LITERATURE REVIEW (12-16 pages) This essay is basically synthesizing 10 sources to support the new research im proposing that needs to be done. The sources needs to be categorized into 3/4 subheadings

Journal of Health Psychology Additional services and information for         http://hpq.sagepub.com/cgi/alerts Email Alerts:   http://hpq.sagepub.com/subscriptions Subscriptions:   http://www.sagepub.com/journalsReprints.nav Reprints:   http://www.sagepub.com/journalsPermissions.nav Permissions:   What is This?   - Oct 22, 2013 OnlineFirst Version of Record >> at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from at UNIV OF SURREY on November 6, 2013 hpq.sagepub.com Downloaded from Journal of Health Psychology0(0) 1 –9© The Author(s) 2013Reprints and permissions: sagepub.co.uk/journalsPermissions.navDOI: 10.1177/1359105313504236hpq.sagepub.com Introduction There is currently much debate concerning the impact of branded cigarette boxes on the uptake and maintenance of smoking. In April 2012, the UK government launched a consulta - tion on the introduction of plain packaging for all tobacco products which met with high levels of public support (Action on Smoking and Health, 2012a). It is argued by the proponents of plain packaging that even though tobacco com - panies have been restricted in their sources for advertising, the cigarette box has become iconic to each manufacturer and remains a vehicle for advertising and an object through which smok - ers express their identity (Action on Smoking and Health, 2010). In light of these debates, this article explores changes in advertising since 1950 in the United Kingdom and the strategies used by tobacco companies to promote their product, with a focus on the use of the box and the meanings associated with smoking. The possibility of branded cigarette boxes functioning to encourage smoking has been explicitly addressed in health interventions, most notably by the Australian government which The cigarette box as an advertising vehicle in the United Kingdom: A case for plain packaging Michaela Dewe, Jane Ogden and Adrian Coyle Abstract This research aimed to study tobacco advertising between 1950–2003 an\ d to evaluate the role of the cigarette box in advertising. Tobacco company advertisements ( n = 204) were coded for content and meanings used to promote their product. There was a significant shift from cigarettes being displayed to the cigarette box only. Changes in advertising and the meanings evoked were unrelated to changes in smoking behaviour. It is argued that the cigarette box has absorbed t\ he meanings associated with smoking and has become an effective vehicle for advertising. It is also \ argued that this can only be minimised with plain packaging.

Keywords advertising, content analysis, plain packaging, smoking, tobacco University of Surrey, UK Corresponding author:Michaela Dewe, School of Psychology, University of Surrey, Guildford, Surrey GU2 7XH, UK.

Email: [email protected] 504236 HPQ 0 0 10.1177/1359105313504236 Journal of Health PsychologyDewe et al.

2013 Article 2 Journal of Health Psychology 0(0) introduced compulsory plain packaging for tobacco products in December 2012. ‘Plain packaging’ requires the removal of branding including colours, imagery and logos and only allows the brand name to be printed on the ciga - rette box in a standardised font. In July 2013, the UK government announced a delay in imple - menting a plain packaging policy until the impact of the implementation of the Australian policy is known. Yet, recent research by Wakefield et al.

(2013) has found that the introduction of plain packaging there has resulted in smo kers perceiv - ing their cigarettes as less satisfying and of lower quality than previously. The study also found that smokers were more likely to have thought about quitting in the last week and now rated quitting as a higher priority in their lives. The researchers argued that the early results sug - gested that plain packaging is having an effect and is associated with making smoking less appealing. Across its history, cigarette advertising – like the advertising of any product – has been oriented towards increasing the appeal and con - sumption of the product. Advertising per se has been defined as ‘any form of paid communica - tion by an identified sponsor aimed to inform and/or persuade target audiences about an organization, product, service or idea’ (Fennis and Stroebe, 2010: 2). There has been much work undertaken within psychology to under - stand how advertising functions. According to Fennis and Stroebe (2010), it operates in a com - plementary way on a societal/collective level, facilitating competition between brands, and on an individual level, endeavouring to persuade audience members. Persuasion can be seen here as ‘any change in beliefs and attitudes that results from exposure to a communication’ (Petty and Cacioppo, 1986: 5). There are sev - eral standard theories of persuasion that can be applied to advertising, including the cognitive response model (Greenwald, 1968; Petty et al., 1981), dual process models (Chen and Chaiken, 1999; Petty and Cacioppo, 1986) and the infor - mation processing model (McGuire, 1969).

