsat.2013.01.011 Issabakhsh, M., Sánchez-Romero, L. M., Le, T. T. T., Liber, A. C., Tan, J., Li, Y., et al. (2023). Machine learning application for predicting smoking cessation among US adults: An analysis of waves 1-3 of the PATH study. PLOS ONE , 18 (6), e0286883. https://doi.org/10.1371/journal.pone.0286883 Jamal, M., Willem Van der Does, A. J., Cuijpers, P., & Penninx, B. W. J. H. (2012). Association of smoking and nicotine dependence with severity and course of symptoms in patients with depressive or anxiety disorder. Drug and Alcohol Dependence , 126 (1–2), 138–146. https://doi.org/10.1016/j.drugalcdep.2012.05.001 Jayakumar, N., Chaiton, M., Zhang, B., Selby, P., & Schwartz, R. (2020). Sex Differences in Use of Smoking Cessation Services and Resources: A Real-World Study. Tobacco Use Insights , 13 , 1179173X2090150. https://doi.org/10.1177/1179173X20901500 Jeong, B. Y., Lim, M. K., Yun, E. H., & Oh, J.-K. (2019). User characteristics of national smoking cessation services in Korea: who chooses each type of tobacco cessation program? BMC Health Services Research , 19 (1), 14. https://doi.org/10.1186/s12913-018-3817-z Johnson, J. M., & Khoshgoftaar, T. M. (2019). Survey on deep learning with class imbalance. Journal of Big Data , 6 (1), 27. https://doi.org/10.1186/s40537-019-0192-5 Kalafi, E. Y., Nor, N. a. M., Taib, N. A., Ganggayah, M. D., Town, C., & Dhillon, S. K. (2019). Machine Learning and Deep Learning Approaches in Breast Cancer Survival Prediction Using Clinical Data. Folia Biologica , 65 (5–6), 212–220. Khodadadi, A., Ghanbari Bousejin, N., Molaei, S., Kumar Chauhan, V., Zhu, T., & Clifton, D. A. (2023). Improving Diagnostics with Deep Forest Applied to Electronic Health Records. Sensors (Basel, Switzerland) , 23 (14), 6571. https://doi.org/10.3390/s23146571 Kingsbury, J. H., Parks, M. J., Amato, M. S., & Boyle, R. G. (2016). Deniers and Admitters: Examining Smoker Identities in a Changing Tobacco Landscape. Nicotine & Tobacco Research , 18 (11), 2130–2137. https://doi.org/10.1093/ntr/ntw110 Kuhn, M. (2008). Building Predictive Models in R Using the caret Package. Journal of Statistical Software , 28 (5), 1–26. https://doi.org/10.18637/jss.v028.i05 Lai, C.-C., Huang, W.-H., Chang, B. C.-C., & Hwang, L.-C. (2021). Development of Machine Learning Models for Prediction of Smoking Cessation Outcome. International Journal of Environmental Research and Public Health , 18 (5), 2584. https://doi.org/10.3390/ijerph18052584 Levinson, A. H., Campo, S., Gascoigne, J., Jolly, O., Zakharyan, A., & Tran, Z. V. (2007). Smoking, but not smokers: Identity among college students who smoke cigarettes. Nicotine & Tobacco Research , 9 (8), 845–852. https://doi.org/10.1080/14622200701484987 Levy, D. T., Yuan, Z., Luo, Y., & Mays, D. (2018). Seven years of progress in tobacco control: an evaluation of the effect of nations meeting the highest level MPOWER measures between 2007 and 2014. Tobacco Control , 27 (1), 50–57. https://doi.org/10.1136/tobaccocontrol-2016-053381 Leyro, T. M., Crew, E. E., Bryson, S. W., Lembke, A., Bailey, S. R., Prochaska, J. J., et al. (2016). Retrospective analysis of changing characteristics of treatment-seeking smokers: implications for further reducing smoking prevalence. BMJ Open , 6 (6), e010960. https://doi.org/10.1136/bmjopen-2015-010960 Lindson-Hawley, N., Hartmann-Boyce, J., Fanshawe, T. R., Begh, R., Farley, A., & Lancaster, T. (2016). Interventions to reduce harm from continued tobacco use. Cochrane Database of Systematic Reviews , 2016 (12). https://doi.org/10.1002/14651858.CD005231.pub3 Liu, Y., Wang, Y., & Zhang, J. (2012). New Machine Learning Algorithm: Random Forest. In B. Liu, M. Ma, & J. Chang (Eds.), Information Computing and Applications (Vol. 7473, pp. 246–252). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-34062-8_3 Louppe, G. (2015, June 3). Understanding Random Forests: From Theory to Practice. arXiv. http://arxiv.org/abs/1407.7502. Accessed 27 February 2024 McCrabb, S., Twyman, L., Palazzi, K., Guillaumier, A., Paul, C., & Bonevski, B. (2019). A cross sectional survey of internet use among a highly socially disadvantaged population of tobacco smokers. Addiction Science & Clinical Practice , 14 (1), 38. https://doi.org/10.1186/s13722-019-0168-y Moreno-Coutiño, A., & Villalobos-Gallegos, L. (2017). Psychometric Properties of the Fagerström Test for Nicotine Dependence in a Sample of Mexican Smokers. Journal of Addictions Nursing , 28 (1), 27–33. https://doi.org/10.1097/JAN.0000000000000156 Park, J., Luu Ngoc, Shin, S., Oh, J.-K., Yun, E. H., Lee, D., & Lim, M. K. (2019). Influence of new tobacco control policies and campaigns onQuitline call volume in Korea. Tobacco Induced Diseases , 17 (March). https://doi.org/10.18332/tid/104674 Paz-Ballesteros, W. C., Zavala-Arciniega, L., Gutiérrez-Torres, D. S., Ponciano-Rodríguez, G., & Reynales-Shigematsu, L. M. (2019). Evaluación de la dependencia física y psicológica al tabaco en fumadores mexicanos adultos, Encodat 2016. Salud Pública de México , 61 (2, Mar-Abr), 136. https://doi.org/10.21149/9858 Posit team. (2023). RStudio: Integrated Development Environment for R. Boston, MA: Posit Software, PBC. http://www.posit.co/ Pulakka, A., Halonen, J. I., Kawachi, I., Pentti, J., Stenholm, S., Jokela, M., et al. (2016). Association Between Distance From Home to Tobacco Outlet and Smoking Cessation and Relapse. JAMA Internal Medicine , 176 (10), 1512. https://doi.org/10.1001/jamainternmed.2016.4535 R Core Team. (2023). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/ Radovic, A., & Badawy, S. M. (2020). Technology Use for Adolescent Health and Wellness. Pediatrics , 145 (Supplement_2), S186–S194. https://doi.org/10.1542/peds.2019-2056G Reitsma, M. B., Kendrick, P. J., Ababneh, E., Abbafati, C., Abbasi-Kangevari, M., Abdoli, A., et al. (2021). Spatial, temporal, and demographic patterns in prevalence of smoking tobacco use and attributable disease burden in 204 countries and territories, 1990–2019: a systematic analysis from the Global Burden of Disease Study 2019. The Lancet , 397 (10292), 2337–2360. https://doi.org/10.1016/S0140-6736(21)01169-7 Reynales Shigematsu, L. M., Zavala-Arciniega, L., Paz-Ballesteros, W. C., Gutiérrez-Torres, D. S., García-Buendía, J. C., Rodríguez-Andrade, M. A., et al. (2017). Encuesta Nacional de Consumo de Drogas, Alcohol y Tabaco 2016-2017: Reporte de Tabaco. Ciudad de México, México: INPRFM, INSP, CONADIC, SS. https://drive.google.com/file/d/1Iktptvdu2nsrSpMBMT4FdqBIk8gikz7q/view Sharma, M., Suman, L., Srivastava, K., Suma, N., & Vishwakarma, A. (2021). Psychometric properties of Fagerstrom Test of Nicotine Dependence: A systematic review. Industrial Psychiatry Journal , 30 (2), 207. https://doi.org/10.4103/ipj.ipj_51_21 Solberg, L. I., Boyle, R. G., McCarty, M., Asche, S. E., & Thoele, M. J. (2007). Young adult smokers: are they different? The American Journal of Managed Care , 13 (11), 626–632. Song, Y., Zhao, L., Palipudi, K. M., Asma, S., Morton, J., Talley, B., et al. (2016). Tracking MPOWER in 14 countries: results from the Global Adult Tobacco Survey, 2008–2010. Global Health Promotion , 23 (2_suppl), 24–37. https://doi.org/10.1177/1757975913501911 SS, CONASAMA, & INSP. (2023). . Encuesta Global de Tabaquismo en Adultos (GATS): Hoja de Resumen. (p. 2). México: INSP. https://www.insp.mx/avisos/presentacion-de-resultados-de-la-gats-2023 Sun, S., Yu, H., Ling, J., Yao, D., Chen, H., & Liu, G. (2023). The influence of health literacy and knowledge about smoking hazards on the intention to quit smoking and its intensity: an empirical study based on the data of China’s health literacy investigation. BMC Public Health , 23 (1), 2355. https://doi.org/10.1186/s12889-023-17292-1 Szerman, N., Parro, C., Pinet, C., Martínez-Raga, J., Basurte, I., & Saiz, P. (2022). TOBACCO USE DISORDER AND DUAL DISORDERS Joint statement by the Spanish Psychiatry Society and the Spanish Dual Disorders Society. Actas Espanolas De Psiquiatria , 50 (Supplement), 77–138. Taylor, D. S., Medaglio, D., Jurkovitz, C. T., Patterson, F., Zhang, Z., Gbadebo, A., et al. (2020). Evaluation of a Systems-Based Tobacco Cessation Program Using Bedside Volunteers. Nicotine & Tobacco Research , 22 (3), 440–445. https://doi.org/10.1093/ntr/nty252 Taylor, G. M. J., Dalili, M. N., Semwal, M., Civljak, M., Sheikh, A., & Car, J. (2017). Internet-based interventions for smoking cessation. Cochrane Database of Systematic Reviews , 2017 (9). https://doi.org/10.1002/14651858.CD007078.pub5 Thakur, S. S., Poddar, P., & Roy, R. B. (2022). Real-time prediction of smoking activity using machine learning based multi-class classification model. Multimedia Tools and Applications , 81 (10), 14529–14551. https://doi.org/10.1007/s11042-022-12349-6 US Public Health Service Office of the Surgeon General, & National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health. (2020). Chapter 6, Interventions for Smoking Cessation and Treatments for Nicotine Dependence. In Smoking Cessation: A Report of the Surgeon General [Internet] . Washington (DC): US Department of Health and Human Services. https://www.ncbi.nlm.nih.gov/books/NBK555596/ Wetherill, R. R., Rao, H., Hager, N., Wang, J., Franklin, T. R., & Fan, Y. (2019). Classifying and characterizing nicotine use disorder with high accuracy using machine learning and resting‐state fMRI. Addiction Biology , 24 (4), 811–821. https://doi.org/10.1111/adb.12644 WHO. (2019). WHO Report on the Global Tobacco Epidemic, 2019 (p. 109). Geneva: World Health Organization. https://apps.who.int/iris/rest/bitstreams/1239531/retrieve WHO. (2021). WHO global report on trends in prevalence of tobacco use 2000-2025 (No. 4ta. edición.) (p. 150). Geneva. https://apps.who.int/iris/rest/bitstreams/1263754/retrieve WHO. (2023). World No Tobacco Day. https://www.who.int/campaigns/world-no-tobacco-day. Accessed 9 January 2024 WHO FCTC. (2003). WHO Framework Convention on Tobacco Control . Geneva, Switzerland: WHO Press. https://iris.who.int/bitstream/handle/10665/42811/9241591013.pdf Williams, R. M., Eyestone, E., Smith, L., Philips, J. G., Whealan, J., Webster, M., et al. (2022). Engaging Patients in Smoking Cessation Treatment within the Lung Cancer Screening Setting: Lessons Learned from an NCI SCALE Trial. Current Oncology , 29 (4), 2211–2224. https://doi.org/10.3390/curroncol29040180 World Medical Association. (2013). World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. JAMA , 310 (20), 2191. https://doi.org/10.1001/jama.2013.281053 Wu, Y., & Fang, Y. (2020). Stroke Prediction with Machine Learning Methods among Older Chinese. International Journal of Environmental Research and Public Health , 17 (6), 1828. https://doi.org/10.3390/ijerph17061828 Yarkoni, T., & Westfall, J. (2017). Choosing Prediction Over Explanation in Psychology: Lessons From Machine Learning. Perspectives on Psychological Science , 12 (6), 1100–1122. https://doi.org/10.1177/1745691617693393 Tables Table I. Sociodemographic Characteristics of Online Tobacco Self-Assessment Questionnaire Users, 2021-2022. Online Tobacco Self-Assessment Questionnaire n % Gender Man 7,534 53.1 Woman 6,153 43.4 Non-binary 495 3.5 LGBT+ Yes 1,331 9.4 No 12,851 90.6 Indigenous Yes 473 3.3 No 13,709 96.7 Request for services Yes 3,423 24.1 ‍ No 10,759 75.9 Age (Mean ± SD) 29.9 ± 11.8 Note: % = percentage within column. Table II. Indicators of Nicotine Dependence in Online Tobacco Self-Assessment Questionnaire Users, 2021-2022. Online Tobacco Self-Assessment Questionnaire n % 1. How many cigarettes per day do you smoke? 