Predictors Of Smoking Behavior Research Articles

Utilizing Machine Learning to Forecast Smoking Behavior

Abstract: The examination of smoking habits, extensively studied over time, has posed challenges in accurately predicting and thoroughly analyzing its determinants. Previous research efforts struggled to precisely predict smoking behavior due to the presence of continuous target variables, which hindered the application of crucial feature selection techniques like mutual information. This study aims to tackle these hurdles through an innovative approach that integrates data preprocessing, feature engineering, and advanced machine learning methods. To address the issue of continuous target variables, our methodology involves categorizing smoking behavior into discrete groups, enabling the utilization of feature selection techniques such as mutual information scores. Logistic regression, Gaussian Naive Bayes, and Random Forest Classifier models are utilized to achieve highly accurate predictions of smoking behavior. The Select KBest method is employed to evaluate the importance of features based on mutual information scores. The investigation delves into various health markers, including BMI, haemoglobin levels, and cholesterol, offering comprehensive insights into their influence on smoking habits. Furthermore, Principal Component Analysis (PCA) is implemented to reduce dimensionality while preserving essential information from the dataset. Through this novel approach and a steadfast commitment to ethical data collection practices, our objective is to advance the comprehension of smoking behavior, surmounting past challenges, and providing valuable insights for public health initiatives and smoking cessation endeavors. The paper assesses outcomes using specified algorithms and parameters, presenting a comparative analysis to enhance the clarity and reliability of our findings.

Predicting Smoking Status of a Person using Machine Learning

Abstract: The study of smoking behavior, a topic extensively researched over time, has presented difficulties in accurately predicting and thoroughly analyzing its determinants. Previous studies struggled to predict smoking behavior accurately, mainly due to the presence of continuous target variables, hindering the application of vital feature selection methods like mutual information. This research aims to address these challenges through an innovative approach that incorporates data preprocessing, feature engineering, and advanced machine learning techniques. To overcome the issue of continuous target variables, our methodology involves categorizing smoking behavior into discrete groups, allowing the use of feature selection methods such as mutual information scores. Logistic regression, Gaussian Naive Bayes, and Random Forest Classifier models are employed in this study to achieve highly accurate predictions of smoking behavior. The Select KBest method is utilized to assess the significance of features based on mutual information scores. The research explores various health indicators, including BMI, haemoglobin levels, and cholesterol, providing comprehensive insights into their impact on smoking behavior. The principal component analysis, or PCA, is another technique used to lower multiplicity. while retaining essential information from the dataset. Through this innovative approach and a rigorous commitment to ethical data collection practices, our goal is to advance the understanding of smoking behavior, overcoming previous challenges, and offering valuable insights for public health initiatives and smoking cessation efforts. The study evaluates results using specified algorithms and parameters, presenting a comparative analysis to enhance the clarity and robustness of our findings.

Implicit beliefs and automatic associations in smoking

Background and objectivesDual process models of addiction suggest that controlled, goal-directed processes prevent drug-use, whereas impulsive, stimulus-driven processes promote drug-use. The most frequently used measure of automatic smoking-related processes, the implicit association test (IAT), has yielded mixed results. We examine the validity of two alternative implicit measures: 1) the affect misattribution procedure (AMP), a measure of automatic evaluations, and 2) the relational responding task (RRT), a measure of implicit beliefs. MethodsSmokers and non-smokers performed smoking-related versions of the AMP and the RRT and filled in questionnaires for smoking dependence. Smokers participated in two sessions: once after they just smoked, and once after being deprived for 10 h. Smokers also kept a smoking diary for a week after the second session. ResultsWe found significant differences between smokers and non-smokers on the RRT, t (86) = 2.86, p = .007, d = 0.61, and on the AMP, F (1, 85) = 6.22, p = .015, pƞ2 = 0.07. Neither the AMP nor the RRT were affected by the deprivation manipulation. Smoking dependence predicted smoking behavior in the following week; the AMP and RRT did not explain additional variance. LimitationsPossibly, our manipulation was not strong enough to affect the motivational state of participants in a way that it changed their implicit cognitions. Future research should examine the sensitivity of implicit measures to (motivational) context. ConclusionsWe found limited evidence for the validity of the smoking-AMP and the smoking-RRT, highlighting the need for a critical view on implicit measures.