Ultimately persuasive advertising leads to the consumer forming positive associations about specific products or brands which promote consumption. The Framework Convention on Tobacco Control, an initiative of the World Health Organization (WHO, 2001), has defined tobacco advertising as any form of commercial and non-commercial communication with the aim or direct or indirect effect of promoting a tobacco product, including advertising which, while not specifically men - tioning the tobacco product, uses brand names, trademarks, emblems, or other distinctive fea - tures of tobacco products. (p. 4) Although tobacco advertising has been per - vasive in the Western world and far-reaching in its scope (Wakefield et al., 2005), since the 1990s, tobacco companies have been increas - ingly restricted in their potential advertising outlets. For example, in line with the UK Broadcasting Acts of 1990 and 1996, they were prevented from advertising on television and radio and since the Tobacco Advertising and Promotion Act in 2002, which came into effect in the United Kingdom in 2003, they have been unable to advertise via billboards or in print.

Furthermore, from 2005, they were banned from advertising through either direct mail or sponsorship. Since this time, tobacco compa - nies moved their focus to ‘point-of-sale’ adver - tising to promote their products (Action on Smoking and Health, 2012b). This, however, has now also been limited following research demonstrating that adolescents were particu - larly susceptible to this approach and laws have been implemented in the United Kingdom, meaning that since April 2012, large stores can no longer have tobacco products on display (MacKintosh et al., 2012). These measures will be extended to smaller stores from 2015. With potential advertising outlets having been systematically limited over the past 20 years in the United Kingdom, the cigarette box has been left as the last significant vehicle for the promotion of smoking. Indeed, cigarette packaging has been recognised as a prominent Dewe et al. 3 form of tobacco marketing in the United Kingdom (Hammond et al., 2013). Research by Ford et al. (2012) has suggested that branded packaging is key at the point of sale and post- sale, influencing the perceived attributes of the product, and that it is an important promotional tool for advertisers. Furthermore, this research has suggested that the introduction of plain packaging for tobacco products in the United Kingdom would further restrict the capability of the tobacco companies to influence consum - ers through promotional activities using their boxes. Indications are that plain packaging of tobacco products leads to a significant reduc - tion in the positive associations with smoking such as glamour and sophistication (Hammond et al., 2013). The current research explores the associations with the cigarette box that were formed through its use in advertising which may have been carried forward by the box itself since the advertising ban. Supporters of the call for plain packaging argue that in line with the WHO definition cited earlier, cigarette boxes are a form of advertising and have been described as the principal way in which tobacco companies now advertise their products (Hammond et al., 2012), functioning as ‘silent salesmen’ (Mitchell and Studdert, 2012). In support of this, research has high - lighted a series of changes made by tobacco companies to the packaging used, including changes in size, design and whether the price is printed on the packets. Such changes have been argued to illustrate the use of the cigarette box to attract attention and promote the products they contain (Moodie and Hastings, 2011).

Likewise, a review of internal tobacco industry documents indicated that following increasing legislation that has limited the possibilities for advertising, tobacco companies themselves view the packet as being crucial for creating a brand image and enhancing the impact of their brand where cigarettes are on sale (Wakefield et al., 2002). Furthermore, research indicates a rise between 2002 and 2006 in the number of adolescents who were aware of changes to the design of cigarette packets, suggesting that this age group may be vulnerable to this source of information (Action on Smoking and Health, 2010). This is of particular concern given that there are currently around 150,000 smokers aged 11–15 years in the United Kingdom with two-thirds of smokers starting before the age of 18 years (Action on Smoking and Health, 2012b). It is also argued that the compulsory introduction of plain packaging would remove this final outlet for the tobacco companies and that non-branded packaging would lower the appeal of cigarettes and cause a subsequent decrease in both smoking uptake and mainte - nance (Moodie and Ford, 2011; Thrasher et al., 2011). In contrast, the tobacco industry has argued that packaging has no impact on the consump - tion of their products. For example, an online campaign entitled ‘Hands off our packs! Say no to plain packaging’, run by the Freedom Organisation for the Right to Enjoy Smoking Tobacco (Forest, n.d.), has advanced several reasons why plain packaging should not be endorsed. First, it is said that plain packaging will have no impact on smoking prevalence.

