Recent Presentation Abstracts

Recent Presentations by the Faculty of the Penn/Georgetown TTURC--Presentation Abstracts

 

Physical activity and smoking in adolescents: A pilot study.

Audrain J, Lerman C, Main D

The 30-day prevalence of cigarette smoking among U.S. high school students increased from 27.5% in 1991 to 36.4% in 1997. Understanding factors that deter smoking are key to informing prevention efforts. A few studies have found that participation in sports/exercise is negatively related to smoking. However, the type or level of activity has not been adequately defined. The goal of this pilot investigation was to determine whether the type (e.g., sports teams, strength training) and/or intensity (i.e., hard, moderate) of physical activity is related to adolescent smoking behavior. Self-report physical activity items were derived from the Youth Risk Behavior Survey. The sample included 38 male and female high school students (M age 16; 54% white). Results revealed that adolescents who did not smoke in the past 30 days were more likely to engage in hard activity over the past week than adolescents who smoked in the past 30 days (t (36)=3.28, p=.002; 5 days versus 2.5 days). Although marginally significant, adolescents who did not smoke in the past 30 days engaged in two times the strength training activity compared to adolescents who smoked in the past 30 days (t (36)=1.99, p=.054; 3.2 days versus 1.6 days). Days engaged in moderately intense activity and the number of sports teams played in the past 12 months were unrelated to smoking (p values >.2). The preliminary results suggest that youth smoking prevention programs may want to evaluate whether strategies to increase more intense physical activity has an impact on smoking prevention.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2000.

Which adolescents are most receptive to tobacco industry marketing? Implications for counter-advertising campaigns.

Audrain J, Lerman C, Tercyak KP, Shields AE, Bush A, Espinel CF

This study sought to identify adolescents most receptive to tobacco advertising based on individual differences in novelty-seeking personality and other key variables. Confidential self-report surveys were completed by 1 ,071 high school freshmen at 5 public high schools. The survey included validated measures of novelty-seeking personality, smoking habits, peer and family smoking, and tobacco advertising recep- tivity. Multiple logistic regression analysis was used to evaluate the independent asso- ciations of these variables and demographics with receptivity to tobacco advertising. Of the ninth graders, 44% had moderate to high levels of advertising receptivity and 54% had minimal to low levels of receptivity. Higher levels of receptivity were associ- ated with ever smoking (OR = 2.59, CI = 1.99-3.39) and novelty-seeking personality (OR = 2.14, CI = 1.57-2.93). The association of novelty-seeking personality and tobacco advertising receptivity was most pronounced among adolescents who had never had a puff of a cigarette. Counter-advertising messages should consider indi- vidual differences in novelty-seeking, since novelty-seekers may be most receptive to tobacco industry promotional campaigns.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2002.

Applying a behavioral economic framework to understanding adolescent smoking.

Audrain-McGovern J, Tercyak KP, Epstein L, Wileyto, EP

Adolescents’ choice to smoke may depend on alternative reinforcers or substitutes for smoking, complementary activities to smoking, and individual differences in reinforcer value. The goal of this study was to determine the influence of these variables on smoking among 1, 012 adolescents. Substitutes were school involvement, academic performance, physical activity, and sports team participation; complements were peer smoking and substance use; delayed discounting assessed individual differences in reinforcer value. Logistic regression showed that adolescents who had ever smoked had higher levels of complementary reinforcers, lower levels of substitute reinforcers, and higher levels of delayed discounting than adolescents who had never smoked. Current smokers were more likely to have higher levels of complementary reinforcers accompanied by higher levels of delayed discounting. Results suggest adolescents who smoke may have fewer alternative reinforcers that protect against smoking, more complementary reinforcers that promote smoking, and greater discounting of future rewards.

Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Recruiting adolescents into genetic studies of smoking behavior.

