‘The secret shame’: a content analysis of online news reporting of a celebrity admitting smoking while pregnant

Key words: advocacy, media, tobacco.

On 6 February 2013 Australian radio and television personality Chrissie Swan made a public announcement that she had been smoking during her pregnancy. This announcement followed photographs of her smoking in her car that were subsequently being offered for sale to media outlets. The announcement generated extensive publicity and discussion in Australia. Swan is popularly described as an ‘everywoman’, appealing to women and most notably to mothers and she had previously been the focus of controversy in Australia for her parenting opinions and public comments.^1,2

The evidence of harm to mothers and their babies associated with maternal smoking has been extensively documented and maternal smoking remains a significant cause of infant death and illness.^3,4 Despite the range of smoking and pregnancy programs offered in Australia, in 2010 just over one in 10 (11.7%) Australian women reported smoking during their pregnancy, with rates disproportionately high among socioeconomically disadvantaged women.⁵ These figures may well be a significant underestimation of actual smoking rates.⁶ The stigmatisation of pregnant smokers may influence pregnant women to falsely report their smoking status, undermine their cessation efforts^7–9 and deter antenatal healthcare providers from intervening with clients.¹⁰

This paper reports on the findings of a content analysis of online news stories generated by Swan’s announcement. Studies show that news stories about celebrities and their health issues can increase the community’s uptake of important preventive health behaviours such as screening and searching for health information.^11,12 However, the important factual health information in these stories is found to be overshadowed by more newsworthy ‘human interest’ storylines.^13,14

Studies of online news reporting are increasingly relevant today¹⁵ as the number of Australians seeking their news from online news sources is growing. As at May 2013, 65% of adult Australians (11.39 million) read news online.¹⁶ Analysing news media coverage can usefully monitor the standing of tobacco control issues in the community and inform advocacy and communication strategies within a comprehensive tobacco control program.^17–21 The aim of this research is to identify the main themes in the reporting of the ‘Swan pregnant and admitting smoking’ story by online news media.

The dataset was limited to online coverage of the ‘Swan pregnant and admitting smoking’ story on Australian news sites located using Internet search engines Google News (http://www.google.com.au) and Google Blogs (http://www.google.com.au/blogsearch?hl=en). These sites included a diverse range of Australian national and regional newspapers, weekly magazines, non-print news and entertainment sites and blogs. Online news and blog sites included The Age, The Sydney Morning Herald, Daily Telegraph, Yahoo7 News, 9News Entertainment, Women’s Day and Mamamia. Both news and blog sites were included because some Australian blog sites, for example Mamamia.com.au, have evolved into mass-reach media portals with large readership and influential content.²² The dataset was limited to the story’s coverage during the period 6 February 2013 to 18 February 2013 inclusive using the keywords: ‘Chrissie Swan smoking’, and ‘Chrissie Swan pregnant and smoking’. This time period limitation was imposed as the media’s interest was concentrated in the first few days following Swan’s public announcement and then significantly diminished thereafter. All online news items identified by the search were recorded chronologically onto a spreadsheet.

Content analysis is an appropriate research method for investigating text data.²³ The research study applied a Framework Method to the content analysis²⁴ to systematically categorise pieces of news content within the dataset into themes and codes.^15,19,25 A preliminary examination of the content of 10 news items identified 21 themes and each theme was allocated a code. Shorthand titles and brief descriptions were allocated to each code. An analysis of the remaining news items identified theme similarities and further refining of the codes and their themes was undertaken. Nine distinct themes emerged. Within one of the themes, ‘societal judgment of pregnant women’, two different and opposing viewpoints were identified. As this theme was critical to this study, two sub-codes were created to represent the following two sub-themes: (1) ‘critical of judgement’ and (2) ‘advocate judgement’. A code guide describing the final set of nine themes and two sub-themes was developed.

Each news item was then read in-depth and codes representing the themes were entered against pieces of content within each news item. As each news item generally contained several themes, most news items were allocated more than one code. News items that only included an embedded link to another online news item and/or audio or video of the celebrity’s announcement and no other content were not coded.

To address the risk of the researcher inadvertently influencing the data coding, an interrater reliability assessment was undertaken. Six individual coders coded 40 pieces of content from the dataset. A kappa statistic was calculated to test interrater reliability.²⁶ Descriptive statistical analysis was used to assess the frequency of each content theme.

A total of 124 online news items was found. The dataset included content published through national and regional online news media, radio, television and entertainment websites and blogs. More than half (58.10%; n = 72) of the news items reported the story in the first two days following the announcement. The result of the kappa test for agreement was 0.80 (a kappa value of 0.61–0.80 is in the substantial agreement range).²⁶

Table 1 shows the frequency of the nine content themes found in the online news reporting of the ‘Swan pregnant and admitting smoking’ story. A brief description of the types of content represented by each theme is included under each theme in this table. For example, the theme ‘celebrity story’ encompasses: celebrity Chrissie Swan pregnant and smoking; media/public interested in celebrities’ lives; photographs taken of a celebrity smoking; and a celebrity bidding war to obtain photos.

Table 1.  The frequency of content themes found in online news items reporting the ‘Swan pregnant and admitting smoking’ story 6–18 February 2013 (n = 124 new items)

Illustrative quotes from analysed news items are included to explain and contextualise themes.

