[Skip to Content]
[Skip to Content Landing]
February 7, 2011

Depictions of Tobacco Use in 2007 Broadcast Television Programming Popular Among US Youth

Author Affiliations

Author Affiliations: Department of Research and Evaluation, American Legacy Foundation (now known as Legacy), Washington, DC (Drs Cullen, Vallone, and Healton and Mss Sokol, Slawek, and Allen); and Department of Sociomedical Sciences, Columbia University, New York, New York (Dr Healton).

Arch Pediatr Adolesc Med. 2011;165(2):147-151. doi:10.1001/archpediatrics.2010.276

Objectives  To determine the quantity of tobacco use in network television programming popular among US youth and to examine variation in tobacco depictions by TV Parental Guidelines system rating and television network.

Design  A content analysis was conducted of broadcast network television programming popular among youth. Nielsen viewership rating data were used to identify a sample of top-rated television series for youth aged 12 to 17 years during the fall 2007 television season. Depictions of tobacco use per television episode were examined by TV Parental Guidelines rating and television network. χ2 testing was used to examine differences in proportions of tobacco depictions across television episode ratings and networks.

Setting  Data collection and analysis were conducted at the American Legacy Foundation (now known as Legacy).

Subjects  Broadcast television viewers in 2007.

Main Outcome Measure  Tobacco use depictions on broadcast television were examined.

Results  Forty percent of television episodes examined had at least 1 depiction of tobacco use. Of these depictions, 89% were of cigarettes. Among episodes rated TV-PG (ie, parental guidance suggested) (N = 73), 50% showed 1 or more incidents of cigarette use, in contrast to 26% of TV-14 (ie, parents strongly cautioned) episodes. The percentage of episodes with any tobacco use depictions was highest on the FOX network (44%; n = 32), followed closely by The CW (CBS–Warner Brothers) (41%; n = 30).

Conclusions  Substantial tobacco use was observed in television shows popular among youth. It is projected that almost 1 million youth were exposed to tobacco depictions through the programming examined. Tobacco use on television should be a cause for concern, particularly because of the high volume of television viewing among younger audiences.

In 2004, 78% of middle school students and 87% of high school students reported seeing actors using tobacco on television or in movies.1 Studies have documented the prevalence of tobacco imagery in movies,1-7 as well as the association between exposure to tobacco imagery in movies and initiation of smoking among youth.8-12 Based on these findings, the National Cancer Institute recently concluded that “the total weight of evidence. . . indicates a causal relationship between exposure to movie smoking depictions and youth smoking initiation.”13(p11-12) In fact, Dalton et al12,13 estimate that exposure to smoking in movies accounts for approximately 52% of all smoking initiation. Because movies and television are not mutually exclusive media channels, the body of evidence pertaining to movies is relevant to television as well, particularly because most movies are shown on television after airing in cinemas. In addition, movie trailers (the advertisements used to promote movies playing in theaters) contain depictions of tobacco use and commonly air on television. A 2006 study14 found that 95% of youth aged 12 to 17 years in the United States were exposed to images of tobacco use on television in the context of a movie trailer during a 1-year period between August 2001 and July 2002. Exposure to these trailers has been shown to increase the attractiveness of smoking among youth who have experimented with cigarettes.15

Several studies16-20 have found a relationship between television viewing and smoking initiation. One study16 used longitudinal data to demonstrate that television viewing was associated with subsequent initiation; another study17 noted a quadratic relationship, with risk for initiation increasing markedly with more hours of television viewing. However, these studies do not document the number or prevalence of tobacco depictions in the televised content, nor do they make any claim regarding a causal association between televised tobacco depictions and youth smoking initiation. Other investigations have documented the prevalence of televised tobacco depictions within popular television series, such as The Simpsons,21,22 and across multiseries samples that are designed to represent popular television more broadly.23-27 The most recent data on US youth are from an Office of National Drug Control Policy study28 conducted in 1998. The study found tobacco depictions in 20% and tobacco-related dialogue in 23% of the sampled episodes, which were drawn from the 20 top-rated situation comedies or dramas popular among US youth aged 12 to 17 years. The study also found tobacco imagery in 6% of programs rated TV-G (defined as appropriate for a general audience), 20% of programs rated TV-PG (ie, parental guidance suggested), and 24% of programs rated TV-14 (ie, parents strongly cautioned).

