Media Campaigns and Tobacco Control
Michael D. Slater
Ohio State University
School of Communication
Media-based efforts to promote non-use of tobacco products have become an increasingly prominent feature of the tobacco control landscape over the past decade, with aggressive and well-funded efforts having taken place under the auspices of the American Legacy Foundation as well as in California, Massachusetts, and Florida, among other states. These campaigns typically make considerable use of highly visible and dramatic advertising in television and other media. They are sometimes referred to as “counter-advertising,” because such campaigns are seen in the public health community as efforts to provide even a modest counterweight to the very extensive marketing efforts of tobacco companies in the United States, which spent $8.24 billion on advertising and promotion in 1999 (FTC 2001). It should be noted that, like tobacco company marketing efforts, such campaigns often are not confined to the use of advertising, but also involve a wide range of educational, promotional, and social marketing activities in a comprehensive program.
In this appendix, we briefly review highlights of laboratory and field research concerning effects of media and advertising efforts to influence tobacco-related attitudes and behavior (see Agostinelli and Grube 2002; Friend and Levy 2002; and Wakefield et al. 2003b for more detailed reviews). We focus in particular on evaluations of the recent major efforts by Massachusetts, Florida, and California (because these states had the most ambitious and the best-evaluated media efforts), and on the national efforts of the American Legacy Foundation. Based on such evaluations and on meta-analytic work previously published, we suggest effect size ranges that might be anticipated from appropriate national campaign efforts, as well as noting the particular concerns and obstacles associated with successful implementation of such efforts. Issues regarding media campaigns and smoking cessation are addressed briefly at the end of this section.
It should be noted that the different authorities cited report effect sizes in one of two ways: as absolute percent change or as relative percent change. For example, a change from 20 percent youth smoking prevalence to 18 percent prevalence would be described as a 2 percent change effect size. Others would instead report this effect size as a 10 percent reduction in prevalence. This is referred to as relative percent change.
EVALUATIONS OF RECENT MAJOR MEDIA-BASED EFFORTS
In 1988, California passed a citizen-supported initiative to raise cigarette taxes and use funds to support a comprehensive tobacco control initiative. Funds for the program fluctuated between
$54 million and $140.7 million over the next 10 years depending largely on the extent of political support from the governor and the legislature (Independent Evaluation Consortium 2001; Pierce et al. 1998b). An ambitious media campaign began in 1990 that addressed secondhand smoke issues, youth prevention (largely through ads critical of the tobacco industry), and smoking cessation (Independent Evaluation Consortium 2001); only a minority of media expenditures were directed at youth.
Evaluations of the California effort suggested that the campaign had an impact on smoking prevalence in the first 3 years of the effort. Adult smoking prevalence dropped from 22.7 percent to 18 percent from 1989 to 1993, a rate of decline that was about double that of the United States as a whole (Pierce et al. 1998b). According to Friend and Levy (Friend and Levy 2002), similar effect estimates were based on analyses by the Centers for Disease Control and Prevention (CDC 1996). Subsequently, prevalence rates flattened, with no evidence of declines, suggesting that the effects occurred in the first 4 years of the campaign or that the effects were dampened due to reductions in expenditures by California.
Complicating the overall picture, based on self-reported exposure to the campaign, several studies did not find clear evidence for reductions in prevalence associated with campaign exposure (Pierce et al. 1998a; Popham et al. 1994), although another study suggested that the media campaign influenced decisions to quit smoking in California (Popham et al. 1993). It also should be noted that this California campaign was not primarily youth-focused, in contrast to the Massachusetts and Florida efforts (see below).
Massachusetts has been engaged in an ambitious tobacco control program since January 1993, which includes increased taxes that have been used to fund an extensive paid media campaign. Amounts available for these efforts were as high as $43 million in 1995, but have declined since. Strong population-based evidence exists for the effectiveness of this comprehensive effort in reducing adult smoking prevalence (Biener et al. 2000). Media efforts included television, radio, and billboard antismoking advertising, directed primarily at youth. The effectiveness of the media component of this effort was evaluated based on a 4-year longitudinal panel survey of 592 youth (ages 12–15 years at baseline), conducted from 1993 to 1997. Campaign impact was assessed by measuring recall of campaign advertising and comparing progression to established smoking among nonsmoking youth who recalled campaign advertisements to those who did not. A variety of control variables were applied, including parental smoking at home, hours of television watched, and so forth. Results suggested that younger adolescents (ages 12–13 years) reporting exposure to television advertisements at baseline were only about half as likely to progress to regular smoking as were the unexposed (odds ratio [OR] = .49, p < .05). No such differences were found for older adolescents (OR = .94) and no effects of radio or billboard exposure were found for either group (Biener and Siegel 2000).
