Predictors Of Cigarette Smoking Research Articles

Reprint of: Clustering of behavioral economic biases in decision-making and risk for cigarette smoking and other substance use in women and men

BackgroundLow loss aversion (LA) and high delay discounting (DD) are behavioral-economic decision-making biases that independently predict cigarette smoking and other risky substance use. Here we examine (1) whether low-LA and high-DD co-occur, (2) does co-occurrence increase the odds of current smoking and other substance use compared to only low-LA, high-DD, or neither; and (3) potential gender differences in these associations. MethodData are from five studies with U.S. adults who currently smoked or never-smoked cigarettes recruited using online convenience sampling matching on gender and education. Participants completed identical sociodemographic, substance use (cigarette, other drugs, alcohol), and LA (hypothetical 50–50 gambles) and DD (monetary-choice questionnaire) measures. LA and DD scores were dichotomized as low and high using Receiver-Operating-Characteristic Curve logistic regression. ResultsLA and DD each independently predicted substance use and with few exceptions were not influenced by gender. Low-LA compared to high-LA predicted two-fold greater odds of co-occurring high-DD (AOR = 2.120, 95%CI:1.749–2.571, p < .0001). Similarly, high-DD compared to low DD predicted two-fold greater odds of low-LA (AOR = 2.118, 95%CI:1.747–2.568, p < .0001). Among those with co-occurring low-LA and high-DD, odds of substance use were 5–10 times greater than those exhibiting neither, and 2–3 times greater than those exhibiting only low-LA or high-DD. ConclusionsLow-LA and high-DD cluster in women and men such that exhibiting one of these decision-making biases doubles the odds of exhibiting the other. These results demonstrate reliable clustering of low-LA and high-DD and a striking increase in risk for substance use relative to having only one or neither decision-making bias.

Clustering of behavioral economic biases in decision-making and risk for cigarette smoking and other substance use in women and men

BackgroundLow loss aversion (LA) and high delay discounting (DD) are behavioral-economic decision-making biases that independently predict cigarette smoking and other risky substance use. Here we examine (1) whether low-LA and high-DD co-occur, (2) does co-occurrence increase the odds of current smoking and other substance use compared to only low-LA, high-DD, or neither; and (3) potential gender differences in these associations. MethodData are from five studies with U.S. adults who currently smoked or never-smoked cigarettes recruited using online convenience sampling matching on gender and education. Participants completed identical sociodemographic, substance use (cigarette, other drugs, alcohol), and LA (hypothetical 50–50 gambles) and DD (monetary-choice questionnaire) measures. LA and DD scores were dichotomized as low and high using Receiver-Operating-Characteristic Curve logistic regression. ResultsLA and DD each independently predicted substance use and with few exceptions were not influenced by gender. Low-LA compared to high-LA predicted two-fold greater odds of co-occurring high-DD (AOR = 2.120, 95%CI:1.749–2.571, p < .0001). Similarly, high-DD compared to low DD predicted two-fold greater odds of low-LA (AOR = 2.118, 95%CI:1.747–2.568, p < .0001). Among those with co-occurring low-LA and high-DD, odds of substance use were 5–10 times greater than those exhibiting neither, and 2–3 times greater than those exhibiting only low-LA or high-DD. ConclusionsLow-LA and high-DD cluster in women and men such that exhibiting one of these decision-making biases doubles the odds of exhibiting the other. These results demonstrate reliable clustering of low-LA and high-DD and a striking increase in risk for substance use relative to having only one or neither decision-making bias.

Prevalence and predictors of cigarette smoking among people with HIV in Western Jamaica

ABSTRACT With highly active antiretroviral therapy, HIV infection has become a treatable chronic disease. However, modifiable risk factors such as cigarette smoking continue to impact the morbidity and mortality of people with HIV (PWH). We assessed the prevalence and factors associated with cigarette smoking and motivation to quit among PWH in Western Jamaica. A cross-sectional study was conducted in which 392 adults seeking HIV care at health facilities in Western Jamaica completed an interviewer-administered questionnaire. Current smoking prevalence among participants was 17.4%. Current smoking was significantly associated with being male (OR = 2.99), non-Christian/non-Rastafarian (OR = 2.34), living or working with another smoker (aOR =1.86), being moderate to severely depressed (OR = 3.24), having an alcohol drinking problem (OR = 1.84), and never being asked by a healthcare provider if they smoked (OR = 3.24). Among the PWH who currently smoke, 36.7% are moderately to highly dependent on nicotine. One-third of people who smoke (33.8%) started smoking for the first time after HIV diagnosis, while 66.2% initiated smoking before; 88% were willing to quit smoking. These findings provide baseline information for designing and implementing a comprehensive smoking cessation program that considers the needs of PWH in Jamaica, with the potential of becoming a replicable model for other HIV-specialized healthcare settings in the Caribbean.

