Evidence of the dose effects of an antitobacco counteradvertising campaign

Prev Med. 2002 Nov;35(5):511-8. doi: 10.1006/pmed.2002.1100.


Objective: The objectives were to assess the cumulative effects of exposure to multiple antitobacco advertisements shown over a 22-month period on smoking uptake, and determine if there is evidence of a dose effect and how this effect operates through response to the campaign's major message theme and antitobacco attitudes.

Methods: A follow-up telephone survey of persons ages 12-20 years was conducted after 22 months of the Florida "truth" antitobacco media campaign. Logistic regression analyses were used to estimate adjusted odds ratios for the likelihood that time-one nonsmokers would remain nonsmokers at time two by levels of confirmed advertisement awareness, self-reported influence of the campaign's message theme, and anti-tobacco industry manipulation attitudes. Separate cohorts are analyzed and controls include gender and time-one susceptibility.

Results: The likelihood of nonsmokers remaining nonsmokers increases as the number of ads confirmed, the self-reported influence of the campaign's major message theme, and the level of antitobacco attitudes increases. The pattern to these relationships holds within cohorts of young and older youth and for a cohort that has aged into the early young adult years. Considering all variables simultaneously suggests that ad confirmation operates through its effects on the influence of the message theme and antitobacco industry manipulation attitudes.

Conclusions: There is evidence of a dose effect; however, considering only ad confirmation underestimates this. Antitobacco campaigns that target youth can have effects at least through the early young adult ages. The uniqueness of the Florida campaign may limit the generalization of reported results.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Advertising*
  • Child
  • Florida / epidemiology
  • Follow-Up Studies
  • Health Knowledge, Attitudes, Practice
  • Humans
  • Likelihood Functions
  • Logistic Models
  • Models, Psychological
  • Odds Ratio
  • Outcome Assessment, Health Care*
  • Smoking / epidemiology
  • Smoking Prevention*
  • Social Marketing*