Limitations of this study include the limited control over possible variables that might be confounded with exposure at baseline. Moreover, measures of campaign message recall or recognition are inherently subject to problems of endogeneity, in which propensity to attend to and recall such messages may be related—positively or negatively—to a possible interest in smoking (Slater 2004). However, the prospective design, lack of consistency of results for older versus younger adolescents, and lack of evidence for effects due to susceptibility differences or differential attrition argue against attributing results to confounding relationships. It should also be noted that the effects of the campaign on adults, as opposed to youth only, are not assessed in
this analysis. Friend and Levy (2002) also reexamined these data, adjusting for price effects, and estimated a 6 percent (relative percent) change effect size attributable to this campaign (Friend and Levy 2002).
Further evidence for the impact of antismoking advertising in Massachusetts is provided in a quasi-experimental evaluation of advertisements directed against light cigarettes (Kozlowski et al. 2000). Randomly sampled respondents in Massachusetts (n = 500) were less likely than respondents from elsewhere in the United States (n = 501) to believe that “light” cigarettes might reduce the risks of health problems (32 percent versus 49 percent, p < .05). Within Massachusetts, smokers who reported having seen the ads were less likely to believe that “light” cigarettes decreased health risks (26 percent for exposed versus 44 percent for unexposed smokers, p < .05). The advertising also appeared effective in increasing knowledge concerning filter vents. Behavioral or sales data are not reported.
Florida initiated its $25 million “truth” campaign in 1998, a campaign that, at first, focused on attacking the tobacco industry but was later obliged to refocus message strategies (Zucker et al. 2000). It should be emphasized that the “truth” campaign not only was an industry attack effort, but also can be understood as a campaign that endeavored to provide youth with a distinctive attitude of independence and control with respect to smoking and tobacco products. The so-called “truth” brand is designed to appeal to “edgy,” trend-setting youth who are influential with their peers and who also may be at risk for smoking (Farrelly et al. 2002). Smoking rates after the first year dropped from 18.5 percent to 15 percent among middle school youth, and from 27.4 percent to 25.2 percent among high school students, based on independently-collected data from Monitoring the Future (Friend and Levy 2002). Friend and Levy (2002) calculate the overall decline in youth prevalence to be a 5 percent (relative percent) change (Friend and Levy 2002).
Evidence also supports dose–response effects as estimated from advertisement recognition self-reports (n = 1820) (Sly et al. 2001a; Sly et al. 2001b). It appears that smoking rates remained lower among Florida youth than among youth nationally in the 2001–2002 school year (excluding youth from states with similar comprehensive tobacco control efforts), and that abstention from smoking was well predicted by awareness of the “truth” campaign whereas these rates were comparable prior to the campaign effort (Niederdeppe et al. 2004). It does not appear that possible confounding effects of tax and price changes in states under study were controlled in analyses of results, however, somewhat decreasing confidence in the precision of these results. Evaluators argue that the sizes of the Florida prevalence declines were too great to be attributable to the price increase, especially in 1998, the first year of the campaign (Farrelly et al. 2002).
National truth® Campaign
The American Legacy Foundation launched a national antismoking advertising effort, an extension of the “truth” campaign described above, in February 2002. An initial evaluation was conducted using nationally representative samples of adolescents, with a baseline pretest survey (n = 6,897) and a follow-up survey (n = 10,692) 10 months into the campaign (Farrelly et al. 2002). Results suggested statistically significant shifts in attitudes opposed to smoking and decreases in attention to smoke. Dose–response analyses indicated that greater self-reported recognition of truth® campaign advertising was associated with greater agreement with critical statements about cigarette companies, with an interest in taking a stand against smoking, and with the statement that not smoking was a way to express independence. Recognition of the so-called
truth advertisements was negatively related to agreement that smoking helped young people look cool or fit in (all p < .05 or better). Dose–response effects on intentions to smoke were in the predicted direction but were only marginally significant (p < .10).