A hybrid machine learning approach in predicting smoking behaviour: The case of Turkey

This study aims to analyze the smoking behaviour of people aged 15 and older in Turkey using supervised and unsupervised machine learning methods. In this study, C4.5 and Random Forest (RF) were trained to predict smoking behaviour, and an apriori algorithm was used to detect associations. Sensitivity, specificity, accuracy, positive predicted value (PPV), and f-measure were used to compare the performances of the supervised models. The Turkey Health Interview Survey 2019 was used with a sample size of 17084 to predict smoking behaviour and determine the factors affecting smoking. Data analysis and performance evaluation were performed with R programming language by RStudio. By association rules, gender, age, and alcohol consumption are the most representative attributes of smoking behaviour. Associations were determined on smoking, non-smoking and quit-smoking behaviour. Also, it has been seen that the RF algorithm has better results than the C4.5 algorithm. It’s preferred to use the RF model, which had better performance with an accuracy of 0.909, a specificity of 0.965, a sensitivity of 0.782, a PPV of 0.908, and an f-measure of 0.840 for predicting smoking behaviour. This study contributes to the literature covering the most comprehensive national health survey data and using machine learning methods on this data in Turkey. Also, it indicates that machine learning methods can be used to analyze such datasets.

Smoking and healthcare expenditure reductions associated with the California Tobacco Control Program, 1989 to 2019: A predictive validation.

Previous research used data through 2008 to estimate a model for the effect of the California Tobacco Control Program (CTCP) that used cumulative real per capita tobacco control expenditure to predict smoking behavior (current adult smoking prevalence and mean cigarette consumption per current smoker). Predicted changes in smoking behavior due to the CTCP were used to predict its effect on health care expenditure. This research updates the model using the most recently available data and estimates CTCP program effect through 2019. The data used in the previous research were updated, and the original model specification and a related predictive forecast model were re-estimated. The updated regression estimates were compared to those previously published and used to update estimates of CTCP program effect in 2019 dollars. There was no evidence of structural change in the previously estimated model. The estimated effect of the CTCP program expenditures on adult current smoking prevalence and mean consumption per adult current smoker has remained stable over time. Over the life of the program, one additional dollar per capita of program expenditure was associated with a reduction of current adult smoking prevalence by about 0.05 percentage point and mean annual consumption per adult current smoker by about 2 packs. Using updated estimates, the program prevented 9.45 (SE 1.04) million person-years of smoking and cumulative consumption of 15.7 (SE 3.04) billion packs of cigarettes from 1989 to 2019. The program produced cumulative savings in real healthcare expenditure of $544 (SE $82) billion using the National Income and Product Accounts (NIPA), and $816 (SE $121) billion using the Center for Medicare and Medicaid Services (CMS) measure of medical costs. During this time, the CTCP expenditure was $3.5 billion. A simple predictive model of the effectiveness of the CTCP program remained stable and retains its predictive performance out-of-sample. The updated estimates of program effect suggest that CTCP program has retained its effectiveness over its 31-year life and produced a return on investment of 231 to 1 in direct CMS medical expenditure.

Prevalence and predictors of smoking behavior in Azorean adolescents

1. World Health Organization. Tobacco. WHO; 2022. Accessed December 29, 2022. https://www.who.int/en/news-ro... Google Scholar

Development and validation of a questionnaire based on the protection motivation theory to predict cigarette smoking preventive behavior in the Iranian university students.

Prevention of smoking is very important at a young age and during the student period. Because of the lack of a questionnaire based on the prevention of smoking behavior, this study was conducted to design and standardize a tool according to the protection motivation theory (PMT). This was an analytical cross-sectional study where information and opinions of experts and then designing tools and assessing the face validity in the first phase and content validity, construct validity, test-retest reliability, and internal consistency of questions in the second phase were examined. The initial themes and structures were determined based on the literature Review and the opinion of experts. Three hundred students of Yazd Universities were selected by cluster sampling and included in the study. All analyses were performed in SPSS and AMOS software version 24 with a significance level of less than 5%. The final version of the scale had 39 questions in eight dimensions. The overall Cronbach's alpha value was 0.89, and for each dimension, it was more than 0.7. The content validity ratio and content validity index for the whole tool were 0.87 and 0.94, respectively. The results of confirmatory factor analysis showed that the latent construct model had good fit (X2/df = 4.31; RMSEA = 0.07; CFI = 0.94; NFI = 0.92; TLI = 0.91, P < 0.001). In this study, a valid and reliable tool based on PMT to predict smoking behavior in students was introduced. This tool can be used in interventional and etiological studies. It also has questions related to the coronavirus disease 2019 (Covid 19) virus and can be used in the coronavirus pandemic.

