TOBACCO
Newspaper coverage of tobacco control in New Zealand Rob McGee,1 Sophie Bang,1 Louise Marsh1
T
he media represents a powerful source of influence in modern society, and media coverage can determine what we perceive as the norm, and what we view as being ‘good’, ‘bad’, ‘important’ or ‘insignificant’.1 The types of events and issues published in newspapers are those journalists and news editors have deemed for various reasons to be newsworthy2 and, in a sense, the media can define reality in public discourse. As one journalist noted, “if you don’t exist in the media, to all intents and purposes, you don’t exist”.3 Consequently, the media may have a powerful role in determining the effectiveness of public health policies such as tobacco control policy through its influence on individuals and policymakers.1 From a public health perspective, coverage can help inform the public of health issues and government actions to address those issues. Tobacco control initiatives such as the plain packaging of cigarettes can be portrayed in specific ways, and this can greatly influence their public perception and their palatability.4 Previous Australian and US research indicates that although coverage may vary considerably across a year, tobacco events are seen as newsworthy; reporting of them is frequent with one article reported on average every four days per newspaper; and reporting is generally sympathetic to tobacco control objectives.5,6 Studies have shown that enhancing media coverage of tobacco control initiatives is one way of influencing their effectiveness, as it shapes the values held by the community.4 Furthermore, recent Australian research suggests that news stories about tobacco and its health effects can
Abstract Objective: The New Zealand (NZ) government has proposed that the country be ‘Smokefree’ by 2025. This paper examines how NZ newspapers have portrayed tobacco control initiatives to achieve this goal. Method: We examined tobacco-related articles from NZ newspapers published between 1 November 2011 and 31 October 2012, using a coding frame to capture smoking themes, portrayal of actions and degree of support expressed for various pro- and anti- tobacco control objectives. Results: A total of 537 separate articles were obtained from national and regional newspapers. Six themes appeared in more than 5% of the total articles: Smokefree 2025; smokefree areas; pricing; tobacco industry; plain packaging; and cessation. Overall, articles on tobacco smoking were three times more likely to emphasise actions consonant with tobacco control objectives rather than against them, and to report them with a positive rather than negative tone. Conclusions: NZ tobacco control advocates can take heart from the nature and extent of coverage of tobacco control initiatives. Opportunities for further media advocacy are discussed. Key words: tobacco control, media, advocacy, smokefree, 2025
positively influence smokers’ perceptions of the harms of tobacco and act as significant prompts to quit.7 Each year, about 5,000 people die in New Zealand (NZ) through smoking and 300 New Zealanders die from exposure to second-hand smoke, and half of all long-term smokers will die from a smoking-related disease.8 To reduce this toll, the NZ government made a commitment in March 2011 to a goal of a Smokefree NZ by 2025.9 This goal aims to have less than 5% of the population identifying as current smokers by 2025 and, to achieve this goal, 40,000 smokers need to successfully quit smoking annually leading up to 2025, assuming no new smokers start.9
achieve this goal. These are outlined in ‘Smokefree Aotearoa/New Zealand 2025 Logic’,9 and include: 1) protecting children from exposure to tobacco smoking and its promotion by supporting plain packaging, health warnings on packs, and smokefree cars and communities; 2) reduction in demand and supply achieved through policy changes such as increases in taxation and further regulating and restricting the sale of tobacco; and 3) support for smoking cessation both at the individual level through supporting smokers with Quitline and nicotine replacement therapy, and at a national level through promoting the Smokefree 2025 goal.
