The Evidence
The Cheat Criticizes The Liars, The Liars Refute The Cheat

FORCES - Evidence by topic - Back to: Proving the lies of the anti-tobacco cartel: The Evidence


Isn't this touching" The anti-tobacco gang in Massachusetts (Connelly) are taking on the anti-smoking cartel in California (Glantz-Goldman).While we are pleased that at least JAMA takes its distance from the Glantz-Goldman pack of deceptions (... sorry, we meant to say "study"), we have to remember that the whole anti-tobacco enterprise is bases solely on twisted data, assumption, and mathematical models twisted to achieve a social and political end.

It follows that any disagreement between operatives of the cartel, though enjoyable and useful, is still a disagreement between false assumptions and corrupted politics.

[Letter, Sep 16 JAMA.1998;280:964-965]

Evaluating Antismoking Advertising Campaigns

To the Editor.—Ms Goldman and Dr Glantz[1] conclude that antismoking messages concerning "industry manipulationand secondhand smoke are the most effective strategies for . . .reducing cigarette consumption." The evidence, however, does notwarrant such a conclusion.

Without supporting citations, the authors classify the Massachusettstobacco control campaign as "a more youth-oriented approach" thanCalifornia's more confrontational advertisements. This conclusionappears to be based solely on the authors' own categorization ofadvertisements contained in focus group research reports.

Goldman and Glantz confound 2 different concepts in the analysis ofmedia campaigns: target audience and content of message. Young peopleare a target audience, while issues of industry manipulation andsecondhand smoke are particular themes. As shown in theTable, anti-tobacco industry and environmentaltobacco smoke (ETS) themes constituted 32% of all youth-targetedmessages and 37% of all messages in the Massachusetts tobacco controlcampaign.[2]Goldman and Glantz compute that Massachusetts'overall antismoking media spending amounted to $2.42 per capita. Hadthe authors relied on the rating-point data in the Table, they wouldhave computed that Massachusetts' per capita spending on industrymanipulation and secondhand smoke alone was $0.90, an amount 80%greater than California's overall per capita spending of $0.50.

Goldman and Glantz use data on total cigarette consumption per capitato assess the cost-effectiveness of the advertising campaigns in eachstate. Children consume only 2% to 3% of all cigarettes sold. Theauthors thus appear to be evaluating Massachusetts' "moreyouth-oriented approach" solely by examining adult cigarette use.From 1993 to 1996, cigarette smoking among Massachusetts students ingrades 7 through 12 remained unchanged[3]but increased inCalifornia[4]and the rest of the United States.

The authors' analysis of consumption trends per capita does not appearto consider possible influences other than media spending and cigaretteprice. By 1996, California prohibited smoking in all public placesexcept bars,[4]while only 50% of the Massachusettspopulation were covered by similar local laws.[5]Thisomission is important in view of the authors' conclusion that mediamessages emphasizing the dangers of ETS are effective.

Comparing the experience of 7 years in California (1989-1996) with 3years in Massachusetts (1993-1996), the authors conclude thatCalifornia's per capita consumption declined more rapidly than that ofMassachusetts. However, in an analysis of both states during 1990 to1996,[6]we reported a larger decline in Massachusetts.

Goldman and Glantz offer no quantitative analyses of focus groupreports that they reviewed. No sample sizes or statistical tests arereported. The methodological basis for the effectiveness scale reportedin their summary table (ie, "highly effective," "moderatelyeffective," "effective," and "not effective") is neitherdocumented nor validated.

We urge legislators and public health officials who make policydecisions about antismoking campaigns to rely on sounder, morescientific evidence than that presented by the authors.

Gregory N. Connolly, DMD, DPH
Massachusetts Tobacco Control Program

Jeffrey E. Harris, MD, PhD
Massachusetts Institute of Technology


1. Goldman LK, Glantz SA. Evaluation of antismokingadvertising campaigns. JAMA. 1998;279:772-777.

2. Arnold Communications. Gross Rating Points Analysisof Massachusetts Tobacco Education Campaign, Fiscal Years 1995 and1996. Boston, Mass: Arnold Communications; April 20, 1998.

3. Briton NJ, Clark TW, Baker AK, et al. Adolescent Tobacco Use in Massachusetts: Trends Among Public School Students,1984-1996. Boston, Mass: Health and Addictions Research Inc; May1997.

