PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Tob Regul Sci. Author manuscript; available in PMC 2017 April 19.
Published in final edited form as:
Tob Regul Sci. 2016 April; 2(2): 106–122.
doi:  10.18001/TRS.2.2.2
PMCID: PMC5397111
NIHMSID: NIHMS818895

Multiple Levels of Influence That Impact Youth Tobacco Use

Patricia A. Cavazos-Rehg, PhD, Associate Professor, Melissa J. Krauss, MPH, Shaina J. Sowles, MPH, Edward L. Spitznagel, PhD, Professor, Richard Grucza, PhD, Associate Professor, Frank J. Chaloupka, PhD, Professor, and Laura J. Bierut, MD, Professor

Abstract

Objective

Multiple levels of influence interplay to impact youth tobacco use. We work towards understanding important policy and environmental strategies that are associated with youth tobacco use behaviors.

Methods

We utilized data from participants of the Monitoring the Future (MTF) study and linked national data from multiple sources to assess correlates of youth tobacco use behaviors across individual, family, school, community, and state-level policy influences.

Results

Higher cigarette prices had the strongest association with youth tobacco use behaviors. Demographic and socio-economic characteristics at the individual, familial, and community/school-levels were associated with youth tobacco use behaviors.

Conclusions

In the present study, we confirm that higher cigarette prices could help to reduce youth tobacco use behaviors. Several states are still lagging behind in terms of their low cigarette tax and they should enact tax policies to reduce youth tobacco use.

Keywords: smoking, nicotine, adolescent, tobacco

Despite great progress on curbing tobacco use in the United States (US), it is still the leading cause of preventable morbidity and mortality.1,2 Whereas cigarette smoking in the past month decreased from 28% in 1975 to 8% in 2014, the increasing use of non-cigarette tobacco products among youth (eg, e-cigarettes, hookah) is worrisome and their use may lead to cigarette smoking and/or poly-tobacco use behaviors.39 Because the vast majority of adult regular smokers (90%) initiate tobacco use by age 18 years, it is important for tobacco prevention efforts to target the prevention and reduction of tobacco use among adolescents.10

Formative biological and social transitions occur in adolescence, and researchers have put forth extensive effort toward understanding both the risk and protective factors that impact tobacco use behaviors during this critical time of development.1117 There is consensus that social forces from one’s family and peers interact in a dynamic and complex way to influence youth tobacco use behaviors.15,1821 Even more distal influences (ie, further from the individual), such as anti-tobacco policies and anti-tobacco efforts delivered in schools, like-wise can influence youth tobacco use behaviors.2227 Thus, a multidimensional ecological framework that accounts for the complexity in overlapping and interacting determinants of tobacco use behaviors is a more comprehensive model than ones that focus on a specific aspect (eg, SES) of a particular life domain (eg, family).2830

The aforementioned youth tobacco use behaviors are impacted by proximal levels of influence including one’s socio-demographic characteristics (eg, sex, age, racial/ethnic minority status) and family status.31,32 For instance, young people living in single-parent households and/or with lower socioeconomic status have increased risk for tobacco use.3235 In addition, the average US student will spend 180 days in a school setting, meaning that for approximately half the year, adolescents have the opportunity to interact with peers at school on a daily basis.36 Therefore, school characteristics, although more distal than individual and familial factors, are important predictors of tobacco use because they depict the social surroundings where youth spend most of their daily lives.3739

Anti-tobacco control policies and campaigns delivered at a population level are desirable for their potential to reach a mass audience despite their distal relation to youth.4042 For instance, higher cigarette prices and smoke-free indoor air policies have been well documented as effective strategies for preventing the uptake of tobacco use as well as reducing the cigarette smoking behaviors of youth.4345 Similarly, media campaigns delivered through television, radio, print, and billboards are costly for reaching a mass population but can be a worthwhile expense for reducing cigarette smoking uptake among young people.4650 Correspondingly, mass media campaigns are considered to be an integral component of a comprehensive tobacco control program.5153

There is a vast amount of literature that examines specific risk and protective factors for tobacco use behaviors.15,1821 Yet, understanding how multiple levels of influence interplay to impact youth tobacco use is challenging because of the unique and numerous data sources that must be linked to analyze different levels of the social hierarchy simultaneously. As such, there is a paucity of research utilizing an ecological approach despite its value for comprehensively explaining youth tobacco use behaviors. In response, the present study utilizes nationally representative data to study the integrated effects of individual, family, school, community, and state-level influences on youth cigarette and smokeless tobacco use behaviors. In essence, we aimed to work towards understanding important policy and environmental strategies for reducing tobacco use behaviors among young people using a comprehensive and theoretically relevant approach. Based on extant evidence, we hypothesized that the multiple levels of influence we examined in our models would exhibit varying levels of associations with youth tobacco use behaviors. We expected to find strong associations between individual and family-level variables and tobacco use because they are proximate mechanisms. We also expected to find effects of varying strengths among school, community, and state-levels and tobacco use behaviors given their distal levels of influence.

METHODS

The present study utilized data from 2000–2006 because these are the years when data at a national level on 2 primary independent variables were available (ie, anti-tobacco media campaigns and school tobacco policies).

Data Source and Respondents

We utilized data on 8th, 10th, and 12th grade students from the Monitoring the Future (MTF) study, an ongoing national study (since 1975) of the substance use behaviors of adolescents conducted by the Institute for Social Research at the University of Michigan and sponsored by the National Institute on Drug Abuse.54 Students complete one of 4 different surveys for 8th and 10th graders or one of 6 different surveys for 12th graders. The forms are dispersed to participants in an ordered sequence that guarantees equally random subsamples.

Dependent Variable

The dichotomous dependent variables included cigarette smoking and smokeless tobacco use in the last 30 days and daily use of cigarettes. Current cigarette smoking was assessed by an item that queried participants on frequency of cigarette smoking in the past 30 days. Participants who responded positively to any cigarette smoking in the past 30 days were defined as a current cigarette user. Similar methods were used to define a current smokeless tobacco user. Participants who endorsed at least 1 cigarette per day in the past 30 days were defined as daily cigarette smokers.

