Published online April 3, 2006
PEDIATRICS Vol. 117 No. 4 April 2006, pp. 1216-1225 (doi:10.1542/peds.2005-0808)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Audrain-McGovern, J.
Right arrow Articles by Cuevas, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Audrain-McGovern, J.
Right arrow Articles by Cuevas, J.
Related Collections
Right arrow Adolescent Medicine

How Do Psychological Factors Influence Adolescent Smoking Progression? The Evidence for Indirect Effects Through Tobacco Advertising Receptivity

Janet Audrain-McGovern, PhD, Daniel Rodriguez, PhD, Vaishali Patel, BA, Myles S. Faith, PhD, Kelli Rodgers, BA and Jocelyn Cuevas, BA

Department of Psychiatry, University of Pennsylvania, Philadelphia, Pennsylvania


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
OBJECTIVES. To determine whether novelty seeking and depressive symptoms had mediated or indirect effects on adolescent smoking progression through tobacco advertising receptivity.

METHODS. More than 1000 adolescents were monitored from 9th grade to 12th grade and completed annual surveys that measured demographic characteristics, smoking behavior, tobacco advertising receptivity, novelty-seeking personality, depressive symptoms, family and peer smoking, alcohol use, and marijuana use.

RESULTS. Latent growth modeling indicated that novelty seeking had a significant indirect effect on smoking progression through baseline tobacco advertising receptivity. For each 1-SD increase in novelty seeking, the odds of being more receptive to tobacco advertising increased by 12% (ie, being in a specific category or higher), which in turn resulted in an 11% increase in the odds of smoking progression from 9th grade to 12th grade. The indirect effect from depressive symptoms to smoking progression did not reach significance.

CONCLUSIONS. These findings may inform future research on other factors that influence tobacco advertising receptivity, as well as programs aimed at preventing adolescent smoking initiation and progression.

Key Words: smoking • tobacco advertising • adolescents

Abbreviations: CI—confidence interval • OR—odds ratio • CES-D—Center for Epidemiologic Studies Depression Scale • CFI—comparative fit index • WRMR—weighted root mean residual • RMSEA—root mean square error of approximation • LGM—latent growth curve modeling

ALMOST ONE FOURTH of adolescents are regular smokers.1 Of adults who have ever smoked regularly, the majority began smoking during adolescence and progressed to a regular habit by age 18.2 Given that adolescence is a critical period for the initiation of smoking experimentation and the progression to a regular habit, it is important to identify and to intervene regarding factors that influence smoking progression.

Receptivity to tobacco advertising and promotions has been shown to predict adolescent smoking initiation and progression.3–6 Receptivity to tobacco advertising, expressed by owning a cigarette promotional item, increases the likelihood of smoking progression threefold, and the loss of a promotional item or unwillingness to use one decreases the likelihood of smoking progression by ~40%.7 Therefore, receptivity seems to be a proximal influence on smoking behavior among adolescents. Investigations of the factors that may influence adolescents’ receptivity to tobacco advertising and thus their likelihood of progressing to a regular habit could provide useful information for smoking prevention interventions, including tobacco counter-advertising campaigns.

Variability in temperament, such as novelty seeking, between adolescents may influence receptivity to tobacco advertising and thus the likelihood of progressing to a regular habit.8,9 Novelty seeking is characterized by a tendency to seek out new and exciting stimuli, to engage in sensation-seeking, impulsive, and risk-taking behavior, and to be sensitive to reward.10 Studies have shown that this personality dimension predicts tobacco use during adolescence11,12 and early onset of smoking among adolescent boys.13 Indeed, a recent study of longitudinal smoking patterns from 14 to 18 years of age found that adolescents who were high in novelty seeking were ~15% to 20% more likely to be members of a trajectory involving regular smoking than one involving never smoking.6

Adolescents high in novelty seeking also tend to be more receptive to tobacco industry promotional campaigns.14 This relationship can be conceptualized through activation theory, which posits that (1) there are biologically based individual differences in the need for arousal or stimulation (eg, novelty seeking) and (2) the potential for a stimulus, such as a tobacco advertisement, to attract and to sustain attention is determined by how well its features coincide with the individual’s need for stimulation.15,16 Messages will be attended to and create arousal and recall if they match an individual’s need for stimulation.16 The heightened receptivity to tobacco advertising among youths high in novelty seeking may be attributable to their greater need for stimulation and rewarding experiences. Although not yet investigated, novelty seeking may be a key factor in receptivity to tobacco advertising, which in turn increases the likelihood that an adolescent will progress in smoking behavior across time. Structural equation models support the notion that novelty seeking affects substance use indirectly through other variables that are more proximal to use.12,17 Tobacco advertising receptivity could be considered more proximal to smoking than the dispositional variable, novelty seeking.

Depression is another factor that may influence an adolescent’s receptivity to tobacco advertising and subsequent progression in smoking behavior. Depression has been shown to be a predictor of smoking initiation18 and to be associated with nicotine dependence among adolescents.19 Teens who smoke regularly are almost twice as likely as teens who smoke occasionally to report high levels of depression.20 Among adolescents who have peers who smoke, depression has been shown to predict the onset of smoking experimentation.21 Therefore, depression may make an adolescent more vulnerable to social influences to smoke, including tobacco advertising. In fact, cross-sectional research indicated that adolescents who were high in depressive symptoms and tobacco advertising receptivity were twice as likely to have ever smoked, compared with adolescents who were high in tobacco advertising receptivity but low in depressive symptoms.22 The images and messages conveyed in tobacco advertisements (eg, fun, happiness, and success) may be especially appealing to depressed adolescents. Theories of self-image would suggest that smoking portrayals in tobacco advertisements may be consistent with a depressed adolescent’s ideal self-image23–26 and that smoking uptake would then make an adolescent’s self-image more consistent with his or her ideal self-image. Depressive symptoms may influence how receptive an adolescent is to tobacco promotions, which in turn influences whether an adolescent experiments with smoking and then progresses to a regular habit.