Modifying outcome expectancies and behavioral reinforcers to induce quit attempts among young adult smokers

Research has shown that outcome expectancies predict smoking behavior, and expectancy challenge interventions can reduce smoking. This study tested the hypothesis that supplementing expectancy challenge with a behavioral activation intervention promoting increased exposure to alternative reinforcers would help increase motivation to quit and induce actual quit attempts within the following month among young adult smokers. Smokers, aged 18–35 (N = 159, of whom 93 provided one-month follow-up data) and not required to be interested in quitting at the point of enrollment, were randomized to (a) the combined cognitive-behavioral intervention, (b) a cognitive-only expectancy challenge, or (c) a neutral reading (control) task. There were no significant between-group effects on motivation, but the conditions differed significantly in likelihood of leading to a quit attempt. Post hoc comparisons showed the combined condition participants (52 % of completers) to be significantly more likely than control group participants (25 %) to make a quit attempt, with the cognitive-only group (43 %) intermediate and not significantly different from the other conditions. These data provide preliminary evidence of utility for the brief cognitive-behavioral intervention in promoting smoking cessation attempts; more research is needed to test various possible explanations of how and for whom the intervention is effective, as well as whether it would significantly exceed expectancy challenge alone in larger samples.

An Assessment of Racial and Ethnic and Sexual Identity Centrality and Smoking Behavior

ABSTRACT Although evidence indicates that identity centrality, or identity importance, can serve as a positive coping mechanism regarding well-being, less is known if it can also buffer against health risk behaviors like cigarette smoking. This study uses an intercategorical intersectional approach using data from 1,571 Black and Latino/a sexual and gender minority adults in the Social Justice Sexuality Project to assess the relationship between sexual and racial and ethnic identity centrality and smoking patterns. Relative risk ratios from multinomial logistic regressions highlight three findings. First, there is no evidence of a significant association between identity centrality and smoking behavior nor evidence of a significant interaction effect between racial and ethnic and sexual identity centrality. Once models were adjusted for education, the association between centrality and smoking was no longer significant. Second, results indicate that education, gender identity, familial support and outness were significant predictors of smoking behaviors. Third, results suggest that there are significant differences across the intersection of race and ethnicity and sexual identity in relative risk of smoking. In addition, findings highlight elevated risk of engaging in more casual behaviors of smoking as opposed to heightened smoking behavior among both Black and Latino/a sexual minority adults.

Studying the Utility of Using Genetics to Predict Smoking-Related Outcomes in a Population-Based Study and a Selected Cohort.

The purpose of this study is to examine the predictive utility of polygenic risk scores (PRSs) for smoking behaviors. Using summary statistics from the Sequencing Consortium of Alcohol and Nicotine use consortium, we generated PRSs of ever smoking, age of smoking initiation, cigarettes smoked per day, and smoking cessation for participants in the population-based Atherosclerosis Risk in Communities (ARIC) study (N = 8638), and the Collaborative Genetic Study of Nicotine Dependence (COGEND) (N = 1935). The outcomes were ever smoking, age of smoking initiation, heaviness of smoking, and smoking cessation. In the European ancestry cohorts, each PRS was significantly associated with the corresponding smoking behavior outcome. In the ARIC cohort, the PRS z-score for ever smoking predicted smoking (odds ratio [OR]: 1.37; 95% confidence interval [CI]: 1.31, 1.43); the PRS z-score for age of smoking initiation was associated with age of smoking initiation (OR: 0.87; 95% CI: 0.82, 0.92); the PRS z-score for cigarettes per day was associated with heavier smoking (OR: 1.17; 95% CI: 1.11, 1.25); and the PRS z-score for smoking cessation predicted successful cessation (OR: 1.24; 95% CI: 1.17, 1.32). In the African ancestry cohort, the PRSs did not predict smoking behaviors. Smoking-related PRSs were associated with smoking-related behaviors in European ancestry populations. This improvement in prediction is greatest in the lowest and highest genetic risk categories. The lack of prediction in African ancestry populations highlights the urgent need to increase diversity in research so that scientific advances can be applied to populations other than those of European ancestry. This study shows that including both genetic ancestry and PRSs in a single model increases the ability to predict smoking behaviors compared with the model including only demographic characteristics. This finding is observed for every smoking-related outcome. Even though adding genetics is more predictive, the demographics alone confer substantial and meaningful predictive power. However, with increasing work in PRSs, the predictive ability will continue to improve.