Primary Independent Variables. Individual, Familial, and School-level Characteristics

We controlled for student-level demographic variables including sex, age, race/ethnicity, parents’ educational attainment (neither parent achieved a high school diploma versus having at least one parent who completed and/or achieved more than a high school diploma), and number of parents that currently live in the home (none/one/both). We also controlled for type of school (public/private), school size (based on the number of students from the targeted grade eligible for the survey), percent of students receiving free or reduced cost lunch, and percent of students who are black or Hispanic, and population density.

Primary Independent Variables. State-level Policies

State excise tax per pack of cigarettes

We utilized a historical compilation of annual state cigarette excise tax data from “The Tax Burden on Tobacco” to measure increases in cigarette prices.55 For this analysis, the reported results were adjusted for inflation to reflect dollars from 2012. This adjustment was made using the Bureau of Labor Statistics (www.bls.gov/cpi/) consumer price index.

Smoke-free indoor air (SFA) policies

We used state-specific policy data from the ImpacTeen State-Level Tobacco Legislative Database to examine SFA policies, PUP penalties, and youth access to tobacco laws. For SFA policies, 3 key policies are considered that include those covering private worksites, restaurants, and bars. For this rating system, there are 3 key levels for each policy that range from no restrictions (score 0) to a complete ban (score 2) resulting in composite scores ranging from 0 to 6.

Possession/use/purchase (PUP) penalties

PUPs include consequences such as fines, tobacco cessation classes, and community service and/or driver license suspension. States were coded for the presence or absence of a state law prohibiting minors from (1) possessing, (2) using, and/or (3) purchasing tobacco products. A PUP index was calculated as the sum of the number of possession, use, and/or purchase laws in each state in a given year (possible values: 0 = no PUP laws; 1 = 1 PUP law present; 2 = 2 PUP laws present; 3 = all 3 PUP laws present).

Youth access to tobacco laws

For youth access to tobacco laws, we used a rating system developed by Alciati et al,56 which measures the extensiveness of state tobacco control youth access laws. Nine state laws that prohibit youth access to tobacco were coded and summed to establish the total youth access to tobacco score. The composite score ranges from 0 to 31 with a higher score indicating stricter policies.

Synar retailer violation rate (%)

We obtained state retailer violation rates from Substance Abuse and Mental Health Services Administration for the years 2000–2006 as a way to quantify compliance with laws prohibiting the sale of tobacco to youth. Lower rates are indicative of fewer retailer violations statewide.

Acres of tobacco harvested

Although not a state policy, we included the number of acres of tobacco harvested per state as a covariate in our models to account for a culture that may be supportive of tobacco use. These data were obtained from the US Department of Agriculture, National Agricultural Statistics Service.57 The number of tobacco acres harvested per state was only available for the years 1997, 2002, and 2007; thus, data for the remaining years needed for our study were estimated using linear interpolation.

Primary Independent Variables. School-level Policies

School-level policy variables to deter youth tobacco use came from the Youth, Education, & Society (YES) Surveys of School Principals, a cross-sectional nationally representative study of US secondary schools also participating in MTF. The 2 school-based tobacco policies that were assessed in the current study included monitoring compliance to the school’s tobacco policy and the severity of students’ punishment when caught violating the tobacco policies.

The level of monitoring compliance to the school’s tobacco policy was measured by averaging 7 items that assess students’ compliance in school (5 items) and at school-sponsored events (2 items). School administrators indicate on a 5-point scale (1 = not at all strictly to 5 = very strictly) how strictly they monitor tobacco use during the school day in school buildings, parking lots, playing fields, school grounds, and school buses, in addition to school-sponsored sporting and social events.

School administrators also reported on items related to the actions that they take when students were found to be in violation of the tobacco use policy. Using similar methods published elsewhere,58 schools were assigned scores according to severity of their punishments to misbehaviors, taking into consideration whether the student’s violation was the first, second, or third (or more) time. Fifteen possible disciplinary actions were rated on a 7-point scale where 1 = least severe and 7 = most severe. Data on the severity of consequences for violating school tobacco policies were only available through 2003.

Primary Independent Variables. Anti-tobacco Media Campaigns

Commercial ratings purchased from Nielson Media Research were used to examine average audience exposure to anti-tobacco campaigns that are representative of those that were shown on any of the Nielsen Monitor-Plus monitored network and cable television distributors nationally and for local spot, clearance, and syndication television across the largest 75 media markets (called designated market areas) in the US for the years 2000–2006. These 75 markets accounted for most of the viewing households in the US (78%). We used Nielson’s Targeted Rating Points (TRPs) to specifically examine exposure in adolescents who are ages 12–17 over a 4-month period. The Nielsen data were linked to individual-level data from the MTF based on state and county Federal Information Processing Standard (FIPS) code of the respondents’ schools. TRP values estimate the reach and frequency of advertising to a targeted group. One hundred (100) TRPs are equal to one exposure. A depreciated sum of mean exposure to anti-tobacco advertising over 4 months (survey month plus the 3 preceding months) was calculated using existing methods.59 For the purposes of analysis, we created 2 dummy variables (reference group = 0 TRPs): (1) greater than zero but less than one mean exposure; and (2) one or more mean exposures.