Second, it is said that to suggest that consumers select their cigarette brand on the basis of col - our is an insult to their intelligence. Third, it is contended that by removing the use of branding and trademarks for tobacco products, plain packaging will ‘eliminate the well known advantages trademarks create in a free market society’ (Philip Morris Limited, 2008: 31). In this debate about the impact of branded cigarette boxes and the (proposed compulsory) introduction of plain packaging, it is unclear what, if any, positive associations were formed in relation to the cigarette box through its use in cigarette advertising. This study addresses this and evaluates changes in advertising strategy between 1950 when the first research was pub - lished concerning the health effects of smoking (Doll and Hill, 1950) and 2003 when cigarette advertising in print and on billboards was banned in the United Kingdom. The research focuses on how the cigarette box has been used in cigarette advertising and what associations 4 Journal of Health Psychology 0(0) are created in relation to the box in order to dis - cern whether and, if so, how the cigarette box operates as a form of advertising for the tobacco industry. The study also explores the relation - ship between changes in the use of the cigarette box and its associated meanings and existing data on changes in smoking behaviour.

Method A total of 40 UK tobacco print advertisements from each decade (1950s–2000s) were selected from an online advertising archive that con - tains over 1500 tobacco advertisements using a random number generator. The original sample contained both pro-smoking and anti-smoking advertisements. Of all, 36 were omitted from this analysis as they were anti-smoking adver - tisements, leaving a sample of 204 advertise - ments for analysis (39 from the 1950s, 38 from the 1960s, 33 from the 1970s, 38 from the 1980s and 28 from the 1990s and 2000s). A comprehensive coding framework was devel - oped and refined through analysis of a sub - group of advertisements to address aspects of content and meaning. The advertisements were grouped by the decade in which they were pub - lished and then analysed using content analysis (Rose, 2012), a method that has previously been suggested for analysing tobacco advertisements (Anderson et al., 2006). The advertisements were coded for their content in terms of whether they displayed a cigarette that was being smoked, whether they displayed a cigarette that was not being smoked and whether they displayed a cigarette box but no cigarette. They were also coded for how they promoted their product and the meanings being evoked. A total of 18 codes were used here:

aroma, convenience, cost, fear, flavour, friend - ship, happiness, health, humour, image, lady - like, pleasure, popularity, promotions, quality, quantity, relaxing and size. In addition, changes in smoking prevalence data from 1950 to the present day were obtained from Cancer Research UK (2012) and were mapped onto the analytic outcomes. The results were analysed in terms of changes in content across the dec - ades, associations between changes in content and changes in smoking prevalence and changes in meaning across the decades.

Results Changes in the content of advertisements across the decades (1950–2003) Changes in the content of advertisements from 1950 to 2003 are shown in Figure 1. Chi-square goodness-of-fit tests were con - ducted on the data. The results showed a sig - nificant association between decade and advertisements with a cigarette being smoked (χ2(5) = 39.953, p < .01) with percentages indi - cating that fewer advertisements used a ciga - rette being smoked to promote their product over time. There were, however, brief increases in the 1970s and 2000s (64.1% in the 1950s, 36.8% in the 1960s, 48.5% in the 1970s, 15.8% in the 1980s, 3.6% in the 1990s and 17.9% in the 2000s). The results also showed a signifi - cant association between decade and advertise - ments showing a cigarette not being smoked (χ2(5) = 56.468, p < .01). Figure 1 shows an overall decline in the use of a cigarette not being smoked over the decades apart from the 2000s which showed a brief increase (82.1% in the 1950s, 65.8% in the 1960s, 54.5% in the 1970s, 34.2% in the 1980s, 3.6% in the 1990s 0 10 20 30 40 50 60 70 80 90 1950s1960 s1970 s1980s1 990s2000 s Cigare e Smoked Cigare e No t Smoked Cigare e Box, No Cigare e Figure 1. Changes in the content on smoking advertisements from 1950 to 2003. Dewe et al. 5 and 21.4% in the 2000s). There was also a sig - nificant relationship between decade and whether advertisements showed a cigarette box but no cigarette ( χ2(2) = 22.399, p < .01). The percentages for ‘box only’ showed a more complex pattern but generally there was an increase across the decades, with a particular increase between the 1980s and 1990s and a decrease in the 2000s (7.7% in the 1950s, 5.3% in the 1960s, 15.2% in the 1970s, 13.2% in the 1980s, 42.9% in the 1990s and 32.1% in the 2000s).