Loss aversion predicts cigarette smoking status across levels of sociodemographic characteristics.

Loss aversion (LA) is a tendency to be more sensitive to potential losses relative to similar gains. Low LA is associated with increased risk for cigarette smoking and use of other substances. Previous studies of LA and smoking risk controlled for potentially confounding influences of sociodemographic characteristics associated with smoking risk. The present study replicates these earlier observations while also examining the generality of the association between low LA and smoking risk within different levels of each of the five sociodemographic risk factors for smoking (age, educational attainment, gender, income, race/ethnicity). Parallel analyses were conducted using delay discounting (DD) as a positive control; DD is a decision-making bias regarding the rate at which rewards lose value with increasing delay to receipt. Participants were recruited using standard crowdsourcing methods and completed a sociodemographics questionnaire, a hypothetical gamble task measure of LA, and a monetary choice measure of DD. Low LA was associated with increased risk of cigarette smoking after accounting for the influence of sociodemographic characteristics and DD. Similarly, high DD was associated with increased risk of cigarette smoking after accounting for the influence of sociodemographic characteristics and LA. Further analyses showed that associations of LA with smoking risk or DD with smoking risk generally although not always remained significant within varying levels of the sociodemographic characteristics of interest. These results provide support for low LA as a reliable risk factor for smoking that has generality within and across sociodemographic characteristics and closely parallels associations observed with DD and smoking risk. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

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.

Predictors of cigarette smoking in pregnant women with substance use disorders

Background Cigarette smoking is common among pregnant women with substance use disorders (SUD) and may contribute to more adverse health consequences for the infant than alcohol and illicit drug use. However, most studies focused on stopping illicit drug use and paid little attention to cigarette smoking in pregnant women with SUD. Purpose To identify predictors of current smoking among pregnant women with SUD, given past-month psychological distress, alcohol use and illicit drug use, the receipt of past-year mental health and substance use treatment controlling for potential confounders. Methods Secondary analysis of cross-sectional data from the National Survey on Drug Use and Health (NSDUH) 2015–2019 was conducted. The NSDUH included 3,540 pregnant women aged 18–44 years; among them were 195 lifetime smokers with SUD. Multiple logistic regression modeling was used to examine the probability of prenatal smoking. Results Sixty-one percent of pregnant women with SUD reported current cigarette smoking. The likelihood of prenatal smoking increased with a higher level of past-month psychological distress (Adjusted Odds Ratio [AOR] 1.14; 95% Confidence Interval [CI]: 1.02–1.28), past-month illicit drug use (AOR: 5.68; 95% CI: 1.59–20.21), and past-year substance use treatment receipt (AOR: 5.73; 95% CI: 1.88–17.45). Conclusion The receipt of substance use treatment markedly increased the probability of smoking in pregnant women with SUD. Treatment and policy initiatives are required to address and integrate cigarette smoking within other substance use treatment modalities for pregnant women with SUD.

Negative Affect and Cigarette Cessation in Dual Users of Cigarettes and Electronic Nicotine Delivery Systems

Objectives: Dual use of cigarettes and electronic nicotine delivery systems (ENDS) is increasingly common in adult smokers, who often report using ENDS to quit smoking. Elevated negative affect is an established predictor of increased difficulty quitting smoking combustible cigarettes but has not yet been examined in the context of cigarette cessation for dual users. Method: This study examined whether mood-related factors predict cigarette smoking cessation among dual users (N = 364) over 12-months. Self-reported cigarette smoking at 12 months, with abstinence defined as no smoking for the past 7 days, was the primary outcome variable. Logistic regression included baseline levels of depression symptoms (CES-D), anxiety symptoms (MASQ), and negative affect expectancies for smoking, with baseline nicotine dependence for cigarettes (NDSS), motivation to quit, age, race/ethnicity, rate of cigarette smoking at baseline, and ENDS usage at baseline and 12 months as covariates. Interactions between CES-D, MASQ, and negative affect expectancies were examined. We predicted that negative affect, especially for smokers who had high negative affect expectancies for smoking, would be negatively associated with quitting. Results: Contrary to expectations, negative affect constructs did not predict quitting. Baseline nicotine dependence for cigarettes, gender, and race/ethnicity significantly predicted the likelihood of cigarette cessation. Higher rates of ENDS use, higher motivation, and lower negative affect smoking expectancies were significantly correlated with quitting cigarettes. Conclusion: In this non-treatment seeking sample of dual users, negative affect did not predict cigarette cessation over and above nicotine dependence for cigarettes, gender, and race/ethnicity.