Statistical Analysis

Logistic regression was used to model the likelihood of each dichotomous tobacco outcome (current cigarette use, daily cigarette use, current use of smokeless tobacco) in 2000 through 2006. Participants that were included in the analysis were those students participating in the MTF survey 2000–2006 whose schools also participated in the YES survey, and who were in the top 75 designated market areas (thus had anti-smoking advertising data). The models were built in 2 steps. First state tobacco policies (including smoke-free air score, cigarette excise tax, PUP index, youth access to tobacco score, and Synar retailer violation rates), level of school tobacco policy monitoring, and exposure to anti-smoking advertising, were each examined separately as independent variables, adjusting for individual, familial, and school characteristics as well as state tobacco acres harvested and survey year. Results from this first stage of model development are discussed only briefly, and results are shown in Appendix A. Second, full models were run including all covariates. Each grade level (8th, 10th, and 12th grades) was modeled separately because effects of factors influencing substance use may differ between older and younger teens.60,61 In addition, past MTF studies have examined substance use outcomes separately by grade level;6064 therefore, we opted to utilize a similar approach. Separate models also were run for 2000 through 2003, including the above independent variables, as well as severity of punishment for violation of the school’s tobacco policy (this variable was not available after 2003). Parameter estimates and standard errors were calculated using the SAS (Version 9.2, SAS Institute, Cary, NC) procedure “surveylogistic,” employing state (the highest level of nesting) as the clustering unit and applying sampling weights to adjust for differential selection probabilities due to the survey’s multi-stage complex sample design. Adjusted odds ratios and 95% confidence intervals are reported. Analyses were restricted to students for whom both school tobacco policy data and Nielsen data on anti-smoking advertising exposure were available (8th grade weighted N = 55,520, 10th grade weighted N = 51,614, and 12th grade weighted N = 46,514).

RESULTS

Demographic Characteristics and Tobacco Use

Characteristics of the 8th, 10th, and 12th graders for 2000–2006 in our sample are presented in Table 1. Past 30-day cigarette use was lowest among 8th graders (10%), followed by 10th graders (17%), and 12th graders (26%). The percent of current smokers who smoked daily was approximately 46%, 54%, and 62% for 8th, 10th, and 12th grade, respectively. Rates of smokeless tobacco use were much less than rates of current cigarette use (smokeless tobacco use 8th graders 4%, 10th graders 5%, 12th graders 7%). A little over half of the sample for each grade was female, the majority was non-Hispanic white, most lived in a house with 2 parents, and over 70% had at least one parent with education higher than high school. Additional demographic characteristics and school characteristics are shown in Table 1.

Table 1
Characteristics of Participants, 2000–2006

Median tobacco policy monitoring scores were approximately 4.5 for all grades and the median state cigarette excise tax students were exposed to was $0.69 to $0.77 for the 3 grades. Medians for exposure to other state tobacco policies are also shown in Table 1. Regarding exposure to anti-smoking advertising, across grades nearly one-fourth of students were in media markets with no exposure (8th graders 24%, 10 graders 25%, 12th graders 19%). Over half of the students were in media markets with greater than zero but less than one mean exposure (8th graders 56%, 10th graders 62%, 12th graders 62%), and only 1 in 5 were in markets with one or more mean exposures.

Eighth Grade Multivariable Models

Models that examined tobacco use as a function of each policy and anti-smoking advertising variable separately (adjusting for individual, familial and school characteristics) among 8th graders are shown in Appendix A, Table S1. Higher cigarette excise taxes were significantly associated with decreased odds of current and daily cigarette use as well as current smokeless tobacco use. Stronger SFA policies were associated with decreased odds of daily cigarette use. Greater exposure to anti-smoking advertising was associated with decreased likelihood of current cigarette and smokeless tobacco use.

Results from full multivariable logistic regression models including all independent variables are provided in Table 2. Regarding individual-level demographic variables, male students had decreased odds of current cigarette use but increased odds of smokeless tobacco use. Participants who were age 14 or older had increased odds of all 3 tobacco use outcomes. Black students had decreased odds of all current and daily cigarette use, but Hispanic students had increased odds of current cigarette use. Family-related factors also impacted tobacco use. Students with at least one parent with more than a high school degree were at decreased risk for all 3 tobacco outcomes, whereas those without both parents in the home were more likely to use tobacco.

Table 2
Eighth Grade Multivariable Logistic Regression Models Predicting Tobacco Use, 2000–2006

Regarding school/community characteristics, students whose school had a higher percent of students receiving free/reduced cost lunch were more likely to use tobacco, and those in schools with higher percentages of black or Hispanic students had decreased odds of all 3 tobacco use outcomes. Students in public schools had greater odds of daily smoking. Students in small or medium sized schools (versus large schools) had decreased odds of current smoking and daily smoking, and students in SMSAs had decreased odds for smokeless tobacco use.

The only state tobacco policy significantly associated with reduced tobacco use among 8th graders was cigarette taxes. Whereas higher cigarette taxes appeared protective of current and daily cigarette use, this level did not reach statistical significance. However, higher cigarette taxes were associated with decreased odds of current use of smokeless tobacco. Living in a state with a $1 higher cigarette tax reduced the odds of smokeless tobacco use by approximately 30%. School monitoring for tobacco policies was not significantly associated with any of the outcomes.

Being in media markets with >0 mean exposure to anti-smoking advertising was significantly protective of current use of cigarettes and current use of smokeless tobacco. Compared to 8th grade students in media markets with no exposure to anti-smoking advertising, exposure to < 1 TRP was associated with 15% reduced odds of current cigarette use, and 31% reduced odds of current smokeless tobacco use. Exposure to ≥ 1 TRP reduced the odds even more so, with 24% reduced odds for current cigarette use and 45% reduced odds for smokeless tobacco use.

Tenth Grade Multivariable Models

When we examined each policy and anti-smoking advertising variable in separate models, higher PUP index was associated with increased odds of smokeless tobacco use (results shown in Appendix A, Table S2) among 10th graders. No other policies or advertising exposure were significantly associated with tobacco use in these models.

Results of full multivariable models among 10th graders including all independent variables are shown in Table 3. Variables with strong associations with tobacco use were individual, familial, and school-level characteristics. Tenth grade boys were less likely than girls to use cigarettes, but more likely to use smokeless tobacco. Those who were 16 or older had increased odds for all 3 tobacco outcomes. Black or Hispanic students were less likely to use tobacco. Students with at least one parent with more than a high school education had decreased odds of all 3 tobacco outcomes, whereas those without both parents in the home were more likely to use tobacco. Those in public schools were more likely to be daily smokers, and those in schools with a higher percentage of students receiving free/reduced cost lunch were more likely to use smokeless tobacco. Students whose school had a higher percentage of Hispanic students, and those living in a SMSA were less likely to use tobacco.