A number of tobacco control initiatives have been identified as necessary to
In this paper we examine the portrayal of Smokefree 2025 tobacco control initiatives in
1. Preventive and Social Medicine, Otago University, New Zealand Correspondence to: Professor Rob McGee, Preventive and Social Medicine, Otago University, Dunedin School of Medicine, Dunedin 9054, New Zealand; e-mail: [email protected] Submitted: July 2013; Revision requested: August 2013; Accepted: January 2014 The authors have stated they have no conflict of interest. Aust NZ J Public Health. 2014; 38:265-9; doi: 10.1111/1753-6405.12216
2014 vol. 38 no. 3
Australian and New Zealand Journal of Public Health © 2014 The Authors. ANZJPH © 2014 Public Health Association of Australia
265
McGee, Bang and Marsh
the print media. As far as we are aware, there have been no published papers examining how the print media is portraying the path to a Smokefree NZ. While the portrayal of tobacco issues in the media has been addressed prior to Smokefree 2025, this has primarily involved tobacco imagery in the visual media.10 Newspapers here have retained a steady readership even during a tough market.11 If anything, newspapers are perhaps even more accessible through easy access to newspaper articles online. Smokefree 2025 is a national goal requiring broad action on several fronts, and it is meaningful to examine how often and how sympathetically this goal and associated tobacco control initiatives have been portrayed by the NZ media.
centres of Auckland, Hamilton, Wellington, Christchurch and Dunedin, and the 15 daily papers with circulation figures of 10,000 or more as well as the three weekly papers.13
Coding of articles
At the same time, newspapers have the ability to present the tobacco industry’s version of events and industry reaction to these policy proposals. Smokefree initiatives will not only have an impact on industry profits in NZ, but will serve as a model for other jurisdictions to follow suit, perhaps in much the same way that Australia’s plain packaging is to be implemented in this country. Applying ‘scream test’ logic – the more effective the tobacco control activity, the greater the industry’s response – would suggest that the tobacco industry will fight these proposals as vigorously as possible.12 Consequently, it is also important to examine how newspapers frame tobacco industry perceptions and arguments.
Method Sources of articles The data for this analysis consisted of tobacco-related articles from NZ newspapers published over the 12-month period between 1 November 2011 and 31 October 2012. The articles were found through three newspaper databases: 1) Newztext Plus – The Knowledge Basket;2) Factiva;and 3) Otago Daily Times website. Articles were obtained from eighteen different newspapers: NZ Herald; Dominion Post; Otago Daily Times; Timaru Herald; Waikato Times; Bay of Plenty Times; Manawatu Standard; Northern Advocate; Taranaki Daily Times; Christchurch Press; Nelson Mail; Daily Post; Marlborough Express; Southland Times; Sunday Star Times; Herald on Sunday; Hawkes Bay Today; and Sunday News. These papers covered all daily newspapers in the five main population
266
Three tobacco-related keywords were used to search for relevant articles on the newspaper databases: smoking, tobacco and cigarettes. For the article to be eligible for inclusion, the main event or issue discussed had to be tobacco related; this eliminated articles that only briefly mentioned one or more keywords. Items retrieved included news articles, editorials from the newspaper editors, opinion pieces submitted by other writers, and letters to the editor. All were coded using a system adapted from the Smith et al. (2005) framework.5,6 This coding frame included three main sets of variables: descriptors, content theme and tone. The article descriptors included the month of publication, type of article, name of newspaper, and the number of appearances, where the same article was published multiple times in different newspapers. The content theme was assessed under 19 themes including: ‘Smokefree 2025’ (which included references to a Smokefree NZ); consumption and smoking prevalence; youth smoking and access; health effects; second-hand smoke; smokefree areas including cars; point of sale display; plain packaging; health warnings on packs; advertising and promotion issues; pricing including tax increases; addiction; prevention and education; cessation including the national Quitline; tobacco industry actions; ‘endgame’; product issues; economic issues including free trade; and other issues not already classified. These categories were adapted from the Smith et al. (2005) framework to give a better reflection of the current tobacco control policies at issue in NZ.5 While more than one theme could be identified for each article, the focus was on the main theme(s) contained in the article; themes that were only briefly mentioned and not elaborated upon were not included in the classification of the articles. The tone of the article was assessed along two dimensions: a) event or action impact; and b) opinion. The first dimension refers to the impact of the event or action described in the article on smokefree objectives. The event in an article could be classified as having a positive, negative, mixed or neutral impact on tobacco control initiatives. Events with a positive impact were those that lined up