This study sought to determine whether 2 psychological factors, namely, novelty seeking and depressive symptoms, influenced adolescent smoking progression because of their effects on adolescent receptivity to tobacco advertising. Cross-sectional research indicated a positive relationship between both of these variables and tobacco advertising receptivity and adolescent smoking.11,14,22 In addition, on the basis of previous research, we expected tobacco advertising receptivity to predict smoking progression.3,4,6 We hypothesized that novelty seeking and depressive symptoms would have mediated or indirect effects on smoking progression through tobacco advertising receptivity. Although the Master Settlement Agreement placed restrictions on tobacco advertising and promotions targeting adolescents, youths are still targeted, they receive significant exposure to tobacco advertising, and their decline in receptivity to tobacco promotions has been limited.27–30


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Participants and Procedures
Participants included 1053 high school students who were enrolled in 1 of 5 public high schools in northern Virginia. More than one half of the sample (52%) was female, and the approximate racial distribution of the sample was 65% white, 11% Hispanic, 10% Asian, 8% black, and 6% other. These adolescents represent a cohort that was monitored for 4 years (9th grade to 12th grade) for evaluation of the social, psychological, and genetic predictors of adolescent smoking adoption. Participation involved the completion of an annual survey and the provision of DNA (buccal swab) in the 9th grade. Adolescents were ineligible to participate in the study if they had a special classroom placement (ie, severe learning disability and/or English as a second language) that might preclude valid survey administration.

Eligible participants were identified through class rosters at the beginning of 9th grade. On the basis of the aforementioned exclusionary criteria, 89% of students (2120 of 2393 students) were eligible to participate in 9th grade. After determination of eligibility, recruitment efforts were initiated by mail. Seventy-two percent of the parents/guardians (1533 of 2120 parents) approached provided a response regarding permission for their adolescent to participate. Of parents who responded, 75% (1151 of 1533 parents) provided written consent for their adolescent to participate. Therefore, a total of 54% of parents (1151 of 2120 parents) of eligible students provided written consent and 18% (382 of 2120 parents) declined. White parents with greater than a high school education were significantly more likely to provide consent, compared with parents with a high school education or less.31 Although these differences in parental consent were small (89% for more than high school education versus 77% for high school education or less), some caution is warranted in generalizing the results of this study.

Study enrollment required both active parental consent and adolescent assent (administrative approval of the study protocol was granted by the Georgetown University and the University of Pennsylvania institutional review board). Of the 1151 students with parental permission to participate, 15 declined (1%) and 13 (1%) were unavailable on baseline survey administration days because of school absence, although they later provided assent.

Because this was a longitudinal study, participants were resurveyed in the autumn and spring of the 10th grade and in the spring of 11th and 12th grades, for a total of 5 data collection waves. The rates of participation in the 3 spring follow-up surveys in the 10th, 11th, and 12th grades were ~96% (1081 students), 93% (1043 students), and 89% (1005 students), respectively. Those lost to follow-up monitoring did not differ significantly from those retained with respect to the primary variables of interest. The primary variables of interest were novelty seeking, depression, tobacco advertising receptivity, and smoking progression from 9th grade to 12th grade. The data presented below were based on 1053 participants with "all available data" (ie, a pairwise missing data strategy was used when data were missing at random, capitalizing on the data that were available for each wave for each participant) for these variables, although this method did not include participants with data missing on the covariates (ie, gender, race, smoking exposure, alcohol use, marijuana use, and novelty seeking).

Measures
Demographic Features
Gender and race were assessed through self-report. Race was dichotomized as white versus nonwhite (0 vs 1) because there were insufficient adolescents in the various racial groups for evaluation of specific racial differences. These demographic variables were examined for characterization of the sample and for use as controlling variables.

Substance Use
Lifetime alcohol use was assessed at baseline with an item that asked, "During your life, on how many days have you had at least 1 drink (not just a sip) of alcohol?"1 The response options ranged from 0 days = 1 to ≥100 days = 7. Marijuana use was assessed at baseline with an item that asked, "During your life, how many times have you used marijuana?"1 The response options ranged from 0 times = 1 to ≥100 times = 7. The effects of these variables were controlled for in the model, because substance use has been shown to be associated with novelty seeking and smoking.6,11

Smoking Exposure
Exposure to smoking from family members and the 4 best male and female friends was evaluated as in previous studies.32 Overall exposure was characterized as no exposure, family exposure only, peer exposure only, or both family and peer exposure.22,32 The effect of smoking exposure was controlled for in the model, because smoking by family members and friends has been shown to influence adolescent smoking and may influence advertising receptivity.3,5,6,33

Novelty-Seeking Personality
A 20-item version of the Temperament and Character Inventory was used to measure novelty seeking.34 The true/false novelty seeking scale included items such as "I often try new things just for fun or thrills" and "I like to think about things for a long time before I make a decision." In the present study, the Kuder-Richardson coefficient of reliability was satisfactory (0.74).

Depressive Symptoms
Depressive symptoms were assessed with the Center for Epidemiologic Studies Depression Scale (CES-D) at baseline. The CES-D is a 20-item self-report measure of depressive symptoms.35 Items on the CES-D are rated with a 4-point Likert scale to indicate how frequently in the past week each symptom occurred (0 = rarely or none of the time; 3 = most of the time). Scores range from 0 to 60, and higher scores indicate a greater degree of depressive symptoms.