Prevalence of Tobacco Use and the Factors Affecting It: A Cross-sectional Analysis of Baseline Data From the Bandare Kong Cohort Study in Southern Iran

Background: Tobacco use (TU) has been increasing in Iran and turned into a major national healthcare concern. Therefore, the present research aimed to explore the prevalence of TU and its predictors in Bandar Kong in the south of Iran. Materials and Methods: The baseline data from the Bandare Kong Cohort Study were used in a prospective study of 4,035 subjects (40-70 years old) in Hormozgan province from 2016 to 2018. Adjusted and crude odds ratios (OR) were used at 95% confidence interval (CI) to determine the predictors of TU. Results: Hookah smoking (HS) was found to be the most prevalent type of smoking (17.60%, 95% CI: 16.41-18.78%), followed by second-hand smoking (17%, 95% CI: 15.81-18.14%) and cigarette smoking (9%, 95% CI: 8.11-9.89%), respectively. Age, gender, education, alcohol consumption, and environmental tobacco smoke (ETS) were the main predictors of cigarette and HS. Marital status, place of residence, and psychological problems were the predictors of HS (OR = 0.1.65, 95% CI: 1.11, 2.45, P = 0.01), and body mass index (BMI) (OR = 0.95, 95% CI: .93, 0.98, P = 0.002) and drug abuse (OR = 4.52, 95% CI: 3.11, 6.55, P &lt; 0.001) were the predictors of cigarette smoking. Conclusion: Male gender, higher age, rural residence, low education, alcohol consumption, drugs, second-hand smoking, and history of depression were among the main risk factors of smoking behavior. The results indicated a considerable need for more educational and preventive programs for each risk factor involved in smoking behavior especially for HS, exposure to ETS, and its consequences in the population of Bandar Kong.

Cohort Study of Cannabis Use History and Perinatal Cigarette Use Among Overweight and Obese Women.

In non-pregnant populations, cannabis use and cannabis use disorder (CUD) have been linked to tobacco use and less successful quit attempts. We compared perinatal cigarette use in women across 3 groups: never used cannabis (No CU group); used cannabis but did not meet CUD criteria (CU group); history of CUD (CUD group). Interviews with 257 pregnant women with overweight/obesity (M age = 28years; 52% white) were conducted for a study of eating behavior in Western Pennsylvania from 2012-2016. Tobacco use was assessed early in pregnancy (< 20weeks gestation), late in pregnancy (34-38weeks gestation) and 6months postpartum. CUD was measured with the Structured Clinical Interview for DSM-IV (SCID). Data relevant to the proposed analyses were available for 252 women. Generalized mixed effect models were used to predict perinatal cigarette use based on cannabis use group, time and their interaction, adjusting for age, race, education, income, parity, and mood/anxiety disorder. Forty-eight percent of participants reported prior cannabis use and 15% were diagnosed with lifetime CUD. History of cannabis use predicted cigarette smoking in early pregnancy (OR 11.12, CI 3.27-37.85), late pregnancy (OR 6.55, CI 1.70-25.27), and 6months postpartum (OR 7.57, CI 2.72-21.07), regardless of CUD. A history of CUD did not appear to confer additional risk for perinatal cigarette use. Given increasing rates of cannabis use among pregnant women, these results highlight the importance of addressing history of cannabis use in conjunction with tobacco use to improve smoking cessation efforts.