with tobacco control objectives; for example, an article about political support for plain packaging would be coded as having a positive impact. On the other hand, an article about the tobacco industry’s 2012 Agree/ Disagree campaign to counteract legislation for plain packaging (see: www.agreedisagree. co.nz) would be coded as having a negative impact. Mixed impact was used for articles that talked about, for example, both political support for plain packaging and the tobacco industry’s actions to counter that support, or discussed events or initiatives that might have an uncertain effect on tobacco control. The opinion dimension assessed the general impression that a reader of the article would be left with. Was the article broadly supportive of, or hostile towards, tobacco control objectives? This could be reflected by either the views of people interviewed or mentioned in the news article, or the explicit views of the writers in opinion pieces or letters. For example, a positive opinion code would be used for articles containing expressions of support for plain packaging, or the views of people that the industry’s Agree/ Disagree campaign was unethical or a waste of money. Conversely, an article containing statements by tobacco retailers that plain packaging will not be effective, or an industry viewpoint on the need for the Agree/Disagree campaign to counteract plain packaging, would be coded as a negative opinion. So, it was possible to code articles for positive event but negative opinion and vice-versa. Similarly, the same article might have both positive and negative opinions expressed. A copy of the coding framework is available from the authors on request.
Reliability of coding The author (SB) coded all the articles and randomly selected 71 articles (13% of all articles) spread throughout the 12 months of reporting, which were then independently scored by RM. Inter-rater agreement on the coding (including descriptors, content theme and tone variables) was assessed by Kappa statistic.14 The majority of the Kappa scores were between 0.66 and 0.96, an acceptable level of agreement; the Kappa score for the ‘other’ category, however, was relatively low at 0.48. In those instances where the primary coder (SB) felt uncertain as to the appropriate code, coding was discussed with the other authors to reach consensus. Descriptive statistics are provided for all variables. A standard Z-test, equivalent to
Australian and New Zealand Journal of Public Health © 2014 The Authors. ANZJPH © 2014 Public Health Association of Australia
2014 vol. 38 no. 3
Media and tobacco
Results Tobacco articles A total of 537 separate articles were obtained from both national and regional newspapers: 60% were news items; 15% opinion pieces or editorials; and 23% letters to the editor. Ninety of these individual articles were repeated two or more times in different newspapers; 56 articles were repeated in one other newspaper; and the remainder were repeated two or more times. Consequently, 693 articles were published in total. Figure 1 shows the number of individual articles and individual plus repeated articles published over the 12 months from November 2011 to October 2012. The amount of tobacco coverage fluctuated over the course of the year, with high numbers of articles in January, May, July and August. The number of repeated articles followed a similar trend. This fluctuation makes it difficult to see the general trend of tobacco coverage over time.
Theme of articles In total, 1,057 themes were identified in the 537 articles. Table 1 shows the distribution of the 19 different tobacco themes in the newspaper articles. Of these, six themes appeared in more than 5% of the total articles: Smokefree 2025; smokefree areas; pricing; tobacco industry; plain packaging; and cessation. These six themes accounted for 73% of all published articles. Additionally, more than 5% of the articles were coded as ‘other’, a category that included stories about prisoners’ responses to the introduction of smokefree prisons; comparisons between tobacco control and alcohol policy, usually with an emphasis on the comparatively weak nature of alcohol policy; and a supermarket chain’s decision to instruct staff to place cigarette packets in a plastic bag before handing them to customers, so that “no one can actually see cigarettes”. Temporal variation in the appearance of themes was also evident. For example, the theme ‘smokefree areas’ peaked in January with 20 of the total 120 articles published or 17% of all articles with smokefree areas as a central theme; ‘cessation’ also peaked in