Tobacco Advertising Receptivity
A 5-item scale was used to assess the purchase, receipt, and use of tobacco promotional items, as well as the recall of brands advertised most often, brands of advertising that attracted the most attention, and brands of favorite advertisements.3 Receptivity was evaluated on the basis of affirmative responses to a sequence of items reflecting progressive levels of receptivity. For example, adolescents who could not name an often-advertised brand and who had never received or were not willing to use a promotional item were classified as having no receptivity. Adolescents who were able to name no more than a frequently advertised brand and who did not have a favorite brand or had never received or used promotional items were classified as having low receptivity. Adolescents who could name an often-advertised brand and who also had a favorite advertisement were classified as having moderate receptivity. Finally, those who had a favorite brand or were willing to use promotional items were labeled as having high receptivity. Tobacco advertising receptivity was measured at all 4 waves.

Smoking Progression
Adolescent smoking practices were assessed with an ordered-categorical variable generated from responses to a series of standard epidemiologic questions regarding smoking that have been used in previous studies.1,32 These questions included the following: "Have you ever tried or experimented with cigarette smoking, even a few puffs?" "Have you smoked a cigarette in the past 30 days?" The 4 ordered categories were as follows: 0, never smoker; 1, puffer (not ever having smoked a whole cigarette); 2, experimenter (smoked ≥1 cigarette but ≤100 cigarettes total in a lifetime); 3, regular smoker (smoked in the past 30 days and smoked >100 cigarettes in a lifetime). Smoking practices were assessed at all 4 waves. Adolescents who had smoked >100 cigarettes in a lifetime but had not smoked in the past 30 days (n = 7) were classified as experimenters.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Statistical Analyses
Univariate statistics were generated to describe the study population in terms of demographic features, smoking exposure, novelty seeking, depressive symptoms, tobacco advertising receptivity, and smoking practices. We conducted associative latent growth curve modeling (LGM) to assess the indirect effects of the exogenous variables (variables for which predictors are not defined), namely, novelty seeking and depressive symptoms, on smoking trend through baseline tobacco advertising receptivity. Briefly, LGM is growth modeling from a latent variable framework. LGM models repeated observed measures (eg, smoking status) on factors (latent/unobserved variables) representing random effects ({eta}) that define the shape (eg, linear, cubic, or quadratic) of development with factor loadings. The random effects are level (intercept) and trend (slope). They are random because level and trend are free to vary, together defining each participant’s unique developmental trajectory based on the observed variables. For instance, one individual may be an experimenter at baseline but progress slowly to regular smoking by 12th grade, whereas another participant may remain an experimenter across all 4 waves. A benefit of LGM is that it provides 2 sources of variability, ie, within-subjects (repeated measures) and between-subjects (average level and trend) variability.

Associative LGM (also known as parallel processes LGM) is an extension of LGM that allows testing of paths among random effects (eg, levels [{eta}0] and trends [{eta}1] for a linear trend) for ≥2 LGM models parallel in time.36,37 In the present study, we modeled 2 LGM models, 1 each for repeated measures of smoking and tobacco advertising receptivity. The use of associative LGM allowed us to test the indirect effects of the exogenous observed variables, ie, novelty seeking and depressive symptoms, on the smoking trend factor, through the tobacco advertising receptivity level factor.

The repeated measures of smoking and receptivity used in study are ordered polytomous variables (ordered categorical variables), which means that higher categories represent qualitatively greater levels of behavior (eg, regular smoking represents greater cigarette use than experimentation). Furthermore, unlike for a continuous measured variable, the spaces between consecutive categories are not necessarily equivalent. Therefore, progression from never smoker to puffer is different from the progression from experimenter to regular smoker. To account for qualitative differences in change between categories in an ordered polytomous variable, such as our measure of smoking, the Mplus38,39 general modeling framework relates observed dependent variables in the model to latent continuous variables. For instance, the 4 repeated, ordered, polytomous, dependent variables of smoking level (y9–y12) are related to continuous variables (y9*–y12*) through the use of threshold parameters ({tau}c), where C represents the number of categories in the ordered polytomous variable, c = 0, ..., C – 1, {tau}0 = –{infty}, and {tau}C = {infty}. In the case of the C = 4 category ordered polytomous variable of smoking, there are C – 1 = 3 thresholds. Thresholds are related to cutoff points when continuous variables are translated into ordered categorical variables (Bengt O. Muthén, Stephen H. C. du Toit, and Damir Spisic, written communication, 1997).40,41 Therefore, y = c for {tau}c < y* ≤ {tau}c+1. It is important to note that threshold values are related inversely to the likelihood of advancing to a higher category, such that the higher the threshold, the lower the likelihood of advancing to the next category. Therefore, thresholds describe the facility with which individuals advance to higher levels of an ordered polytomous variable. Finally, to assess significant changes with time, Mplus holds constant thresholds across time and tests the fit of this model to the data. Therefore, the model takes into consideration the fact that the influence of the covariates differs depending on the point of smoking progression.

Multivariate data analysis in this study was conducted with Mplus version 3.11 software.38 Associative LGM with ≥1 repeated, ordered, categorical variable (eg, smoking) in Mplus is conducted with a proportional odds logistic regression model.38 We exponentiated the logistic regression ß coefficients (log odds), to estimate the odds of progression to higher levels of the criterion variables (level and trend factors) for a unit change in the predictor variable.40,41

The first step in the analysis is a properly fitting model. We evaluated model fit with reference to the model {chi}2, comparative fit index (CFI), root mean square error of approximation (RMSEA), and weighted root mean residual (WRMR). Suggested criteria for model fit are nonsignificant model {chi}2, CFI of >0.95, RMSEA of <0.05 to 0.08, and WRMR of <0.9.42,43 Although there are a variety of methods for estimating model parameters, the present analysis used a weighted least-squares estimation technique in which the diagonal weight matrix uses robust SEs and the {chi}2 test statistic is mean and variance adjusted.38 The weighted least squares estimation technique is the default weight matrix in Mplus 3.11 for modeling with categorical outcome variables.