The objective of this study was to provide current information about the potential exposure of US youth to tobacco imagery via television by quantifying the number of tobacco use depictions in a sample of television programs popular with youth aged 12 to 17 years during the fall 2007 season. We also compare the quantity of tobacco use across television rating categories and television networks.

Sample selection

The sample for this study included new television programming popular among an audience aged 12 to 17 years that aired on broadcast television during prime-time hours in the fall season of 2007 (September 21 through December 31, 2007). Prime time is defined as the part of day during which the greatest television viewership takes place,29 usually 8 to 11 PM. Programming that achieved a top 15 viewership rating among those aged 12 to 17 years according to Nielsen30 viewership data were selected for the sample. Sporting events, movies, awards programming, rerun episodes, and programming on cable television were excluded from the sample. Because of the Hollywood writers' strike in December 2007, two of these series (Family Guy and House) did not air new programming during the last month of the study period. The final study sample included 73 episodes representing the following 8 television series: American Dad, America's Next Top Model, Desperate Housewives, Family Guy, Gossip Girl, Heroes, House, and The Simpsons. All 73 episodes aired on 1 of the following 4 national broadcast television networks: ABC, FOX, NBC, or The CW (CBS–Warner Brothers). These 8 series represent an estimated 61.5 hours of television programming over a 14-week period.


The thumbs up! thumbs down! method for counting tobacco use depictions was used for this study.4,31,32 In brief, this approach consists of counting tobacco use in the following manner: (1) Each instance of on-screen tobacco depiction—regardless of whether the product is lit or how it is depicted in the episode (ie, in an ashtray, on a billboard, in a character's hands, etc)—is counted as a single incident. (2) When 2 characters are shown simultaneously smoking, it is counted as 2 incidents. (3) When a character who is smoking moves his or her cigarette, cigar, or pipe off screen and then back on, it is counted as 2 incidents. (4) When the camera focuses on a character using tobacco, focuses elsewhere, and then refocuses on the character using tobacco, it is counted as 2 incidents. An appearance of a carton of cigarettes is counted as 1 incident, with each visible cigarette in the carton counted as an additional incident.

Using the thumbs up! thumbs down! method, the following 3 forms of tobacco use were analyzed: cigarette, cigar, and pipe use. The number of tobacco incidents is presented as a cumulative count across the series, as well as dichotomously (no tobacco use vs ≥1 incidents of tobacco use per episode) to account for the potential of the thumbs up! thumbs down! approach to produce an overestimation of tobacco use.4 Dichotomous variables for tobacco use depictions were also created for cigarette and cigar use separately (any vs none).

All 73 episodes were analyzed by 2 trained research assistants. A third coder (D.S.) was used to resolve any discrepancies in incident counts. To assess overall agreement in tobacco content counts between the 2 coders, κ interrater reliability statistic was computed for each tobacco variable. Agreement for all forms of tobacco use was high (κ = 0.999, P < .001).

Tv parental guidelines data

All television episodes are categorized according to their rating based on the TV Parental Guidelines system, a national public service classification established by the National Association of Broadcasters, the National Cable Television Association, and the Motion Picture Association of America to help guide parents regarding television content for their children.33 Rating categories, ordered from less to more restrictive, include the following: TV-Y, TV-Y7, TV-G, TV-PG, TV-14, and TV-Mature. All episodes in the study sample were rated TV-PG or TV-14. TV-PG is defined as the following: “Parental Guidance Suggested. This program contains material that parents may find unsuitable for younger children.” In contrast, TV-14 is defined in the following manner: “Parents Strongly Cautioned: This program contains some material that many parents would find unsuitable for children under 14 years of age.”34

Coarse or indecent language, sexual content, and portrayals of violence are the only factors considered in assigning TV Parental Guidelines.33 These ratings can be found on the television screen in real time for the first 15 seconds of each episode, as well as in TV Guide, Reader's Digest, and online. In addition, the Federal Communications Commission33 has made these ratings accessible via the V-chip technology.

