HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Electronic Letters to:

RESEARCH AND PRACTICE: Matthew C. Farrelly, Kevin C. Davis, M. Lyndon Haviland, Peter Messeri, and Cheryl G. Healton Evidence of a Dose—Response Relationship Between "truth" Antismoking Ads and Youth Smoking Prevalence Am J Public Health 2005; 95: 425-431 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

Electronic letters published:

Re: Impact of the “truth” Campaign on Cigarette Smoking

Matthew C Farrelly, Kevin Davis   (19 May 2005)

Impact of the "truth" Campaign on Cigarette Smoking

Joel M. Moskowitz, Ph.D.   (31 March 2005)

Re: Impact of the “truth” Campaign on Cigarette Smoking 19 May 2005
Matthew C Farrelly, Health Economist RTI International, Kevin Davis Send letter to journal: Re: Re: Impact of the “truth” Campaign on Cigarette Smoking mcf{at}rti.org Matthew C Farrelly, et al.	The commentary by Dr. Moskowitz on the Farrelly et al (2005) paper on the association between exposure to the truth campaign and youth cigarette smoking raises some important issues worthy of further discussion. However, some of the study’s main findings were misinterpreted. The focus of this commentary is on two points—the shape of the relationship between truth exposure and the prevalence of youth smoking and the effectiveness of the truth® campaign across grades. With regard to the first point, Dr. Moskowitz correctly notes that we found a statistically significant quadratic relationship between gross rating points (GRPs, a measure of exposure to the campaign) and youth smoking as displayed in Figure 2 of the paper. Dr. Moskowitz is correct that the shape of this curve implies that at extreme levels of exposure, the campaign will have no detectable effects. This result does not mean, however, that the campaign did not have effects. (Had that been the case, the curve would have been a flat straight line.) We agree with this point in principle, however, the statement that the campaign had “no detectable effect on smoking prevalence among…students in most major metropolitan areas” is incorrect. When properly interpreted, Figure 2 suggests that there may not have been detectable effects (odds ratio of 1 or higher) for youth residing in only 2 out of 210 media markets, not most major metropolitan areas. Figure 2 in the paper displays the estimated relationship between cumulative GRPs and the odds of youth smoking for the overall time period of the campaign (2000-2002) and separately by year for 2000, 2001, and 2002 to illustrate the changing relationship over time. The results for the overall time period represent average effects over the years 2000- 2002, averaging the relatively small and statistically insignificant results from 2000 with the larger and statistically significant results for 2001 and 2002. The relevant maximum GRPs for the 2000, 2001, and 2002 models (Columns 2-4 in Table 2, p. 429) are approximately 7,500, 12,500 and 22,000 respectively. To understand the estimated effects for the maximum level of GRPs that occur in 2002, one must apply this level to the results from Column 4 in Table 2 that are specific to 2002. Extending the 2002-specific model results up to 22,000 indicates that there are only 2 media markets out of the 210 nationwide that had GRPs at levels that would suggest no detectable truth effect (odds ratios approximately equal to 1)—not “most major metropolitan areas” as Moskowitz asserts. As Dr. Moskowitz fairly asks, what are the policy implications of this quadratic relationship? Our results suggest that in very oversaturated markets, youth may have tired of the campaign, thus reducing its effectiveness. That result is not all together unsurprising as advertising experts warn of campaign “wear out” due to overexposure to campaign messages. Does this suggest as cumulative exposure grows, more and more youth will reside in markets where truth is no longer effective? This situation will only exist if the campaign strategy and the relationship between youth smoking and truth exposure remain constant over time. Preliminary results from models that include 2003-2004 data from Monitoring the Future suggest the campaign continues to be effective. Dr. Moskowitz’ second main point is that truth was only effective among youth in the 8th grade when the models were stratified by grade. When estimated separately by grade, we found a statistically significant association via the quadratic specification for 8th grade students only. However, as we noted in the paper, we found that the association between truth and youth smoking was marginally significant (p<0.07) among 12th graders in an alternative linear specification. Preliminary results using data through 2004 suggest that this linear relationship for 12th graders has become statistically significant (p<0.05). Finally, it is important to note that we found a statistically significant relationship between truth exposure and youth smoking among all grades combined.
Impact of the "truth" Campaign on Cigarette Smoking 31 March 2005
Joel M. Moskowitz, Ph.D. Director, Center for Family and Community Health, School of Public Health, UC Berkeley Send letter to journal: Re: Impact of the "truth" Campaign on Cigarette Smoking jmm{at}berkeley.edu Joel M. Moskowitz, Ph.D.	&#65279;The evaluation of the "truth" anti-smoking media campaign by Farrelly et al. (2005) raises important questions about its impact on smoking prevalence. This study investigated the relationship between exposure to "truth" television advertising and 30-day smoking prevalence among U.S. youth in grades 8, 10 and 12. Exposure was measured by cumulative gross rating points (GRP) – the percentage of the target audience reached by the campaign times the frequency of exposure. The authors failed to find a significant linear relationship between GRP and smoking prevalence (p. 428). They tested for a curvilinear relationship by adding a quadratic term (GRP-squared) to their statistical models. In 2 of the 6 final models, both terms for GRP were significant (Table 2, p. 429). The relationship between "truth" advertising and smoking prevalence was U- shaped. Figure 2 censors the rightmost portion of this relationship by truncating GRP at 15,000 (p. 430). The optimal GRP was approximately 10,000. Figure 1 reveals that many students were exposed to GRP’s greater than 15,000, and that the maximum GRP was 22,389 (p. 427). The theoretical rationale for inclusion of the GRP-squared term was to test whether the campaign had “diminishing returns" (p. 428). This would suggest an L-shaped relationship between campaign advertising and smoking prevalence; not the U-shaped relationship found. The results suggest that the campaign had no detectable effect on smoking prevalence among those who resided in media markets that received higher levels of exposure which included students in most major metropolitan areas. Yet, the paper obscured this finding and failed to address its policy implications. Did an overdose of “truth” render the campaign ineffective? Or were the models improperly specified to estimate campaign effects? When examined by grade level, the effect of “truth” advertising on smoking prevalence was significant only for students in grade 8 in media markets with moderate exposure (Table 2). That the campaign’s impact did not sustain through high school suggests that “truth” advertising was no more effective than school-based, smoking prevention programs (e.g., Wiehe et al., 2005). References Farrelly, MC, Davis, KC, Haviland, ML, Messeri, P, Healton, CG. Evidence of a dose-response relationship between “truth” antismoking ads and youth smoking prevalence. Am J Public Health. 2005; 95:425-431. Wiehe, SE, Garrison, MM, Christakis, DA, Ebel, BE, Rivara, FP. A systematic review of school-based smoking prevention trials with long-term follow-up. J Adol Health. 2005; 36:162-169.