10 or less 11,271 79.5 11 to 20 1,800 12.7 21 to 30 669 4.7 31 or more 442 3.1 2. Do you smoke more frequently during the first hours after waking than during the rest of the day? Yes 3,292 23.2 No 10,890 76.8 3. How soon after you wake up do you smoke your first cigarette? Within 5 minutes 1,418 10.0 Within 6 - 30 minutes 1,498 10.6 Within 31 - 60 minutes 1,479 10.4 After 60 minutes 9,787 69.0 4. Which cigarette would you hate most to give up? The first one in the morning 4,796 30.9 All others 9,796 69.1 5. Do you find it difficult to refrain from smoking in places where it is forbidden? Yes 2,658 18.7 No 11,524 81.3 6. Do you smoke if you are so ill that you are in bed most of the day? Yes 3,190 22.5 ‍ No 10,992 77.5 Cronbach’s α 0.689 Note: % = percentage within column. Additional Declarations The authors declare no competing interests. Supplementary Files SupplementaryTableIMLTobacco12.odt Spatial and Time Use Pattern of Online Tobacco Self-Assessment Questionnaire, 2021-2022. SupplementaryTableIIMLTobacco12.odt Comparison of Sociodemographic Characteristics of Online Tobacco Self-Assessment Questionnaire Users in Train versus Test Set, 2021-2022. SupplementaryTableIIIMLTobacco12.odt Comparison of Indicators of Nicotine Dependence in Online Tobacco Self-Assessment Questionnaire Users in Train versus Test Set, 2021-2022. SupplementaryTableVIMLTobacco12.odt Comparison of Spatial and Time Use Pattern of Online Tobacco Self-Assessment Questionnaire in Train versus Test Set, 2021-2022. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4014199","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":276341934,"identity":"0f534b16-f97d-4e6b-bb7e-cc754e1f6ae3","order_by":0,"name":"Norberto Francisco Hernández-Llanes","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEElEQVRIiWNgGAWjYDACCQjF2MDAfADMYmNIgAkS1MKWQLIWHgOoEAEt5tLNxz4X1NyR7ec/8/HDx5zDeXzsyQ9vMLbdkddt4E7DptdyzrHk2TOOPTOe2XB2s+TMbYeL2XieGVswtj0z3HaAd7MBFi0GN3KMmXnYDiduONi7jZl32+HENokEMwnGtsOMQC0bH2DVkv+Zmeff4cT9h3meQbWkfwNpsQdq2XAAuy3MzLxtQFvYeNigWnLAtiTitiXNmJm377DxjDNsxkC/pCe28bwptkg4dzh522Fcfkl+zMzz7bBsf//hhx8+brNOnN+evvHGh7LDttuO927DH0EoIAFEMBOvfhSMglEwCkYBKgAAG3Js+iA1j9sAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-5918-3137","institution":"Programa de Maestría y Doctorado en Ciencias Médicas, Odontológicas y de la Salud. UNAM.","correspondingAuthor":true,"prefix":"","firstName":"Norberto","middleName":"Francisco","lastName":"Hernández-Llanes","suffix":""},{"id":276342320,"identity":"c44fe163-566e-483f-b4e6-067125475eb8","order_by":1,"name":"Ricardo Sánchez-Domínguez","email":"","orcid":"https://orcid.org/0000-0001-6221-6091","institution":"Centros de Integración Juvenil AC","correspondingAuthor":false,"prefix":"","firstName":"Ricardo","middleName":"","lastName":"Sánchez-Domínguez","suffix":""},{"id":276342321,"identity":"a1eedd32-86eb-4db1-863e-18fdae5226f8","order_by":2,"name":"Sofía Alvarez-Reza","email":"","orcid":"https://orcid.org/0000-0001-7380-5297","institution":"Centros de Integración Juvenil AC","correspondingAuthor":false,"prefix":"","firstName":"Sofía","middleName":"","lastName":"Alvarez-Reza","suffix":""},{"id":276342322,"identity":"3637036c-6c29-48be-b72d-9ea1a2f85dae","order_by":3,"name":"Carmen Fernández-Cáceres","email":"","orcid":"https://orcid.org/0009-0000-1851-194X","institution":"Centros de Integración Juvenil AC","correspondingAuthor":false,"prefix":"","firstName":"Carmen","middleName":"","lastName":"Fernández-Cáceres","suffix":""},{"id":276342323,"identity":"87e13b3e-35f0-443a-87ac-3f0401aa96de","order_by":4,"name":"Rodrigo Marín-Navarrete","email":"","orcid":"https://orcid.org/0000-0002-6084-8199","institution":"Centros de Integración Juvenil AC","correspondingAuthor":false,"prefix":"","firstName":"Rodrigo","middleName":"","lastName":"Marín-Navarrete","suffix":""}],"badges":[],"createdAt":"2024-03-04 19:38:05","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4014199/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4014199/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":52105579,"identity":"690e042c-06f5-4e86-aca5-44fa0900dabf","added_by":"auto","created_at":"2024-03-06 19:28:58","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":44637,"visible":true,"origin":"","legend":"\u003cp\u003eParticipant´s Flowchart.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/7622ee04d8b08fe38d45e99e.png"},{"id":52106203,"identity":"8805c802-90bf-422b-a4a2-b15d69ab35f1","added_by":"auto","created_at":"2024-03-06 19:36:58","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":33034,"visible":true,"origin":"","legend":"\u003cp\u003eResampling Method effect on the Predictive Ability of the Random Forest Algorithm.\u003c/p\u003e\n\u003cp\u003eNote for Figure 2: Preproc = Preprocessed data (recoded, centered, and standardized); Down = Downsampling; Hyperp =Hyperparameter Tunning; Up = Upsampling; SMOTE = Synthetic Minority Over-sampling; ROSE = Random Oversampling; AUC = ROC Area Under the Curve.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/9d16aef8263986da7df62b60.png"},{"id":52107165,"identity":"d6f10d70-ab07-4af6-9270-456a33f27202","added_by":"auto","created_at":"2024-03-06 19:52:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":34343,"visible":true,"origin":"","legend":"\u003cp\u003eRelative importance of predictors of cessation service use among users of the Tobacco Self-Assessment Questionnaire.\u003c/p\u003e\n\u003cp\u003eNote for Figure 3: month_may = May; state_sin = state of Sinaloa; ftnd_ = Fagerström Test of Nicotine Dependence; month_mar = March; state_chp = state of Chiapas; state_df = Mexico City; state_mex = State of Mexico; day_thu = Thursday; day_mon = Monday; state_tla = state of Tlaxcala.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/90ec1757067fe28f26ade455.png"},{"id":52107731,"identity":"25ea85af-fca4-4613-b0a0-5c9e82d785f9","added_by":"auto","created_at":"2024-03-06 20:00:59","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":403995,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/89c55bbe-0ca1-4864-baac-58e32ad5d27e.pdf"},{"id":52106206,"identity":"88034e36-7b2c-4e41-8d9a-af7c7718e22d","added_by":"auto","created_at":"2024-03-06 19:36:58","extension":"odt","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":22000,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatial and Time Use Pattern of Online Tobacco Self-Assessment Questionnaire, 2021-2022.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"SupplementaryTableIMLTobacco12.odt","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/0f19ede228570056eb1b7113.odt"},{"id":52106760,"identity":"808d3ef1-8bc0-48d2-a7f6-53fa99821068","added_by":"auto","created_at":"2024-03-06 19:44:58","extension":"odt","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":22915,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparison of Sociodemographic Characteristics of Online Tobacco Self-Assessment Questionnaire Users in Train versus Test Set, 2021-2022.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"SupplementaryTableIIMLTobacco12.odt","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/550a0b18de8e9b4c14e69e94.odt"},{"id":52105574,"identity":"c489b3c7-dc51-4f93-b87d-e3189e1770f6","added_by":"auto","created_at":"2024-03-06 19:28:58","extension":"odt","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":31409,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparison of Indicators of Nicotine Dependence in Online Tobacco Self-Assessment Questionnaire Users in Train versus Test Set, 2021-2022.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"SupplementaryTableIIIMLTobacco12.odt","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/f20fc6e942e22ba770022d05.odt"},{"id":52105577,"identity":"eee3bd3c-b929-4b3f-b243-3c93436e2d34","added_by":"auto","created_at":"2024-03-06 19:28:58","extension":"odt","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":31547,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparison of Spatial and Time Use Pattern of Online Tobacco Self-Assessment Questionnaire in Train versus Test Set, 2021-2022.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"SupplementaryTableVIMLTobacco12.odt","url":"https://assets-eu.researchsquare.com/files/rs-4014199/v1/b52420d29a8ff20e2dfe9ccc.odt"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eMachine Learning Analysis to Identify Factors Associated with Requesting Tobacco Cessation Services Among Users of an Online Self-Diagnostic Questionnaire in Mexico.