218 Australian and New Zealand Journal of Public Health 2016 VOL. 40 NO. 3 © 2016 Public Health Association of Australia T elevision viewers, including youth, are exposed to considerable tobacco imagery on television. 1,2 This imagery typically does not show adverse health effects of tobacco, and may normalise smoking by encouraging the misperception that smoking is prevalent. While the relationship between tobacco imagery on television and youth smoking is not as well established as it is for movies, 3 television viewing among children and adolescents has been shown to be associated with smoking initiation, 4,5 regular smoking 5 and persistence into young adulthood. 6 New Zealand has had a ban on tobacco marketing on television for many years. Despite this, evidence from a decade ago showed that tobacco smoking was shown on children’s television 7 and in popular films; 8 that one in four NZ television programs contained tobacco imagery, with only 15% of scenes portraying anti-tobacco messages. 9 Most scenes portrayed tobacco as neutral or positive, with many depictions of smoking appearing incidental to the storyline. This research examines changes in the frequency and contexts of tobacco imagery on NZ television from 2004 to 2014. Methods Following our earlier research, 9 all television programming (including movies, advertising and trailers) shown on the three major free- to-air channels (state-owned TV1 and TV2, and privately owned TV3), were recorded Little change in tobacco imagery on New Zealand television: 10 years on Louise Marsh, 1 Rob McGee, 1 Lindsay Robertson, 1 Matthew Ward, 2 Rebecca Llewellyn 1 1. Cancer Society Social and Behavioural Research Unit, University of Otago, New Zealand 2. Department of Media, Film and Communication, University of Otago, New Zealand Correspondence to: Dr Louise Marsh, Cancer Society Social and Behavioural Research Unit, Department of Preventive and Social Medicine, University of Otago, PO Box 913, Dunedin 9054, New Zealand; e-mail: louise.marsh@otago.ac.nz Submitted: November 2015; Revision requested: December 2015; Accepted: December 2015 The authors have stated they have no conflict of interest. Aust NZ J Public Health. 2016; 40:218-20; doi: 10.1111/1753-6405.12527 Abstract Objectives: To examine changes in the frequency and contexts of tobacco imagery on New Zealand television since 2004. Methods: A content analysis of 73 hours of prime time evening television in 2014, including programs, advertisements and trailers, was coded for tobacco imagery. Imagery was defined as being either neutral/pro-tobacco or anti-tobacco. Results: Of the 93 programs coded, 29% had at least one scene with tobacco imagery. Of the 71 scenes with tobacco imagery, 59 were judged as showing neutral/pro-tobacco imagery, while 12 showed anti-tobacco imagery. No significant change in the number of programs containing tobacco imagery, or the type of imagery, was found since 2004, but there were fewer scenes that contained imagery. Conclusions: There has been little change in the amount of tobacco imagery over the past decade. Given the potential for tobacco imagery to promote smoking among young people while reinforcing the habit among those who are trying to quit, action needs to be taken. Implications: More could be done to counterbalance pro-tobacco imagery by promoting the Quitline and anti-tobacco media campaigns, and encouraging producers of local TV programs to consider the depiction of tobacco imagery in a way that reflects declining tobacco use. Keywords: media, public policy, tobacco, smoking, youth from 6.00pm to 9.30pm for one week in October 2014 using eTV (http://www.etv. org.nz/v4/). This amounted to 73.5 hours of television. A music channel, C4, was recorded in the 2004 study, but is no longer available. The primary analysis was of scenes containing tobacco imagery, and these scenes were coded in accordance with our previous work to identify ‘pro/neutral’ and ‘anti- tobacco’ imagery. 9 The coding was carried out by MW and RL, all identified imagery was then reviewed by LM and RM, and any discrepancies in coding were identified and final agreement was reached. Descriptive statistics identified the number of programs, trailers and advertisements over the period. The number of scenes with tobacco imagery, a description of the scene, and the context of the imagery was recorded. Tobacco imagery depictions were compared by program characteristics using chi-square tests; if the expected values were less than five, Fishers Exact test was used. A test for the difference between proportions was used to compare tobacco imagery between 2004 and 2014. Analysis was undertaken using Stata10.1. ADDICTIVE BEHAVIOURS Brief Report