Descriptive Statistics
The average level of depressive symptoms at baseline was 13.8 (SD: 9.09). The proportion of adolescents scoring above the clinical cutoff for depressive symptoms was consistent with the literature, and the average depression score was within 0.5 SD of the published normative values.44 The average level of novelty seeking at baseline was 10.80 (SD: 3.87), which was within 0.5 SD of the published normative values.34 For smoking exposure at baseline, 36% had neither family nor peer exposure, 10% had exposure to family smoking only, 35% had peer exposure only, and 19% had exposure to both family and peer smoking. The transitions in smoking and tobacco advertising receptivity between 9th grade and 12th grade are presented in Table 1.


View this table:
[in this window]
[in a new window]
  		TABLE 1 Transition in Tobacco Advertising Receptivity and Smoking Between 9th Grade and 12th Grade

 
Associative LGM
Measurement models in associative LGM are multifactor models without directional paths among the factors; these are generally the best-fitting models because they allow between-factor paths to be estimated freely. On the basis of individual measurement models to determine the interindividual growth forms, tobacco advertising receptivity was linear, whereas the smoking trend was nonlinear, with the factor loadings from the trend factor for baseline and first through third follow-up surveys being 0, 1, 1.86, and 2.46, respectively. The associative measurement model fit the data well [{chi}2(15,N=1136) = 18.61, P = .23, RMSEA = 0.01, CFI = 1.00, WRMR = 0.51]. Factor variances for smoking and receptivity level and trend were significant, indicating significant developmental heterogeneity in smoking status and receptivity at baseline and their rates of acceleration across time (P < .0001).

Associative LGM With Directional Paths and Covariates
We fit an associative LGM model with the 2 parallel processes, ie, smoking progression and tobacco advertising receptivity, to the data. We assessed the direct and indirect effects of novelty seeking and depressive symptoms on smoking trend, with tobacco advertising receptivity level as the mediating mechanism. Mediation is a stage-sequential process through which the effects of one stage in a series affect succeeding stages.45 We controlled for the following covariates: gender, race, baseline peer and family smoking exposure, and baseline alcohol and marijuana use. The associative LGM model fit the data fairly well [{chi}2(26,N=1053) = 46.69, P = .008, RMSEA = 0.03, CFI = 1.00, WRMR = 0.63]. A graphical representation of the associative LGM model with standardized path coefficients for the significant effects ({gamma} values) is presented in Fig 1. Table 2 presents the nonstandardized path coefficients, SEs, and test statistics for the hypothesized effects in the model.


Figure 1
View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 1 Associative LGM model with standardized path coefficients. Only significant paths are shown. Rectangles represent measured variables; circles represent latent (unobserved/unmeasured) variables. Receptivity indicates tobacco advertising receptivity; depression, depressive symptoms (9th grade); alcohol, lifetime alcohol use (9th grade); marijuana, lifetime marijuana use (9th grade); smoking exposure, exposure to family and peer smoking (female = 1; male = 0; nonwhite = 1; white = 0).

 

View this table:
[in this window]
[in a new window]
  		TABLE 2 Nonstandardized Path Coefficients ({gamma})

 
Direct Effects of Novelty Seeking and Depressive Symptoms on Tobacco Advertising Receptivity Level and on Smoking Trend
Novelty seeking had a significant positive effect (odds ratio [OR]: 1.06; 95% confidence interval [CI]: 1.02–1.09) on tobacco advertising receptivity at baseline. For each 0.5-SD (SD: 3.87) increase in novelty seeking, the odds of a higher level of receptivity to tobacco advertising at baseline increased by 6%. Depressive symptoms also had a significant positive effect (OR: 1.07; 95% CI: 1.02–1.12) on tobacco advertising receptivity at baseline. More specifically, for each 0.5-SD (SD: 13.8) increase in depressive symptoms at baseline, the odds of being more receptive to tobacco advertising increased by 7%. We used 0.5 SDs of depressive symptoms and novelty seeking to estimate their effects in this study to increase the interpretability of our findings.40 Neither novelty seeking nor depressive symptoms had significant direct effects on smoking trend.

The absence of direct effects of baseline depressive symptoms and novelty seeking on smoking trend could suggest that the effects of the independent variables (eg, novelty seeking) on the dependent variable (smoking trend) are delayed, meaning that the variables exert their effects only after a time lag and sequentially after affecting a mediator such as tobacco advertising receptivity.45,46 A delayed effect is likely to occur in a study such as the present study, in which the dependent process is a change in the odds of smoking progression over time and there is substantial time for mediating effects to exert their influence. The direct effect criterion proposed by Baron and Kenny47 is more applicable to situations in which change occurs immediately, such as in experimental designs.46 The criteria for mediation proposed by Collins et al45 are presented below.

Effects of Tobacco Advertising Receptivity Level on Smoking Trend
Baseline tobacco advertising receptivity had a significant positive effect on smoking trend (OR: 1.11; 95% CI: 1.02–1.19). More specifically, for each 1-unit increase in tobacco advertising receptivity at baseline (ie, for each increase in the level of tobacco advertising receptivity), the odds of progressing to a higher level of smoking increased 11%. It is important to reiterate the nonlinear fixed relationship between time and smoking trend, represented by the progressively increasing factor loading. Therefore, to understand the relationship between tobacco advertising receptivity at baseline, smoking progression, and time, one must multiply the log odds (ß) by the factor loadings (0, 1, 1.86, and 2.46) to obtain the estimated odds of progression in smoking for each 1-year increase; only the change from year 1 to year 2 was linear.