Changes in content of advertisements across the decades and smoking prevalence Changes in advertising content and smoking prevalence across the decades are shown in Figure 2. Smoking prevalence among females has consistently been lower than among males, but this gender gap has closed in recent years.

Smoking prevalence declined slightly follow - ing the Tobacco Advertising and Promotion Act ban in 2003, but has levelled off since then, remaining stable for both males and females since 2006. Smoking prevalence declined most sharply between 1970 and 1990. There is no obvious association between changes in adver - tising strategy and smoking prevalence.

Changes in the meanings of advertisements across the decades The percentages of meaning codes displayed by the advertisements in each decade are presented in Table 1. The results showed that overall, the four most common meanings were ‘quality’ (e.g.

‘555 cigarettes are the same superlative quality the world over’ and ‘A new cigarette gifted with Embassy quality’), ‘flavour’ (e.g. ‘Filters the smoke but not the taste’ and ‘Come to where the fla - vour is’), ‘pleasure’ (e.g. ‘Whatever the pleas - ure Player’s complete it’ and ‘The simple pleasure of Benson and Hedges Sovereign.

Anytime’) and ‘cost’ (e.g. ‘Now! Benson and Hedges pleasure at every day prices!’ and ‘I like these better than Benson’s Hedges! Yes, Sir, at £3.99 we’re cuttin’ edge’). The four least com - mon codes were ‘fear’, ‘being ladylike’, ‘friendship’ and ‘convenience’. ‘Health’ was present in the top four codes in the 1980s. There was a general downward trend over time in the range and number of meaning codes present in the advertisements. Figure 2. Content of advertisements since 1950 and changes in smoking prevalence for men and women. 6 Journal of Health Psychology 0(0) Changes in the four most common codes (‘quality’, ‘flavour’, ‘pleasure’ and ‘cost’) are compared to the shift in the use of the cigarette box only in Figure 3. The results showed that over time, the use of the meanings associated with pleasure, cost and flavour declined. The invocation of quality in advertisements similarly declined but showed an increase in prevalence in parallel to the increased use of the box only rather than the cigarette.

Discussion The findings from this study indicate that the tobacco companies increasingly used the ciga - rette box in their UK advertising until the Tobacco Advertising and Promotion Act ban in 2003, illustrated by the decline in the use of either a cigarette being smoked or not being smoked and the increase in the use of the box only. The results also suggest that this shift from cigarette to box was unrelated to smoking prevalence. Furthermore, the results indicate that the key meanings used in advertising across the decades were flavour, quality, pleasure and cost, with health being important in the 1980s, and that although an overall decline in their use was seen over time, the shift from cigarette to box only was reflected in an increased presenta - tion of images relating to quality. As noted earlier, proponents of plain packag - ing argue that the cigarette box is a key source of advertising and that the introduction of plain packaging would reduce smoking behaviour (Thrasher et al., 2011). This is refuted by tobacco companies for various reasons, includ - ing the contention that consumers do not make smoking choices on the basis of packaging (Philip Morris Limited, 2008). The results of this study indicate that the box is indeed a source of advertising and that this source was used for several decades before current restric - tions were introduced. Moreover, the results indicate that the tobacco companies are fully cognisant of the power of the cigarette box as a form of advertising as they used it to promote their product even when other advertising foci Table 1. The percentage counts of meanings evoked by smoking advertisements in each decade from 1950 to 2003.