Table 3
Tenth Grade Multivariable Logistic Regression Models Predicting Tobacco Use, 2000–2006

None of the state tobacco policies were significantly associated with any of the tobacco use outcomes. Students in states with greater number of harvested tobacco acres were at increased odds of current cigarette use and current smokeless tobacco use. Living in media markets with > 0 TRP exposure to anti-smoking advertising was not significantly associated with any of the tobacco use outcomes, nor was school tobacco policy monitoring.

Twelfth Grade Multivariable Models

When we examined each policy and anti-smoking advertising variable in separate models (shown in Appendix A, Table S3), higher cigarette taxes and SFA scores were associated with reduced odds of all 3 tobacco use outcomes. In addition, higher Synar retailer violation rates were significantly associated with increased odds of all 3 tobacco use outcomes.

Table 4 provides results from 12th grade models including all independent variables. Twelfth grade males were slightly more likely to be current users of cigarettes, and were much more likely to use smokeless tobacco. Those who were 18 or older had increased odds of current cigarette or smokeless tobacco use. Black and Hispanic students were less likely to use cigarettes or smoke daily; black students also were less likely to use smokeless tobacco. Not having both parents in the home increased odds of current cigarette use and daily smoking, and having more educated parents decreased the odds of these outcomes and smokeless tobacco use. Students in schools with a higher percentage of Hispanic students had decreased odds of all 3 tobacco outcomes. Those in small schools (vs large) were more likely to be daily smokers, and those in medium size schools were more likely to use smokeless tobacco.

Table 4
Twelfth Grade Multivariable Logistic Regression Models Predicting Tobacco Use, 2000–2006

Higher cigarette excise taxes were significantly protective of current cigarette use and current use of smokeless tobacco. Living in a state with $1 higher cigarette tax reduced the odds of these outcomes by 16% to 33%. Higher smoke-free air scores were associated with reduced odds of smokeless tobacco use (1 point higher score associated with 10% reduced odds of use). Unexpectedly, a higher youth access to tobacco score was associated with slightly increased odds of smokeless tobacco use. School tobacco policy monitoring and living in media markets with >0 TRP for exposure to anti-smoking advertising were not significantly associated with any of the tobacco use outcomes.

Severity of Consequences for Violating School Tobacco Policies

Data on the severity of consequences for violating school tobacco policies were only available through 2003. Median severity of consequences scores during 2000–2003 were 4.9 (inter-quartile range [IQR] 4.3–5.9) for the 8th graders, 4.7 (IQR 4.0–5.9) for 10th graders, and 4.9 (IQR 4.2–5.6) for 12th graders. Models restricted to years 2000–2003 included the same independent variables as models presented in Tables 24, but also including the score for severity of consequences for violating school tobacco policies as a predictor. In each grade, there were no statistically significant associations between school severity of consequences and any of the tobacco outcomes.

DISCUSSION

This study makes important contributions to tobacco research and tobacco control policy by using linked national data from multiple sources for a simultaneous examination of the relative contributions of proximal and distal factors that are known to be associated with youth tobacco use behaviors. Additionally, our study spans the entire spectrum of adolescence by examining data from middle-school through high-school. Our analyses showed that Caucasian youth have the highest risk for tobacco use with the exception of 8th graders where Hispanics had the most risk. Related, we found that schools with a higher percentage of minority students tended to be a protective factor for youth tobacco use behaviors. For the most part, our findings corroborate the well-documented findings that report African-American and Hispanic young adults use tobacco less and initiate tobacco use later in life compared to their Caucasian counterparts.65,66 Still, despite their delayed onset of tobacco use versus Caucasian youth, members of minority groups have demonstrated consistently higher smoking-related illness and mortality rates.67 Therefore, whereas our findings make important progress towards signaling who is at risk for youth tobacco use and when this risk occurs, reducing youth tobacco use behaviors across all individual-level characteristics should stay a public health priority.

Correspondingly, familial-level factors were also strong correlates of tobacco use behaviors, specifically parental educational attainment and number of parents in the household. In addition, tobacco use was higher at times when participants attended schools with more students receiving free or reduced cost lunch. It is widely recognized that one’s life circumstances can profoundly impact one’s tendency to engage in risk behaviors and it may be that the family and school characteristics that we measured are indicative of home-life and/or community/school conditions (eg, socio-economic status).26 Thus, our findings support the contention that tobacco control advocates should work towards reaching young people whose social and/or economic situations may be influencing their decisions to engage in tobacco use.

Our study found that higher cigarette prices were also strongly associated with reduced youth tobacco use behaviors even after controlling for influential proximal and distal factors but this occurred only among 12th graders. Our findings corroborate other studies that have similarly found that the cigarette use behaviors of older teens are more responsive to price when compared to younger teens.6870

Existing studies have posed numerous reasons to explain this price sensitivity including their limited access to disposable income and short-history of tobacco use potentially creating a response to even small price increases in cigarettes.7172 In any case, our results carry important implications for several states that are lagging behind in terms of their low cigarette tax. For instance, state-level cigarette taxes vary widely with a range from 17 cents per pack in Missouri to $4.35 in New York. In addition, our findings are relevant for targeting high school seniors who are at a critical period in their development when tobacco use behaviors often transition into addictive patterns.

We found that higher smoke-free indoor air (SFA) policy scores reduced the odds of tobacco use, but only in models that did not include other state and school policies or exposure to anti-smoking advertising. When including all of these covariates, significance was lost, perhaps because the effects were confounded by cigarette taxes, which tend to have a stronger effect.73 That effect notwithstanding, continued support for smoke-free air legislation is encouraged because of its success in reducing secondhand smoke exposure and modifying social norms regarding the permissibiliy of smoking.74

We found fewer significant associations between other tobacco control policies and youth tobacco use. In the case of PUP tobacco laws, they have been criticized for a number of reasons including their endorsement by tobacco companies and the low likelihood of detection among youth.75 Nevertheless, efforts that strive to prevent and decrease tobacco use at a population-level are relevant to sustain because of the messages they may send to youth about tobacco-related harms and their benefits to society (ie, smoke-free air laws).