\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTobacco use is a major preventable health risk factor, responsible for more than 7.7\u0026nbsp;million deaths per year and 200\u0026nbsp;million years of life lost worldwide (Reitsma et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The global prevalence of tobacco use has declined in the last 20 years to approximately 22.3% in 2020 and is estimated to decline further to 20.4% in 2025 (WHO \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis decline has been proposed to be due to the coordinated and sustained implementation of cost-effective interventions envisioned in MPOWER strategies at the global level (Levy et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Song et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), which were promoted under the Framework Convention on Tobacco Control (FCTC) of the World Health Organization (WHO FCTC \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Among these, the provision of tobacco cessation services is an effective intervention to maintain abstinence among people with tobacco dependence problems (Apollonio et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) or to reduce the amount of tobacco consumed (Lindson-Hawley et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCessation services encompass various treatments to help people stay tobacco-free and prevent relapse, which can range from brief counseling, behavioral interventions, and pharmacological treatment, which can be provided at the first, second, or third level of healthcare (Garc\u0026iacute;a-G\u0026oacute;mez et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAlthough the beneficial health effects of smoking cessation occur at any age, early quitting significantly improves the health profile of a population by dramatically reducing the risk of death (Di Cicco et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). For example, quitting at the age of 50 increases life expectancy by six years and quitting before the age of 65 reduces the risk of having a heart attack by 18.2% (Reitsma et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In addition, the use of cessation services reduces the costs of chronic disease at the individual, family, and societal levels, as well as the social and economic productivity (WHO \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAs a result, there is a high demand for smoking cessation services among smokers. In the United States, 70% of the people who smoke want to quit and more than half of them try it every year (Babb et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), while in Mexico, according to the Global Adult Alcohol Use Survey (GATS), 71.3% of smokers have thought or planned to quit, and 53.6% of smokers have tried to quit in the last 12 months, although 90.2% have tried without professional support (SS et al. 2023).\u003c/p\u003e \u003cp\u003eHowever, only 30% of the global population has access to smoking cessation services (WHO \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), so offering online or internet-based cessation (I-BC) services has become an attractive option for healthcare providers. Approximately one third of treatment requests over the past decade have been through online services, with an increase of 43.7% during 2017 (Graham et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and it has been reported that I-BC interventions may be effective in reducing smoking (G. M. J. Taylor et al. \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eI-BC also has led to greater data availability, which requires more robust analytical tools such as machine learning (ML) (Yarkoni and Westfall \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). ML uses algorithms to train a model that predicts the value of a variable of interest based on elements or features that act as inputs or predictors, and by considering all available information, provides greater accuracy and increases the generalization of results in healthcare (Fu et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Graham et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eDuring the past decade, ML modeling have been increasingly used in tobacco research to predict smoking, nicotine dependence, or urge to use, among others (Dumortier et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Fu et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Thakur et al. \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Wetherill et al. \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eAlthough there is plenty evidence on the factors associated with the demand for cessation services, including being female, early-onset tobacco use, multiple substance use, high tobacco use, not belonging to an Indigenous community, previous attempts to quit, or low perceived support in smoking quitting (Al Wahaibi et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Allen et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Borsari et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Chin et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2012\u003c/span\u003e)(Al Wahaibi et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Allen et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Borsari et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Chin et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), the use of ML for identify predictors of request for cessation services is scarce.\u003c/p\u003e \u003cp\u003eOne study found that among adult smokers in the United States, increased use of electronic cigarettes in the past 30 days, decreased use of cigarettes in the past 30 days, smoking initiation after age 18, use of other substances, and higher body mass index were associated with increased use of cessation services using an ensemble model (random forest plus bagging) (Issabakhsh et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)(. Another study reported that predictors of smoking cessation included age, severity of use (measured by duration of use, number of cigarettes per day, and levels of exhaled CO2), treatment use, and nicotine dependence using an artificial neural network model (Lai et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis shows a promising outlook for the use of I-BC and ML for evaluation of service utilization. However, we found no evidence on the request of cessation services by users of online screening questionnaire, therefore this study aims to analyze predictors associated with the request for tobacco cessation among users of an online self-diagnosis questionnaire as first contact.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis is a cross-sectional, descriptive, and correlational study of records of users of an online tobacco self-assessment questionnaire in Mexico.\u003c/p\u003e \u003cp\u003e \u003cem\u003eParticipants\u003c/em\u003e:\u003c/p\u003e \u003cp\u003eGeneral population who used the online tobacco self-assessment questionnaire from the Centers for Youth Integration (CIJ) (freely available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.cij.gob.mx/Autodiagnostico/frmtabaco.asp\u003c/span\u003e\u003cspan address=\"http://www.cij.gob.mx/Autodiagnostico/frmtabaco.asp\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were included.\u003c/p\u003e \u003cp\u003eInclusion criteria:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eIndividuals over 18 years old.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eRecords dated between February 8th, 2021, and June 16th, 2022.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003eExclusion criteria:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eRecords with missing data.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eData Collection\u003c/em\u003e:\u003c/p\u003e \u003cp\u003eUpon entering the website, users answer questions about tobacco dependence symptoms and sociodemographic information, then they receive automatic feedback about their risk level of tobacco use. Users also receive several options of smoking cessation services, including voluntarily leaving contact information (such as email or phone number) or providing them with the address from the nearest treatment center.\u003c/p\u003e \u003cp\u003e \u003cem\u003eInstruments\u003c/em\u003e:\u003c/p\u003e \u003cp\u003eNicotine dependence: the Spanish version of the Fagerstr\u0026ouml;m Test of Nicotine Dependence (FTND) (Beco\u0026ntilde;a and V\u0026aacute;zquez \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e1998\u003c/span\u003e; Heatherton et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e1991\u003c/span\u003e) was used. The psychometric properties of the questionnaire vary depending on the type of population in which it was validated, although its performance is lower in occasional smokers, users of electronic cigarettes, and smokers without smoke (Sharma et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In the Mexican population, although it showed low reliability, it had a high correlation with physiological measures of tobacco use and with the reported number of cigarettes consumed (Moreno-Couti\u0026ntilde;o and Villalobos-Gallegos \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), hence it is considered a good indicator of the tobacco use severity.\u003c/p\u003e \u003cp\u003eSociodemographic data were collected: age, gender (woman, man and non-binary), member of the LGBT\u0026thinsp;+\u0026thinsp;or Indigenous community and state of residence. In addition, all records were automatically identified by the system, as date and time.\u003c/p\u003e \u003cp\u003e \u003cem\u003eEthical Procedures\u003c/em\u003e:\u003c/p\u003e \u003cp\u003eAll procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2013 (World Medical Association \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) .\u003c/p\u003e \u003cp\u003eAll users of the online tobacco self-assessment questionnaire consented to provide clinical and sociodemographic data, as the questionnaire does not collect any personally identifiable information (such as name, address, email, or IP address).\u003c/p\u003e \u003cp\u003eFor those users who voluntarily provide contact information (phone, email) for cessation services request, the system deidentifies the records (removes all fields containing personal information) before generating the databases for analysis.\u003c/p\u003e \u003cp\u003e \u003cem\u003eOutcome variable\u003c/em\u003e:\u003c/p\u003e \u003cp\u003eThe outcome variable is the request for tobacco cessation services, which for the present study was operationalized as all those users who left their contact details on the online questionnaire to receive cessation services in the CIJ attention network.