Meaning code 1950s 1960s 1970s 1980s 1990s 2000s Total Rank Aroma 0 7.9 9.1 0 3.6 0 20.6 12 Convenience 7.7 5.3 0 0 0 0 13 15 Cost 12.8 50 36.4 5.3 7.1 10.7 77.3 4 Fear 2.6 0 0 0 0 0 2.6 18 Flavour 38.5 34.2 39.4 28.9 3.6 7.1 151.7 2 Friendship 5.1 5.3 0 0 0 0 10.4 16 Happiness 7.7 2.5 0 2.6 3.6 10.7 27.1 11 Health 5.1 2.6 9.1 15.8 0 0 32.6 10 Humour 17.9 0 3 0 7.1 25 53 6 Image 2.6 5.3 6.1 0 0 0 14 13 Ladylike 0 2.6 3 0 0 0 5.6 17 Pleasure 43.6 34.2 15.2 2.6 3.6 0 99.2 3 Popularity 20.5 13.2 6.1 2.6 0 3.6 46 7 Promotions 0 21.1 12.1 2.6 0 7.1 42.9 8 Quality 66.7 50 18.2 2.6 21.4 3.6 162.5 1 Quantity 0 21.1 0 2.6 3.6 7.1 34.4 9 Relaxing 5.1 5.3 3 0 0 0 13.4 14 Size 17.9 18.4 15.2 18.4 0 7.1 77 5 Dewe et al. 7 were available to them. This research can be said to build on previous studies that have examined the use of branded packaging (Ford et al., 2012; Hammond et al., 2013) and pro - vides further evidence that the cigarette box is indeed an advertising vehicle in its own right. In addition, the results also indicate that the mean - ings associated with cigarettes were transferred from the cigarette onto the box in the years prior to the 2003 ban on print and billboard advertising, particularly those meanings associ - ated with quality. The downward trend in the number and range of meaning codes used in the advertisements throughout the decades can be interpreted as pointing to the key meanings that the tobacco industry wished to be associated with their products. Theories of persuasion provide insights into how changes in tobacco companies’ advertising foci may have oriented towards influencing consumption behaviour. At a very basic level, a strong case or argument in favour of a product needs to be created in order to be persuasive.

(Potential) Consumers are more likely to be persuaded by advertising that focuses exclu - sively on the merits of a particular product (Fennis and Stroebe, 2010). The focus over time on specific key positive features that were ascribed to cigarettes can be seen as a refine - ment of persuasion endeavours. The focus on quality in association with the cigarette box functioned as a catch-all ‘bottom line’ argument for brands in the face of decreasing advertising possibilities in context and content. Furthermore, it has been claimed that a combination of infor - mation and image can increase the likelihood of an advertisement persuading an audience to purchase a product: an image can act as a reminder cue for positive ‘informational’ claims (Ajzen, 2005; Fishbein and Ajzen, 1975). In tobacco advertising, it is possible that the increased use of the cigarette box led to the box becoming a reminder cue for the positive claims that were included in the advertisement. With the increase in the use of the box in advertising, the positive associations from the advertise - ments became linked with the box itself and the box carried forward these associations. The shift from the cigarette to the box was not reflected in any obvious decline in smoking, suggesting that the box was as effective at pro - moting smoking as the cigarette itself. Figure 3. Percentages of the top four most frequently utilised meaning codes and the use of the cigarette box only. 8 Journal of Health Psychology 0(0) This provides some evidence in support of the call for compulsory plain packaging as the box cannot be considered a neutral object that has no impact on consumer choices. Instead the box can be viewed as a form of branded pack - aging that is oriented towards persuading the target audience to purchase the product. After the recent bans on advertising, the box contin - ues to function as a mobile marketing tool that is presented to potential consumers by existing consumers as they smoke their cigarettes. The introduction of compulsory plain packaging of cigarettes would eliminate this advertising medium and its persuasive function.

Funding This research received no specific grant from any funding agency in the public, commercial or not-for- profit sectors.

References Action on Smoking and Health (2010) The smoke filled room: How big tobacco influences health policy in the UK. Available at: http://www.ash.

org.uk/SmokeFilledRoom (accessed 7 January 2013). Action on Smoking and Health (2012a) Public sup - port plain standardised packaging of cigarettes, August. Available at: http://www.ash.org.uk/ media-room/press-releases/:public-support- plain-standardised-packaging-of-cigarettes (accessed 7 January 2013). Action on Smoking and Health (2012b) ASH fact sheet on: Tobacco advertising and promotion in the UK, May. Available at: http://www.ash.org.

uk/files/documents/ASH_124.pdf (accessed 7 January 2013). Ajzen I (2005) Attitudes, Personality and Behavior (2nd edn). Maidenhead: Open University Press/ McGraw-Hill. Anderson SJ, Dewhirst T and Ling PM (2006) Every document and picture tells a story: Using inter - nal corporate document reviews, semiotics, and content analysis to assess tobacco advertising.