This study also examined the impact of Phillip Morris USA’s “Think. Don’t Smoke” campaign, which began in 1998. Unsurprisingly, this campaign did not negatively impact perceptions of the tobacco industry and in some cases improved such perceptions. Exposure to the “Think. Don’t Smoke” advertisements was associated with agreement that not smoking was a way to express independence. However, the greater the self-reported exposure to the “Think. Don’t Smoke” campaign, the greater was the intention to smoke in the next year (p < .02). From a methodological perspective, using the same method (self-reported recognition of television anti-smoking advertising) and finding opposite results with two campaigns reduce concerns that such self-reporting might be confounded with preexisting attitudes or behavior about tobacco use.
A subsequent study provides more compelling evidence of the impact of the national truth® campaign on youth smoking (Farrelly et al. 2005). Unlike evaluation studies described above, this study examines actual variability in exposure to the ads that can occur in each major media market. Variability is substantial because the truth® campaign places its ads on cable channels. Cable penetration, or the number of households in which cable is viewed, varies substantially between markets in the United States. To the extent that variables that might influence both cable penetration and youth smoking prevalence can be controlled, differences in youth smoking prevalence by market that can be associated with market-level exposure differences can be especially persuasive. Such estimates are an excellent complement to other study designs because they are not dependent on self-reported exposure and thus they are not subject to possible confounding due to propensity to pay attention to and recall antismoking advertising because of existing attitudes and smoking-related behavior.
Results, utilizing multilevel modeling and national data (n = 50,000) from the Monitoring the Future study of adolescent substance use, indicated a significant relationship at the media market level, in which media markets receiving greater doses of the truth® campaign had at aggregate levels less smoking prevalence than markets receiving smaller doses of the campaign. Additional analyses were conducted to estimate effect size on youth smoking prevalence after accounting for other influences such as price on changes in prevalence. These estimates suggested that, of the 7.8 percent change in youth smoking prevalence between 2000 and 2002 (from 25.8 percent to 18 percent), an absolute percent change of 1.64 percent (or a relative percent of about 6 percent) could be attributed to the national truth® campaign. It is also likely that this analysis produced a conservative effect size estimate, because individual variability in exposure to campaign messages was ignored in this design in favor of market-level variability in advertising intensity.
One possible concern with the Farrelly and collgues (2005) evaluation is the close relationship of the evaluators to the truth® campaign (Farrelly et al. 2005). Another regards limitations on the assessment of exogenous factors and other issues affecting the nature of the relationship between market-level exposure differences and effects. These limitations have been in many respects addressed in recent evaluations of state-sponsored anti-tobacco advertising efforts (Emery et al. 2005). This research used a similar multi-level approach, examining market-level differences in expected exposure to anti-tobacco advertising rather than individual self-reports of exposure, and associating this with Monitoring the Future data. An innovative element of this research was the inclusion of controls for expected exposure to tobacco-related advertising as well as other tobacco control policies. Significant effects for exposure to state-sponsored anti-tobacco ads were found on probability of being a smoker (OR = .74, 95 percent confidence interval [CI]
= .63–.88). Wakefield and colleagues (2003a) used this same approach to analyzing effects of industry-sponsored youth antismoking ads (Wakefield et al. 2003a). Consistent with the results discussed earlier, this research found no evidence for the impact of these ads in reducing youth smoking or related variables, and there was evidence among 10th–12th graders for increased likelihood of smoking and smoking intentions as well as and reduced perceived harm as a function of exposure to industry-sponsored antismoking ads.
Possible Additional Beneficial Effects of Media Campaigns for Tobacco Control Efforts
This effect size estimate focuses on effects of a campaign on youth uptake. These, in some respects, may underestimate the utility of such campaigns for tobacco control efforts. For example, impact on uptake among young adults typically is not measured. There is also evidence that such campaigns may encourage cessation attempts among adult smokers (Netemeyer et al. 2005). Such effects are not incorporated in these estimates.
In addition, the visibility of such campaigns is likely to maintain or increase the salience of tobacco control as a public priority—a phenomenon known as agenda setting (McLeod et al. 1991). Such salience is typically a necessary prerequisite for the willingness of public officials to pursue tobacco control policies in legislation, regulation, and enforcement. The possible impact of media campaigns on the public opinion climate for tobacco control policy has largely been unresearched and is not accounted for in the estimates made here.