Smoking Behavior among Female Worker in Indonesia Does Education Matter?

Smoking behavior in women has a risk of fertility problems, the reproductive system, and many otherhealth problems. This study was aimed at analyzing the effect of education level on smoking behavioramong female workers in Indonesia. The samples used were a female worker. The sample size was 51,815respondents. Apart from education, other independent variables analyzed were the place of residence, age,marital, wealth, and pregnant status. Determination of determinant by binary logistic regression. The resultsof the study informed that female workers with primary education were 0.456 times more likely than noeducation to become smokers. Female workers with secondary education are 0.428 times more likely thanno education to become smokers. Female workers with higher education are 0.219 times more likely thanno education to become smokers. The results of this analysis indicate that the higher the education level, thelower the possibility of female workers becoming smokers. Apart from the education level, 4 other variableswere also found as predictors of smoking behavior among female workers, namely the place of residence,age, wealth, and pregnant status. It was concluded that the education level was a predictor of smokingbehavior among female workers in Indonesia. The higher the education level, the lower the possibility offemale workers becoming smokers.

Smoking Behavior and Moral Identity in Malaysian Youth: Is Gender a Moderator or a Predictor?

The role of gender in psychosocial development and health outcome is undeniable. This study aims to investigate the role of gender in two conditions, whether it as a moderator to the link between moral identity (internalization and symbolization) and smoking behavior; or if it is a single predictor or a co-predictor with moral identity in predicting smoking behavior. To examine this, 388 youths (61% female) who met the inclusion criteria participated in this online cross-sectional study. Participants were asked to rate their emotional responses ranging from unpleasant to pleasant whilst viewing images of smoking behavior. Participants were also instructed to provide their socio-demographic information and fill in the Moral Identity Scale. Gender had no moderation effect on the link between moral identity (symbolization and internalization) and smoking behavior, instead, gender (whether alone or in combination with internalization) was shown to directly predict smoking behavior. The implications for social health interventions and programs are highlighted by the findings.

Personality Traits, Identity style and Depression as predictors of smoking behavior among university students

Present work was aimed to find the contribution of personality traits, identity style and depression in predicting cigarette smoking. Data was collected from male students of universities located in Peshawar through Beck Depression Inventory – II (BDI – II), NEO-Five Factor Inventory (NEO-FFI) and Identity Style Inventory, Revised (ISI3). A total of 300 male students were involved, out of which n=150 were smokers whereas n=150 were nonsmokers. Of the total n=150 smokers, 75 were hostelers 75 were day scholars and same as for non-smokers. The outcome indicated that those who smoked had high levels of neuroticism and they also scored high on the Beck Depression Inventory. The outcome further revealed that persons having diffused identity style had likelihood of being involved in smoking behavior. Further results exposed that the hostelers were more vulnerable to smoking behavior as compared to the day scholars.

