Cumulative research indicates that cigarette advertising and promotional activities and images of smoking in films have made young people smoke (Lovato et al. 2011). Experimental studies have used quasi-experimental or random designs to better manage risk factors and influences that could confuse the effect of film images on behavior. A recent review summarized the results of eight experimental studies on the effects of smoking on viewers’ beliefs about smoking or reactions to film. The results also suggest that showing young people an anti-smoking ad before watching a movie depicting smoking weakens the favorable attitude among teenagers .

Likewise, a meta-analysis of random effects from the six longitudinal studies in Figure 5.12 produced a grouped RR of 1.76 (95% CI; 1.31–2.37) for the same equation. A meta-analysis of arbitrary effects of the seven studies related to later smoke phases yielded a grouped OR of 1.82 (95% CI; 1.45–2.30). Since OR is an approach to RR, a meta-analysis of random effects from the 17 studies provided a general estimate of the risk of smoking based on high exposure to smoking in films to make it 1.93 (95% CI; 1 , 64–2.27) . In addition, the risks are attributable to the population for the four studies ak bullets that provided such estimates (Dalton et al. 2003, 2009; Sargent et al. 2005; Titus-Ernstoff et al. 2008) produced a fraction of the risk attributable to the general population of 0.44 for adolescent smoking due to smoking exposure in the films . Due to widespread exposure to smoking in films, and because film shows are not viewed with the same skepticism as marketing messages, some authors suggest that smoking films may represent a greater part of the start of smoking young people than traditional cigarette advertising (Glantz 2003; Sargent and Hanewinkel 2009 ; Sargent et al. 2009a).

However, the downward trend in film consumption is consistent with an impact on downward trends in adolescent smoking ”(p. 2212). A second content analysis examined trends based on film classification (Worth et al. 2007), which is important because adolescents are more exposed to films with a youth rating (Sargent et al. 2007b). Overall, the percentage of the top 100 blockbusters representing smoking fell from 91% in 1996 to 63% in 2005.

Tobacco manufacturers have consistently stated that the goal of billions of dollars in cigarette marketing is to attract and retain current adult smokers for their cigarette brands (Tye et al. 1987). In addition, companies deny that marketing campaigns aim to increase demand for cigarettes among existing smokers or to encourage young people to smoke (Cummings et al. 2002). However, the economic value of the amount of the brand change taking place does not justify the amount of the marketing costs (Tye et al. 1987; Siegel et al. 1994). Since most brands are owned by some tobacco manufacturers, most changing brands would have no substantial impact on a company’s profits.

So, Including peer influence only as an independent variable in studies of the direct effect of cigarette advertising on adolescent smoking will underestimate the full effect of cigarette advertising and other influences from the proto-smoking media, such as on-screen smoking exposure in movies (Wills et al. 2007, 2008; Ling et al. 2009). There are strong and consistent indications that marketing affects adolescents’ smoking behavior, including brand selection, smoking and general cigarette smoking (Lovato et al. 2003; DiFranza et al. 2006; Goldberg 2008; NCI 2008). This section discusses empirical data from econometric studies, brand preference studies and studies on changes in smoking in adolescents and smoking their cigarettes. Some studies have analyzed the relationship between advertising and promotion costs and general cigarette consumption, while others have analyzed the relationship between these expenditures and brand preference. A meta-analysis of random effects from the four cross-sectional studies of early smoking among early adolescents, summarized in figure 5.12, yielded a grouped OR of 2.32 (95% CI; 1.98–2.73) for teenagers who smoke in the upper quartile of film smoke exposure compared to the quartile with lower exposure.

As a general rule, teens don’t expect to smoke in the future and discount the power of nicotine addiction when they project their future smoking status. Furthermore, these studies show that even slight changes in firm commitments to abstain from tobacco use increase the risk of smoking in adolescents. In this sense, tobacco ads can indirectly influence real smoking behaviors by diminishing teens’ intentions to abstain from tobacco. There is strong and consistent evidence that advertising and promotion influence factors that directly lead to adolescent tobacco use, including the start of cigarette smoking, as well as its continuation (USDHHS 1989, nineteen ninety six, 1998, 2000; Lynch and Bonnie 1994; Federal Register 1996; Lovato et al. 2003; NCI 2008).

A Norwegian study found that even in the presence of advertising bans, limited exposure to tobacco marketing predicted both current smoking and future smoking intent (Braverman and Aarø 2004). In Norway, surveys were conducted in adolescents aged 13 to 15 years in 1990 and 1995 and at 15 years, despite an advertising ban, 50% of adolescents in each cohort reported exposure to other types of tobacco marketing in the form of tobacco-related paraphernalia, imported newspapers, and television broadcasts from other countries. After controlling potential disruptive factors, teenagers exposed to tobacco marketing were much more likely to be smokers or expect to smoke at the age of 20 than unexposed teenagers. This study establishes a clear link between early exposure and the current and future state of smoking, even if most forms of advertising are limited.

Several cross-sectional studies have found relationships between tobacco marketing exposure and smoking experiments (Schooler et al. 1996; Redmond 1999). In addition, in California, frequent self-reported retail cigarette marketing exposure was independently associated with a significant increase in smoking prospects (Henriksen et al. 2004a; Feighery et al. 2006). In New Zealand, where advertising of tobacco retail is prohibited, a national cross-sectional study found that a higher frequency of visits to tobacconists was linked to an increased risk of sensitivity to smoking and experiments in young people aged 14 to 15 (Paynter et al. 2009). In Canada, the highest levels of cigarette advertising promotions and the lowest prices in school neighborhood stores were related to a higher prevalence of smoking in those schools (Lovato et al. 2007).

Studies on the effectiveness of tobacco-sponsored ads have more often used individual industry-sponsored ad assessments, compared to anti-smoking ad assessments sponsored by public health organizations or other corporate advertising in a forced exposure environment. In these studies, young people are exposed to a range of advertisements individually or in groups and then asked to review each ad immediately after viewing (p. Eg Henriksen and Fortmann 2002; Niederdeppe et al. 2005; Wakefield et al. 2005a; Donovan et al. 2006). Some studies have also required young people to select the ad they considered to be the most effective or to indicate measurements of smoke-related beliefs, attitudes and intentions after exposure (p. Eg Henriksen et al. 2006; Pechmann and Reibling 2006). Others have added follow-up measures to remove and process cognitive advertisements (p. E.g. Terry-McElrath et al. 2005). Of the seven studies summarized in Table 5.10, they all show that advertisements for tobacco-sponsored youth prevention performed poorly in terms of increased knowledge, perceived effectiveness and influence on smoke intent compared to anti-smoking ads sponsored by public health organizations (the seventh study had alcohol-related ads as control ). Advertisements with this type of message generally do not work well compared to advertisements with the serious health effects of smoking and marketing and promotion practices in the industry (Farrelly et al. 2002, 2005; NCI 2008).