Exposure to anti-tobacco media campaigns was associated with reduced likelihood of cigarette smoking and smokeless tobacco use behaviors, but these effects were limited to the youngest students (8th graders) in our sample. It may be that the tobacco attitudes of younger students are more malleable versus their older counterparts, and that the 8th graders responded more readily to the anti-tobacco media messages.76 On the other hand, our findings may be due to the low exposure to anti-tobacco media campaigns across all youth in our study. A related factor may be that exposure to anti-tobacco campaigns are captured on the population level, not the individual level, which limits the precision of the ratings. Similarly, we found that the level of monitoring of the school tobacco-related policies and the severity of punishment for students who were caught violating these policies have minimal to no impact on reducing youth tobacco use behaviors. A significant barrier hindering the efficacy of school-based and media campaigns that deliver anti-tobacco messages is their inability to compete with other, more salient environmental influences, namely pro-tobacco messages delivered by the media and peers.7780 Still, anti-tobacco campaigns, whether delivered in schools or via the media, continue to be an essential part of a comprehensive tobacco control program; it is important to refine the salient school and media-based strategies that have the potential to impact youth tobacco use behaviors.

Among the limitations of this project is our use of cross-sectional data, which leaves us unable to make direct causal inferences. In addition, although the implications of our study are significant for tobacco control efforts, we acknowledge that our data are relatively dated, and therefore, do not account for more temporal changes in the tobacco culture, including an emergence of new tobacco that may appeal to young people. Nevertheless, in the present study, we utilize an ecological approach to work towards comprehensively explaining youth tobacco use behaviors, and in doing so, we highlight how such rigorous analysis could be valuable for understanding how multiple levels of influence interplay to impact youth tobacco use. Additionally, MTF is a school-based survey that does not include data from high school dropouts or adolescents schooled at home. We further acknowledge that in-school surveys can underestimate the substance use of certain populations but note that our findings will be highly relevant for the majority of youth in this country (~90%) and for school-based prevention programs. In addition, measures of enforcement beyond the Synar retailer violation rate are important to consider, but are not currently available, and were therefore, omitted from analyses. Nielsen data only account for 78% of the US markets, which could omit US regions where tobacco use behaviors among young people is prevalent (eg, rural areas), and it is possible that omitting these areas from our analysis could account for the null findings. Likewise, although we evaluate multiple levels of influence, it is beyond the scope of any study to examine every known determinant of youth tobacco use. Therefore, it is possible that unmeasured variables are influencing our models and future studies should continue to explore the causal pathways to tobacco use behaviors (for eg, understanding how higher cigarette prices reduce smokeless tobacco use behaviors). Similarly, we focus our analyses on state and school-level policies and programs where prevention initiatives are usually decided upon and implemented versus at more regional levels.

Despite the limitations, the implications that are derived are significant for tobacco use prevention efforts. Our findings show that multiple levels of influence are important to consider for potentially impacting youth tobacco use, especially demographic and socio-economic characteristics at the individual, familial, and school-levels. Communities have the potential to reinforce tobacco-related norms of families who reside there, and we found that community characteristics, namely racial/ethnic makeup, may also influence youth tobacco use behaviors. In addition, from the numerous tobacco control policies that we examined, higher cigarette prices had the strongest association with youth tobacco use behaviors, and this has important implications for the states in this country where cigarette taxes remain considerably low. Furthermore, anti-tobacco media effects demonstrate success with reducing youth tobacco use but we found that 25% of students had no exposure to anti-tobacco media and overall exposure was low.

IMPLICATIONS FOR TOBACCO REGULATION

This study takes important steps towards delineating the multiple levels of influence that are strong correlates of youth tobacco use behaviors and should be considered as part of a comprehensive tobacco control agenda. Increasing taxes on cigarettes and other tobacco products should be enacted but do fall outside US Food and Drug Administration (FDA) authority. Thus, we conclude by acknowledging a number of available options to the FDA that could limit the appeal of tobacco products to young people. For instance, the use of graphic warning labels and plain packaging are effective strategies for reducing tobacco use in other countries,8184 but they have not yet been implemented in the US. In addition, limiting the marketing novel tobacco products (ie, hookah and electronic cigarettes) is essential for preventing their initiation among young people. Furthermore, public health campaigns addressing the harms of tobacco use, especially the use of emerging tobacco products, should continue to remain a high priority.

Supplementary Material

01

Acknowledgments

This research was supported by National Institutes of Health Grants R01DA039455 (PCR) and R01DA032843 (PCR).

Footnotes

Human Subjects Statement

This study was reviewed and approved by the Washington University Human Research Protection Office.

Conflict of Interest Statement

One of the authors, Dr Bierut, is listed as an inventor on Issued US Patent 8, 080, 371, “Markers for Addiction,” covering the use of certain SNPs in determining the diagnosis, prognosis, and treatment of addiction. All other authors of this article declare they have no conflicts of interest.

Contributor Information

Patricia A. Cavazos-Rehg, Washington University School of Medicine, Department of Psychiatry, St. Louis, MO.

Melissa J. Krauss, Washington University School of Medicine, Department of Psychiatry, St. Louis, MO.

Shaina J. Sowles, Washington University School of Medicine, Department of Psychiatry, St. Louis, MO.

Edward L. Spitznagel, Washington University School of Medicine, Division of Biostatistics, St. Louis, MO.

Richard Grucza, Washington University School of Medicine, Department of Psychiatry, St. Louis, MO.

Frank J. Chaloupka, University of Illinois at Chicago, Department of Economics, Chicago, IL.

Laura J. Bierut, Washington University School of Medicine, Department of Psychiatry, St. Louis, MO.