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData Analysis:\u003c/h2\u003e \u003cp\u003eThe total sample was randomly divided into a train and test set (split ratio 7:3). The variables were pre-processed (recoded, centered, and standardized) before performing the analyzes, obtaining a total of 119 features to perform the analyzes.\u003c/p\u003e \u003cp\u003eDescriptive statistics were calculated for all measured characteristics. For predictive analysis, we used the Random Forest (RF) algorithm. RF is an ensemble algorithm that averages the results of multiple decision trees (Khodadadi et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). It was used a 10-fold Cross Validation for internal validation and hyperparameter tunning for improving the predictive models. Due to the unbalanced nature of the group of interest (Johnson and Khoshgoftaar \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), the model was trained under four oversampling conditions: upsampling, downsampling, SMOTE and ROSE.\u003c/p\u003e \u003cp\u003eThe model with the highest predictive capacity in the test set defined by the area under the ROC curve (ROC AUC), was selected (Bradley \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). Performance metrics, including Sensitivity, Specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV) were obtained to assess the best model (Graham et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Issabakhsh et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) and identify the variables that contribute the most to the model. The study followed the recommendations of the MINIMAR standard for studies using artificial intelligence in medicine (Hernandez-Boussard et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eStatistical analyzes were performed in R version 4.3 programming language (R Core Team, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) using the Caret library version 6.0\u0026ndash;94 (Kuhn \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) and the R Studio IDE version 2022.12.0 (Posit Team \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e displays the records selected for analysis. Among all tobacco questionnaire users, 4,173 (22.7%) were excluded from the analysis because they were underage, and three records (0.01%) were excluded for incomplete information. As a result, a total of 14,182 records are available for analysis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e******************************** Insert Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e here *********************************\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;1 shows the sociodemographic characteristics of the questionnaire users. Almost half of the users were male. Approximately 9% of the users identified as LGBT+, 3% as indigenous, and 24% request for cessation services. The average age of the users was 30 years (29.9\u0026thinsp;\u0026plusmn;\u0026thinsp;11.8).\u003c/p\u003e \u003cp\u003e******************************** Insert Table I here ********************************\u003c/p\u003e \u003cp\u003eTable II shows the indicators of tobacco dependence, as defined by the FTND. The electronic version of the FTND had a reliability of 0.689, measured by Cronbach's alpha. Almost 80% of users smoke less than one pack per day, however, 3% smokes more than 31 cigarettes per day.\u003c/p\u003e \u003cp\u003eAdditionally, 23.2% smoke their first cigarette in the morning, approximately 30% smoke within the first hour of waking up, and 20% have difficulty quitting smoking in prohibited areas and 22% continue to smoke despite being sick.\u003c/p\u003e \u003cp\u003e******************************** Insert Table II here *********************************\u003c/p\u003e \u003cp\u003eInformation on the pattern of use can be found in Supplementary Table I. The data show that most visits to the questionnaire occurred in May (77.5%). Furthermore, more than half of the users of the self-diagnosis tool (51.8%) were located in the states of Aguascalientes (4.8%), Mexico City (17.3%), State of Mexico (12.6%), Jalisco (4%), Nuevo Leon (4.8%) and Sinaloa (8.3%). Furthermore, half of the users (52.4%) accessed the questionnaire between 13:00 and 20:00 hrs. Most visits (88.2%) occurred during weekdays.\u003c/p\u003e \u003cp\u003eNo significant differences were found between the training and validation samples regarding sociodemographic and clinical variables. However, there were significant differences in the proportion of users of the state of Tlaxcala and Quintana Roo (see Supplementary Tables II to IV).\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e displays the results of the predictive modeling. The area under the ROC curve ranged from 0.506 for ROSE resampling to 0.737 for downsampling. Sensitivity ranged from 0.411 to 1.000, Specificity ranged from 0.012 to 0.935, PPV from 0.234 to 0.682, and NPV from 0.840 to 1.000.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e******************************** Insert Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e here *********************************\u003c/p\u003e \u003cp\u003eThe downsampling technique was used to obtain the best predictive model (sensitivity\u0026thinsp;=\u0026thinsp;0.786, specificity\u0026thinsp;=\u0026thinsp;0.688, PPV\u0026thinsp;=\u0026thinsp;0.445, NPV\u0026thinsp;=\u0026thinsp;0.910, ROC AUC\u0026thinsp;=\u0026thinsp;0.737), so it was used to evaluate the importance of predictors.\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e shows the relative importance of the first 20 main characteristics, age and the month of May were the variables with the greatest relative importance for the predictive model, both having a relative importance greater than 80%.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOther variables that contribute to the model's predictive capacity include those related to clinical characteristics of nicotine dependence, like tobacco consumption after wake up in the morning; sociodemographic variables such as gender, and variables related to spatial and temporal usage patterns (such as being from the states of Sinaloa, Chiapas, Mexico City, or the State of Mexico, and using the platform on the 30th and 31st, Mondays and Thursdays, and in March). These variables, although of lesser importance, still play a role in the model's accuracy.\u003c/p\u003e \u003cp\u003e******************************** Insert Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e here *********************************\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study aimed to analyze predictors associated with the request for smoking cessation services among users of an online tobacco self-assessment questionnaire as a first point of contact.\u003c/p\u003e \u003cp\u003eThe use of the Random Forest algorithm with downsampling technique proved to be effective in predicting the use of tobacco cessation services (ROC AUC\u0026thinsp;=\u0026thinsp;0.737), correctly selecting 78.6% of users and correctly rejecting 68.8% of non-users. Random Forest is one of the most widely used ensemble algorithms (Fern\u0026aacute;ndez-Delgado et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) because it reduces the generalization error (error generated when predicting over new samples), is robust in collinearity situations, and reduces overfitting (Fang et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Liu et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Loupe, 2015).\u003c/p\u003e \u003cp\u003eAnother advantage of the Random Forest algorithm for estimating the importance of predictors is that they are weighted by individual random trees, so it behaves as an \"average\" of the individual importance of each tree (Loupe, 2015).\u003c/p\u003e \u003cp\u003eIn terms of the major predictors, our study found that age was the first significant predictor of seeking cessation services. Our findings are consistent with the literature, which suggests that older individuals, who are more likely to develop tobacco dependence, are more likely to seek formal treatment from a specialized professional compared to younger individuals (Hughes \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Research also suggests that young people may have a harder time quitting smoking due to their preference for quitting without assistance, such as nicotine replacement therapies, and instead opt for methods like \"cold turkey\" quitting or reducing cigarette smoking in certain situations (Dugas et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Solberg et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFurthermore, young adults who occasionally smoke cigarettes often label themselves as social smokers rather than smokers (Choi et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Kingsbury et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) rather than smokers (Choi et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), and may perceive smoking as harmless, deny addiction, and lack motivation to quit (Levinson et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). We recommend including other measures in future work that assess not only physical dependence, but also consumption pattern and other characteristics related to tobacco addiction. This suggestion has been made by other authors (Paz-Ballesteros et al. \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), as it is known that occasional smokers may not be detected by the FTND (Sharma et al. \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe second most important predictor was the month of May, and related to it, days 30 (5th importance level) and 31 (17th importance level) as predictors of service seeking. These dates are related to the World No Tobacco Day campaign, which will be commemorated in Mexico and worldwide on May 31, 2023 (WHO \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eEvidence has shown that campaigns that promote education and health services effectively increase the use of cessation services. For example, in the United States, such campaigns generated more than 1.3\u0026nbsp;million additional calls to helplines (US Public Health Service Office of the Surgeon General and National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health 2020). Similar results were found in Korea, where the demand for treatment increased 5.5 and 3.1 times during the months in which campaigns were promoted in 2010 and 2011, respectively, compared to the average demand during other months (Park et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOther predictor like gender female (15th importance level) is also a well-documented predictor in the literature, and our study is consistent with evidence that being female predicts seeking cessation services (Abdullah \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Jayakumar et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Jeong et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Williams et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAmong the clinical predictors of the FTND, item 3 \"How soon after you wake up do you smoke your first cigarette?