EFFECTS OF ALTERNATIVE MESSAGE AND DELIVERY STRATEGIES: EVIDENCE FROM CONTROLLED EXPERIMENTS AND FIELD STUDIES
Several studies have examined the effectiveness of alternative message strategies for anti-smoking efforts (see Agostinelli and Grube 2003; Kelder et al. 2002; Pechmann and Reibling 2000 for reviews, including discussion of possible psychological mechanisms for the effects of antismoking advertisements). For example, a field study using panel data (n = 618) from Massachusetts found evidence supporting the use of strong negative emotion over normative influence or humor messages (Biener et al. 2004). In a controlled lab study (n = 1,667) conducted in secondary schools in California, results indicated that message themes associated with endangering others, the negative life circumstances of smokers, and refusal skills outperformed control messages with respect to post-test smoking intentions. Messages about health consequences of smoking, about the tobacco industry’s marketing tactics and how it profits from a product that causes disease and death, and about the cosmetic effects of smoking were not significantly different from controls, although all tended in the desired direction (Pechmann and Reibling 2006). Another study by the same group conducted in California (n = 2,194) found that ads of the type used by the Massachusetts campaign that focused on true stories of the negative impact of smoking reduced intent to smoke 35 percent. Effects of advertisements that focused on criticism of the tobacco industry did not have a statistically significant effect on intention to smoke (Pechmann and Reibling 2006). The authors attributed their findings to possible wear out of the industry criticism strategy among California youth in the study, as well as the novelty and effectiveness of the ads portraying true stories of negative impacts.
Such experiments that compare strategy effectiveness should be interpreted with caution. While in principle they test strategies against one another, in practice they must test selected examples of those strategies. Differences in executional quality may substantially influence results. Moreover, controlled experimental tests, while relatively rigorous, also exclude the social influence processes of discussion among youth that might have a substantial impact on reception of advertisements. In any case, given the above findings and the professional wisdom in the advertising community concerning wear out of a given advertising strategy, it is likely that ongoing campaign efforts would be well advised to explore at least several strategies. A given strategy may lose its novelty and effectiveness over time and need to be replaced, perhaps to be returned to in somewhat different ways in a few years.
There is also some evidence regarding ineffective strategies. As Green and colleagues (2002) point out, portraying tobacco use as attractive but not permitted (known in psychology as a “forbidden fruit” strategy) may be likely to make smoking seem more rather than less desirable (Cummings and Clarke 1998; Green et al. 2002; Malone et al. 2002). Green and colleagues (2002) suggest this might be an explanation for boomerang effects found by Farrelly and colleagues (2002) in the Philip Morris “Think. Don’t Smoke” campaign, and might be anticipated as a consequence of Lorillard’s “Tobacco Is Whacko If You Are a Teen” campaign (Farrelly et al. 2002).
It is encouraging that several strategies appear to have shown effectiveness, given that advertisements critical of the industry have come under considerable legal and political pressure (Healton 2001). However, there is no guarantee that any antismoking campaign will show comparable effects. Each of the major strategies reported above was developed after extensive formative research and message testing. Moreover, there are antismoking campaigns in the literature that show no evidence of impact, although often these involve rather brief or low-intensity efforts. As noted above, findings by Farrelly and Colleagues (2002) suggest that a high production quality, nationally-distributed, paid campaign (produced by Phillip Morris) failed to show any evidence of positive effects and indeed appeared to have increased intentions to smoke, underscoring the dependence of campaigns on well-conceived, well-executed, and well-tested strategies (Farrelly et al. 2002). Any assumptions about positive effects of such initiatives must assume comparable care and skill in development and implementation. Uncertainties about successful implementation include effects of political and legal pressures constraining advertising content, such as pressures that have precluded the use of ads critical of the tobacco industry in several campaigns or that, as in Arizona, have limited expenditures on the campaign (Bialous and Glantz 1997).
Delivery in Conjunction with School-Based Prevention Efforts
The foregoing discussion has focused on the independent effects of media campaigns, particularly using television advertising. Many of these efforts have taken place in conjunction with various school- and community-based intervention activities, including school prevention curricula. Several studies have provided evidence suggesting that there can be a synergistic effect of school-based media and media advertising or other media messaging (Flynn et al. 1992; Flynn et al. 1994; Perry et al. 1992); there is also evidence suggesting that such effects can be additive at least in the cases of alcohol and marijuana, with school and local community media efforts influencing prevalence above and beyond the effects of school prevention curricula (Slater 2004). Such strategies remain options for states and communities in conjunction with state or national media efforts.