Audrain J, Tercyak KP, Goldman P, Bush A

The goal of this study is to describe the process of establishing a longitudinal cohort to study genetic, psychological, and social predictors of adolescent smoking. Parents of eligible adolescents were approached for their consent via mail. Among the parents who responded, 75% (1, 151) provided written consent. Demographic and smoking history data were compared between parents who agreed to have their teens participate in the project and those who declined. Compared to parents who consented to their teens' participation, parents who declined were less educated (89% had greater than a high school education compared to 69% of those who did not provide consent), less likely to be Caucasian (68% versus 48%), and less likely to report having ever even experimented with smoking (71% versus 600;.). The most frequently reported reasons parents gave for declining.consent included lack of inter- est, confidentiality concerns, time, and that their teen does not smoke. A logistic regression model predicting consent to participate revealed a significant race by edu- cation interaction, indicating that among Caucasian parents, those with an education beyond high school were over two times more likely to provide consent compared to Caucasian parents with a high school education or less (OR = 2.43; CI = 1.37 -4.32, p=.003). Suggestions to promote study accrual are provided.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2002.

Psychosocial risk factors associated with ever and current smoking among adolescents.

Audrain J, Tercyak KP, Goldman P, Smith A, Herrera J

The overall prevalence of current cigarette smoking among U.S. high school students was about 35% in 1999. Understanding factors that are associated with ever smoking and current smoking are key to informing prevention and interven. tion efforts. To determine factors associated with different levels of smoking, 990 high school freshmen (63% white; 52% female) completed a baseline survey. The survey measured smoking habits, school involvement, physical activity, sub- stance use, and novelty seeking and reward dependence personality. Two com- parisons of interest were evaluated in logistic regression models. ever smokers versus never smokers; and (2) current smokers versus nonsmokers. Results revealed that compared to never smokers, ever smokers had lower grades (OR=1.54), greater lifetime alcohol (OR=3.06) and marijuana use (OR=12.56), and higher levels of novelty seeking (OR=1.10). Compared to nonsmokers, cur- rent smokers were less likely to play on sports teams (OR=.55), had lower grades (OR=1.76), greater lifetime alcohol (OR=2.27) and marijuana use (OR=8.88), higher levels of novelty seeking (OR=1.10, CI =1.02- 1.18), and lower levels of reward dependence (OR=.91 ). These results suggest adolescents who smoke have fewer sources of non-substance related reinforcing experiences. These results also suggest that adolescents who smoke may derive little satisfaction from areas of life that may protect against the use of tobacco.

Poster Presentations at the Society for Research on Nicotine and Tobacco Annual Meeting, 2001.

The impact of poor self-control and good self-control on adolescent smoking.

Audrain-McGovern J, Tercyak KP, Moss H

This study investigated the impact of exposure to others who smoke, good and poor self-control, and their interactions, on adolescent smoking. Study participants were 1,060 10th graders who completed self-report measures of family and peer smoking, good self-control, poor self-control, and cigarette smoking (ever and current). The results of multivariate logistic regression analyses indicated that smoking exposure, good self-control, and poor self-control were independent predictors of ever smoking. Smoking exposure and good self-control were also independent predictors of current smoking. There was no evidence to suggest that good self-control and poor self-control moderated the effect of smoking exposure on ever or current smoking. However, there was a significant interaction between good and poor self-control that influenced the odds of an adolescent being a current smoker. Specifically, good self-control appeared to buffer the negative effects of poor self-control, such that adolescents were 70% less likely to currently smoke. Future research should investigate the relationship between self-control variables, smoking, and other environmental influences across time. Youth smoking prevention and intervention program outcomes may potentially improve by addressing self-control behaviors.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Tobacco treatment evaluation: Achieving transdisciplinary integration.