4. Tobacco Education and Research Oversight Committee(TEROC). Toward a Tobacco Free California: Renewing theCommitment, 1997-2000. Sacramento: California Dept of HealthServices; July 31, 1997.

5. Abt Associates. Independent Evaluation of theMassachusetts Tobacco Control Program. Cambridge, Mass: AbtAssociates; 1997. Third Annual Report, January 1994 to June 1996.

6. Harris JE, Connolly GN, Davis B, Brooks D. Cigarettesmoking before and after an excise tax increase and an antismokingcampaign—Massachusetts, 1993-1996. MMWR Morb Mortal Wkly Rep.1996;45:966-970.

(JAMA. 1998;280:964)

In Reply.—Drs Connolly andHarris[1]criticize us for describing the Massachusettsantitobacco media campaign as "a more youth-oriented approach" thanthe one California used. Yet, they use the fact that the Massachusettscampaign was more youth oriented to explain why we saw faster declinesin per capita consumption of cigarettes in California than inMassachusetts. They are correct in stating that youth consume only 2%to 3% of cigarettes, which is another reason that a youth-focusedprogram is a mistake. It is, after all, the adults whom tobacco kills.Moreover, while it is preferable to prevent tobacco use altogether, tostop smoking in young adulthood avoids the long-term healthconsequences.[2,3]

Connolly and Harris are incorrect when they state that we confusetarget audience and message. The table in our article clearlydistinguishes between effectiveness of messages for youth and adults.The fact is that anti-tobacco industry and secondhand smoke messagesare the best ones for both youths and adults, albeit for differentreasons.

We agree that the efforts in California to create smoke-free workplacescontributed to the greater cost-effectiveness of the California programcompared with Massachusetts. The California media campaign wasexplicitly designed to support this effort. The fact that theanti-tobacco industry and secondhand smoke advertisements weretargeted at adults as well as youth contributed to this success. (As wenoted in our article, one obtains similar measures of the relativeeffectiveness of the 2 states' programs when the calculations arebased on total expenditures.) While we concluded that the more aggressive California campaign was more cost-effective at reducingcigarette consumption than the Massachusetts campaign, we did notconclude that the Massachusetts campaign was ineffective. Smokingdeclined in Massachusetts faster than in the rest of the United States,but not as fast as in California per dollar spent per capita.

We recognize the difficulties associated with comparingCalifornia and Massachusetts, including those Connolly and Harrisdescribe. We also recognize the limitations of focus group research.Indeed, we discussed all these issues in our article.

Connolly and Harris note that teen smoking remained stable inMassachusetts from 1993 to 1996, whereas it increased in California andthe rest of the country during the same period. The statement regardingCalifornia is correct but misleading. Teen smoking remained stable inCalifornia from 1991 to 1993, at around 9%, then increased to around11% in 1994, where it has remained since that time.[4]Thislarge increase in teen smoking in California coincided with a politicaldecision by the Wilson administration to tone down the anti-tobaccoindustry advertisements and refocus the program more directly onyouth.[5]The fact that teen smoking increased when the statetoned down the effective strategies of frontal attacks on the tobaccoindustry and secondhand smoke supports our contention that thesestrategies are effective.

Effective tobacco control programs should be directed at a generalmarket with a youth component and include strong anti-tobacco industryand secondhand smoke messages in order to maximizecost-effectiveness.

Lisa K. Goldman, MPP
Stanton A. Glantz, PhD
University of California
San Francisco


1. Doll R, Peto R, Wheatley K, Gray R, Sutherland I.Mortality in relation to smoking: 40 years' observations on maleBritish doctors. BMJ. 1994;309:901-911.

2. Lightwood J, Glantz S. Short-term economic and healthbenefits of smoking cessation: myocardial infarction and stroke.Circulation. 1997;96:1089-1096.

3. Tobacco Education and Research Oversight Committee(TEROC). Toward a Tobacco Free California: Renewing theCommitment, 1997-2000. Sacramento: California Dept of HealthServices; July 31, 1997.

4. Balbach E, Glantz S. Tobacco control advocates mustdemand high quality media campaigns. Tob Control. In press.

5. [Note: Reference not found in reference section.]

(JAMA. 1998; 280:964-965)

Continue letters
Table of Contents

FORCES is supported solely by the efforts of the readers. Please become a member or donate what you can.

Contact Info
Forces Contacts
Media Contacts
Links To Archived Categories

The Evidence
Inside Forces
About Forces
Book case