Effects of Coping Skills and Personality Traits on Nicotine Dependence

Objectives: Objectives of this study were: 1) to compare coping skills and personality between smokers and non-smokers, 2) to examine correlations among coping styles, personality, and nicotine dependence, and 3) to determine predictors of nicotine dependence. Methods ： A descriptive correlation design was used. Data were collected from 137 smokers and 60 nonsmokers between September 2013 and February 2017 in Incheon, South Korea. The Eysenck Personality Inventory, the Nicotine Dependence Syndrome Scale, and the revised Multidimensional Coping Scale were used. Results ： Smokers were less likely to cope actively (p<0.001) and calm their emotions (p=0.013). However, they were more likely to express their emotions directly (p=0.031) than non-smokers. Moreover, smokers scored higher on personality dimensions of criminality, empathy, impulsiveness, and addiction but scored lower on the lie personality trait than non-smokers. Nicotine dependence was positively correlated with personality traits (such as addiction [r=0.329, p<0.001], criminality [r=0.314, p<0.001], and impulsiveness [r=0.366, p<0.001]) and coping skills such as perseverance (r=0.589, p<0.001). Smokers in the family, addiction personality trait, active coping, emotional expression, positive interpretation, downward comparison, and acceptance explained 73.2% of cigarette smoking. Conclusion： Results of this study revealed that smokers used more unhealthy coping skills with more vulnerable traits than non-smokers. In addition, coping skills and vulnerable personality traits could predict cigarette smoking.

Predicting Tobacco Use Based on the Theory of Planned Behavior in Adolescent Girls

Introduction: Tobacco use poses a serious threat to global health. The tobacco epidemic is spreading in low- and middle-income countries, where the tobacco industries mainly target young people and women. Objective: This study aims to determine the predictors of tobacco use based on the theory of planned behavior (TPB) in female high school students. Materials and Methods: This cross-sectional study was conducted on 340 female high school students aged 14-18 years in Rasht City, Iran, who were selected using a multi-stage sampling method. The data collection tool was the TPB questionnaire for tobacco use in Persian. The obtained data were analyzed using the Mann-Whitney U test, Kolmogorov-Smirnov test, Spearman correlation test and logistic regression analysis. Results: The TPB constructs of attitude (OR = 0.91, 95% CI: 0.85-0.97, P = 0.007) and behavioral intention (OR=0.75, 95% CI; 0.64-0.88, P = 0.001) could predict cigarette smoking. Also, constructs of perceived behavioral control (OR = 0.92, 95%CI; 0.87-0.97, P = 0.007), subjective norms (OR = 0.83, 95%CI; 0.70-0.97, P = 0.02), and behavioral intention (OR = 0.88, 95% CI; 0.78 - 0.99, P = 0.03) could predict hookah smoking in girls students. Conclusion: Considering the predictive power of the TPB constructs in tobacco use by adolescent girls, it is possible to plan to reduce its rate in this group by a focus on the TPB constructs.

Genetically Predicted Cigarette Smoking in Relation to Risk of Polycystic Ovary Syndrome.

BackgroundEvidence from observational studies has suggested a link between cigarette smoking and the risk of polycystic ovary syndrome (PCOS). However, it remains uncertain whether the observed relationship is causal or due to biases inherent in observational studies. Therefore, we adopted two-sample Mendelian randomization (MR) design to assess the potential causal association between smoking and the risk of PCOS.MethodsSummary level data of PCOS was obtained from a genome-wide association study (GWAS) meta-analysis including 4,138 cases and 20,129 controls of European ancestry. Single-nucleotide polymorphisms (SNPs) associated with smoking initiation (n=360) were selected and used as genetic instrumental variables (IVs). MR analysis was performed using inverse-variance weighted (IVW) method, supplemented with the likelihood-based method, weighted median method, MR pleiotropy residual sum and outlier (MR-PRESSO) test, and MR-Egger regression.ResultsGenetically predicted smoking initiation was associated with an increased risk of PCOS in the primary analysis (odds ratio (OR) =1.38, 95% confidence interval (CI) =1.12–1.69). MR-Egger regression did not detect the horizontal pleiotropy. Sensitivity analyses using alternative MR methods and restricted IVs produced similar results.ConclusionOur study provided evidence to support a potential causal association between smoking initiation and an increased risk of PCOS, providing a better understanding of the etiology and prevention of PCOS. Further studies are warranted to clarify the underlying biological mechanisms of smoking in the development of PCOS.

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.

The smoking paradox: exploring why attitudes toward cigarette smoking are a weak predictor of cigarette smoking

Objective An attitude change perspective represents a central line of research in the smoking literature. This line of research focuses on influencing individuals’ attitudes toward smoking in order to alleviate the negative outcomes of smoking and to achieve positive smoking behaviours. This line of research, however, does not explain why attitudes toward cigarette smoking are a weak predictor of cigarette smoking, an indication of a smoking paradox. The current study introduces two domain-general factors to explore this smoking paradox. Design We used an online survey and collected data from American subjects (n = 328) via a crowdsourcing platform, Amazon Mechanical Turk, in 2018. Results We found that the association between attitude toward smoking and the likelihood of smoking is significantly weaker when individuals perceive a low probability of risk and/or believe that this probable risk will only occur in the distant future, after many years. These results suggest that smoking may not be consistent with attitude because individuals perceive low probability of risk and/or they temporally discount the occurrence of this probable risk. Conclusion The results suggest that novel strategies that incorporate the concepts of perceived risk probability and temporal discounting of the probable risk are needed for smoking cessation.