Baker T, Hatsukami D, Lerman C, O’Malley S, Shields A, Fiore M

Transdisciplinary research involves the use and integration of measures and methodologies across diverse response systems and levels of analysis (e.g., subcellular to social) and optimally, integration is achieved at a theoretic as well as a methodologic level. Transdisciplinary research on tobacco cessation treatment poses special challenges since treatments may be evaluated across many dimensions or facets. For instance, treatments may be evaluated with respect to their attractiveness to the target population, their impacts on process measures, as well as their impacts on proximal and ultimate outcomes. Moreover, these effects may be influenced by individual differences and contextual factors. This complexity allows ample opportunity for the integration of diverse research disciplines, with the integration yielding new measurement models, new constructs to be targeted, or new synthetic theories. In this presentation a model is offered that can be used to guide and appraise treatment evaluation, and explore opportunities for transdisciplinary collaboration and integration. The individual talks will address how different Transdisciplinary Tobacco Use Research Centers (TTURC’s) attempt to adopt a transdisciplinary approach to tobacco treatment research. Dorothy Hatsukami, University of Minnesota TTURC, will address research aimed at harm reduction treatment approaches. Caryn Lerman, University of Pennsylvania, TTURC will address genetic predictors and mechanisms of response to bupropion treatment. Stephanie O’Malley, Yale University, will address naltrexone treatment as it affects smoking and alcohol use and their reciprocal relations. Finally, Timothy Baker, University of Wisconsin TTURC, will discuss attempts to validate a transdisciplinary model of nicotine dependence using treatment outcome data.

Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003

Gender differences in smoking relapse in a behavioral counseling + placebo-controlled Bupropion trial.

Collins BN, Niaura R, Wileyto EP, Patterson F, Brown RA, Audrain J, Hawk L, Epstein L, Lerman C

Last year we presented preliminary data from this double-blind, placebo-controlled randomized trial of bupropion. Preliminary data suggested gender differences in ces- sation at end of 10-week treatment (EOT) and a non-significant trend suggesting a potential interaction with buproprion. The present analyses include 886 smokers' data recruited from three sites. Controlling for site, both logistic and survival analysis mod- els demonstrated significant main effects for gender, drug, and pre-treatment nicotine dependence (FTND) a gender x drug interaction; and a gender x drug x FTND inter- action. The 2-way interaction suggests that through EOT, the drug effect is more pro- nounced for females (buproprion reduces relapse rates) than for males (males do well on drug or placebo). The 3-way interaction suggests the following. For low-to- moderate nicotine dependent smokers, females show the greatest drug effects whereas for high nicotine dependent smokers, males demonstrate the highest drug effects. This suggests that females benefit from buproprion, particularly if they demonstrate low-to-moderate pre-treatment FTND. Conversely, men appear to ben- efit most from buproprion when they have high pre-treatment FTND. Clinical implica- tions of these data will be presented..

Society for Research on Nicotine and Tobacco Annual Meeting, 2002.

Effects of Bupropion on depression symptoms in highly dependent smokers.

Collins BN, Wileyto EP, Rukstalis M, Pinto A, Hawk L, Niaura R, Epstein L, Lerman C

Depression symptoms are frequent in smokers and can increase the likelihood of smoking relapse. Bupropion, an anti-depressant, is an effective smoking cessation aid. Although initial studies failed to find an effect of bupropion on clinical depression symptoms in smokers, other studies have suggested its effects on negative affect are important for promoting abstinence. This analysis examined the effect of bupropion on non-clinical depression symptoms (using the CES-D) among 497 smokers participating in a placebo-controlled randomized trial. Linear regression models of changes in depression symptoms at the end of treatment and 6-month follow-up revealed significant treatment x dependence interaction effects (p<.05); however, the pattern of results was different at each stage. During the treatment phase, all groups exhibited decreases in depression symptoms, but the effects of bupropion were significant only for highly dependent smokers. During the post-treatment phase, depression symptoms in all groups remained relatively unchanged, with the exception of the highly dependent smokers on bupropion who showed a significant increase in symptoms. These results suggest bupropion attenuates depression symptoms in the short-term among highly dependent smokers. Longer-term treatment may be needed for this group to address depression symptoms and prevent relapse.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Structural disparities between smokers and neversmokers in select brain regions.