ESTIMATING EFFECT SIZES IN THE ABSENCE OF RANDOMIZED, CONTROLLED TRIALS
Randomly controlled trials (RCTs) are generally not feasible in these national and regional efforts. It is more costly to buy advertising for individual markets in random assignment than to do a national media buy; therefore, the costs of a fully deployed RCT would be comparable to a national media effort. Moreover, such an RCT would demonstrate in principle the effectiveness of antismoking advertising but would not necessarily result in a campaign that then can be rolled out to a full national population. By the time the RCT is completed, it would be necessary to revise ads anyway given changes in youth culture (clothes, music, visual style, etc.). An RCT is highly desirable to demonstrate the effectiveness of an intervention strategy in the absence of strong evidence of effectiveness. In the presence of such evidence, it may be hard to justify the costs of an RCT and the delay in providing intervention approaches with good evidence of impact.
Strengths of the evidence in favor of the effectiveness of these media campaigns involve triangulated results—the consistency of results across different analytic and evaluation design strategies, different advertising executions and strategies, and implementation in multiple locations. As cited above, there are a variety of quasi-experimental studies comparing behavioral outcomes, dose–response studies looking at the relationship between self-reported exposure to the campaign and behavior or behavioral intent, and market-level studies associating media markets that give greater exposure to antismoking campaigns but lower levels to smoking among youth. One type of study that is feasible, but has not yet been conducted, which would further reinforce such triangulation, would be a controlled quasi-experimental test in only a few markets, using a crossover design rather than a large number of community replicates to control for community differences (Palmgreen et al. 2001).
Many field studies have been conducted, several state-wide and others on a community-wide or regional basis, in addition to the major campaigns discussed above. Most of those on which adequate evaluation data are available have been incorporated in a major meta-analysis of media behavior change campaign effects by Snyder and colleagues (2004). This meta-analysis incorporates a total of 17 studies relative to smoking with a total (n) of almost 80,000; 13 of the studies, with a total (n) of more than 75,000, include change data (Snyder et al. 2004).
The Snyder and colleagues (2004) estimate of 6 percent absolute change effect size for prevention campaigns (4 percent for smoking cessation campaigns) is somewhat higher than the major studies described above (only California data, from the campaigns described above, were included in this meta-analysis, representing one of the studies and less than 15 percent of the total (n) (Snyder et al. 2004). As with any meta-analysis, this effect size may be inflated due to ineffective campaigns that remain unpublished. Moreover, these data also include smaller, controlled trials that are likely to be relatively more intensive, less subject to measurement error, and to show larger effect sizes than one finds in field evaluations. These effect sizes are for the duration of a campaign and are not annualized.
Friend and Levy (2002) estimate a 6 percent relative effect for a major, highly publicized youth prevention media effort, consistent with findings for the American Legacy Foundation truth® campaign (Farrelly et al. 2005) and with several of the regional studies mentioned above (Friend and Levy 2002). It would seem therefore that such an effect size estimate for continuing national campaigns that adapt proven strategies, utilize appropriate development and testing strategies for youth tobacco use prevention (Pechmann and Reibling 2000), and are funded well
enough to provide substantial levels of exposure (e.g., equivalent to or greater than that achieved by the recent national truth campaign effort) is reasonable.
However, effects over an extended period are hard to estimate. Campaigns gain increased awareness and penetration over time, but the youth most prone to influence may be reached early. Youth may also habituate to messages, and the effectiveness of given strategies may wane unless they are skillfully revised and updated. Impact on the most receptive youth in a cohort and campaign novelty may mean that effects tend to be stronger in the first year or two. However, the usual target ages of 12–16 years means that the cohort is replaced every 5 years, reducing the impact of habituation and an increasingly resistant core of smoking-susceptible youth. Perhaps a cycle of reduced effect in years 3 to 5 of an effort, followed by a return to greater effectiveness with a new cohort of younger teens might be a reasonable hypothesis, but few or no hard data exist with respect to this issue. It also may be that what continued media efforts do is sustain reductions after the initial impact of a campaign, slight, continued increments may be possible, but the primary effect of continuing is preventing movement of prevalence back to original levels. (Levy and Friend 2002).