Estimation of national tobacco exposure among the target audience

The proportion of youth nationwide exposed to tobacco as a result of viewing the episodes examined in this analysis was estimated using 2 data sources. First, the estimated number of youth aged 12 to 17 years was obtained from the 2000 US Census.35 Second, Nielsen30 data estimating the number and percentage of youth watching the study episodes were obtained. A mean percentage was calculated by adding all the percentages and then dividing the aggregate sum by 73. Finally, the mean percentage of youth watching these 73 episodes was multiplied by the total number of individuals aged 12 to 17 years in the US population, resulting in the number of youth exposed to tobacco as a direct result of the programming examined in this study.

Statistical analysis

Statistical analysis was performed using commercially available software (SAS version 9.1; SAS Institute, Cary, North Carolina).36 Descriptive statistics include the median and interquartile range as measures of dispersion, as well as the absolute count of tobacco use depictions. Two-sided χ2 testing was used to compare TV Parental Guidelines categories and television networks by tobacco use (dichotomized as none vs ≥1 depictions), with episode as the unit of analysis. For contingency tables with at least 20% of cells containing fewer than 5 observations, Fisher exact test P values are reported.


Of 73 episodes examined, 40% (n = 29) included at least 1 depiction of tobacco use, 33% (n = 24) showed 1 or more depictions of cigarette use, and 8% (n = 6) demonstrated at least 1 depiction of cigar use. Most tobacco depictions were of cigarettes, amounting to 271 total incidents in 61.5 total hours of programming analyzed, for a mean of 4.4 incidents per hour (Table 1). Given the low incidence of pipe use, data are not shown for this product (data are available on request from author J.C.). Two of the series, Heroes and Gossip Girl, amounting to 20 episodes (or more than one-quarter of the study sample), did not include any depictions of tobacco use.

Table 1. 
Tobacco Depictions Among 73 Episodes
Tobacco Depictions Among 73 Episodes

More than one-quarter (27% [n = 20]) of episodes examined in this study were rated TV-PG, in contrast to 73% (n = 53) of episodes rated TV-14 (Table 2). When comparing the proportion of episodes with any tobacco use, significantly more episodes rated TV-PG contained at least 1 tobacco depiction vs programming rated TV-14 (60% vs 32%, P = .03). While exactly half (50% [n = 10]) of episodes rated TV-PG had at least 1 cigarette depiction compared with just over one-quarter (26% [n = 14]) of episodes rated TV-14, this difference was not statistically significant (P = .09). In contrast, twice as many episodes rated TV-14 (9% [n = 5]) had 1 or more cigar depictions vs those rated TV-PG (5% [n = 14]), but this difference was also not statistically significant (P = .99). The median number of cigarette use depictions was 1 (interquartile range, 0-4) among episodes rated TV-PG compared with a median of 0 (interquartile range, 0-1) among episodes rated TV-14.

Table 2. 
Tobacco Depictions by TV Parental Guidelines Rating Among 73 Episodes
Tobacco Depictions by TV Parental Guidelines Rating Among 73 Episodes

Most analyzed episodes aired on FOX (44% [n = 32]) network, followed by The CW (30% [n = 22]), NBC (15% [n = 11]), and ABC (11% [n = 8]) (Table 3). All episodes shown on NBC were rated TV-14, yet none contained tobacco depictions. In contrast, more than half of the programming on The CW contained tobacco use depictions; episodes on this network were evenly distributed across rating categories TV-PG and TV-14.

Table 3. 
Tobacco Depictions by Television Network Among 73 Episodes
Tobacco Depictions by Television Network Among 73 Episodes

Significant differences were noted in the distribution of TV Parental Guidelines across television networks (P = .003). Television programming that aired on NBC was exclusively rated TV-14 (100%, n = 1), while most episodes on ABC (75%) and FOX (81%) were rated TV-14, and episodes on The CW were almost evenly distributed across the 2 ratings categories, with 45% rated TV-14 and 55% rated TV-PG.

The greatest proportion of episodes with at least 1 tobacco depiction was observed on FOX (44%), followed by The CW (41%), yet a higher absolute count of tobacco depictions was observed on The CW (n = 157), followed by FOX (n = 143). When comparing the proportions of episodes with 1 or more tobacco use depictions across all networks, significantly greater proportions of FOX and The CW episodes contained tobacco use (P = .004) than was observed in ABC and NBC programming. Nationally, the estimated number of youth aged 12 to 17 years who were exposed to tobacco use as a result of the television programming examined in this study was 940 000.