Franklin T, Gray J, Listerud J, Aleman D, Lerman C, O’Brien C, Childress AR

Neuroimaging techniques offer the opportunity to search for the brain substrates of addiction and relapse vulnerability. Here we used voxel based morphometry (VBM) to determine whether, and where, structural disparities exist between the brains of smokers (N=18) and a group of demographically-matched neversmokers (N=15; P<.001, uncorrected). This method utilizes statistical parametric software (SPM’99) and high-resolution structural magnetic resonance images (MRIs) to provide a measure of the difference between groups of neuronal tissue within a particular voxel or cluster of voxels. Less gray matter was observed in smokers in the midbrain (region of the pedunculopontine tegmental nucleus [PPTg] and ventral tegmental area [VTA]), the anterior cingulate and the left dorsolateral prefrontal cortex. More white matter was observed in a similar area of the midbrain, in the left dorsolateral prefrontal cortex and in the area of the right amygdala. The VTA and PPTg are part of a neuronal circuit activated by self-administered nicotine and other reinforcing substances. Deficiencies in neural matter in these regions could result in low dopaminergic tone in the downstream nucleus accumbens, the substrate strongly implicated as the final common pathway in drug addiction. Whether these deficits are a result of nicotine exposure, pre-existing, or a consequence of both is not clear. Nonetheless, individuals with low dopaminergic tone may continue to smoke in an attempt to normalize dopaminergic tone and thus may be more susceptible to nicotine dependence.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Genetic, shared and nonshared environmental influences on selected smoking behaviors in twins from the Midus Survey.

Hardie T, Moss H, Lynch K, Lerman C

Essential to the improvement in treatment efforts is an understanding of the etiology of smoking behaviors. In order to understand the heritable and environmental influences on smoking behavior, we used biometrical methods to analyze responses to smoking-related questions by twin pairs from the Midlife Development in the United States (MIDUS) epidemiological study. MIDUS is a collaborative, interdisciplinary investigation of midlife development. Respondents were drawn from a nationally representative random-digit-dial sample, aged 25-74, selected from telephone banks. Results of our biometrical analyses of monozygous (n=222) and dizygous (n= 329) twins indicated that 76% of the variability in having ever smoked on a regular basis was due to additive genetic effects and 24% to nonshared environmental effects. Similarly, current regular smoking was explained by 73% additive genetic effects and 27% nonshared/unique environmental effects. These observations confirmed previous reports of the heritability of smoking behavior as approximately 70%. We also examined models for the age of the very first cigarette, which suggested a additive gene effect and effects of the nonshared/unique environment. Despite being the best explanatory model, the model fit was suboptimal. Similarly, a model examining the greatest number of cigarettes smoked in a smoking year was best explained by dominant genetic influences and the nonshared/unique environment, however the model fit was suboptimal. Finally, we tested a model to evaluate the variability in making a quit attempt. The best fit model explained 99% of the variability in responses as being due to the effects of the unshared/unique environment, with no influence by genetic or shared environmental factors. These findings demonstrate the importance of extrafamilial environmental factors in attempting to quit smoking. Thus, interventions, and public health efforts have the potential to have potent effects on smoking behavior.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

ADHD symptoms and smoking patterns among participants in a smoking cessation program.

Kaufmann VG, Rose C, Audrain J, Bush A, Crystal-Mansour S, Tercyak KP, Hawk LW, Niaura R, Epstein L, Lerman C

Previous research has suggested an increased liability to smoking among indi-viduals with attention-deficit hyperactivity disorder (ADHD). This link is thought tobe attributable, in part, to nicotine's effects on attention and performance. Weexamined the association of ADHD symptoms of inattention and hyperactivity withsmoking behavior in a sample of 226 male and female smokers ages 18 and olderwho were enrolled in a smoking cessation program. Prior to treatment, they com-pleted measures of ADHD symptoms and standardized measures of smoking pat-terns. Hierarchical linear regression models were used to characterize the smok-ing patterns associated with ADHD inattention and hyperactivity symptoms, con-trolling for potential confounder variables. Smoking for stimulation purposes andthe urge to smoke to minimize withdrawal symptoms were the primary patternsassociated with ADHD inattention symptoms, while hyperactivity symptoms werenot associated with smoking patterns. Consistent with a self-medication hypothe-sis, these results suggest that smokers with frequent symptoms of inattentionmay use nicotine as a stimulant drug to help manage these symptoms. Theheightened urge to smoke to alleviate withdrawal symptoms further suggests thatthese individuals may be more prone to relapse following smoking cessation.Future studies of the role of inattention symptoms in response to smoking treat-ment are warranted.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2001.

Differential response to nicotine replacement therapies in obese and non-obese smokers.