In projecting the effects of continued media efforts, a 6 percent relative effect would decrease youth prevalence from 18 percent presently to about 17 percent. Conversely, the removal of national and major statewide media efforts would likely negate within a few years the approximately 2 percent absolute prevalence decrease that appears reasonably and conservatively attributable to the antismoking media efforts conducted to date. Clearly, use of evaluation designs that track effects in more detail over time, involving strategies such as rolling cross-sections and time-series analyses, would help inform such estimates (Hornik 2002; Palmgreen et al. 2001; Slater 2004). In the absence of such data, it seems both reasonable and conservative to assume a difference in youth prevalence in absolute numbers of at least 3 percent (17 percent to 20 percent). It is also quite possible that the absence of major media efforts, given the continued relative visibility of tobacco marketing efforts, might lead over time to continued erosion of recent tobacco control gains among youth. Such effects, however, are difficult to quantify with confidence. Finally, these assessments do not include effects of the campaigns on adult uptake or on encouraging adult cessation (either directly through campaign efforts targeting cessation, or indirectly as a result of exposure to prevention-oriented messages).
Media campaign effect sizes are modest. At the same time, it should be recalled that these interventions are assessed against an entire population and that these effects are found not merely in controlled test studies, but also in state-wide and national implementations. Therefore, such interventions may be quite cost-effective. In state campaigns described above, per capita costs per year varied from $0.50 to about $2.00 (Friend and Levy 2002). Costs per capita for the truth® campaign ranged from $1.54 to $2.92, depending on the year (information provided by the American Legacy Foundation). Moreover, if, as these estimates suggest, the absence of ongoing national or statewide media campaigns would mean about a 3 percent absolute difference in youth prevalence relative to continuing such efforts, then that would represent almost an 18 percent relative increase in youth prevalence relative to what would be anticipated with continuing media efforts.
At the same time, it must be emphasized again that a media campaign is not a vaccine. Campaign effectiveness, as discussed above, is highly dependent on conceptualization, testing, and execution as well as adequate funding to achieve necessary levels of audience exposure (Hornik 2002). Effects of a national effort may be considerably better or worse than our estimates depending on these factors, and boomerang effects of such campaigns are possible (Hornik et al.
2001; Pechmann and Slater 2005). Uncertainties would be minimized if the effort is directed by organizations and associated advertising agencies, with an empirical record of successful impact on youth smoking behavior via media efforts and with a policy commitment toward appropriate development, testing, support for adequate levels of exposure, and careful evaluation.
MEDIA AND SMOKING CESSATION
With the exception of the California campaign, the recent major media-based tobacco control efforts have not focused on encouraging smoking cessation efforts. As noted above, there was some evidence for the impact of the California smoking cessation campaign (Popham et al. 1993). Snyder and colleagues (2004) found an absolute effect size of 4 percent for such campaign. Again, the Snyder and colleagues (2004) meta-analysis included controlled studies that may have increased effect sizes relative to what might be found in large-scale field evaluations (Snyder et al. 2004). The outcome measures for what constituted cessation or cessation attempts also varied between studies. Nonetheless, these are encouraging findings, especially given the point made elsewhere in this report regarding the potential of smoking cessation programs and the importance of increasing quit attempts in order to realize this potential, especially if cessation programs are made less expensive and more widely available.
Given that the available data on smoking cessation media campaigns are less complete than those for youth prevention efforts, it may be premature to recommend a national, large-scale media campaign. The Snyder and colleagues (2004) meta-analysis suggested that cessation campaign effects were about two-thirds the size of prevention campaign effects on uptake (Snyder et al. 2004). However, many of the studies included in this meta-analysis were conducted prior to the availability of more recently developed cessation technologies and generally did not also increase access to these technologies. The combination of increased access and increased marketing has substantial potential, as noted in the appendix on smoking cessation. The potential is such that large-scale trials, supported either by states or by the National Institutes of Health, would be worthwhile as the basis for such a recommendation for national cessation media and social marketing efforts in the not-too-distant future.
Agostinelli G, Grube JW. 2002. Alcohol counter-advertising and the media. A review of recent research. [Review]. Alcohol Research and Health. 26(1):15-21.