\" (5th importance level), item 1 \"How many cigarettes per day do you smoke?\" (6th importance level) and item 4 \"Which cigarette would you hate most to give up?\" (8th importance level) were found to be the most important predictors of request for cessation services.\u003c/p\u003e \u003cp\u003eThese indicators have been consistently identified in the scientific literature as being associated with excessive tobacco use and nicotine dependence. Items 1 and 3 composes up the Heaviness of Smoking Index, an indicator of severity of smoking (Heatherton, et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1989\u003c/span\u003e), which is a predictor of quitting (Borland, et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Chaiton, et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAn issue that requires further attention in future research is the spatial pattern of self-assesstment questionnaire use, which may be closely related to the social determinants of tobacco use (Brady \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In Mexico, the most recent national household tobacco survey (Reynales Shigematsu et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) reported subnational differences in patterns of tobacco consumption and demand for cessation services; for example, the prevalence of tobacco consumption was higher than the national average in states such as Aguascalientes, Mexico City, and Mexico State, and our study found a higher proportion of users who expressed a desire to quit smoking in Mexico City and Mexico State, which is consistent with the aforementioned findings.\u003c/p\u003e \u003cp\u003e \u003cem\u003e4.1 Limitations of the study\u003c/em\u003e:\u003c/p\u003e \u003cp\u003eOne of the main limitations of the study is that the online questionnaire did not measure many variables that the literature has related consistently to tobacco use and that could explain both the severity of use and the request of cessation services, such as socioeconomic status, schooling, type of job, or tobacco availability (Brady \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Pulakka et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOther variables such as internet access and health literacy could also impact IB-C services request. Research has shown a negative correlation between Internet use and smoking severity (McCrabb et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), while health literacy has been found to have a positive impact on smoking cessation intentions and behaviors (Sun et al. \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe co-occurrence of psychiatric symptomatology and smoking is also related with the severity of tobacco use and the need for treatment (Szerman et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). For example, depressive symptoms are strongly related to frequency, severity, and level of tobacco dependence (Bakhshaie et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Leyro et al. \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). As a result, people experiencing these symptoms could require more care due to the cumulative effect of physical and psychological distress (Jamal et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2012\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFuture studies should also consider measurement of social determinants of health such as income inequality, barriers, and facilitators to access to services that could impact predictors and outcomes of both tobacco use and care seeking (Alegr\u0026iacute;a et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Brady \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eI-BC has become an attractive option for providing health care services to population, as it allows to reach a larger number of people at a lower cost. However, it is necessary to understand the characteristics and needs of online tool users, as they differ from traditional users.\u003c/p\u003e \u003cp\u003eBesides, the use of web-based screening tools has increased access to mental health and substance use treatment, but it also poses new challenges for analyzing the large amounts of data generated, so the use of machine learning/artificial intelligence algorithms will become increasingly widespread.\u003c/p\u003e \u003cp\u003eIn this paper, the Random Forest algorithm with downsampling was shown to be a highly effective predictive modeling tool. It was shown to accurately identify potential users (AUC ROC\u0026thinsp;=\u0026thinsp;0.737), correctly identifying 78.6% of those who requested cessation services.\u003c/p\u003e \u003cp\u003eBy using ML to identify predictors associated with request of tobacco cessation services, such as age, massive campaigns like World No Tobacco Day (May 31th), gender, or social determinants like location, it is possible to develop and implement targeted prevention campaigns and strategies that allow us to reach a larger clinical population.\u003c/p\u003e \u003cp\u003eThe results presented here could be used to guide the use of these tools in the planning and delivery of mental health services that are sensitive and adapted to the needs of the population seeking care.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbdullah, A. S. M. (2004). Which smokers use the smoking cessation Quitline in Hong Kong, and how effective is the Quitline? \u003cem\u003eTobacco Control\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(4), 415\u0026ndash;421. https://doi.org/10.1136/tc.2003.006460\u003c/li\u003e\n\u003cli\u003eAl Wahaibi, N., Al Lawati, A., Al Ruqeishy, F., Al Khatri, A., Al-Farsi, Y., Juma, T. M. A., et al. (2019). The characteristics and patterns of utilization of healthcare services among Omanis with substance use disorders attending therapy for cessation. \u003cem\u003ePLOS ONE\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), e0210532. https://doi.org/10.1371/journal.pone.0210532\u003c/li\u003e\n\u003cli\u003eAlegr\u0026iacute;a, M., Alvarez, K., Cheng, M., \u0026amp; Falgas-Bague, I. (2023). Recent Advances on Social Determinants of Mental Health: Looking Fast Forward. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e, \u003cem\u003e180\u003c/em\u003e(7), 473\u0026ndash;482. https://doi.org/10.1176/appi.ajp.20230371\u003c/li\u003e\n\u003cli\u003eAllen, A. M., Yuan, N. P., Wertheim, B. C., Krupski, L., Bell, M. L., \u0026amp; Nair, U. (2019). Gender differences in utilization of services and tobacco cessation outcomes at a state quitline. \u003cem\u003eTranslational Behavioral Medicine\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(4), 663\u0026ndash;668. https://doi.org/10.1093/tbm/iby083\u003c/li\u003e\n\u003cli\u003eApollonio, D., Philipps, R., \u0026amp; Bero, L. (2016). Interventions for tobacco use cessation in people in treatment for or recovery from substance use disorders. \u003cem\u003eCochrane Database of Systematic Reviews\u003c/em\u003e, \u003cem\u003e2017\u003c/em\u003e(1). https://doi.org/10.1002/14651858.CD010274.pub2\u003c/li\u003e\n\u003cli\u003eBabb, S., Malarcher, A., Schauer, G., Asman, K., \u0026amp; Jamal, A. (2017). Quitting Smoking Among Adults United States, 2000\u0026ndash;2015. \u003cem\u003eMMWR. Morbidity and Mortality Weekly Report\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e(52), 1457\u0026ndash;1464. https://doi.org/10.15585/mmwr.mm6552a1\u003c/li\u003e\n\u003cli\u003eBakhshaie, J., Zvolensky, M. J., \u0026amp; Goodwin, R. D. (2015). Cigarette smoking and the onset and persistence of depression among adults in the United States: 1994\u0026ndash;2005. \u003cem\u003eComprehensive Psychiatry\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e, 142\u0026ndash;148. https://doi.org/10.1016/j.comppsych.2014.10.012\u003c/li\u003e\n\u003cli\u003eBeco\u0026ntilde;a, E., \u0026amp; V\u0026aacute;zquez, F. L. (1998). The Fagerstr\u0026ouml;m Test for Nicotine Dependence in a Spanish Sample. \u003cem\u003ePsychological Reports\u003c/em\u003e, \u003cem\u003e83\u003c/em\u003e(3_suppl), 1455\u0026ndash;1458. https://doi.org/10.2466/pr0.1998.83.3f.1455\u003c/li\u003e\n\u003cli\u003eBorland, R., Yong, H.-H., O\u0026rsquo;Connor, R. J., Hyland, A., \u0026amp; Thompson, M. E. (2010). The reliability and predictive validity of the Heaviness of Smoking Index and its two components: Findings from the International Tobacco Control Four Country study. Nicotine \u0026amp; Tobacco Research, 12(Supplement 1), S45\u0026ndash;S50. https://doi.org/10.1093/ntr/ntq038\u003c/li\u003e\n\u003cli\u003eBorsari, L., Storani, S., Malagoli, C., Filippini, T., Tamelli, M., Malavolti, M., et al. (2018). Impact of Referral Sources and Waiting Times on the Failure to Quit Smoking: One-Year Follow-Up of an Italian Cohort Admitted to a Smoking Cessation Service. \u003cem\u003eInternational Journal of Environmental Research and Public Health\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(6), 1234. https://doi.org/10.3390/ijerph15061234\u003c/li\u003e\n\u003cli\u003eBradley, A. P. (1997). The use of the area under the ROC curve in the evaluation of machine learning algorithms. \u003cem\u003ePattern Recognition\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e(7), 1145\u0026ndash;1159. https://doi.org/10.1016/S0031-3203(96)00142-2\u003c/li\u003e\n\u003cli\u003eBrady, K. T. (2020). Social Determinants of Health and Smoking Cessation: A Challenge. \u003cem\u003eAmerican Journal of Psychiatry\u003c/em\u003e, \u003cem\u003e177\u003c/em\u003e(11), 1029\u0026ndash;1030. https://doi.org/10.1176/appi.ajp.2020.20091374\u003c/li\u003e\n\u003cli\u003eChaiton, M. O., Cohen, J. E., McDonald, P. W., \u0026amp; Bondy, S. J. (2007). The Heaviness of Smoking Index as a predictor of smoking cessation in Canada. Addictive Behaviors, 32(5), 1031\u0026ndash;1042. https://doi.org/10.1016/j.addbeh.2006.07.008\u003c/li\u003e\n\u003cli\u003eChin, D. L., Hong, O., Gillen, M., Bates, M. N., \u0026amp; Okechukwu, C. A. (2012). Occupational Factors and Smoking Cessation among Unionized Building Trades Workers. \u003cem\u003eWorkplace Health \u0026amp; Safety\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e(10), 445\u0026ndash;452. https://doi.org/10.1177/216507991206001006\u003c/li\u003e\n\u003cli\u003eChoi, Y., Choi, S. M., \u0026amp; Rifon, N. (2010). \u0026ldquo;I Smoke but I Am Not a Smoker\u0026rdquo;: Phantom Smokers and the Discrepancy Between Self-identity and Behavior. \u003cem\u003eJournal of American College Health\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(2), 117\u0026ndash;125. https://doi.org/10.1080/07448481.2010.483704\u003c/li\u003e\n\u003cli\u003eDi Cicco, M. E., Ragazzo, V., \u0026amp; Jacinto, T. (2016). Mortality in relation to smoking: the British Doctors Study. \u003cem\u003eBreathe\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(3), 275\u0026ndash;276. https://doi.org/10.1183/20734735.013416\u003c/li\u003e\n\u003cli\u003eDong, J., Feng, T., Thapa-Chhetry, B., Cho, B. G., Shum, T., Inwald, D. P., et al. (2021). Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care. \u003cem\u003eCritical Care\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(1), 288. https://doi.org/10.1186/s13054-021-03724-0\u003c/li\u003e\n\u003cli\u003eDugas, E. N., Wellman, R. J., Kermack, A., Tremblay, M., \u0026amp; OʼLoughlin, J. (2016). Reasons Young Smokers Do Not Use NRT Even When it is Available Free-of-Charge: An Exploratory Study: \u003cem\u003eThe Canadian Journal of Addiction\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(2), 14\u0026ndash;21. https://doi.org/10.1097/02024458-201606000-00003\u003c/li\u003e\n\u003cli\u003eDumortier, A., Beckjord, E., Shiffman, S., \u0026amp; Sejdić, E. (2016). Classifying smoking urges via machine learning. \u003cem\u003eComputer Methods and Programs in Biomedicine\u003c/em\u003e, \u003cem\u003e137\u003c/em\u003e, 203\u0026ndash;213. https://doi.org/10.1016/j.cmpb.2016.09.016\u003c/li\u003e\n\u003cli\u003eFang, G., Annis, I. E., Elston-Lafata, J., \u0026amp; Cykert, S. (2019). Applying machine learning to predict real-world individual treatment effects: insights from a virtual patient cohort. \u003cem\u003eJournal of the American Medical Informatics Association\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(10), 977\u0026ndash;988. https://doi.org/10.1093/jamia/ocz036\u003c/li\u003e\n\u003cli\u003eFern\u0026aacute;ndez-Delgado, M., Cernadas, E., Barro, S., \u0026amp; Amorin, D. (2014). Do we Need Hundreds of Classifiers to Solve Real World Classification Problems? Journal of Machine Learning Research, 15, 3133\u0026ndash;3181.\u003c/li\u003e\n\u003cli\u003eFu, R., Kundu, A., Mitsakakis, N., Elton-Marshall, T., Wang, W., Hill, S., et al. (2021). Machine learning applications in tobacco research: a scoping review. \u003cem\u003eTobacco Control\u003c/em\u003e, tobaccocontrol-2020-056438. https://doi.org/10.1136/tobaccocontrol-2020-056438\u003c/li\u003e\n\u003cli\u003eGarc\u0026iacute;a-G\u0026oacute;mez, L., Hern\u0026aacute;ndez-P\u0026eacute;rez, A., No\u0026eacute;-D\u0026iacute;az, V., Riesco-Miranda, J. A., \u0026amp; Jim\u0026eacute;nez-Ruiz, C. (2019). Smoking Cessation Treatments: Current Psychological and Pharmacological Options. \u003cem\u003eRevista de investigaci\u0026oacute;n Cl\u0026iacute;nica\u003c/em\u003e, \u003cem\u003e71\u003c/em\u003e(1), 1410. https://doi.org/10.24875/RIC.18002629\u003c/li\u003e\n\u003cli\u003eGraham, S., Depp, C., Lee, E. E., Nebeker, C., Tu, X., Kim, H.-C., \u0026amp; Jeste, D. V. (2019). Artificial Intelligence for Mental Health and Mental Illnesses: an Overview. \u003cem\u003eCurrent Psychiatry Reports\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(11), 116. https://doi.org/10.1007/s11920-019-1094-0\u003c/li\u003e\n\u003cli\u003eHeatherton, T. F., Kozlowski, L. T., Frecker, R. C., \u0026amp; Fagerstrom, K.-O. (1991). The Fagerstrom Test for Nicotine Dependence: a revision of the Fagerstrom Tolerance Questionnaire. \u003cem\u003eAddiction\u003c/em\u003e, \u003cem\u003e86\u003c/em\u003e(9), 1119\u0026ndash;1127. https://doi.org/10.1111/j.1360-0443.1991.tb01879.x\u003c/li\u003e\n\u003cli\u003eHeatherton, T. F., Kozlowski, L. T., Frecker, R. C., Rickert, W., \u0026amp; Robinson, J. (1989). Measuring the heaviness of smoking: using self-reported time to the first cigarette of the day and number of cigarettes smoked per day. British Journal of Addiction, 84(7), 791\u0026ndash;799. https://doi.org/10.1111/j.1360-0443.1989.tb03059.x\u003c/li\u003e\n\u003cli\u003eHernandez-Boussard, T., Bozkurt, S., Ioannidis, J. P. A., \u0026amp; Shah, N. H. (2020). MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care. \u003cem\u003eJournal of the American Medical Informatics Association\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(12), 2011\u0026ndash;2015. https://doi.org/10.1093/jamia/ocaa088\u003c/li\u003e\n\u003cli\u003eHughes, J. R. (2013). An updated algorithm for choosing among smoking cessation treatments. \u003cem\u003eJournal of Substance Abuse Treatment\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(2), 215\u0026ndash;221. https://doi.org/10.1016/j.jsat.2013.01.011\u003c/li\u003e\n\u003cli\u003eIssabakhsh, M., S\u0026aacute;nchez-Romero, L. M., Le, T. T. T., Liber, A. C., Tan, J., Li, Y., et al. (2023). Machine learning application for predicting smoking cessation among US adults: An analysis of waves 1-3 of the PATH study. \u003cem\u003ePLOS ONE\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(6), e0286883. https://doi.org/10.1371/journal.pone.0286883\u003c/li\u003e\n\u003cli\u003eJamal, M., Willem Van der Does, A. J., Cuijpers, P., \u0026amp; Penninx, B. W. J. H. (2012). Association of smoking and nicotine dependence with severity and course of symptoms in patients with depressive or anxiety disorder. \u003cem\u003eDrug and Alcohol Dependence\u003c/em\u003e, \u003cem\u003e126\u003c/em\u003e(1\u0026ndash;2), 138\u0026ndash;146. https://doi.org/10.1016/j.drugalcdep.2012.05.001\u003c/li\u003e\n\u003cli\u003eJayakumar, N., Chaiton, M., Zhang, B., Selby, P., \u0026amp; Schwartz, R. (2020). Sex Differences in Use of Smoking Cessation Services and Resources: A Real-World Study. \u003cem\u003eTobacco Use Insights\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e, 1179173X2090150. https://doi.org/10.1177/1179173X20901500\u003c/li\u003e\n\u003cli\u003eJeong, B. Y., Lim, M. K., Yun, E. H., \u0026amp; Oh, J.-K. (2019). User characteristics of national smoking cessation services in Korea: who chooses each type of tobacco cessation program? \u003cem\u003eBMC Health Services Research\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(1), 14. https://doi.org/10.1186/s12913-018-3817-z\u003c/li\u003e\n\u003cli\u003eJohnson, J. M., \u0026amp; Khoshgoftaar, T. M. (2019). Survey on deep learning with class imbalance. \u003cem\u003eJournal of Big Data\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(1), 27. https://doi.org/10.1186/s40537-019-0192-5\u003c/li\u003e\n\u003cli\u003eKalafi, E. Y., Nor, N. a. M., Taib, N. A., Ganggayah, M. D., Town, C., \u0026amp; Dhillon, S. K. (2019). Machine Learning and Deep Learning Approaches in Breast Cancer Survival Prediction Using Clinical Data. \u003cem\u003eFolia Biologica\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e(5\u0026ndash;6), 212\u0026ndash;220.\u003c/li\u003e\n\u003cli\u003eKhodadadi, A., Ghanbari Bousejin, N., Molaei, S., Kumar Chauhan, V., Zhu, T., \u0026amp; Clifton, D. A. (2023). Improving Diagnostics with Deep Forest Applied to Electronic Health Records. \u003cem\u003eSensors (Basel, Switzerland)\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(14), 6571. https://doi.org/10.3390/s23146571\u003c/li\u003e\n\u003cli\u003eKingsbury, J. H., Parks, M. J., Amato, M. S., \u0026amp; Boyle, R. G. (2016). Deniers and Admitters: Examining Smoker Identities in a Changing Tobacco Landscape. \u003cem\u003eNicotine \u0026amp; Tobacco Research\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(11), 2130\u0026ndash;2137. https://doi.org/10.1093/ntr/ntw110\u003c/li\u003e\n\u003cli\u003eKuhn, M. (2008). Building Predictive Models in R Using the caret Package. \u003cem\u003eJournal of Statistical Software\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(5), 1\u0026ndash;26. https://doi.org/10.18637/jss.v028.i05\u003c/li\u003e\n\u003cli\u003eLai, C.-C., Huang, W.-H., Chang, B. C.-C., \u0026amp; Hwang, L.-C. (2021). Development of Machine Learning Models for Prediction of Smoking Cessation Outcome. \u003cem\u003eInternational Journal of Environmental Research and Public Health\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(5), 2584. https://doi.org/10.3390/ijerph18052584\u003c/li\u003e\n\u003cli\u003eLevinson, A. H., Campo, S., Gascoigne, J., Jolly, O., Zakharyan, A., \u0026amp; Tran, Z. V. (2007). Smoking, but not smokers: Identity among college students who smoke cigarettes. \u003cem\u003eNicotine \u0026amp; Tobacco Research\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(8), 845\u0026ndash;852. https://doi.org/10.1080/14622200701484987\u003c/li\u003e\n\u003cli\u003eLevy, D. T., Yuan, Z., Luo, Y., \u0026amp; Mays, D. (2018). Seven years of progress in tobacco control: an evaluation of the effect of nations meeting the highest level MPOWER measures between 2007 and 2014. \u003cem\u003eTobacco Control\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(1), 50\u0026ndash;57. https://doi.org/10.1136/tobaccocontrol-2016-053381\u003c/li\u003e\n\u003cli\u003eLeyro, T. M., Crew, E. E., Bryson, S. W., Lembke, A., Bailey, S. R., Prochaska, J. J., et al. (2016). Retrospective analysis of changing characteristics of treatment-seeking smokers: implications for further reducing smoking prevalence. \u003cem\u003eBMJ Open\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(6), e010960. https://doi.org/10.1136/bmjopen-2015-010960\u003c/li\u003e\n\u003cli\u003eLindson-Hawley, N., Hartmann-Boyce, J., Fanshawe, T. R., Begh, R., Farley, A., \u0026amp; Lancaster, T. (2016). Interventions