Predicting smoking behavior: intention and future self-continuity among Austrians

ABSTRACT Smoking is a major public health problem in Austria, but relevant research and intervention is limited. Based on the Theory of Planned Behavior (TPB), the present study aimed to test the model utility in an Austrian sample. As future self-continuity is likely to impact on health behavior, we also hypothesized an extended TPB with future self-continuity could further explain the variance in smoking. Using a prospective design, 94 current smokers (74.5% women; M age = 24.27 years; 61.7% daily smokers) from a university in South Austria completed the baseline and follow-up survey (one month after the baseline). Consistent with the TPB, intention and perceived behavioral control (PBC) significantly predicted smoking behavior; affective attitude and PBC were significantly associated with smoking intention. In contrast, cognitive attitude and subjective norm were unrelated to smoking intention. As hypothesized, the TPB explained 42% variance of smoking behavior and 31% variance of smoking intention. Participants' future self-continuity further explained the variance of smoking behavior. Our study demonstrates the utility of the TPB in understanding Austrian smoking behavior. The role of psychological perception of future self among smokers has been highlighted. Future smoking intervention may target PBC, affective attitude, as well as a life-span perspective.

Anxiety in social interactions and nicotine dependence in nicotine-dependent men: The role of metacognitions about smoking

Anxiety in social interactions is an important factor in cigarette use and nicotine dependence. Metacognitions about smoking have been found to predict smoking behavior and may help understand the relationship between anxiety in social interactions and nicotine dependence. In the current study, we evaluated the direct effect of anxiety in social interactions on nicotine dependence and its indirect effect through metacognitions (controlling for anhedonia and depression) in nicotine-dependent men (n = 388). Participants completed measures of anxiety in social interactions [the Social Interaction Anxiety Scale (SIAS)], anhedonia [the Snaith HamiltonPleasure Scale (SHAPS)], metacognitions about smoking [e.g., theMetacognitions aboutSmoking Questionnaire (MSQ)] nicotine dependence [the Fagerström Test for Nicotine Dependence (FTND)], and clinical factors related to smoking including depressive symptoms [e.g., the Beck Depression Inventory-II (BDI-II)]. As expected, after controlling for depressive symptoms and anhedonia, anxiety in social interactions indirectly affected nicotine dependence through negative metacognitions about smoking, but not positive metacognitions. These findings are discussed in relation to the metacognitive model of addictive behaviors.

Coffee and cigarettes: Examining the association between caffeinated beverage consumption and smoking behaviour among youth in the COMPASS study.

In adults, coffee, sugar-sweetened beverage (SSB) and high energy drink consumption have been related to increases in risky behaviour, including smoking. However, these associations are not well understood during adolescence. The purpose of this study was to examine the association between beverage consumption and smoking behaviour among Canadian adolescents. Using data from the COMPASS study (2016-2017; n = 46,957), four models were developed to investigate whether beverage consumption explained variability in smoking behaviour in adolescence (age = 15.7 ± 1.2 yrs); 1) smoking status; 2) e-cigarette use status; 3) days smoking cigarettes per month; and 4) days using an e-cigarette per month. Models were adjusted for demographic factors. Logistic (models 1 and 2) and ordinal logistic (models 3 and 4) were used for analysis. An association between the frequency of SSBs, coffee/tea or high energy drinks consumption and smoking behaviour was identified in all models. Greater beverage consumption was associated with being a current smoker (OR = 2.46 (2.02, 2.99)), former smoker, (OR = 2.50 (1.53, 4.08)), and currently using an e-cigarette (OR = 4.66 (3.40, 6.40)). Higher beverage consumption was also associated with more days smoking/using an e-cigarette per month (OR = 2.67 (1.92, 3.70) and 3.45 (2.32, 5.12), respectively). High energy drink consumption on 4 or 5 days of the school week was the best predictor of smoking behaviour in all models. Given the health consequences of smoking and e-cigarette use and their association with SSB, high energy drinks and coffee consumption, policy initiatives to prevent smoking initiation and limit access to these beverages needs ongoing attention and implementation.