References

1. Samet JM. Tobacco smoking the leading cause of preventable disease worldwide. Thorac Surg Clin. 2013;23(2):103–112. [PubMed]
2. Thun MJ, Carter BD, Feskanich D, et al. 50-Year trends in smoking-related mortality in the United States. N Engl J Med. 2013;368(4):351–364. [PMC free article] [PubMed]
3. Johnston LD, Miech RA, O’Malley PM, et al. Use of alcohol, cigarettes, and number of illicit drugs declines among U.S. teens. Ann Arbor, MI: University of Michigan News Service; 2015. [Accessed June 29]. Available at: http://www.monitor-ingthefuture.org.
4. Yamin CK, Bitton A, Bates DW. E-cigarettes: a rapidly growing internet phenomenon. Ann Intern Med. 2010;153(9):607–609. [PubMed]
5. Severson HH, Forrester KK, Biglan A. Use of smokeless tobacco is a risk factor for cigarette smoking. Nicotine Tob Res. 2007;9(12):1331–1337. [PubMed]
6. Schuster RM, Hertel AW, Mermelstein R. Cigar, cigarillo, and little cigar use among current cigarette-smoking adolescents. Nicotine Tob Res. 2013;15(5):925–931. [PMC free article] [PubMed]
7. Richardson A, Xiao H, Vallone DM. Primary and dual users of cigars and cigarettes: profiles, tobacco use patterns and relevance to policy. Nicotine Tob Res. 2012;14(8):927–932. [PubMed]
8. Maher JE, Bushore CJ, Rohde K, et al. Is smokeless tobacco use becoming more common among U.S. male smokers? Trends in Alaska. Addict Behav. 2012;37(7):862–865. [PubMed]
9. Lund KE, McNeill A. Patterns of dual use of snus and cigarettes in a mature snus market. Nicotine Tob Res. 2013;15(3):678–684. [PMC free article] [PubMed]
10. US Centers for Disease Control and Prevention. Current tobacco use among middle and high school students--United States, 2011. MMWR Morb Mortal Wkly Rep. 2012;61(31):581–585. [PubMed]
11. Agaku IT, Ayo-Yusuf OA, Vardavas CI, et al. Predictors and patterns of cigarette and smokeless tobacco use among adolescents in 32 countries, 2007—2011. J Adolesc Health. 2014;54(1):47–53. [PMC free article] [PubMed]
12. Van Ryzin MJ, Fosco CM, Dishion TJ. Family and peer predictors of substance use from early adolescence to early adulthood: an 11-year prospective analysis. Addict Behav. 2012;37(12):1314–1324. [PMC free article] [PubMed]
13. Tanski SE, Stoolmiller M, Gerrard M, et al. Moderation of the association between media exposure and youth smoking onset: race/ethnicity, and parent smoking. Prev Sci. 2012;13(1):55–63. [PMC free article] [PubMed]
14. Hiscock R, Bauld L, Amos A, et al. Socioeconomic status and smoking: a review. Ann NY Acad Sci. 2012;1248:107–123. [PubMed]
15. Villanti A, Boulay M, Juon HS. Peer, parent and media influences on adolescent smoking by developmental stage. Addict Behav. 2011;36(1–2):133–136. [PubMed]
16. Goldade K, Choi K, Bernat DH, et al. Multilevel predictors of smoking initiation among adolescents: findings from the Minnesota Adolescent Community Cohort (MACC) study. Prev Med. 2012;54(3–4):242–246. [PMC free article] [PubMed]
17. Pierce JP, White VM, Emery SL. What public health strategies are needed to reduce smoking initiation? Tob Control. 2012;21(2):258–264. [PubMed]
18. Liao Y, Huang Z, Huh J, et al. Changes in friends’ and parental influences on cigarette smoking from early through late adolescence. J Adolesc Health. 2013;53(1):132–138. [PMC free article] [PubMed]
19. Kelly AB, O’Flaherty M, Connor JP, et al. The influence of parents, siblings and peers on pre-and early-teen smoking: a multilevel model. Drug Alcohol Rev. 2011;30(4):381–387. [PubMed]
20. Ferguson CJ, Meehan DC. With friends like these…: peer delinquency influences across age cohorts on smoking, alcohol and illegal substance use. Eur Psychiatry. 2011;26(1):6–12. [PubMed]
21. Mahabee-Gittens EM, Xiao Y, Gordon JS, et al. Continued importance of family factors in youth smoking behavior. Nicotine Tob Res. 2012;14(12):1458–1466. [PMC free article] [PubMed]
22. Farrelly MC, Loomis BR, Han B, et al. A comprehensive examination of the influence of state tobacco control programs and policies on youth smoking. Am J Public Health. 2013;103(3):549–555. [PubMed]
23. Wilson LM, Avila Tang E, Chander G, et al. Impact of tobacco control interventions on smoking initiation, cessation, and prevalence: a systematic review. J Environ Public Health. 2012;961724:1–36. [PMC free article] [PubMed]
24. Huang J, Chaloupka FJ. The Impact of the 2009 Federal Tobacco Excise Tax Increase on Youth Tobacco Use (No. w18026) Cambridge, MA: National Bureau of Economic Research; 2012.
25. Farrelly MC, Loomis BR, Kuiper N, et al. Are tobacco control policies effective in reducing young adult smoking? J Adolesc Health. 2014;54(4):481–486. [PMC free article] [PubMed]
26. Tauras JA, Huang J, Chaloupka FJ. Differential impact of tobacco control policies on youth sub-populations. Int J Environ Res Public Health. 2013;10(9):4306–4322. [PMC free article] [PubMed]
27. Lipperman-Kreda S, Grube JW, Friend KB. Contextual and community factors associated with youth access to cigarettes through commercial sources. Tob Control. 2014;23:39–44. [PMC free article] [PubMed]
28. Kandel DB, Gebre-Egziabher K, Schaffran C, et al. Racial/ethnic differences in cigarette smoking initiation and progression to daily smoking: a multilevel analysis. Am J Public Health. 2004;94(1):128–135. [PubMed]
29. Unger JB, Cruz T, Baezconde-Garbanati L, et al. Exploring the cultural context of tobacco use: a transdisciplinary framework. Nicotine Tob Res. 2003;5(Suppl 1):S101–S117. [PubMed]