2014 vol. 38 no. 3
Figure 1: Number of articles, and repeated articles, on tobacco published between November 2011 and October 2012. 140 140 120 120 100 100 80 80 60 60 40 40 20 20 0 0
Num,ber of appearances Num,ber of appearances
a chi-square test, was used to assess the difference between the proportions of positive and negative responses for event and opinion variables, with p
Newspaper coverage of tobacco control in New Zealand Rob McGee,1 Sophie Bang,1 Louise Marsh1
T
he media represents a powerful source of influence in modern society, and media coverage can determine what we perceive as the norm, and what we view as being ‘good’, ‘bad’, ‘important’ or ‘insignificant’.1 The types of events and issues published in newspapers are those journalists and news editors have deemed for various reasons to be newsworthy2 and, in a sense, the media can define reality in public discourse. As one journalist noted, “if you don’t exist in the media, to all intents and purposes, you don’t exist”.3 Consequently, the media may have a powerful role in determining the effectiveness of public health policies such as tobacco control policy through its influence on individuals and policymakers.1 From a public health perspective, coverage can help inform the public of health issues and government actions to address those issues. Tobacco control initiatives such as the plain packaging of cigarettes can be portrayed in specific ways, and this can greatly influence their public perception and their palatability.4 Previous Australian and US research indicates that although coverage may vary considerably across a year, tobacco events are seen as newsworthy; reporting of them is frequent with one article reported on average every four days per newspaper; and reporting is generally sympathetic to tobacco control objectives.5,6 Studies have shown that enhancing media coverage of tobacco control initiatives is one way of influencing their effectiveness, as it shapes the values held by the community.4 Furthermore, recent Australian research suggests that news stories about tobacco and its health effects can
Abstract Objective: The New Zealand (NZ) government has proposed that the country be ‘Smokefree’ by 2025. This paper examines how NZ newspapers have portrayed tobacco control initiatives to achieve this goal. Method: We examined tobacco-related articles from NZ newspapers published between 1 November 2011 and 31 October 2012, using a coding frame to capture smoking themes, portrayal of actions and degree of support expressed for various pro- and anti- tobacco control objectives. Results: A total of 537 separate articles were obtained from national and regional newspapers. Six themes appeared in more than 5% of the total articles: Smokefree 2025; smokefree areas; pricing; tobacco industry; plain packaging; and cessation. Overall, articles on tobacco smoking were three times more likely to emphasise actions consonant with tobacco control objectives rather than against them, and to report them with a positive rather than negative tone. Conclusions: NZ tobacco control advocates can take heart from the nature and extent of coverage of tobacco control initiatives. Opportunities for further media advocacy are discussed. Key words: tobacco control, media, advocacy, smokefree, 2025
positively influence smokers’ perceptions of the harms of tobacco and act as significant prompts to quit.7 Each year, about 5,000 people die in New Zealand (NZ) through smoking and 300 New Zealanders die from exposure to second-hand smoke, and half of all long-term smokers will die from a smoking-related disease.8 To reduce this toll, the NZ government made a commitment in March 2011 to a goal of a Smokefree NZ by 2025.9 This goal aims to have less than 5% of the population identifying as current smokers by 2025 and, to achieve this goal, 40,000 smokers need to successfully quit smoking annually leading up to 2025, assuming no new smokers start.9
achieve this goal. These are outlined in ‘Smokefree Aotearoa/New Zealand 2025 Logic’,9 and include: 1) protecting children from exposure to tobacco smoking and its promotion by supporting plain packaging, health warnings on packs, and smokefree cars and communities; 2) reduction in demand and supply achieved through policy changes such as increases in taxation and further regulating and restricting the sale of tobacco; and 3) support for smoking cessation both at the individual level through supporting smokers with Quitline and nicotine replacement therapy, and at a national level through promoting the Smokefree 2025 goal.