Lerman C, Audrain J, Patterson F, Kaufmann V, Wileyto EP, Kucharski S, Jepson C

Obesity and smoking are both major risk factors for cardiovascular disease and cancer. Despite this, surprisingly little research has investigated nicotine dependence and treatment outcome among obese smokers. There is some evidence to suggest that obesity and nicotine addiction have a common genetic basis that relates to the drive for reinforcement (from food or drugs of abuse). Thus, we compared the efficacy of two forms of nicotine replacement therapy that differ in terms of their reinforcing properties, among 72 obese (body mass index (BMI)>30) and 183 non-obese smokers. The study design was an open-label randomized clinical trial comparing transdermal nicotine (TN) to nicotine nasal spray (NS). Both treatments were delivered over an 8-week period in conjunction with behavioral smoking cessation group counseling. Baseline assessments included: demographic factors, smoking history, and depression symptoms. Smoking status was assessed and verified at the end of treatment (EOT) and at 6-month follow-up. Among non-obese individuals, verified point-prevalence abstinence rates at EOT were 46% for TN and 28% for NS (p=.01). For obese individuals, the EOT abstinence rates were 35% for TN and 43% for NS (p>.10). In a logistic regression model of abstinence at EOT (controlling for smoking rate, sex, race, and education), there was a significant group (BMI < vs > 30) by treatment interaction effect (p=.03). Data collection for the six-month follow-up and analyses of mediating mechanisms (e.g., usage, percent replacement) are underway. These initial results indicate a clear advantage for TN over NS for non-obese individuals. Among obese smokers, response to TN and NS is variable, and at least some obese individuals may have a better treatment outcome with NS.

Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Genetics and response to Buproprion.

Lerman C, Epstein L, Wileyto EP, Audrain J, Hawk L, Pinto A, Roth D, Shields PG, Niaura R

Animal and human twin studies have established that smoking, in part, is heritable. Studies have found associations of smoking with genes involved in the regulation of dopamine and serotonin; however, this has not been replicated in all studies. These results have laid the foundation for pharmacogenetic studies that examine whether specific genetic markers can be used to predict treatment response. Research on bupropion has found an association between the dopamine transporter gene and smoking. Based on data from 405 Caucasian smokers who participated in two placebo- controlled trials, we examined whether this and other genes moderate the effective- ness of bupropion. Since one hypothesized mechanism is the reduction of nicotine reinforcement, we conducted analyses in the total sample, as well as a subset of 301 early lapsers, with end-of-treatment 7-day point prevalence the primary endpoint. Comparisons of treatment effects by dopamine transporter genotype do not support an overall effect. However, among participants who lapsed, the effect of bupropion v. placebo was small, but significant, only in smokers who had at least one 9 allele. No significant effect was found lor the moderating effect of the CYP2B6 gene, which codes for an enzyme that metabolizes bupropion. However, among females, absti- nence rates were 23% for placebo, 54% for bupropion group with wild type genotype, and 63% for bupropion group with a mutation (poor metabolizers). Although this effect is small and of questionable clinical significance, it is intriguing because females have been found to metabolize bupropion more rapidly than males. These data will be aug- mented with additional genotypes and longer term follow-up data.

Society for Research on Nicotine and Tobacco Annual Meeting, 2002

Morning plasma luteinizing hormone concentrations in pre- adolescent males predicts regular smoking in mid-adolescence.

Moss H, Cornelius M, White L, Yao, J

Objective: The onset of puberty is marked by a surge in circulating concentrations of the gonadotropins, luteinizing hormone (LH) and follicle-stimulating hormone (FSH). Prior research has suggested that early onset of puberty can be associated with behavioral deviancy, aggression, and conduct disturbances. Consequently, we examined early morning fasting plasma FSH, LH, testosterone concentrations and pubertal staging in 10-12 year old boys (n= 264) who were evaluated for tobacco use behaviors five years later in mid-adolescence. Method: Fasting plasma samples were collected at 7:30 AM from the boys after an overnight stay at the research center, and assayed for FSH, LH, and testosterone concentrations. A nurse evaluated pubertal staging. At the five-year follow-up assessment, subjects were categorized as being “regular smokers”, “occasional smokers” or “non-smokers” based on the frequency of smoking behavior. Logistic regression analysis was conducted to determine the predictive effects of illicit paternal substance abuse, gonadotropins, testosterone and pubertal stage on mid-adolescent smoking status. Results: Regular smoking in mid-adolescence was significantly associated with greater morning plasma LH concentrations five years earlier (OR= 4.20, p<.001). Effects of baseline age, FSH, testosterone, paternal substance abuse or pubertal stage were non-significant. Conclusion: The mechanism through which elevated plasma LH is associated with the liability for smoking behavior is unclear. However, the results are consistent with an earlier onset of pubertal maturational processes being associated with adolescent smoking risk.

Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003

Individual differences in nicotine intake per cigarette.

Patterson F, Benowitz N, Shields P, Kaufmann V, Jepson C, Wileyto EP, Kucharski S, Lerman C

The increase in levels of blood nicotine that occurs from smoking a single cigarette, sometimes referred to as a “nicotine boost,” is an individualized measure of how much nicotine has been extracted from smoking a cigarette. As such, nicotine boost is a key indicator of tobacco smoke exposure and a possible risk factor for tobacco-related disease. This study investigated the demographic, smoking status, and psychological predictors of nicotine boost in the context of an ongoing clinical trial of nicotine replacement therapy. Participants were 95 male and 95 female treatment-seeking smokers who reported smoking at least 10 cigarettes a day. To assess nicotine boost, participants provided a blood sample before and after smoking one of their own brand cigarettes. Positive affect (mood) was a significant positive predictor (p= .03) of nicotine boost, controlling for baseline cotinine levels and cigarette brand (FTC) nicotine delivery. Among males, significant predictors included smoking rate (p=.005) and positive affect (p= .02). Among females, only race predicted boost (p= .001), with African Americans exhibiting higher boost levels than Caucasians. These results suggest that positive mood states may lead to increases in overall nicotine intake, particularly among men. Future studies examining the role of nicotine boost could lead to individually tailored harm-reduction interventions.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Effects of repeated smoking of Eclipse cigarettes.

Pickworth WB, Malson JL, Lee EM, Moolchan ET

In April 2000, R J R Reynolds introduced a new smoking device, Eclipse., that the company claimed delivered lower levels of smoke than conventional cigarettes. Eclipse. uses a carbon fuel-element to vaporize substances in the rod that are inhaled by the user. In this report, the effects of Eclipse. on smoking topography, sub- stance delivery factors, physiologic subjective and biochemical markers of smoking were compared to conventional cigarettes. Ten adult smokers (8 men) smoked ad lib on four occasions: Eclipse. twice,through the mouthpiece of the topography unit and handheld; own brand twice, through the mouthpiece of the topography unit and hand- held. Measures were collected before and up to 1 hr alter smoking. Sessions were separated by at least 24 hr. There were no apparent differences between smoking through the topography mouthpiece and conventional (hand-held)smoking. Compared to their own brand of cigarette, the Eclipse. took longer to smoke (366 vs 292 sec) and needed more puffs (14.8 vs 10.8) and caused a larger increase in exhaled CO (7.3 vs. 4.2 ppm). Puff volume (90.7 vs. 63.0 ml) and puff velocity (81.6 vs 58.2 mVsec) were greater alter the Eclipse. than the usual brand; interpuff inter- val and puff duration were similar. Cardiovascular effects were similar alter Eclipse and the usual brand. Subjects rated the Eclipse. as less satisfying (3.1 vs 5.5) and tasteful (2.6 vs. 4.9) and the subjects liked the Eclipse. less (2.4 vs. 5.2) than their usual brand. The results of this study indicate that acute exposure to the Eclipse. device exposes the user to significant quantities of CO and possibly other harmful components of tobacco smoke. The results further validate the use of topography as a quantitative index of smoke exposure.

Society for Research on Nicotine and Tobacco Annual Meeting, 2002.

The impact of depression and physical activity on smoking progression.