Indirect Effects of Novelty Seeking and Depressive Symptoms on Smoking Trend Through Tobacco Advertising Receptivity Level
Indirect effects indicate the presence of mediation. Mediation is a stage-sequential process in which the effects of the initial step in a series (eg, novelty seeking) affect a succeeding step (eg, tobacco advertising receptivity level), which then affects the following step (eg, smoking trend) in the series.45 For tobacco advertising receptivity level to be a mediator in the relationship between novelty seeking and smoking trend, for instance, the following 3 criteria must be met. First, the probability of an individual undergoing the sequence of independent variable to mediator to outcome variable would be greater for individuals with higher levels of the independent variable (novelty seeking). Second, having a higher level of the independent variable would increase the likelihood of the mediator variable (tobacco advertising receptivity level). Third, for every level of the independent variable, the mediator would have to increase the likelihood of the dependent variable (smoking trend).45 Consistent with this definition of mediation, we hypothesized that novelty seeking would result in an increased level of tobacco advertising receptivity, which in turn would result in an increased likelihood of smoking progression. We also hypothesized that tobacco advertising receptivity level would mediate the relationship between depression and smoking progression.

The results supported the possibility of mediated relationships between novelty seeking and smoking trend and between depressive symptoms and smoking trend through the effects of baseline tobacco advertising receptivity. To bolster support for this conclusion, we tested for the significance of these specific indirect effects (see ref 46 for a discussion of testing indirect effects). Novelty seeking had a significant indirect effect on smoking trend through tobacco advertising receptivity level (ß = .003, z = 2.026, P = .043), indicating that higher novelty seeking scores were associated with higher levels of tobacco advertising receptivity at baseline, which in turn was associated with increasing odds of smoking progression from 9th grade to 12th grade. This finding, taken together with the significant direct effects reported earlier, supports a mediated effect of novelty seeking on smoking trend through baseline tobacco advertising receptivity.45 More specifically, a 0.5-SD increase in novelty seeking resulted in a 6% increase in the odds of receptivity to tobacco advertising (ie, being in a specific category or higher) at baseline. A 1-unit increase (eg, a single-level increase, such as moving from moderate to high level of receptivity) in baseline tobacco advertising receptivity in turn resulted in an 11% increase in the odds of smoking progression from 9th grade to 12th grade. Although the individual paths from depression to tobacco advertising receptivity level and from tobacco advertising receptivity level to smoking trend were significant, the analysis indicated that the indirect effect from baseline depressive symptoms to smoking trend did not reach significance (ß = .001, z = 1.856, P = .064).


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
This study sought to determine whether 2 psychological factors, namely, novelty seeking and depressive symptoms, influenced adolescent smoking progression through their effects on adolescent receptivity to tobacco advertising. We hypothesized that novelty seeking and depressive symptoms would have mediated or indirect effects on smoking progression through tobacco advertising receptivity. Higher novelty seeking scores were associated with a 12% increase in the odds of receptivity to tobacco advertising (ie, being in a specific category or higher), which in turn was associated with an 11% increase in the odds that an adolescent would progress in smoking uptake from the 9th grade to the 12th grade. Depressive symptoms did not have a significant indirect effect on smoking progression through tobacco advertising receptivity.

These findings suggest that increased receptivity to tobacco advertising is one factor that promotes smoking initiation and progression among adolescents high in novelty seeking and it should be a target in youth smoking prevention programs. The heightened receptivity to tobacco advertising among adolescents high in novelty seeking may be attributable to their greater need for stimulation and rewarding experiences.10 Tobacco advertisements and promotional campaigns appeal to these characteristics by highlighting stimulating and adventurous activities and convey that excitement and rewards are associated with smoking.48–50 Adolescent exposure to such tobacco advertising and promotional campaigns has not diminished significantly despite the Master Settlement Agreement.27,30 Counter-advertising approaches tailored to these characteristics of novelty seekers may influence tobacco advertising receptivity and subsequent smoking progression.51 Tailoring messages to adolescents who are high in novelty-seeking type traits to promote receptivity to the message has been shown to be effective with anti-marijuana use campaigns.52

Contrary to our hypothesis, tobacco advertising receptivity did not mediate the relationship between depression and smoking progression. Although the individual paths from depression to tobacco advertising receptivity and from tobacco advertising receptivity to smoking progression were significant, the overall indirect effect only approached significance. Several potential explanations can be offered regarding why this effect was not significant. Quite simply, tobacco advertising receptivity may not mediate or may be a very weak mediator of the relationship between depression and smoking progression. Other variables, such as physical activity, may explain the relationship between depression and smoking progression. Physical activity has been shown to be associated negatively with depression and with smoking,53–55 whereas depression has been shown to be associated positively with adolescent smoking.56,19 Alternatively, depression was treated as a time-invariant covariate, because only baseline (9th grade) depression scores were included in the model. Baseline depression scores might not be related consistently to depression in subsequent years, which could affect the observed relationship with tobacco advertising receptivity and smoking progression.57 In addition, depression had a large SE, indicating heterogeneity and the possibility of subpopulations of adolescents with respect to depressive symptoms.58

The size of the effects of novelty seeking and depressive symptoms on tobacco advertising receptivity suggest that there are other variables that are important in influencing tobacco advertising receptivity. Family and peer smoking was included in the model as a controlling variable. Although we did not have a specific hypothesis about its impact on tobacco advertising receptivity, there was a significant path from smoking exposure to baseline tobacco advertising receptivity. This significant path indicated that, for each level of peer and family smoking exposure, the odds of an increased level of tobacco advertising receptivity increased 15% (OR: 1.15; 95% CI: 1.08–1.22). This in turn resulted in an increase in the odds of smoking progression, such that, for each 1-level increase in receptivity, there was a 10% increase in the odds of progression to a higher level of smoking (OR: 1.10; 95% CI: 1.02–1.19). The overall indirect effect from smoking exposure to smoking trend through baseline receptivity was significant (P < .05). Studies showed that tobacco advertising undermines parental discouragement from smoking.59 Parents who smoke may offer even less discouragement to smoke, either explicitly or implicitly through their own smoking behavior, thereby facilitating receptivity to tobacco marketing. Adolescents who have parents who smoke may also have greater access to cigarette promotional items.33 Although mechanisms through which peer smoking influences adolescent smoking require additional investigation, one aspect of peer smoking influence may be to increase an adolescent’s receptivity to tobacco marketing strategies.60 Another variable that may influence tobacco advertising receptivity and the subsequent progression in smoking is an adolescent’s perceived ability to quit. In fact, one study found that >50% of adolescents who believed they could quit at any time and who were willing to use a cigarette promotional item progressed to a regular smoking habit.5