Demographics and medical disorders associated with smoking: a population-based study

BackgroundFew studies have investigated factors associated with smoking behaviors. In this population-based study, we investigated demographics and medical comorbid diseases to establish a prediction model for smoking behaviors by using the National Health Interview Survey (NHIS) and National Health Insurance Research Database (NHIRD).MethodsWe enrolled individuals aged ≥40 years who had participated in the NHIS in 2001, 2005, and 2009. We identified the smoking behaviors of the study participants in the NHIS. Smoking behaviors were divided into ever smokers (current smokers and ex-smokers) and nonsmokers (never smokers).We defined medical comorbid disorders of the study participants by using medical claim data from the NHIRD. We used multivariable logistic regression models to calculate the adjusted odds ratio and 95% confidence interval for variables associated with smoking. The significant variables in the multivariable model were included in the receiver operating characteristic curves (ROC) to predict the sensitivity and specificity of the model.ResultsIn total, 26,375 participants (12,779 men and 13,596 women) were included in the analysis. The prevalence of smoking was 39.29%. The mean ages of the 16,012 nonsmokers were higher than those of the 10,363 smokers (57.86 ± 12.92 years vs. 53.59 ± 10.82 years). Men outnumbered women among smokers (68.18% vs. 31.82%). Male sex, young age and middle age, being insured categories, residence in suburban areas, and chronic obstructive pulmonary disease (COPD) were independent factors associated with smoking. The area under the ROC curve of these significant factors to predict smoking behaviors was 71.63%.ConclusionSex, age, insured categories, residence in suburban areas, and COPD were associated with smoking in people.

Prediction of Smoking Behavior From Single Nucleotide Polymorphisms With Machine Learning Approaches.

Smoking is a complex behavior with a heritability as high as 50%. Given such a large genetic contribution, it provides an opportunity to prevent those individuals who are susceptible to smoking dependence from ever starting to smoke by predicting their inherited predisposition with their genomic profiles. Although previous studies have identified many susceptibility variants for smoking, they have limited power to predict smoking behavior. We applied the support vector machine (SVM) and random forest (RF) methods to build prediction models for smoking behavior. We first used 1,431 smokers and 1,503 non-smokers of African origin for model building with a 10-fold cross-validation and then tested the prediction models on an independent dataset consisting of 213 smokers and 224 non-smokers. The SVM model with 500 top single nucleotide polymorphisms (SNPs) selected using logistic regression (p<0.01) as the feature selection method achieved an area under the curve (AUC) of 0.691, 0.721, and 0.720 for the training, test, and independent test samples, respectively. The RF model with 500 top SNPs selected using logistic regression (p<0.01) achieved AUCs of 0.671, 0.665, and 0.667 for the training, test, and independent test samples, respectively. Finally, we used the combined logistic (p<0.01) and LASSO (λ=10−3) regression to select features and the SVM algorithm for model building. The SVM model with 500 top SNPs achieved AUCs of 0.756, 0.776, and 0.897 for the training, test, and independent test samples, respectively. We conclude that machine learning methods are promising means to build predictive models for smoking.

275. Exposure to Smokers and Positive Influencing Behavior as Predictors of Smoking Behavior Among Adolescents

275. Exposure to Smokers and Positive Influencing Behavior as Predictors of Smoking Behavior Among Adolescents

Development and Validation of Smoking Abstinence Self-efficacy Scale among Iranian Female Adolescents

Background: Among different factors, self-efficacy is one of the most suitable components for predicting smoking behavior among adolescents. Therefore, it is essential to design a native tool to measure this high-risk behavior in adolescents and implement prevention programs in this age group. Aim: This study aimed to design and validate a smoking abstinence self-efficacy questionnaire in female adolescents. Method: In this methodological study, a comprehensive literature review was conducted, and the available similar questionnaires were used to prepare the items. In total, 351 female schoolchildren were selected by convenience sampling in Mashhad, Iran, between October 2017 and January 2018. Face and content validity and construct validity were applied for psychometric evaluation of the developed tool. The reliability of the tool was tested by internal consistency. Results: The final version of the smoking abstinence self-efficacy scale in Iranian female adolescents includes 16 items after a psychometric evaluation. The content validity of the tool was obtained at 0.99. Moreover, exploratory factor analysis explored three factors of emotional, social, and high-risk situations explaining 60% of the variance. Cronbach's alpha coefficient of the questionnaire was estimated at 0.94. Implications for Practice: The smoking abstinence self-efficacy scale is a valid and reliable tool among Iranian female adolescents derived from the native context of this group of people which can examine smoking abstinence self-efficacy in three dimensions of emotional, social, and high-risk situations. Moreover, it can identify adolescents with different levels of smoking abstinence self-efficacy.