Rodriguez D, Audrain-McGovern J

Research supports a negative relationship between physical activity and smoking, and a positive relationship between depression and smoking. Depression has also been found to relate negatively to physical activity. These relationships have not been evaluated longitudinally, particularly in the context of ethnicity and gender. We assessed the longitudinal relationships of smoking progression and physical activity with depression, gender, and ethnicity as predictors of rate of change in both. Participants were 1060 adolescents participating in a longitudinal study of the genetic and non-genetic determinants of smoking adoption. Parallel process linear growth modeling was conducted with data collected at three time points (grades 9-10). Results supported the fit of the model to the data. Smoking progression increased and physical activity decreased over time. Depression predicted change in both physical activity and smoking progression, indicating that higher levels of depression related to greater smoking progression and less physical activity. Physical activity did not predict smoking progression. Males and white adolescents exhibited higher smoking progression and physical activity. The results for physical activity on smoking progression are puzzling. Because Physical Education (PE) was mandatory for 9th and 10th graders, it is possible that participation in PE masked the effects of voluntary physical activity on smoking progression. Future studies should account for different types of physical activity. If these preliminary findings are replicated, it may help inform youth prevention and intervention efforts by suggesting the need to address depression and physical activity in smoking progression, as well as gender and ethnic differences.

Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

The effect of puff volume on carbon monoxide exposure in 1mg tar and light cigarettes.

Strasser A, Kozlowski L

When the filter vents of low-yield, heavily-ventilated cigarettes (e.g., 1mg tar) are blocked, smokers produce greater carbon monoxide (CO) boosts than when filter vents are not blocked. Prior research found that vent blocking on Light cigarettes does not increase CO boost.When smoking 1mg tar cigarettes, some smokers produce relatively larger CO boosts compared to other smokers. Smoking topography variables (puff volume, puff velocity, and puff duration) were investigated as potential sources of individual differences in CO boost.Participants completed four cigarette conditions: Light unblocked, Light blocked, 1mg tar unblocked, 1mg tar blocked under controlled smoking conditions (8 puffs, 45 second interpuff interval), and had smoking topography and CO boost measured.CO boost for the unblocked and blocked Light cigarettes were 4.5 ppm and 6.8 ppm. CO boost for the unblocked and blocked 1mg tar cigarettes were 0.9 ppm and 3.4 ppm. Main effects (cigarette type and blocking) were statistically significant, but the interaction effect was not significant meaning the difference in CO boost between unblocked and blocked did not differ by cigarette type. This differs from results in previous research (e.g., Sweeney and Kozlowski, 1998; Sweeney et al., 1999). Smokers took significantly larger, faster, and longer puffs on 1mg tar cigarettes compared to the Light cigarettes; and on unblocked cigarettes compared to blocked cigarettes. Smoking topography measures did not explain individual differences in CO boost.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Comorbid relationship among depression, substance use, and level of smoking during adolescence.

Tercyak KP, Audrain J, Goldman P, Smith A, Herrera J, Zack S

Depression may be associated with increased risk in smoking. This studyexam- ined the relationship among depression, substance use (lifetime exposure to al- cohol, marijuana), and level of smoking in adolescents. Participants were 986 high school freshmen. Level of smoking was: (a) never smoker= never tried or experimented with smoking, even a few puffs (n = 599; 61%); (b) ever smoker = smoked at least a partial or whole cigarette but not within the past 30 days (n = 281; 28%); and (c) current smoker = smoking at least one cigarette within the past 30 days (n = 106; II %). Logistic regression models tested two contrasts: (I) never smokers vs. ever smoker/current smokers; (2) ever smokers vs. current smokers. Models controlled for the effects of gender, race, and history of depres- sive symptoms and had p values < .05. Compared to never smokers, a positive lifetime history of alcohol (OR = 3.84, CI = 2.80, 5.27) and marijuana (OR = 14.88, CI = 8.83,25.08) nse, as well as greater depressive symptomatology (OR = 1.67, CI= 1.15,2.42), were associated with an increase in the risk of being an ever/current smoker. Compared to ever smokers, current smokers were more likely to have a positive lifetime history of marijuana use (OR = 4.31, CI = 2.57, 7.22) and greater depressive symptoms (OR = 1.83, CI = 1.02,3.29). Pediatric smoking and substance use may be attempts to alleviate depressive symptoms; smoking interventions should address these symptoms and their impact u

Society of Behavioral Medicine Annual Meeting, 2001

The influence of dopamine receptor genes (DRD2, DRD3 & DRD4) on smoking status: Smokers and Non-Smokers Study (SANS).

Vandenbergh D, Grant M, O’Connor R, Strasser A, Severns V, Vogler G, Kozlowski L

Smoking and smoking-related behaviors have long been considered multi-factorial, polygenic outcomes. The difficulty of smokers in cessation is attributed to nicotine and nicotine’s action results in activation of the mesolimbic dopamine reward pathway.
DRD2, DRD3, and DRD4 encode proteins that respond to the neurotransmitter dopamine in the brain. DRD2 and DRD3 have simple tandem repeat polymorphisms that confer four and six common alleles respectively. The DRD4 has a variable number of tandem repeats (VNTR) in the third cytoplasmic loop. A sub-sample of adult volunteers (recruited by random-digit dialing) provided DNA from buccal swabs that was subsequently genotyped for the three dopamine receptor genes of interest. Analysis of variance was utilized to determine effects of DRD2, DRD3, and DRD4 on various smoking-related behaviors. Preliminary analysis indicates that the DRD2-80 allele is associated with trouble concentrating during withdrawal, while the 86 allele is associated with trouble sleeping during withdrawal. For DRD3, the 125 allele is higher with increasing Heaviness of Smoking Index; the 131 allele is associated with age of initiation,
and the 139 allele is associated with feeling down/depressed during withdrawal. Finally, DRD4 shows no statistically significant association to smoking status or smoking-related behaviors in this sample. Sample size is insufficient to analyze gene-interaction effects due to the large number of alleles at DRD2 and DRD3.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.

Do small lapses predict relapse to smoking behavior under buproprion treatment?

Wileyto EP, Audrain J, Collins B, Kaufman V, Pinto A, Niaura R, Hawk L, Epstein L, Lerman C

While buproprion is known to be an effective aid in smoking cessation, less is known about its mode of action. This analysis tested whether buproprion prevents small lapses «7 days, followed by a return to abstinence) from leading to full relapses (7 contiguous days of smoking). Participants were 303 smokers who enrolled in a placebo-controlled clinical trial of buproprion. This question was approached using Cox-regression with time-varying covariates, which allowed us to separate the effects of lapses during treatment and follow-up, and look at interactions of those lapses with drug treatment. Overall, 87% of participants reported a lapse within the 10 week treatment period, while 46% reported a full relapse. Small lapses predicted a relapse (Hazard Ratio = 7.63, P=O.00O). However, lapses that occurred while taking bupro- prion appeared to lead to relapse at the same rate as lapses in the placebo group. Also, lapses that occurred following treatment led to relapse at the same rate as those occurring during treatment. Thus, while buproprion is an effective smoking cessation aid, it does not appear to prevent lapses from leading to relapses

Society for Research on Nicotine and Tobacco Annual Meeting, 2002.

Time to event analysis of Bupropion clinical trial data.

Wileyto EP, Niaura R, Audrain J, Kaufmann V, Epstein L, Lerman C

This paper reports an analysis of participants from two clinical trials of bupropion as an aid for smoking cessation, examining the effects of sex, drug, and several baseline survey measures on the pattern of smoking behavior during the trial and follow-up period. 1079 smokers who had passed the 6 month follow up yielded 9556 self-reported transitions, including 3936 lapse or relapse events, and 3491 recoveries from lapses. Separate time to event models (Cox regression) were fitted for each of the transition types, assuming multiple failures and clustering on subject. The important predictors for lapses were Sex (HR 1.22, p=0.004), Treatment Phase (In treatment versus follow-up, HR 1.20, p=0.014), Drug x Treatment Phase (HR 0.80, p=0.047), and the Drug x FTND score (HR 0.73, p=0.018). Results indicate that women lapse more readily than men, and that bupropion uniformly prevents lapses across both sexes. Predictors important for recovery were Drug x Sex (HR 1.27, p=0.030), Treatment Phase (HR 1.49, p=0.000), and Drug x FTND score (HR 0.70, p=0.013). Results highlight asymmetries in factors responsible for lapse versus recovery, that bupropion differentially affects the ability of men and women to recover from lapses, while uniformly preventing lapses.

Poster Presentation at the Society for Research on Nicotine and Tobacco Annual Meeting, 2003.