A number of tobacco control initiatives have been identified as necessary to
In this paper we examine the portrayal of Smokefree 2025 tobacco control initiatives in
1. Preventive and Social Medicine, Otago University, New Zealand Correspondence to: Professor Rob McGee, Preventive and Social Medicine, Otago University, Dunedin School of Medicine, Dunedin 9054, New Zealand; e-mail: [email protected] Submitted: July 2013; Revision requested: August 2013; Accepted: January 2014 The authors have stated they have no conflict of interest. Aust NZ J Public Health. 2014; 38:265-9; doi: 10.1111/1753-6405.12216
2014 vol. 38 no. 3
Australian and New Zealand Journal of Public Health © 2014 The Authors. ANZJPH © 2014 Public Health Association of Australia
265
McGee, Bang and Marsh
the print media. As far as we are aware, there have been no published papers examining how the print media is portraying the path to a Smokefree NZ. While the portrayal of tobacco issues in the media has been addressed prior to Smokefree 2025, this has primarily involved tobacco imagery in the visual media.10 Newspapers here have retained a steady readership even during a tough market.11 If anything, newspapers are perhaps even more accessible through easy access to newspaper articles online. Smokefree 2025 is a national goal requiring broad action on several fronts, and it is meaningful to examine how often and how sympathetically this goal and associated tobacco control initiatives have been portrayed by the NZ media.
centres of Auckland, Hamilton, Wellington, Christchurch and Dunedin, and the 15 daily papers with circulation figures of 10,000 or more as well as the three weekly papers.13
Coding of articles
At the same time, newspapers have the ability to present the tobacco industry’s version of events and industry reaction to these policy proposals. Smokefree initiatives will not only have an impact on industry profits in NZ, but will serve as a model for other jurisdictions to follow suit, perhaps in much the same way that Australia’s plain packaging is to be implemented in this country. Applying ‘scream test’ logic – the more effective the tobacco control activity, the greater the industry’s response – would suggest that the tobacco industry will fight these proposals as vigorously as possible.12 Consequently, it is also important to examine how newspapers frame tobacco industry perceptions and arguments.
Method Sources of articles The data for this analysis consisted of tobacco-related articles from NZ newspapers published over the 12-month period between 1 November 2011 and 31 October 2012. The articles were found through three newspaper databases: 1) Newztext Plus – The Knowledge Basket;2) Factiva;and 3) Otago Daily Times website. Articles were obtained from eighteen different newspapers: NZ Herald; Dominion Post; Otago Daily Times; Timaru Herald; Waikato Times; Bay of Plenty Times; Manawatu Standard; Northern Advocate; Taranaki Daily Times; Christchurch Press; Nelson Mail; Daily Post; Marlborough Express; Southland Times; Sunday Star Times; Herald on Sunday; Hawkes Bay Today; and Sunday News. These papers covered all daily newspapers in the five main population
266
Three tobacco-related keywords were used to search for relevant articles on the newspaper databases: smoking, tobacco and cigarettes. For the article to be eligible for inclusion, the main event or issue discussed had to be tobacco related; this eliminated articles that only briefly mentioned one or more keywords. Items retrieved included news articles, editorials from the newspaper editors, opinion pieces submitted by other writers, and letters to the editor. All were coded using a system adapted from the Smith et al. (2005) framework.5,6 This coding frame included three main sets of variables: descriptors, content theme and tone. The article descriptors included the month of publication, type of article, name of newspaper, and the number of appearances, where the same article was published multiple times in different newspapers. The content theme was assessed under 19 themes including: ‘Smokefree 2025’ (which included references to a Smokefree NZ); consumption and smoking prevalence; youth smoking and access; health effects; second-hand smoke; smokefree areas including cars; point of sale display; plain packaging; health warnings on packs; advertising and promotion issues; pricing including tax increases; addiction; prevention and education; cessation including the national Quitline; tobacco industry actions; ‘endgame’; product issues; economic issues including free trade; and other issues not already classified. These categories were adapted from the Smith et al. (2005) framework to give a better reflection of the current tobacco control policies at issue in NZ.5 While more than one theme could be identified for each article, the focus was on the main theme(s) contained in the article; themes that were only briefly mentioned and not elaborated upon were not included in the classification of the articles. The tone of the article was assessed along two dimensions: a) event or action impact; and b) opinion. The first dimension refers to the impact of the event or action described in the article on smokefree objectives. The event in an article could be classified as having a positive, negative, mixed or neutral impact on tobacco control initiatives. Events with a positive impact were those that lined up