30. Flay BR, Petraitis J, Hu FB. Psychosocial risk and protective factors for adolescent tobacco use. Nicotine Tob Res. 1999;1(Suppl 1):S59–S65. [PubMed]
31. Freedman KS, Nelson NM, Feldman LL. Smoking initiation among young adults in the United States and Canada, 1998–2010: a systematic review. Prev Chronic Dis. 2012;9:E05. [PMC free article] [PubMed]
32. Harrell JS, Bangdiwala SI, Deng S, et al. Smoking initiation in youth: the roles of gender, race, socioeconomics, and developmental status. J Adolesc Health. 1998;23(5):271–279. [PubMed]
33. Mathur C, Erickson DJ, Stigler MH, et al. Individual and neighborhood socioeconomic status effects on adolescent smoking: a multilevel cohort-sequential latent growth analysis. Am J Public Health. 2013;103(3):543–548. [PMC free article] [PubMed]
34. Oman RF, Vesely SK, Tolma E, et al. Does family structure matter in the relationships between youth assets and youth alcohol, drug and tobacco use? J Res Adolesc. 2007;17(4):743–766.
35. Brown SL, Rinelli LN. Family structure, family processes, and adolescent smoking and drinking. J Res Adolesc. 2010;20(2):259–273. [PMC free article] [PubMed]
36. Newport JHM. Time in school: how does the U.S. compare? [Accessed December 16, 2014];Center for Public Education. 2011 Available at: http://www.centerforpubliceducation.org/Main-Menu/Organizing-a-school/Time-in-school-How-does-the-US-compare.
37. Alexander C, Piazza M, Mekos D, Valente T. Peers, schools, and adolescent cigarette smoking. J Adolesc Health. 2001;29(1):22–30. [PubMed]
38. Murray DM, Hannan PJ. Planning for the appropriate analysis in school-based drug-use prevention studies. J Consult Clin Psychol. 1990;58(4):458–468. [PubMed]
39. Fuqua JL, Gallaher PE, Unger JB, et al. Multiple peer group self-identification and adolescent tobacco use. Subst Use Misuse. 2012;47(6):757–766. [PMC free article] [PubMed]
40. Ding A. Youth are more sensitive to price changes in cigarettes than adults. Yale J Biol Med. 2003;76(3):115–124. [PMC free article] [PubMed]
41. Townsend J, Roderick P, Cooper J. Cigarette smoking by socioeconomic group, sex, and age: effects of price, income, and health publicity. BMJ. 1994;309:923–927. [PMC free article] [PubMed]
42. Chaloupka FJ, Straif K, Leon ME. Effectiveness of tax and price policies in tobacco control. Tob Control. 2011;20(3):235–238. [PubMed]
43. Botello-Harbaum MT, Haynie DL, Iannotti RJ, et al. Tobacco control policy and adolescent cigarette smoking status in the United States. Nicotine Tob Res. 2009;11(7):875–885. [PMC free article] [PubMed]
44. Carpenter C, Cook PJ. Cigarette taxes and youth smoking: new evidence from national, state, and local Youth Risk Behavior Surveys. J Health Econ. 2008;27(2):287–299. [PubMed]
45. Ross H, Chaloupka FJ. The effect of cigarette prices on youth smoking. Health Econ. 2003;12(3):217–230. [PubMed]
46. Emory KT, Messer K, Vera L, et al. Receptivity to cigarette and tobacco control messages and adolescent smoking initiation. Tob Control. 2015;24(3):281–284. [PMC free article] [PubMed]
47. Farrelly MC, Healton CG, Davis KC, et al. Getting to the truth: evaluating national tobacco countermarketing campaigns. Am J Public Health. 2002;92(6):901–907. [PubMed]
48. Sly DF, Hopkins RS, Trapido E, et al. Influence of a counteradvertising media campaign on initiation of smoking: the Florida” truth” campaign. Am J Public Health. 2001;91(2):233–238. [PubMed]
49. Pierce J, White M, Gilpin E. Adolescent smoking decline during California’s tobacco control programme. Tob Control. 2005;14(3):207–212. [PMC free article] [PubMed]
50. Kandra K, McCullough A, Summerlin-Long S, et al. The evaluation of North Carolina’s state-sponsored youth tobacco prevention media campaign. Health Educ Res. 2013;28(1):1–14. [PubMed]
51. US Centers for Disease Control and Prevention (CDC) Best Practices for Comprehensive Tobacco Control Programs—2007. Atlanta, GA: US Department of Health and Human Services, CDC, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health; 2007. [Accessed December 8, 2015]. Available at: http://www.das.state.ct.us/rfpdoc/DPH06/bids/2009-0919.pdf.
52. Wakefield MA, Durkin S, Spittal MJ, et al. Impact of tobacco control policies and mass media campaigns on monthly adult smoking prevalence. Am J Public Health. 2008;98(8):1443–1450. [PubMed]
53. Wakefield MA, Loken B, Hornik RC. Use of mass media campaigns to change health behaviour. Lancet. 2010;376(9748):1261–1271. [PMC free article] [PubMed]
54. Johnston L, O’Malley P, Bachman J, et al. Monitoring the Future: National Survey Results on Drug Use, 1975–2007; Volume 1, Secondary School Students (NIH Publication No. 08-6418A) Bethesda, MD: National Institute of Drug Abuse; 2008. [Accessed December 8, 2015]. Available at: http://www.monitor-ingthefuture.org/pubs/monographs/vol1_2007.pdf.
55. Orzechowski W, Walker R. [Accessed December 25, 2015];The Tax Burden on Tobacco. Historical Compilation 2011. 2012 Available at: www.2.taxadmin.org/fta/tobacco/papers/tax_burden_2012.pdf.
56. Alciati MH, Frosh M, Green SB, et al. State laws on youth access to tobacco in the United States: measuring their extensiveness with a new rating system. Tob Control. 1998;7(4):345–352. [PMC free article] [PubMed]
57. United States Department of Agriculture. 2002 Census of Agriculture – State Data. Washington, DC: National Agricultural Statistics Service, Department of Agriculture; 2002.
58. Kumar R, O’Malley PM, Johnston LD. School tobacco control policies related to students’ smoking and attitudes toward smoking: national survey results, 1999–2000. Health Educ Behav. 2005;32(6):780–794. [PubMed]