The limitations of this study should be noted. Although 75% of the parents who responded provided consent and the differences between those who provided consent and those who declined were relatively small,31 caution is warranted in generalizing the results of this study, especially in light of the study’s consent rate (54%). Although the sample may not be representative of all adolescents in the United States, the sample was nationally and locally representative with respect to basic demographic characteristics,61,62 and the sample smoking rates were regionally and locally comparable to those found in national surveys. For example, data from our 2003 survey indicated that 10% were daily smokers, compared with ~9% in the 2003 Youth Risk Behavior Survey and ~15% in the 2003 Monitoring the Future Survey. In addition, 15% of the adolescents in our sample were current smokers, compared with 13% in the 2003 Youth Risk Behavior Survey and 24% in the 2003 Monitoring the Future Survey.

Another potential limitation is that the model included only baseline depression scores. Depression at 1 time point might not have offered the best depiction of the relationship between depressive symptoms, tobacco advertising receptivity, and smoking progression, because depressive symptoms might have changed over time. Finally, novelty seeking and depression did not account for all of the variance in tobacco advertising receptivity. Coupled with the effects of family and peer smoking exposure, approximately one third of the variance was accounted for. Therefore, smoking prevention programs aimed at these influences on advertising receptivity would not be addressing all potentially important influences on tobacco advertising receptivity.


   CONCLUSIONS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
This longitudinal study has identified several factors important to tobacco advertising receptivity, and thus subsequent adolescent smoking progression. Little research has focused on factors that predict receptivity to tobacco advertising. This line of inquiry is critical, given the consistent findings indicating that tobacco advertising receptivity leads to smoking progression and the uptake of a regular habit.3–6 These findings may help lay the groundwork for future research into other factors important to tobacco advertising receptivity, which may inform programs aimed at preventing adolescent smoking initiation and progression.


   ACKNOWLEDGMENTS
 
This study was supported by a Transdisciplinary Tobacco Use Research Center grant from the National Cancer Institute and the National Institute on Drug Abuse (grant P50 84718).

We thank the high school faculty members, administrative personnel, and students involved in the research.


   FOOTNOTES
 
Accepted Aug 31, 2005.

Address correspondence to Janet Audrain-McGovern, PhD, Department of Psychiatry and Abramson Cancer Center, University of Pennsylvania, 3535 Market St, Suite 4100, Philadelphia, PA 19104. E-mail: audrain{at}mail.med.upenn.edu