Agostinelli G, Grube JW. 2003. Tobacco counter-advertising: a review of the literature and a conceptual model for understanding effects. [Review]. Journal of Health Communication. 8(2):107-127.
Bialous SA, Glantz SA. 1997. Tobacco Control in Arizona, 1973–1997. Center for Tobacco Control Research and Education.
Biener L, Harris JE, Hamilton W. 2000. Impact of the Massachusetts tobacco control programme: population based trend analysis. British Medical Journal 321(7257):351-354.
Biener L, Ji M, Gilpin EA, Albers AB. 2004. The impact of emotional tone, message, and broadcast parameters in youth anti-smoking advertisements. Journal of Health Communication 9(3):259-274.
Biener L, Siegel M. 2000. Tobacco marketing and adolescent smoking: more support for a causal inference. American Journal of Public Health 90(3):407-411.
CDC (Centers for Disease Control and Prevention). 1996. State Specific Prevalence of Cigarette Smoking—United States. Morbidity and Mortality Weekly Report 45(44).
Cummings KM, Clarke H. 1998. Health Science Analysis Project—Policy Analysis No. 8. Washington, DC: Advocacy Institute.
Emery S, Wakefield MA, Terry-McElrath Y, Saffer H, Szczypka G, O'Malley PM, Johnston LD, Chaloupka FJ, Flay B. 2005. Televised state-sponsored antitobacco advertising and youth smoking beliefs and behavior in the United States, 1999–2000. Archives of Pediatric and Adolescent Medicine 159(7):639-645.
Farrelly MC, Davis KC, Haviland ML, Messeri P, Healton CG. 2005. Evidence of a dose–response relationship between “truth” antismoking ads and youth smoking prevalence. American Journal of Public Health 95(3):425-431.
Farrelly MC, Healton CG, Davis KC, Messeri P, Hersey JC, Haviland ML. 2002. Getting to the truth: evaluating national tobacco countermarketing campaigns.[erratum appears in American Journal of Public Health. 2003 May;93(5):703]. American Journal of Public Health. 92(6):901-907.
Flynn BS, Worden JK, Secker-Walker RH, Badger GJ, Geller BM, Costanza MC. 1992. Prevention of cigarette smoking through mass media intervention and school programs. American Journal of Public Health 82(6):827-834.
Flynn BS, Worden JK, Secker-Walker RH, Pirie PL, Badger GJ, Carpenter JH, Geller BM. 1994. Mass media and school interventions for cigarette smoking prevention: effects 2 years after completion. American Journal of Public Health 84(7):1148-1150.
Friend K, Levy DT. 2002. Reductions in smoking prevalence and cigarette consumption associated with mass-media campaigns. [Review]. Health Education Research 17(1):85-98.
FTC (Federal Trade Commission). 2001. Federal Trade Commission Cigarette Report for 1999.
Green LW, Murphy RL, McKenna JW. 2002. New insights into how mass media works for and against tobacco. Journal of Health Communication. 7(3):245-248.
Healton C. 2001. Who's afraid of the truth? American Journal of Public Health 91(4):554-558.
Hornik R, Maklan D, Orwin R, Cadell D, Judkins D, Barmada C, Yanovitzky I, Moser M, Zador P, Southwell B, Baskin R, Morin C, Jacobsohn L, Prado A, Steele D. 2001. Evaluation of the National Youth Anti-Drug Campaign: Third Semi-Annual Report of Findings. Rockville, MD: Westat; Annenberg School for Communication, University of Pennsylvania.
Hornik RC. 2002. Exposure: theory and evidence about all the ways it matters. Social Marketing Quarterly 8(3):30-37.
Independent Evaluation Consortium. 2001. Interim Report: Independent Evaulation of the California Tobacco Control Prevention & Education Program: Wave 2 Data, 1998; Wave 1 & Wave 2 Data Compaisons, 1996–1998. Rockville, MD: The Gallup Organization.
Kelder SH, Pechmann C, Slater MD, Worden JK, Levitt A. 2002. The National Youth Anti-Drug media campaign. American Journal of Public Health 92(8):1211-1212.
Kozlowski LT, Yost B, Stine MM, Celebucki C. 2000. Massachusetts’s advertising against light cigarettes appears to change beliefs and behavior. American Journal of Preventive Medicine 18(4):339-342.