to reduce harm from continued tobacco use. \u003cem\u003eCochrane Database of Systematic Reviews\u003c/em\u003e, \u003cem\u003e2016\u003c/em\u003e(12). https://doi.org/10.1002/14651858.CD005231.pub3\u003c/li\u003e\n\u003cli\u003eLiu, Y., Wang, Y., \u0026amp; Zhang, J. (2012). New Machine Learning Algorithm: Random Forest. In B. Liu, M. Ma, \u0026amp; J. Chang (Eds.), \u003cem\u003eInformation Computing and Applications\u003c/em\u003e (Vol. 7473, pp. 246\u0026ndash;252). Berlin, Heidelberg: Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-34062-8_3\u003c/li\u003e\n\u003cli\u003eLouppe, G. (2015, June 3). Understanding Random Forests: From Theory to Practice. arXiv. http://arxiv.org/abs/1407.7502. Accessed 27 February 2024\u003c/li\u003e\n\u003cli\u003eMcCrabb, S., Twyman, L., Palazzi, K., Guillaumier, A., Paul, C., \u0026amp; Bonevski, B. (2019). A cross sectional survey of internet use among a highly socially disadvantaged population of tobacco smokers. \u003cem\u003eAddiction Science \u0026amp; Clinical Practice\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 38. https://doi.org/10.1186/s13722-019-0168-y\u003c/li\u003e\n\u003cli\u003eMoreno-Couti\u0026ntilde;o, A., \u0026amp; Villalobos-Gallegos, L. (2017). Psychometric Properties of the Fagerstr\u0026ouml;m Test for Nicotine Dependence in a Sample of Mexican Smokers. \u003cem\u003eJournal of Addictions Nursing\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(1), 27\u0026ndash;33. https://doi.org/10.1097/JAN.0000000000000156\u003c/li\u003e\n\u003cli\u003ePark, J., Luu Ngoc, Shin, S., Oh, J.-K., Yun, E. H., Lee, D., \u0026amp; Lim, M. K. (2019). Influence of new tobacco control policies and campaigns onQuitline call volume in Korea. \u003cem\u003eTobacco Induced Diseases\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(March). https://doi.org/10.18332/tid/104674\u003c/li\u003e\n\u003cli\u003ePaz-Ballesteros, W. C., Zavala-Arciniega, L., Guti\u0026eacute;rrez-Torres, D. S., Ponciano-Rodr\u0026iacute;guez, G., \u0026amp; Reynales-Shigematsu, L. M. (2019). Evaluaci\u0026oacute;n de la dependencia f\u0026iacute;sica y psicol\u0026oacute;gica al tabaco en fumadores mexicanos adultos, Encodat 2016. \u003cem\u003eSalud P\u0026uacute;blica de M\u0026eacute;xico\u003c/em\u003e, \u003cem\u003e61\u003c/em\u003e(2, Mar-Abr), 136. https://doi.org/10.21149/9858\u003c/li\u003e\n\u003cli\u003ePosit team. (2023). RStudio: Integrated Development Environment for R. Boston, MA: Posit Software, PBC. http://www.posit.co/\u003c/li\u003e\n\u003cli\u003ePulakka, A., Halonen, J. I., Kawachi, I., Pentti, J., Stenholm, S., Jokela, M., et al. (2016). Association Between Distance From Home to Tobacco Outlet and Smoking Cessation and Relapse. \u003cem\u003eJAMA Internal Medicine\u003c/em\u003e, \u003cem\u003e176\u003c/em\u003e(10), 1512. https://doi.org/10.1001/jamainternmed.2016.4535\u003c/li\u003e\n\u003cli\u003eR Core Team. (2023). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/\u003c/li\u003e\n\u003cli\u003eRadovic, A., \u0026amp; Badawy, S. M. (2020). Technology Use for Adolescent Health and Wellness. \u003cem\u003ePediatrics\u003c/em\u003e, \u003cem\u003e145\u003c/em\u003e(Supplement_2), S186\u0026ndash;S194. https://doi.org/10.1542/peds.2019-2056G\u003c/li\u003e\n\u003cli\u003eReitsma, M. B., Kendrick, P. J., Ababneh, E., Abbafati, C., Abbasi-Kangevari, M., Abdoli, A., et al. (2021). Spatial, temporal, and demographic patterns in prevalence of smoking tobacco use and attributable disease burden in 204 countries and territories, 1990\u0026ndash;2019: a systematic analysis from the Global Burden of Disease Study 2019. \u003cem\u003eThe Lancet\u003c/em\u003e, \u003cem\u003e397\u003c/em\u003e(10292), 2337\u0026ndash;2360. https://doi.org/10.1016/S0140-6736(21)01169-7\u003c/li\u003e\n\u003cli\u003eReynales Shigematsu, L. M., Zavala-Arciniega, L., Paz-Ballesteros, W. C., Guti\u0026eacute;rrez-Torres, D. S., Garc\u0026iacute;a-Buend\u0026iacute;a, J. C., Rodr\u0026iacute;guez-Andrade, M. A., et al. (2017). \u003cem\u003eEncuesta Nacional de Consumo de Drogas, Alcohol y Tabaco 2016-2017: Reporte de Tabaco.\u003c/em\u003e Ciudad de M\u0026eacute;xico, M\u0026eacute;xico: INPRFM, INSP, CONADIC, SS. https://drive.google.com/file/d/1Iktptvdu2nsrSpMBMT4FdqBIk8gikz7q/view\u003c/li\u003e\n\u003cli\u003eSharma, M., Suman, L., Srivastava, K., Suma, N., \u0026amp; Vishwakarma, A. (2021). Psychometric properties of Fagerstrom Test of Nicotine Dependence: A systematic review. \u003cem\u003eIndustrial Psychiatry Journal\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e(2), 207. https://doi.org/10.4103/ipj.ipj_51_21\u003c/li\u003e\n\u003cli\u003eSolberg, L. I., Boyle, R. G., McCarty, M., Asche, S. E., \u0026amp; Thoele, M. J. (2007). Young adult smokers: are they different? \u003cem\u003eThe American Journal of Managed Care\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(11), 626\u0026ndash;632.\u003c/li\u003e\n\u003cli\u003eSong, Y., Zhao, L., Palipudi, K. M., Asma, S., Morton, J., Talley, B., et al. (2016). Tracking MPOWER in 14 countries: results from the Global Adult Tobacco Survey, 2008\u0026ndash;2010. \u003cem\u003eGlobal Health Promotion\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(2_suppl), 24\u0026ndash;37. https://doi.org/10.1177/1757975913501911\u003c/li\u003e\n\u003cli\u003eSS, CONASAMA, \u0026amp; INSP. (2023). . \u003cem\u003eEncuesta Global de Tabaquismo en Adultos (GATS): Hoja de Resumen.\u003c/em\u003e (p. 2). M\u0026eacute;xico: INSP. https://www.insp.mx/avisos/presentacion-de-resultados-de-la-gats-2023\u003c/li\u003e\n\u003cli\u003eSun, S., Yu, H., Ling, J., Yao, D., Chen, H., \u0026amp; Liu, G. (2023). The influence of health literacy and knowledge about smoking hazards on the intention to quit smoking and its intensity: an empirical study based on the data of China\u0026rsquo;s health literacy investigation. \u003cem\u003eBMC Public Health\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(1), 2355. https://doi.org/10.1186/s12889-023-17292-1\u003c/li\u003e\n\u003cli\u003eSzerman, N., Parro, C., Pinet, C., Mart\u0026iacute;nez-Raga, J., Basurte, I., \u0026amp; Saiz, P. (2022). TOBACCO USE DISORDER AND DUAL DISORDERS Joint statement by the Spanish Psychiatry Society and the Spanish Dual Disorders Society. \u003cem\u003eActas Espanolas De Psiquiatria\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(Supplement), 77\u0026ndash;138.\u003c/li\u003e\n\u003cli\u003eTaylor, D. S., Medaglio, D., Jurkovitz, C. T., Patterson, F., Zhang, Z., Gbadebo, A., et al. (2020). Evaluation of a Systems-Based Tobacco Cessation Program Using Bedside Volunteers. \u003cem\u003eNicotine \u0026amp; Tobacco Research\u003c/em\u003e, \u003cem\u003e22\u003c/em\u003e(3), 440\u0026ndash;445. https://doi.org/10.1093/ntr/nty252\u003c/li\u003e\n\u003cli\u003eTaylor, G. M. J., Dalili, M. N., Semwal, M., Civljak, M., Sheikh, A., \u0026amp; Car, J. (2017). Internet-based interventions for smoking cessation. \u003cem\u003eCochrane Database of Systematic Reviews\u003c/em\u003e, \u003cem\u003e2017\u003c/em\u003e(9). https://doi.org/10.1002/14651858.CD007078.pub5\u003c/li\u003e\n\u003cli\u003eThakur, S. S., Poddar, P., \u0026amp; Roy, R. B. (2022). Real-time prediction of smoking activity using machine learning based multi-class classification model. \u003cem\u003eMultimedia Tools and Applications\u003c/em\u003e, \u003cem\u003e81\u003c/em\u003e(10), 14529\u0026ndash;14551. https://doi.org/10.1007/s11042-022-12349-6\u003c/li\u003e\n\u003cli\u003eUS Public Health Service Office of the Surgeon General, \u0026amp; National Center for Chronic Disease Prevention and Health Promotion (US) Office on Smoking and Health. (2020). Chapter 6, Interventions for Smoking Cessation and Treatments for Nicotine Dependence. In \u003cem\u003eSmoking Cessation: A Report of the Surgeon General [Internet]\u003c/em\u003e. Washington (DC): US Department of Health and Human Services. https://www.ncbi.nlm.nih.gov/books/NBK555596/\u003c/li\u003e\n\u003cli\u003eWetherill, R. R., Rao, H., Hager, N., Wang, J., Franklin, T. R., \u0026amp; Fan, Y. (2019). Classifying and characterizing nicotine use disorder with high accuracy using machine learning and resting‐state fMRI. \u003cem\u003eAddiction Biology\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(4), 811\u0026ndash;821. https://doi.org/10.1111/adb.12644\u003c/li\u003e\n\u003cli\u003eWHO. (2019). \u003cem\u003eWHO Report on the Global Tobacco Epidemic, 2019\u003c/em\u003e (p. 109). Geneva: World Health Organization. https://apps.who.int/iris/rest/bitstreams/1239531/retrieve\u003c/li\u003e\n\u003cli\u003eWHO. (2021). \u003cem\u003eWHO global report on trends in prevalence of tobacco use 2000-2025\u003c/em\u003e (No. 4ta. edici\u0026oacute;n.) (p. 150). Geneva. https://apps.who.int/iris/rest/bitstreams/1263754/retrieve\u003c/li\u003e\n\u003cli\u003eWHO. (2023). World No Tobacco Day. https://www.who.int/campaigns/world-no-tobacco-day. Accessed 9 January 2024\u003c/li\u003e\n\u003cli\u003eWHO FCTC. (2003). \u003cem\u003eWHO Framework Convention on Tobacco Control\u003c/em\u003e. Geneva, Switzerland: WHO Press. https://iris.who.int/bitstream/handle/10665/42811/9241591013.pdf\u003c/li\u003e\n\u003cli\u003eWilliams, R. M., Eyestone, E., Smith, L., Philips, J. G., Whealan, J., Webster, M., et al. (2022). Engaging Patients in Smoking Cessation Treatment within the Lung Cancer Screening Setting: Lessons Learned from an NCI SCALE Trial. \u003cem\u003eCurrent Oncology\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(4), 2211\u0026ndash;2224. https://doi.org/10.3390/curroncol29040180\u003c/li\u003e\n\u003cli\u003eWorld Medical Association. (2013). World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. \u003cem\u003eJAMA\u003c/em\u003e, \u003cem\u003e310\u003c/em\u003e(20), 2191. https://doi.org/10.1001/jama.2013.281053\u003c/li\u003e\n\u003cli\u003eWu, Y., \u0026amp; Fang, Y. (2020). Stroke Prediction with Machine Learning Methods among Older Chinese. \u003cem\u003eInternational Journal of Environmental Research and Public Health\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(6), 1828. https://doi.org/10.3390/ijerph17061828\u003c/li\u003e\n\u003cli\u003eYarkoni, T., \u0026amp; Westfall, J. (2017). Choosing Prediction Over Explanation in Psychology: Lessons From Machine Learning. \u003cem\u003ePerspectives on Psychological Science\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(6), 1100\u0026ndash;1122. https://doi.org/10.1177/1745691617693393\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable I. Sociodemographic Characteristics of Online Tobacco Self-Assessment Questionnaire Users, 2021-2022.