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Grunbaum JA, Kann L, Kinchen S, et al. Youth risk behavior surveillance: United States, 2003. MMWR Surveill Summ. 2004;53(2) :1 –96
  2. Preventing tobacco use among young people: a report of the Surgeon General: executive summary. MMWR Recomm Rep. 1994;43(RR-4) :1 –10
  3. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Berry CC. Tobacco industry promotion of cigarettes and adolescent smoking. JAMA. 1998;279 :511 –515[Abstract/Free Full Text]
  4. Biener L, Siegel M. Tobacco marketing and adolescent smoking: more support for a causal inference. Am J Public Health. 2000;90 :407 –411[Abstract]
  5. Choi WS, Ahluwalia JS, Harris KJ, Okuyemi K. Progression to established smoking: the influence of tobacco marketing. Am J Prev Med. 2002;22 :228 –233[CrossRef][ISI][Medline]
  6. Audrain-McGovern J, Rodriguez D, Tercyak KP, Cuevas J, Rodgers K, Patterson F. Identifying and characterizing adolescent smoking trajectories. Cancer Epidemiol Biomarkers Prev. 2004;13 :2023 –2034[Abstract/Free Full Text]
  7. Sargent J, Dalton M, Beach M. Exposure to cigarette promotions and smoking uptake in adolescents: evidence of a dose-response relation. Tob Control. 2000;9 :163 –168[Abstract/Free Full Text]
  8. Stallings M, Hewitt J, Cloninger C, Heath A, Eaves L. Genetic and environmental structure of the Tridimensional Personality Questionnaire: three or four temperament dimensions? J Pers Soc Psychol. 1996;70 :127 –140[CrossRef][ISI][Medline]
  9. Cloninger C, Svrakic D, Przybeck T. A psychobiological model of temperament and character. Arch Gen Psychiatry. 1993;50 :975 –990[Abstract]
  10. Cloninger C. A systematic method for clinical description and classification of personality variants: a proposal. Arch Gen Psychiatry. 1987;44 :573 –588[Abstract]
  11. Wills TA, Vaccaro D, McNamara G. Novelty seeking, risk taking, and related constructs as predictors of adolescent substance use: an application of Cloninger’s theory. J Subst Abuse. 1994;6 :1 –20[CrossRef][Medline]
  12. Wills TA, Windle M, Cleary SD. Temperament and novelty seeking in adolescent substance use: convergence of dimensions of temperament with constructs from Cloninger’s theory. J Pers Soc Psychol. 1998;74 :387 –406[CrossRef][ISI][Medline]
  13. Masse LC, Tremblay RE. Behavior of boys in kindergarten and the onset of substance use during adolescence. Arch Gen Psychiatry. 1997;54 :62 –68[Abstract]
  14. Audrain-McGovern J, Tercyak KP, Shields AE, Bush A, Espinel CF, Lerman C. Which adolescents are most receptive to tobacco industry marketing? Implications for counter-advertising campaigns. Health Commun. 2003;15 :499 –513[CrossRef][ISI][Medline]
  15. Donohew L, Palmgreen P, Duncan J. An activation model of information exposure. Commun Monogr. 1980;47 :295 –303
  16. Donohew L, Palmgreen P, Lorch E. Attention, need for sensation, and health communication campaigns. Am Behav Sci. 1994;38 :310 –322[Abstract]
  17. Wills TA, Sandy JM, Shinar O. Cloninger’s constructs related to substance use level and problems in late adolescence: a mediational model based on self-control and coping motives. Exp Clin Psychopharmacol. 1999;7 :122 –134[CrossRef][ISI][Medline]
  18. Escobedo LG, Kirch DG, Anda RF. Depression and smoking initiation among US Latinos. Addiction. 1996;91 :113 –119[CrossRef][ISI][Medline]
  19. Fergusson DM, Lynskey MT, Horwood LJ. Comorbidity between depressive disorders and nicotine dependence in a cohort of 16-year-olds. Arch Gen Psychiatry. 1996;53 :1043 –1047[Abstract]
  20. Patton GC, Hibbert M, Rosier MJ, Carlin JB, Caust J, Bowes G. Is smoking associated with depression and anxiety in teenagers? Am J Public Health. 1996;86 :225 –230[Abstract/Free Full Text]
  21. Patton GC, Carlin JB, Coffey C, Wolfe R, Hibbert M, Bowes G. Depression, anxiety, and smoking initiation: a prospective study over 3 years. Am J Public Health. 1998;88 :1518 –1522[Abstract/Free Full Text]
  22. Tercyak KP, Goldman P, Smith A, Audrain J. Interacting effects of depression and tobacco advertising receptivity on adolescent smoking. J Pediatr Psychol. 2002;27 :145 –154[Abstract/Free Full Text]
  23. Aloise-Young PA, Hennigan KM, Graham JW. Role of self-image and smoker stereotype in smoking onset during early adolescence: a longitudinal study. Health Psychol. 1996;15 :494 –497[CrossRef][ISI][Medline]
  24. Aloise-Young PA, Hennigan KM. Self-image, the smoker stereotype and cigarette smoking: developmental patterns from fifth through eighth grade. J Adolesc. 1996;19 :163 –177[CrossRef][ISI][Medline]
  25. Chassin L, Presson CC, Bensenberg M, Corty E, Olshavsky RW, Sherman SJ. Predicting adolescents’ intentions to smoke cigarettes. J Health Soc Behav. 1981;22 :445 –455[CrossRef][ISI][Medline]
  26. Barton J, Chassin L, Presson CC, Sherman SJ. Social image factors as motivators of smoking initiation in early and middle adolescence. Child Dev. 1982;53 :1499 –1511[CrossRef][ISI][Medline]
  27. King C III, Siegel M. The Master Settlement Agreement with the tobacco industry and cigarette advertising in magazines. N Engl J Med. 2001;345 :504 –511[Abstract/Free Full Text]
  28. Hamilton WL, Turner-Bowker DM, Celebucki CC, Connolly GN. Cigarette advertising in magazines: the tobacco industry response to the Master Settlement Agreement and to public pressure. Tob Control. 2002;11(suppl 2) :54 –58
  29. Wakefield M, Chaloupka F. Effectiveness of comprehensive tobacco control programmes in reducing teenage smoking in the USA. Tob Control. 2000;9 :177 –186[Abstract/Free Full Text]
  30. Chung PJ, Garfield CF, Rathouz PJ, Lauderdale DS, Best D, Lantos J. Youth targeting by tobacco manufacturers since the Master Settlement Agreement: the first study to document violations of the youth-targeting ban in magazine ads by the three top US tobacco companies. Health Aff (Millwood). 2002;21 :254 –263[Abstract/Free Full Text]
  31. Audrain J, Tercyak KP, Goldman P, Bush A. Recruiting adolescents into genetic studies of smoking behavior. Cancer Epidemiol Biomarkers Prev. 2002;11 :249 –252[Abstract/Free Full Text]
  32. Choi WS, Pierce JP, Gilpin EA, Farkas AJ, Berry CC. Which adolescent experimenters progress to established smoking in the United States? Am J Prev Med. 1997;13 :385 –391[ISI][Medline]
  33. Albers AB, Biener L. Adolescent participation in tobacco promotions: the role of psychosocial factors. Pediatrics. 2003;111 :402 –406[Abstract/Free Full Text]
  34. Cloninger CR, Przybeck TR, Svrakic DM, Wetzel RD. The Temperament and Character Inventory (TCI): A Guide to Its Development and Use. St Louis, MO: Center for Psychobiology and Personality; 1994
  35. Radloff L. The CES-D scale: a new self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1 :385 –401[CrossRef]
  36. Duncan TE, Duncan SC, Strycker LA, Li F, Alpert A. An Introduction to Latent Variable Growth Curve Modeling: Concepts, Issues, and Applications. Mahwah, NJ: Lawrence Erlbaum Associates; 1999
  37. Muthén BO. Beyond SEM: general latent variable modeling. Behaviormetrika. 2002;29 :81 –117
  38. Muthén LK, Muthén BO. Mplus User’s Guide. 3rd ed. Los Angeles, CA: Muthén & Muthén; 2004
  39. Muthén LK, Muthén BO. Mplus User’s Guide. Los Angeles, CA: Muthén & Muthén; 2001
  40. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed. New York, NY: John Wiley & Sons; 2000
  41. Agresti A. Categorical Data Analysis. 2nd ed. Hoboken, NJ: John Wiley & Sons; 2002
  42. Loehlin JC. Latent Variable Models: An Introduction to Factor, Path, and Structural Equation Analysis. 4th ed. Mahwah, NJ: Lawrence Erlbaum Associates; 2004
  43. Muthén LK, Muthén BO. Mplus. 2nd ed. Los Angeles, CA: Muthen & Muthen; 2002
  44. Radloff LS. The use of the Center for Epidemiologic Studies Depression Scale in adolescents and young adults. J Youth Adolesc. 1991;20 :149 –166[CrossRef][ISI]
  45. Collins LM, Graham JW, Flaherty BP. An alternative framework for defining mediation. Multivariate Behav Res. 1998;33 :295 –312[CrossRef]
  46. Shrout PE, Bolger N. Mediation in experimental and nonexperimental studies: new procedures and recommendations. Psychol Methods. 2002;7 :422 –445[CrossRef][ISI][Medline]
  47. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51 :1173 –1182[CrossRef][ISI][Medline]
  48. Cook BL, Wayne GF, Keithly L, Connolly G. One size does not fit all: how the tobacco industry has altered cigarette design to target consumer groups with specific psychological and psychosocial needs. Addiction. 2003;98 :1547 –1561[CrossRef][ISI][Medline]
  49. Shadel WG, Niaura R, Abrams DB. Adolescents’ reactions to the imagery displayed in smoking and antismoking advertisements. Psychol Addict Behav. 2002;16 :173 –176[CrossRef][ISI][Medline]
  50. Hawkins K, Hane A. Adolescents’ perceptions of print cigarette advertising: a case for counter-advertising. J Health Commun. 2000;5 :83 –96[CrossRef][ISI][Medline]
  51. Farrelly MC, Niederdeppe J, Yarsevich J Youth tobacco prevention mass media campaigns: past, present, and future directions. Tob Control 2003;12(suppl 1) :35 –47
  52. Palmgreen P, Donohew L, Lorch E, Hoyle R, Stephenson M. Television campaigns and adolescent marijuana use: tests of sensation seeking targeting. Am J Public Health. 2001;91 :292 –296[Abstract]
  53. Escobedo LG, Marcus SE, Holtzman D, Giovino GA. Sports participation, age at smoking initiation, and the risk of smoking among US high school students. JAMA. 1993;269 :1391 –1395[Abstract]
  54. Pate RR, Heath GW, Dowda M, Trost SG. Association between physical activity and other health behaviors in a representative sample of US adolescents. Am J Public Health. 1996;86 :1577 –1581[Abstract/Free Full Text]
  55. Field T, Diego M, Sanders C. Adolescent depression and risk factors. Adolescence. 2001;36 :491 –498[ISI][Medline]
  56. Covey LS, Tam D. Depressive mood, the single-parent home, and adolescent cigarette smoking. Am J Public Health. 1990;80 :1330 –1333[Abstract/Free Full Text]
  57. Collins LM, Graham JW. The effect of the timing and spacing of observations in longitudinal studies of tobacco and other drug use: temporal design considerations. Drug Alcohol Depend. 2002;68(suppl 1) :S85 –S96
  58. Rodriguez D, Moss HB, Audrain-McGovern J. Developmental heterogeneity in adolescent depressive symptoms: associations with smoking behavior. Psychosom Med. 2005;67 :200 –210[Abstract/Free Full Text]
  59. Pierce JP, Distefan JM, Jackson C, White MM, Gilpin EA. Does tobacco marketing undermine the influence of recommended parenting in discouraging adolescents from smoking? Am J Prev Med. 2002;23 :73 –81[CrossRef][ISI][Medline]
  60. Altman DG, Levine DW, Coeytaux R, Slade J, Jaffe R. Tobacco promotion and susceptibility to tobacco use among adolescents aged 12 through 17 years in a nationally representative sample. Am J Public Health. 1996;86 :1590 –159[Abstract/Free Full Text]
  61. US Bureau of Census. National Report. 2001a. Available at: http://quickfacts.census.gov/qfd/states/00000/html. Accessed August 29, 2003
  62. US Bureau of Census. State Report. 2001b. Available at: http://quickfacts.census.gov/qfd/states/51/html. Accessed February 8,2005
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics



This article has been cited by other articles:


Home page
 PediatricsHome page
J. R. DiFranza, J. A. Savageau, K. Fletcher, L. Pbert, J. O'Loughlin, A. D. McNeill, J. K. Ockene, K. Friedman, J. Hazelton, C. Wood, et al.
Susceptibility to Nicotine Dependence: The Development and Assessment of Nicotine Dependence in Youth 2 Study
Pediatrics, October 1, 2007; 120(4): e974 - e983.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
H. Zhu, M. Lee, S. Agatsuma, and N. Hiroi
Pleiotropic impact of constitutive fosB inactivation on nicotine-induced behavioral alterations and stress-related traits in mice
Hum. Mol. Genet., April 1, 2007; 16(7): 820 - 836.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
D. Rodriguez, D. Romer, and J. Audrain-McGovern
Beliefs About the Risks of Smoking Mediate the Relationship Between Exposure to Smoking and Smoking
Psychosom Med, January 1, 2007; 69(1): 106 - 113.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
S. Agatsuma, M. Lee, H. Zhu, K. Chen, J. C. Shih, I. Seif, and N. Hiroi
Monoamine oxidase A knockout mice exhibit impaired nicotine preference but normal responses to novel stimuli
Hum. Mol. Genet., September 15, 2006; 15(18): 2721 - 2731.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Audrain-McGovern, J.
Right arrow Articles by Cuevas, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow