TY - JOUR T1 - Continuous tracking of the Australian National Tobacco Campaign: advertising effects on recall, recognition, cognitions, and behaviour JF - Tobacco Control JO - Tob Control SP - ii30 LP - ii39 DO - 10.1136/tc.12.suppl_2.ii30 VL - 12 IS - suppl 2 AU - R J Donovan AU - J Boulter AU - R Borland AU - G Jalleh AU - O Carter Y1 - 2003/09/01 UR - http://tobaccocontrol.bmj.com/content/12/suppl_2/ii30.abstract N2 - Objectives: To relate Australian National Tobacco Campaign advertising to outcome measures such as smokers’ awareness of and reaction to the campaign, and indicators of interest in smoking cessation. Design: Continuous tracking was used to survey random cross sectional samples of the target audience via telephone interviews. Baseline measures were collected preceding each advertising phase, whereafter subjects were interviewed on a weekly basis for the entire period of each phase. Changes in outcomes could thus be inferred on a weekly basis allowing variations in advertising intensity to be monitored for effect. Three phases were evaluated variously in Melbourne, Sydney, and Adelaide. Subjects: A total of 9033 subjects aged 18–40 years were interviewed. Age and sex of the sample were evenly distributed. Results: In general, it was found that the greater the media weight, the greater the recall and recognition mediated by the message of the advertisement and the creative execution—advertisements with a clear figure ground executional format appeared more memorable than those without, and health effects advertisements were more memorable than those encouraging calls to a quitline. The relationship between various communication effects and media weight was limited by the confounding of prior activities in two of the phases. Conclusions: Advertisements with clear figure ground executional formats and those illustrating health effects of smoking have high memorability. Future campaigns that are continuously tracked are recommended to systematically vary media weight, flighting schedules, and advertisement type, so as to maximise information about these variables and their interactions. ER -