59. Szczypka G, Emery S, Wakefield M, Chaloupka FJ. [Accessed December 8, 2015];The adaption and use of Nielsen Media Research commercial ratings data to measure potential exposure to televised smoking-related advertisements. ImpacTeen Research Paper Series. 2003 :29. Available at: http://www.i.impacteen.org/gen-eralarea_PDFs/Nielsenpaper_051403.pdf.
60. Miech R, O’Malley PM, Johnston L, Patrick ME. E-cigarettes and the drug use patterns of adolescents. Nicotine Tob Res. 2015 [Epub ahead of print] [PubMed]
61. Mu KJ, Moore SE, LeWinn KZ. Internet use and adolescent binge drinking: findings from the Monitoring the Future Study. Addict Behav Rep. 2015;2:61–66. [PMC free article] [PubMed]
62. Hasin DS, Wall M, Keyes KM, et al. Medical marijuana laws and adolescent marijuana use in the USA from 1991 to 2014: results from annual, repeated cross-sectional surveys. Lancet Psychiatry. 2015;2(7):601–608. [PMC free article] [PubMed]
63. Stanley LR, Harness SD, Swaim RC, Beauvais F. Rates of substance use of American Indian students in 8th, 10th, and 12th grades living on or near reservations: update, 2009-2012. Public Health Rep. 2014;129(2):156–163. [PMC free article] [PubMed]
64. Bachman JG, O’Malley PM, Johnston LD, et al. Racial/ethnic differences in the relationship between parental education and substance use among U.S. 8th-, 10th, and 12th-grade students: findings from the Monitoring the Future project. J Stud Alcohol Drugs. 2011;72(2):279–285. [PubMed]
65. Chen P, Jacobson KC. Developmental trajectories of substance use from early adolescence to young adulthood: gender and racial/ethnic differences. J Adolesc Health. 2012;50(2):154–163. [PMC free article] [PubMed]
66. Keyes KM, Vo T, Wall MM, et al. Racial/ethnic differences in use of alcohol, tobacco, and marijuana: is there a cross-over from adolescence to adulthood? Soc Sci Med. 2015;124:132–141. [PMC free article] [PubMed]
67. Haiman CA, Stram DO, Wilkens LR, et al. Ethnic and racial differences in the smoking-related risk of lung cancer. N Engl J Med. 2006;354:333–342. [PubMed]
68. Gruber J. Youth Smoking in the US: Prices and Policies (NBER Working Paper No. 7506) National Bureau of Economic Research; 2000. Jan, [Accessed December 8, 2015]. pp. 1–47. Available at: http://www.nber.org/papers/w7506.pdf.
69. Gruber J, Zinman J. Youth smoking in the US: evidence and implications. In: Gruber J, editor. Risky Behavior among Youths: An Economic Analysis. Chicago, IL: University of Chicago Press; 2001. pp. 69–120.
70. Chaloupka FJ, Cummings KM, Morley CP, Horan JK. Tax, price and cigarette smoking: evidence from the tobacco documents and implications for tobacco company marketing strategies. Tob Control. 2002;11(Suppl 1) [PMC free article] [PubMed]
71. Grossman M, Chaloupka FJ. Cigarette taxes. The straw to break the camel’s back. Public Health Rep. 1997;112(4):290–297. [PMC free article] [PubMed]
72. Lewit E, Coate D, Grossman M. The Effects of Government Regulation on Teenage Smoking (NBER Working Paper No. 655) National Bureau of Economic Research; 1981. Apr, [Accessed December 8, 2015]. pp. 1–48. Available at: http://www.nber.org/papers/w0655.pdf.
73. Fichtenberg CM, Glantz SA. Youth access interventions do not affect youth smoking. Pediatrics. 2002;109(6):1088–1092. [PubMed]
74. Tynan M, Babb S, MacNeil A, et al. State smoke-free laws for worksites, restaurants, and bars - United States, 2000–2010. JAMA. 2011;305(21):2167–2169.
75. Wakefield M, Giovino G. Teen penalties for tobacco possession, use, and purchase: evidence and issues. Tob Control. 2003;12(Suppl 1):6–i13. [PMC free article] [PubMed]
76. Aloise-Young PA, Graham JW, Hansen WB. Peer influence on smoking initiation during early adolescence: a comparison of group members and group outsiders. J Appl Psychol. 1994;79(2):281–287. [PubMed]
77. Sussman S, Black DS, Rohrbach LA. A concise history of school-based smoking prevention research: a pendulum effect case study. J Drug Educ. 2010;40(3):217–226. [PubMed]
78. Glantz SA, Mandel LL. Since school-based tobacco prevention programs do not work, what should we do? J Adolesc Health. 2005;36(3):157–159. [PubMed]
79. Wiehe SE, Garrison MM, Christakis DA, et al. A systematic review of school-based smoking prevention trials with long-term follow-up. J Adolesc Health. 2005;36(6):162–169. [PubMed]
80. Pentz MA, Brannon BR, Charlin VR, et al. The power of policy - the relationship of smoking policy to adolescent smoking. Am J Public Health. 1989;79(7):857–862. [PubMed]
81. Mays D, Niaura RS, Evans WD, et al. Cigarette packaging and health warnings: the impact of plain packaging and message framing on young smokers. Tob Control. 2015;24(e1):e87–e92. [PMC free article] [PubMed]
82. Smith CN, Kraemer JD, Johnson AC, Mays D. Plain packaging of cigarettes: do we have sufficient evidence? Risk Manag Healthc Policy. 2015;8:21–30. [PMC free article] [PubMed]
83. White V, Williams T, Wakefield M. Has the introduction of plain packaging with larger graphic health warnings changed adolescents’ perceptions of cigarette packs and brands? Tob Control. 2015;24(Suppl 2):i42–ii49. [PMC free article] [PubMed]
84. Yong H-H, Borland R, Hammond D, et al. Smokers’ reactions to the new larger health warning labels on plain cigarette packs in Australia: findings from the ITC Australia project. Tob Control. 2015 [Epub ahead of print] [PMC free article] [PubMed]