Tobacco Control 15(3): 254–261. Cancer Research UK (2012) Smoking statistics. Available at: http://www.cancerresearchuk.org/ cancer-info/cancerstats/types/lung/smoking/ lung-cancer-and-smoking-statistics (accessed 14 March 2013). Chen S and Chaiken S (1999) The heuristic-systematic model in its broader context. In: Chaiken S and Trope Y (eds) Dual Process Theories in Social Psychology . New York: Guilford Press, pp. 73–97. Doll R and Hill AB (1950) Smoking and carcinoma of the lung. British Medical Journal 221(ii): 739–748. Fennis BM and Stroebe W (2010) The Psychology of Advertising . Hove: Psychology Press. Fishbein M and Ajzen I (1975) Belief, Attitude, Inten - tion, and Behavior: An Introduction to Theory and Research . Reading, MA: Addison Wesley. Ford A, Moodie C and Hastings G (2012) The role of packaging for consumer products: Under - standing the move towards ‘plain’ tobacco packaging. Addiction Research and Theory 20(4): 339–347. Freedom Organisation for the Right to Enjoy Smok - ing Tobacco (Forest) (n.d.) Hands off our packs!

Say no to plain packaging. Available at: http:\\ www.handsoffourpacks.com (accessed 15 Janu - ary 2013). Greenwald AG (1968) Cognitive learning, cognitive response to persuasion, and attitude change. In:

Greenwald AG, Brock TC and Ostrom TM (eds) Psychological Foundations of Attitudes . San Diego, CA: Academic Press, pp. 147–170. Hammond D, Daniel S and White CM (2012, in press) The effect of cigarette branding and plain packaging on female youth in the United King - dom. Journal of Adolescent Health . Hammond D, Daniel S and White CM (2013) The effect of cigarette branding and plain packaging on female youth in the United Kingdom. Journal of Adolescent Health 52(2): 151–157. McGuire WJ (1969) The nature of attitudes and attitude change. In: Lindzey G and Aronson E (eds) Handbook of Social Psychology , vol. 3 (2nd edn). Reading, MA: Addison-Wesley, pp.

136–314. MacKintosh AM, Moodie C and Hastings G (2012) The association between point-of-sale displays and youth smoking susceptibility. Nicotine & Tobacco Research 14(5): 616–620. Mitchell AD and Studdert DM (2012) Plain pack - aging of tobacco products in Australia: A novel regulation faces legal challenge. Journal of the American Medical Association 307(3): 261–262. Moodie C and Ford A (2011) Young adult smokers’ perceptions of cigarette pack innovation, pack Dewe et al. 9 colour, and plain packaging. Australasian Mar - keting Journal 19(3): 174–180. Moodie C and Hastings GB (2011) Making the pack the hero, tobacco industry response to marketing restrictions in the UK: Findings from a long-term audit. International Journal of Mental Health and Addiction 9(1): 24–38. Petty RE and Cacioppo JT (1986) Communication and Persuasion: Central and Peripheral Routes to Attitude Change . New York: Springer. Petty RE, Ostrom TM and Brock TC (eds) (1981) Cognitive Responses in Persuasion . Hillsdale, NJ: Erlbaum. Philip Morris Limited (2008) Philip Morris Limited’s response to the Department of Health consul - tation on the future of tobacco control. Avail - able at: http://www.scribd.com/doc/35165899/ Philip-Morris-Limited%E2%80%99s-Response- to-the-Department-of-Health%E2%80%99s- Consultation-on-the-Future-of-Tobacco-Control (accessed 15 January 2013). Rose G (2012) Visual Methodologies: An Introduc - tion to Researching with Visual Materials (3rd edn). London: SAGE. Thrasher FF, Rousu MC, Hammond D, et al. (2011) Estimating the impact of pictorial health warn - ings and ‘plain’ cigarette packaging: Evidence from experimental auctions among adult smok - ers in the United States. Health Policy 102(1): 41–48. Wakefield M, Hayes L, Durkin S, et al. (2013) Intro - duction effects of the Australian plain packaging policy on adult smokers: A cross sectional study.

British Medical Journal Open 3: e003175. Wakefield M, Morley C, Horan JK, et al. (2002) The cigarette pack as image: New evidence from tobacco industry documents. Tobacco Control 11(Suppl. 1): 73–80. Wakefield M, Szczypk G, Terry-McElrath Y, et al. (2005) Mixed messages on tobacco: Comparative exposure to public health, tobacco company- and pharmaceutical company-sponsored tobacco- related television campaigns in the United States, 1999–2003. Addiction 100(12): 1875–1883. World Health Organization (WHO) (2001) Compila - tion of terms and definitions. Available at: http:// apps.who.int/gb/fctc/PDF/inb3/einb3id1.pdf (accessed 27 March 2013).