This is the first study in almost a decade that documents the incidence and proportion of tobacco depictions in a multiseries sample of television programming popular among US youth. Findings indicate that almost 40% of television programming most popular among youth aged 12 to 17 years includes tobacco depictions. Perhaps the most striking observation was that a higher proportion of tobacco use depictions was found among programs rated TV-PG compared with those rated TV-14, indicating that exposure to tobacco use may skew toward youth of younger ages, resulting in earlier exposure to this behavior. In fact, approximately half of all TV-PG–rated programming included some level of tobacco depictions. Estimates project that 940 000 (almost 1 million) youth were exposed to tobacco depictions on television during this single fall programming season.

This study included 1 reality (or allegedly unscripted) television series, America's Next Top Model. This popular series includes an episode in which there were 122 incidents of cigarette use. This incidence far exceeds that of any other single episode in this sample. As a result, analyses were performed with and without this outlier; tests of differences in the proportion of tobacco use depictions were consistent (data are available from author J.C.).

This study has several important limitations. Because the sample included only new and popular television programming and was reduced as a direct result of the writer's strike during the study interval, this study underestimates the total amount of exposure to tobacco depictions. In fact, Nielsen30 data indicate that the typical US youth aged 12 to 17 years watches on average 3 hours 20 minutes of television per day, or 200 minutes per day. Our study captured tobacco use during 61.5 hours (3690 minutes) of television during the 98-day study period, reflecting a mean of 38 minutes per day. If this estimate is taken as a fraction of the daily mean for a typical US teen, the programming examined in this study represents only 19%, or less than one-fifth, of a US youth's mean daily television consumption.37

Future research is warranted to better understand what, if any, effect exposure to televised tobacco depictions has on youth tobacco use initiation, particularly given the high level of television viewing among youth audiences. It is likely that multiple media channels provide exposure to tobacco depictions and, as such, may stimulate a similar effect. Ideally, samples should include a broad portfolio of programming popular among youth, with particular attention to cable and reality television programming.

Public health officials should urge television networks to eliminate all depictions of tobacco use in youth programming. Until then, TV Parental Guidelines should be adjusted so that programs featuring tobacco use are rated TV-MA (mature audience only).


Current TV Parental Guidelines do not consider any form of substance use, including tobacco, in determining a program rating.33 The strong evidence linking youth exposure to tobacco depictions in movies to initiation of smoking among youth should serve as a rationale for including tobacco use depictions as a criterion for rating programming for young audiences. However, only about one-quarter of parents report frequent use of TV Parental Guidelines ratings, and less than 10% of US households have made use of the V-chip, a tool that allows parents to identify and block television programming that they believe may be harmful to their children.37 Efforts to educate parents must be undertaken to increase the use of the TV Parental Guidelines rating system and the V-chip, in conjunction with modernizing the rating system to incorporate tobacco use depictions. Across the few programs that were sampled, almost 1 million youth were exposed to tobacco depictions on television over a single fall programming season. Public health practitioners and policy makers should discourage any tobacco depictions within youth programming. Further restrictions reducing youth exposure to tobacco use on television programming should be considered given the evidence linking smoking imagery in the media to initiation of youth smoking.

Correspondence: Jennifer Cullen, PhD, MPH, Legacy, 1724 Massachusetts Ave NW, Washington, DC 20036 (jcullen@legacyforhealth.org).

Accepted for Publication: July 1, 2010.

Author Contributions:Study concept and design: Cullen, Sokol, Slawek, Vallone, and Healton. Acquisition of data: Sokol and Slawek. Analysis and interpretation of data: Cullen, Slawek, Allen, Vallone, and Healton. Drafting of the manuscript: Cullen, Sokol, Slawek, Allen, Vallone, and Healton. Critical revision of the manuscript for important intellectual content: Cullen, Sokol, Slawek, Allen, and Vallone. Statistical analysis: Cullen. Administrative, technical, and material support: Sokol, Slawek, Allen, Vallone, and Healton. Study supervision: Cullen, Slawek, Vallone, and Healton.