Levy DT, Friend K. 2002. Examining the effects of tobacco treatment policies on smoking rates and smoking related deaths using the SimSmoke computer simulation model. Tobacco Control 11(1):47-54.
Malone R, Wenger L, Bero L. 2002. High school journalists’ perspectives on tobacco. Journal of Health Communication 7:139-156.
McLeod J, Becker LB, Byrnes JE. 1991. Protess DL, McCombs, eds. Another Look at the Agenda-Setting Function of the press. In Agenda Setting: Readings on Media, Public Opinion, and Policymaking. Hillsdale, NJ: Lawrence Erlbaum Associates. Pp. 47-60.
Netemeyer RG, Andrews JC, Burton S. 2005. Effects of antismoking advertising-based beliefs on adult smokers’ consideration of quitting. American Journal of Public Health 95:1062-1066.
Niederdeppe J, Farrelly MC, Haviland ML. 2004. Confirming “truth”: more evidence of a successful tobacco countermarketing campaign in Florida. American Journal of Public Health 94(2):255-257.
Palmgreen P, Donohew L, Lorch EP, Hoyle RH, Stephenson MT. 2001. Television campaigns and adolescent marijuana use: tests of sensation seeking targeting. American Journal of Public Health 91(2):292-296.
Pechmann C, Reibling ET. 2000. Planning an effective anti-smoking mass media campaign targeting adolescents. Journal of Public Health Management Practice 6(3):80-94.
Pechmann C, Reibling ET. 2006. Antismoking advertisements for youths: an independent evaluation of health, counter-industry, and industry approaches. American Journal of Public Health 96(5):906-913.
Pechmann C, Slater MD. 2005. Social marketing messages that may motivate irresponsible behavior. In Ratneshwar S. Inside Consumption: Consumer Motives, Goals, and Desires. New York: Routledge.
Perry CL, Kelder SH, Murray DM, Klepp KI. 1992. Communitywide smoking prevention: long-term outcomes of the Minnesota Heart Health Program and the Class of 1989 Study. American Journal of Public Health 82(9):1210-1216.
Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Berry CC. 1998a. Tobacco industry promotion of cigarettes and adolescent smoking. Journal of the American Medicial Association 279(7):511-515.
Pierce JP, Gilpin EA, Emery SL, White MM, Rosbrook B, Berry CC, Farkas AJ. 1998b. Has the California tobacco control program reduced smoking? Journal of the American Medicial Association 280(10):893-899.
Popham WJ, Potter LD, Bal DG, Johnson MD, Duerr JM, Quinn V. 1993. Do anti-smoking media campaigns help smokers quit? Public Health Reports 108(4):510-513.
Popham WJ, Potter LD, Hetrick MA, Muthen LK, Duerr JM, Johnson MD. 1994. Effectiveness of the California 1990-1991 tobacco education media campaign. American Journal of Preventive Medicine 10(6):319-326.
Slater MD. 2004. Operationalizing and analyzing exposure: the foundation of media effects research. Journalism and Mass Communication Quarterly 81(1):168-183.
Sly DF, Heald GR, Ray S. 2001a. The Florida “truth” anti-tobacco media evaluation: design, first year results, and implications for planning future state media evaluations. Tobacco Control. 10(1):9-15.
Sly DF, Hopkins RS, Trapido E, Ray S. 2001b. Influence of a counteradvertising media campaign on initiation of smoking: the Florida “truth” campaign. American Journal of Public Health 91(2):233-238.
Snyder LB, Hamilton MA, Mitchell EW, Kiwanuka-Tondo J, Fleming-Milici F, Proctor D. 2004. A meta-analysis of the effect of mediated health communication campaigns on behavior change in the United States. Journal of Health Communication 9 (Suppl 1):71-96.
Wakefield M, Flay B, Nichter M, Giovino G. 2003a. Role of the media in influencing trajectories of youth smoking. Addiction 98 (Suppl 1):79-103.
Wakefield M, Flay B, Nichter M, Giovino G. 2003b. Effects of anti-smoking advertising on youth smoking: a review. Journal of Health Communication 8(3):229-247.
Zucker D, Hopkins RS, Sly DF, Urich J, Kershaw JM, Solari S. 2000. Florida’s “truth” campaign: a counter-marketing, anti-tobacco media campaign. Journal of Public Health Management Practice 6(3):1-6.