with tobacco control objectives; for example, an article about political support for plain packaging would be coded as having a positive impact. On the other hand, an article about the tobacco industry’s 2012 Agree/ Disagree campaign to counteract legislation for plain packaging (see: www.agreedisagree. co.nz) would be coded as having a negative impact. Mixed impact was used for articles that talked about, for example, both political support for plain packaging and the tobacco industry’s actions to counter that support, or discussed events or initiatives that might have an uncertain effect on tobacco control. The opinion dimension assessed the general impression that a reader of the article would be left with. Was the article broadly supportive of, or hostile towards, tobacco control objectives? This could be reflected by either the views of people interviewed or mentioned in the news article, or the explicit views of the writers in opinion pieces or letters. For example, a positive opinion code would be used for articles containing expressions of support for plain packaging, or the views of people that the industry’s Agree/ Disagree campaign was unethical or a waste of money. Conversely, an article containing statements by tobacco retailers that plain packaging will not be effective, or an industry viewpoint on the need for the Agree/Disagree campaign to counteract plain packaging, would be coded as a negative opinion. So, it was possible to code articles for positive event but negative opinion and vice-versa. Similarly, the same article might have both positive and negative opinions expressed. A copy of the coding framework is available from the authors on request.
Reliability of coding The author (SB) coded all the articles and randomly selected 71 articles (13% of all articles) spread throughout the 12 months of reporting, which were then independently scored by RM. Inter-rater agreement on the coding (including descriptors, content theme and tone variables) was assessed by Kappa statistic.14 The majority of the Kappa scores were between 0.66 and 0.96, an acceptable level of agreement; the Kappa score for the ‘other’ category, however, was relatively low at 0.48. In those instances where the primary coder (SB) felt uncertain as to the appropriate code, coding was discussed with the other authors to reach consensus. Descriptive statistics are provided for all variables. A standard Z-test, equivalent to
Australian and New Zealand Journal of Public Health © 2014 The Authors. ANZJPH © 2014 Public Health Association of Australia
2014 vol. 38 no. 3
Media and tobacco
Results Tobacco articles A total of 537 separate articles were obtained from both national and regional newspapers: 60% were news items; 15% opinion pieces or editorials; and 23% letters to the editor. Ninety of these individual articles were repeated two or more times in different newspapers; 56 articles were repeated in one other newspaper; and the remainder were repeated two or more times. Consequently, 693 articles were published in total. Figure 1 shows the number of individual articles and individual plus repeated articles published over the 12 months from November 2011 to October 2012. The amount of tobacco coverage fluctuated over the course of the year, with high numbers of articles in January, May, July and August. The number of repeated articles followed a similar trend. This fluctuation makes it difficult to see the general trend of tobacco coverage over time.
Theme of articles In total, 1,057 themes were identified in the 537 articles. Table 1 shows the distribution of the 19 different tobacco themes in the newspaper articles. Of these, six themes appeared in more than 5% of the total articles: Smokefree 2025; smokefree areas; pricing; tobacco industry; plain packaging; and cessation. These six themes accounted for 73% of all published articles. Additionally, more than 5% of the articles were coded as ‘other’, a category that included stories about prisoners’ responses to the introduction of smokefree prisons; comparisons between tobacco control and alcohol policy, usually with an emphasis on the comparatively weak nature of alcohol policy; and a supermarket chain’s decision to instruct staff to place cigarette packets in a plastic bag before handing them to customers, so that “no one can actually see cigarettes”. Temporal variation in the appearance of themes was also evident. For example, the theme ‘smokefree areas’ peaked in January with 20 of the total 120 articles published or 17% of all articles with smokefree areas as a central theme; ‘cessation’ also peaked in
2014 vol. 38 no. 3
Figure 1: Number of articles, and repeated articles, on tobacco published between November 2011 and October 2012. 140 140 120 120 100 100 80 80 60 60 40 40 20 20 0 0
Num,ber of appearances Num,ber of appearances
a chi-square test, was used to assess the difference between the proportions of positive and negative responses for event and opinion variables, with p