Financial Disclosure: None reported.

Centers for Disease Control and Prevention (CDC), Tobacco use, access, and exposure to tobacco in media among middle and high school students: United States, 2004 [published correction appears in MMWR Morb Mortal Wkly Rep. 2006;55(17):490].  MMWR Morb Mortal Wkly Rep 2005;54 (12) 297- 301PubMedGoogle Scholar
Thompson  KMYokota  F Depiction of alcohol, tobacco, and other substances in G-rated animated feature films.  Pediatrics 2001;107 (6) 1369- 1374PubMedGoogle ScholarCrossref
Polansky  JGlantz  S First-run Smoking Presentations in US Movies 1999-2003.  San Francisco University of California, San Francisco2004;
Mekemson  CGlik  DTitus  K  et al.  Tobacco use in popular movies during the past decade.  Tob Control 2004;13 (4) 400- 402PubMedGoogle ScholarCrossref
Goldstein  AOSobel  RANewman  GR Tobacco and alcohol use in G-rated children's animated films.  JAMA 1999;281 (12) 1131- 1136PubMedGoogle ScholarCrossref
Dalton  MATickle  JJSargent  JDBeach  MLAhrens  MBHeatherton  TF The incidence and context of tobacco use in popular movies from 1988 to 1997.  Prev Med 2002;34 (5) 516- 523PubMedGoogle ScholarCrossref
Charlesworth  AGlantz  SA Smoking in the movies increases adolescent smoking: a review.  Pediatrics 2005;116 (6) 1516- 1528PubMedGoogle ScholarCrossref
Sargent  JDDalton  MABeach  ML  et al.  Viewing tobacco use in movies: does it shape attitudes that mediate adolescent smoking?  Am J Prev Med 2002;22 (3) 137- 145PubMedGoogle ScholarCrossref
Sargent  JDBeach  MLAdachi-Mejia  AM  et al.  Exposure to movie smoking: its relation to smoking initiation among US adolescents.  Pediatrics 2005;116 (5) 1183- 1191PubMedGoogle ScholarCrossref
Distefan  JMPierce  JPGilpin  EA Do favorite movie stars influence adolescent smoking initiation?  Am J Public Health 2004;94 (7) 1239- 1244PubMedGoogle ScholarCrossref
Distefan  JMGilpin  EASargent  JDPierce  JP Do movie stars encourage adolescents to start smoking? evidence from California.  Prev Med 1999;28 (1) 1- 11PubMedGoogle ScholarCrossref
Dalton  MASargent  JDBeach  ML  et al.  Effect of viewing smoking in movies on adolescent smoking initiation: a cohort study.  Lancet 2003;362 (9380) 281- 285PubMedGoogle ScholarCrossref
Davis  RMGilpin  EALoken  BViswanath  KWakefield  MA The Role of the Media in Promoting and Reducing Tobacco Use. Bethesda, MD National Cancer Institute, National Institutes of Health, US Dept of Health and Human Services June2008;http://cancercontrol.cancer.gov/tcrb/monographs/19/m19_complete.pdf. Accessed November 15, 2010
Healton  CGWatson-Stryker  ESAllen  JA  et al.  Televised movie trailers: undermining restrictions on advertising tobacco to youth.  Arch Pediatr Adolesc Med 2006;160 (9) 885- 888PubMedGoogle ScholarCrossref
Hanewinkel  R Cigarette smoking and perception of a movie character in a film trailer.  Arch Pediatr Adolesc Med 2009;163 (1) 15- 18PubMedGoogle ScholarCrossref
Gidwani  PPSobol  ADeJong  WPerrin  JMGortmaker  SL Television viewing and initiation of smoking among youth.  Pediatrics 2002;110 (3) 505- 508PubMedGoogle ScholarCrossref
Gutschoven  KVan den Bulck  J Television viewing and smoking volume in adolescent smokers: a cross-sectional study.  Prev Med 2004;39 (6) 1093- 1098PubMedGoogle ScholarCrossref
Gutschoven  KVan den Bulck  J Television viewing and age at smoking initiation: does a relationship exist between higher levels of television viewing and earlier onset of smoking?  Nicotine Tob Res 2005;7 (3) 381- 385PubMedGoogle ScholarCrossref