Cross-lagged effects of posttraumatic stress disorder symptom severity and cigarette smoking among OEF/OIF/OND veterans.

Veterans with posttraumatic stress disorder (PTSD) are known to smoke cigarettes at elevated levels in comparison to both veterans without PTSD and civilians. This study aims to elucidate how cigarette smoking and PTSD symptoms interact over time. This study examined the directionality and strength of the relationship between average daily cigarette smoking and PTSD symptom severity across three (T1-T3) time points in a large cohort (N = 851) of male and female Operation Enduring Freedom/Operation Iraqi Freedom/Operation New Dawn returning veterans who were either current or former smokers at T1 (mean age = 37.56; standard deviation = 10.10). We used cross-lagged panel analyses to evaluate their temporal relations. The analyses indicated that PTSD symptom severity at T1 significantly predicted cigarette smoking at T2, and this predictive association was maintained from T2 to T3. Conversely, smoking at T1 and T2 did not predict PTSD symptom severity at T2 and T3, respectively. Although effect sizes were small, PTSD symptom severity was cross-sectionally related to smoking at T1 and T2, but not T3. In addition, when analyses were examined by gender, the same results were found except these associations were stronger for women than for men cross-sectionally. Our findings provide some evidence of a longitudinal association between PTSD symptom severity and tobacco use and highlight potential targets of intervention.

The Prototypes of Tobacco Users Scale (POTUS) for Cigarette Smoking and E-Cigarette Use: Development and Validation.

Endorsing prototypes of cigarette smokers predicts cigarette smoking, but less is known about prototypes of users of other tobacco products. Our study sought to establish the reliability and validity of a measure of prototypes of smokers and e-cigarette users. Participants were from a national survey of smokers and non-smokers (n = 1414), a randomized clinical trial (RCT) of adult smokers (n = 2149), and adolescent children of adults in the trial (n = 112). The Prototypes of Tobacco Users Scale (POTUS) has four positive adjectives (cool, sexy, smart, and healthy) and four negative adjectives (disgusting, unattractive, immature, and inconsiderate) describing cigarette smokers and e-cigarette users. Confirmatory factor analyses identified a two-factor solution. The POTUS demonstrated strong internal consistency reliability in all three samples (median α = 0.85) and good test–retest reliability among adults in the RCT (median r = 0.61, 1–4 weeks follow-up). In the RCT, smokers more often agreed with negative prototypes for smokers than for e-cigarette users (mean = 2.03 vs. 1.67, p < 0.05); negative prototypes at baseline were also associated with more forgoing of cigarettes and making a quit attempt at the end of the trial (Week 4 follow-up). The POTUS may be useful to public health researchers seeking to design interventions that reduce tobacco initiation or cessation through the manipulation of tobacco user prototypes.

Theory of reasoned action and antismoking social marketing

There is a range of behavior change models used in social marketing. The past studies suggest thattheories can serve as valuable frameworks for the design and evaluation of health interventions. Thisarticle aims to provide a revision of theoretical models used in social marketing and health promotion,specifically Theory of Reasoned Action, and test this theory on health risk behavior change. Smokingbehavior among youth was selected as a specific context for this study due to importance of the healthof youth for the economic and social development of Kazakhstan and opportunity to prevent or reducesmoking behaviors and subsequently mitigate negative consequences of smoking among youth. A deductiveapproach was undertaken and quantitative survey method was applied. The total sample aftercleaning for incomplete data consisted of 2307 respondents. The study found that the Theory of ReasonedAction successfully predicted cigarette smoking behavior among youth. The TRA could providesolid explanations of intention to uptake cigarette use. The results indicated that attitude toward cigarettesmoking and subjective norm directly influenced intention to use cigarettes. These findings mightusefully be integrated in an intervention program aimed at reducing or preventing smoking initiation,especially among youth. Social marketing health intervention programs are recommended to be directedat the attitude and subjective norms by targeting underlying beliefs.

