摘要：Objectives. We assessed whether an anti-tobacco television advertisement called “Stages,” which depicted a woman giving a brief emotional narrative of her experiences with tobacco use, would be recalled more often and have a greater effect on smoking cessation than 3 other advertisements with different intended themes. Methods. Our data were derived from a sample of 2596 California adult smokers. We used multivariable log-binomial and modified Poisson regression models to calculate respondents’ probability of quitting as a result of advertisement recall. Results. More respondents recalled the “Stages” ad (58.5%) than the 3 other ads (23.1%, 23.4%, and 25.6%; P < .001). Respondents who recalled “Stages” at baseline had a higher probability than those who did not recall the ad of making a quit attempt between baseline and follow-up (adjusted risk ratio [RR] = 1.18; 95% confidence interval [CI] = 1.03, 1.34) and a higher probability of being in a period of smoking abstinence for at least a month at follow-up (adjusted RR = 1.55; 95% CI = 1.02, 2.37). Conclusions. Anti-tobacco television advertisements that depict visceral and personal messages may be recalled by a larger percentage of smokers and may have a greater impact on smoking cessation than other types of advertisements. Globally, one person dies from exposure to tobacco smoke every 6 seconds. 1 Even in the United States, tobacco smoke exposure remains the leading cause of preventable death, with an estimated 480 000 adults dying from tobacco smoke exposure each year. 2 Although the effects of tobacco smoke exposure are most severe among individuals who smoke on a daily basis, nondaily smoking and secondhand smoke exposure can lead to the same negative health consequences that result from daily smoking. 3 Increasing smoking cessation, however, can significantly improve life expectancy; lower the risk of cancer, diabetes, heart disease, and lung disease; and, among women, lower the risk of infertility or low-birthweight children. 2,4–7 Although the benefits of cessation are most prominent among the young, cessation at any age can immediately improve health outcomes. 4 It is recommended that tobacco control programs develop mass-reach health communication interventions to increase cessation among current smokers and reduce smoking initiation among nonsmokers. 8 There are multiple pathways through which mass media are intended to promote cessation, including directly marketing cessation assistance to smokers, changing public opinions about tobacco use to create a social norm against smoking, and increasing interpersonal discussions about tobacco use. 9–11 The majority of evidence indicates that these mass media campaigns are effective in motivating individuals to think about quitting, make quit attempts, obtain help in quitting, and maintain abstinence. 12–31 At the population level, advertisements are also associated with decreases in cigarette consumption and smoking prevalence. 13,15,16,19,32–36 The magnitude of the effect of media campaigns on cessation is estimated to be relatively small; however, when this small effect is applied across an entire population, the reductions can be very meaningful. The United States Centers for Disease Control and Prevention (CDC), for example, recently estimated that its Tips From Former Smokers (TIPS) national ad campaign led to 1.64 million quit attempts and more than 100 000 Americans quitting smoking in just 1 year. 19 Despite this current weight of evidence and the recommendations of Article 12 of the Framework Convention on Tobacco Control, 8 there are several barriers that make it difficult for many countries to develop mass media campaigns. 37 One of the main barriers is the expense of developing and airing advertisements. To provide a perspective, CDC reported receiving $54 million in US federal funding to carry out the TIPS campaign, 19 and since the launch of the California Tobacco Control Program (CTCP) in 1990, approximately $464 million has been spent on mass media campaigns in California, with $13.4 million currently being spent annually. 38 In funding climates that are restrictive, tobacco control programs may be able to reduce costs by focusing on high-impact advertisements. 39 In the past decade, tobacco control advocates have begun promoting the use of mass media that rely on emotional messages, deal with the health consequences of smoking, or use personal narratives to convey their messages. These advertisements are intended to promote behavior change by evoking visceral reactions of sadness, fear, disgust, or anger in what has been referred to as a “fear appeal.” 40,41 Although this concept of “scaring” smokers into quitting is anathema to many public health advocates, 42,43 several researchers have demonstrated that advertisements that deal with the negative consequences of smoking in an emotionally evocative or personal way are effective. They are typically recalled more frequently even when they are aired at lower volumes, 44–46 they reach socioeconomic groups equitably, 34,47–50 and they are typically rated as more effective than advertisements with different themes. 44,46,50–55 In addition, they promote more frequent use of quit lines 12,14,26,29,47 and confer greater effects on quitting behavior. 12,14,26,27,29,34,47–50 Thus far, evaluations of fear appeal advertisements have predominantly relied on group-level measures of exposure rather than individual-level measures of recognition. Exposure, in evaluations of advertisements, typically refers to whether a person has viewed an advertisement, but it does not necessarily imply that the person processed the message or remembered the ad. Exposure is usually estimated via group-level measures (e.g., Nielsen ratings or gross rating points) that assess the average number of people in a given population who are likely to have viewed an advertisement. 56–58 These group-level measures can be very valuable to public health programs in their assessments of how well an advertisement is reaching a population; however, when behavior change resulting from advertisements is assessed at the individual level, these group-level instruments serve only as ecological measures. By contrast, advertisement recall, a measure of an individual’s recognition of an advertisement, may reduce the possibility of ecological bias and serves as an indicator of message processing, an expected direct effect of exposure to mass media campaigns. 59 Although ad recall has long been used in mass media evaluations, 12,60,61 to our knowledge it has not been employed in evaluations of behavior change resulting from fear appeal advertisements. We sought to provide further evidence for the use of advertisements that deal with the health consequences of smoking in a personal and emotional way by assessing recall of 4 anti-tobacco television advertisements developed by the CTCP and demonstrating the effects of recall of these ads on smoking cessation. We tested the following hypotheses: an ad called “Stages” that depicted a woman giving a personal narrative of her negative experiences with tobacco use would be recalled more often than 3 other ads with different intended themes among the current smokers in our cohort (hypothesis 1); recall of “Stages” would be distributed more equitably across individual characteristics in multivariable analyses than would recall of the 3 other ads (hypothesis 2); and recall of “Stages” would have a greater effect on cessation than recall of the 3 other ads (hypothesis 3).