Hancox  RJMilne  BJPoulton  R Association between child and adolescent television viewing and adult health: a longitudinal birth cohort study.  Lancet 2004;364 (9430) 257- 262PubMedGoogle ScholarCrossref
Jackson  CBrown  JDL’Engle  KL R-rated movies, bedroom televisions, and initiation of smoking by white and black adolescents.  Arch Pediatr Adolesc Med 2007;161 (3) 260- 268PubMedGoogle ScholarCrossref
Eslick  GDEslick  MG Smoking and The Simpsons.  Med J Aust 2009;190 (11) 637- 639PubMedGoogle Scholar
Hunt  K Re-evaluating gender and smoking in Thunderbirds 35 years on.  Tob Control 2002;11 (2) 151- 153PubMedGoogle ScholarCrossref
Verma  TAdams  JWhite  M Portrayal of health-related behaviours in popular UK television soap operas.  J Epidemiol Community Health 2007;61 (7) 575- 577PubMedGoogle ScholarCrossref
Hanewinkel  RSargent  JD Exposure to smoking in popular contemporary movies and youth smoking in Germany.  Am J Prev Med 2007;32 (6) 466- 473PubMedGoogle ScholarCrossref
McGee  RKetchel  J Tobacco imagery on New Zealand television 2002-2004.  Tob Control 2006;15 (5) 412- 414PubMedGoogle ScholarCrossref
Kaiser Family Foundation, USC Annenberg Normal Lear Center's Hollywood, Health & Society.  How healthy is prime time? an analysis of health content in popular prime time television programs 2008;http://www.kff.org/entmedia/upload/7764.pdf. Accessed November 15, 2010Google Scholar
Cruz  JWallack  L Trends in tobacco use on television.  Am J Public Health 1986;76 (6) 698- 699PubMedGoogle ScholarCrossref
Christenson  PGHenriksen  LRoberts  DFKelly  MCarbone  SWilson  ABOffice of National Drug Control Policy and Mediascope Macro International Inc, Substance use in popular prime-time television 2000;http://www.nmpreventionnetwork.org/primetimetv.pdf. Accessed November 15, 2010
Shapiro  ME The Museum of Broadcast Communications.  Prime Time 2010;http://www.museum.tv/eotvsection.php?entrycode=primetime. Accessed November 15, 2010Google Scholar
Nielsen, Television. http://en-us.nielsen.com/tab/measurement/tv_research. Updated 2009. Accessed November 12, 2010
Mekemson  CGlantz  SA How the tobacco industry built its relationship with Hollywood.  Tob Control 2002;11 ((suppl 1)) I81- I91PubMedGoogle ScholarCrossref
American Lung Association of Sacramento–Emigrant Trails, The thumbs up! thumbs down! 2004 semi-annual report on tobacco in current movies 2004;http://www.scenesmoking.org/docs/2004report.pdf. Accessed November 15, 2010
Federal Communications Commission, Commission finds industry video programming rating system acceptable; adopts technical requirements to enable blocking of video programming (the “V-chip”) (CS docket No. 97-55, CS docket No. 97-321, ET docket 97-206).  Washington, DC2003;Report GN 98-3. http://www.fcc.gov/Bureaus/Cable/News_Releases/1998/nrcb8003.html. Accessed November 12, 2010
Smith  MS V-chip and TV Ratings: Helping Parents Supervise Their Children's Television Viewing Washington, DC Congressional Research Service, Library of Congress June8 2001;http://assets.opencrs.com/rpts/97-43_20010608.pdf. Accessed November 15, 2010
US Census Bureau, Census 2000: population by age, sex, race, and Hispanic or Latino origin for the United States: 2000 (PHC-T-9) 2001;http://www.census.gov/population/www/cen2000/briefs/phc-t9/index.html. Accessed November 15, 2010
SAS Institute, SAS Software Version 9.1 for Windows.  Cary, NC SAS Institute2009;
Kaiser Family Foundation, Parents and the V-chip.  Menlo Park, CA Kaiser Family Foundation2001;