The association of HIV and cocaine use to cigarette smoking in the context of depression and perceived stress

ABSTRACT Mental well-being can contribute to cigarette smoking and negatively impact disease progression among people living with HIV (PLWH). This study examined potential associations between cocaine use (COC), depression, and HIV status in predicting cigarette smoking; hypothesizing that depression would be highest in cocaine users and predict cigarette smoking. An exploratory analysis including stress was also examined as a potential predictor of cigarette use. More than half of the sample (65%) endorsed smoking at some point in the past, and 52% endorsed being current smokers at the time of the study. Smokers were most likely to be cocaine users (87.1%), cocaine using PLWH (74.3%), or PLWH (36.6%). Neither HIV status (χ 2(1) = 1.5, p = .221), perceived stress (χ 2(1) = 0.75 p = .386), nor depressive symptomatology (χ 2(1) = 1.2, p = .274) were related to smoking. Non-cocaine users were approximately 95.4% less likely to smoke than cocaine users, controlling for all other variables. Overall, cocaine use was the greatest predictor of cigarette smoking and quantity of cigarettes smoked. Perceived stress and depression were not associated with cigarette smoking in the sample. Future interventions targeting cigarette use should include a cocaine-related component to encourage smoking cessation among this high-risk group.

Investigating gateway effects using the PATH study.

A recent meta-analysis of nine cohort studies in youths reported that baseline ever e-cigarette use strongly predicted cigarette smoking initiation in the next 6-18 months, with an adjusted odds ratio of 3.62 (95% confidence interval 2.42-5.41). A recent review of e-cigarettes agreed there was substantial evidence for this "gateway effect". However, the number of confounders considered in the studies was limited, so we investigated whether the effect might have resulted from inadequate adjustment, using Waves 1 and 2 of the Population Assessment of Tobacco and Health study. Our main analyses considered Wave 1 never cigarette smokers who, at Wave 2, had information available on smoking initiation. We constructed a propensity score for ever e-cigarette use from Wave 1 variables, using this to predict ever cigarette smoking. Sensitivity analyses accounted for use of other tobacco products, linked current e-cigarette use to subsequent current smoking, or used propensity scores for ever smoking or ever tobacco product use as predictors. We also considered predictors using data from both waves to attempt to control for residual confounding from misclassified responses. Adjustment for propensity dramatically reduced the unadjusted odds ratio (OR) of 5.70 (4.33-7.50) to 2.48 (1.85-3.31), 2.47 (1.79-3.42) or 1.85 (1.35-2.53), whether adjustment was made as quintiles, as a continuous variable or for the individual variables. Additional adjustment for other tobacco products reduced this last OR to 1.59 (1.14-2.20). Sensitivity analyses confirmed adjustment removed most of the gateway effect. Control for residual confounding also reduced the association. We found that confounding is a major factor, explaining most of the observed gateway effect. However, our analyses are limited by small numbers of new smokers considered and the possibility of over-adjustment if taking up e-cigarettes affects some predictor variables. Further analyses are intended using Wave 3 data which should avoid these problems.

Role of cigarette smoking in the development of ischemic stroke and its subtypes: a Mendelian randomization study.

PurposeNumerous studies have indicated that smokers have an increased risk of developing ischemic stroke. However, less is known about the causal relationship between cigarette smoking and ischemic stroke subtypes. In the present study, we aim to determine whether genetically predicted cigarette smoking was associated with subtypes of ischemic stroke using Mendelian randomization (MR).Patients and methodsWe used summary-level genetic association data from the MEGASTROKE consortium, including 438,847 individuals of European ancestry (34,217 cases of ischemic stroke and 404,630 controls). We used 176 single nucleotide polymorphisms as instrumental variables, which were previously identified to be associated with smoking in the Study of the Social Science Genetic Association Consortium (n=518,633). MR analyses were performed using inverse-variance-weighted method, weighted-median method, and MR-Egger regression.ResultsWe found that genetically predicted smoking was associated with a higher risk of ischemic stroke (odds ratio (OR): 1.24, 95% CI: 1.10–1.39) and large artery ischemic stroke (OR: 1.52, 95% CI: 1.14–2.02), but not with risk of cardioembolic ischemic stroke or small vessel ischemic stroke. Sensitivity analyses using alternative MR approaches produced similar results.ConclusionGenetic predisposition toward smoking is causally associated with a higher incidence of large artery ischemic stroke. Further work is warranted to clarify the underlying mechanism of smoking in the development of large artery ischemic stroke.