\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"333\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"54.49101796407186%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eOnline Tobacco Self-Assessment Questionnaire\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003en\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"45.34534534534534%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eGender\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eMan\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e7,534\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e53.1 \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eWoman\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e6,153\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e43.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eNon-binary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e495\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e3.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"45.34534534534534%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eLGBT+\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e1,331\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e9.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e12,851\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e90.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"45.34534534534534%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eIndigenous\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e473\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e3.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e13,709\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e96.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"45.34534534534534%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eRequest for services\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.32732732732733%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e3,423\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e24.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"6.88622754491018%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026zwj;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"38.622754491017965%\" valign=\"top\"\u003e\n \u003cp\u003eNo \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e10,759\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"27.24550898203593%\" valign=\"top\"\u003e\n \u003cp\u003e75.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"45.34534534534534%\" colspan=\"2\"\u003e\n \u003cp\u003eAge (Mean\u0026nbsp;\u0026plusmn;\u0026nbsp;SD)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"54.65465465465466%\" colspan=\"2\"\u003e\n \u003cp\u003e29.9 \u0026plusmn; 11.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: % = percentage within column.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable II. Indicators of Nicotine Dependence in Online Tobacco Self-Assessment Questionnaire Users, 2021-2022.\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"442\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"43.43891402714932%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eOnline Tobacco Self-Assessment Questionnaire\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003en\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e1. How many cigarettes per day do you smoke?\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003e10 or less \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e11,271\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e79.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003e11 to 20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e1,800\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e12.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003e21 to 30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e669\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e4.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003e31 or more\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e442\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e3.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e2. Do you smoke more frequently during the first hours after waking than during the rest of the day? \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e3,292\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e23.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e10,890\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e76.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e3. How soon after you wake up do you smoke your first cigarette?\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eWithin 5 minutes \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e1,418\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e10.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eWithin 6 - 30 minutes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e1,498\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e10.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eWithin 31 - 60 minutes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e1,479\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e10.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eAfter 60 minutes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e9,787\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e69.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e4. Which cigarette would you hate most to give up?\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eThe first one in the morning\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e4,796\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e30.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eAll others \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e9,796\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e69.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e5. Do you find it difficult to refrain from smoking in places where it is forbidden?\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e2,658\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e18.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e11,524\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e81.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e6. Do you smoke if you are so ill that you are in bed most of the day?\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eYes \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e3,190\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e22.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"4.298642533936651%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026zwj;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"52.262443438914026%\" valign=\"top\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"22.171945701357465%\" valign=\"top\"\u003e\n \u003cp\u003e10,992\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.266968325791854%\" valign=\"top\"\u003e\n \u003cp\u003e77.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"56.56108597285068%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eCronbach\u0026rsquo;s\u0026nbsp;\u0026alpha;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"43.43891402714932%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.689\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: % = percentage within column.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Centros de Integración Juvenil AC","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Smoking, Tobacco Use Cessation, Machine Learning, Random Forest, Predictive model, Oversampling, Feature Selection","lastPublishedDoi":"10.21203/rs.3.rs-4014199/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4014199/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eIntroduction\u003c/strong\u003e: While traditional cessation services exist, internet-based options are gaining popularity. However, understanding online users' unique characteristics compared to traditional users is crucial. This study utilize machine learning (ML), aimed to identify these online users and their needs.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethod\u003c/strong\u003e: Through analyzing 14,182 records of adults who completed online nicotine dependence screening questionnaire, a random forest algorithm plus oversampling was used to predict request services.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: The algorithm accurately identified 78.6% of users and rejected 68.8% of non-users. Notably, age, sex, dependence severity indicators, certain locations, and even specific occasions like World No Tobacco Day, were identified as key factors influencing service request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscussion\u003c/strong\u003e: These findings suggest the effectiveness of random forest algorithm in predicting potential users. Moreover, the predictor variables offer valuable insights for crafting targeted prevention and awareness campaigns, potentially leading to improved campaign effectiveness and ultimately, helping more individuals seeking cessation support.\u003c/p\u003e","manuscriptTitle":"Machine Learning Analysis to Identify Factors Associated with Requesting Tobacco Cessation Services Among Users of an Online Self-Diagnostic Questionnaire in Mexico.","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-06 19:28:53","doi":"10.21203/rs.3.rs-4014199/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e263b270-eda7-43a0-b4dd-f295cc8cd376","owner":[],"postedDate":"March 6th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":29127603,"name":"Health Economics \u0026 Outcomes Research"}],"tags":[],"updatedAt":"2024-03-06T19:28:53+00:00","versionOfRecord":[],"versionCreatedAt":"2024-03-06 19:28:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4014199","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4014199","identity":"rs-4014199","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}