‘Celebrity story’ content was most frequently found (90.32%; n = 112):

Fame can be fantastic – the money, all those choices! – but Chrissie Swan has learned first hand (again) this week about the terrible downsides; including that every time you trip up even just a little bit someone is often there with a camera.²⁷

There has been a frenzy of discussion over the last few days regarding the issue of talk show host Chrissie Swan, being photographed smoking while pregnant.²⁸

I could scarcely believe the fuss about Chrissie Swan who confessed to smoking while pregnant with her third child.²⁹

The frequency of this content was found to be double that of ‘quitting is hard’ content (45.97%; n = 57) and more than three times that of ‘smoking and health’ content (29.03%; n = 36).

The second most frequent theme was ‘societal judgement of pregnant women’ (69.35%; n = 86). Over one-half (56.98%; n = 49) of this content was ‘critical of judgement’ of pregnant women who smoked:

We are all so quick to judge each other.’³⁰

You could be a pregnant woman sitting in a pub with friends – having a soda water for all anybody else knows – but someone will still come up to you and tell you that you shouldn’t be drinking.³¹

The remaining content in this theme supported the view to ‘advocate judgement’ of pregnant women (43.02%; n = 37):

The undeniable difference is when you or I scarf [sic] down a donut and bottle of rum (why not?) it is having an influence on our own state of health. A pregnant mom is an entirely different affair as the consequences fall on another party who has no control over the situation and will deal with the consequences for the rest of their life.³²

Any woman who smokes while she is pregnant is either a bogan,^* a fuckwit, or both. So that gets my opinion out of the way before discussing the Chrissie Swan controversy.³³

Swan’s negative opinions of pregnant women who smoke and of being labelled a pregnant smoker were quoted in around one-half (51.35%; n = 19) of ‘advocate judgement’ content:

I knew it was wrong that there is so much terrible judgment that only awful people and bad parents and idiots and bogans smoke during pregnancy – and I didn’t feel like I belonged in any of those categories – so I kept it all under wraps and dealt with it how I could.³⁴

It’s a horrifying situation and people are rightly disgusted.³⁵

I realised the whole of Australia would want to hang me.³⁶

The ‘quitting is hard’ theme was found in less than one-half of news items (45.97%; n = 57). This theme represented issues such as the difficulties of quitting smoking for Swan and for other smokers, the power of addiction, the frequency of relapse and the need for useful quit support.

Here’s the truth. Obviously I know it’s wrong. I’m not an idiot, no smoker wants to smoke – especially when they are pregnant – but it is clearly an addiction and a very serious one because it can cloud your judgment and make the unthinkable somehow okay.³⁴

Shaming smokers, like shaming heavy drinkers such as the two high-profile women journalists who have recently come out and spoken or written about their heavy, unhealthy, alcohol use, gets us nowhere. Supporting them to overcome the dependency, even if they are a celebrity, and even if they’re pregnant, sends the message it’s OK to talk about your issue, and to ask for help.²⁷

Only 36 (29.03%) news items included ‘smoking and health’ content:

Mothers who smoke during pregnancy increase their risk of having a miscarriage or a premature birth, their baby is [sic] has an increased risk of a low birthweight or death from sudden infant death syndrome.³⁵

Just over one in 10 (13.71%; n = 17) news items reported specific ‘quit-smoking referral information’ (Quitline, website www.quitnow.gov.au).

The Quitline 131 848 is a free confidential telephone-based service primarily designed to help smokers quit smoking.³⁴

One-fifth (20.16%; n = 25) of news items mentioned a ‘positive outcome’ from the story, primarily suggesting that the Swan story had raised awareness about smoking and pregnancy.

Only 17 (13.71%) news items included content from a public health or tobacco control expert.

It was not surprising that the study found a high volume of online news stories and blogs as the public’s interest in celebrities’ personal lives and behaviours drives news and entertainment media coverage. However, the study found the dramatic reporting of a celebrity smoking when pregnant dominated the story and there was very little media coverage that communicated positive health promoting messages, relevant health information or referred to tobacco control experts and health professionals.

Swan’s decision to initially frame her smoking as primarily a moral failing set the tone for most of the coverage. The story quickly devolved to an argument about whether pregnant women, like Swan, should be judged by their behaviours. The analysis suggests there was a missed opportunity to highlight the circumstances that influence pregnant women to continue to smoke, and to promote sensitive and non-judgemental messages that support pregnant women who smoke to address their behaviour. For example, there was potential to extend the life of the story by collaborating with Swan to publicise her quit-smoking pathway and promote positive quit-smoking messages that focus on pregnant women’s well being and the importance of staying quit at postpartum. Whilst caution is advised for recruiting celebrities in health promotion campaigns,³⁷ smoking and pregnancy health promotion programs could harness the public’s interest in celebrity pregnancies to communicate important public health information and positive messages designed to influence pregnant women and their partners’ smoking behaviour. The immediate rise in bookings for mammogram screening and the subsequent dramatic increase in mammogram screening rates by Australian women following the 2005 announcement of celebrity Kylie Minogue’s diagnosis of breast cancer demonstrates the power of ‘unplanned media events’ on public health outcomes.¹² Similar increases in cervical screening rates by British women were observed following media coverage of a cancer diagnosis for the late celebrity Jade Goody.¹¹ Public criticism, blaming and even demonisation of individuals recruited to high profile health promotion campaigns are risks that should be assessed when considering this type of strategy.

Finally, while the study’s finding of a sizeable number of news items critical of judgement or the ‘public policing’ of pregnant women was encouraging, there was a lack of alternative positive messages about quitting, quit support and referral information for pregnant women. This lack should not be a gauge of the newsworthiness of smoking and pregnancy health information nor a measure of the public health community’s efforts to engage with the media in the story. Instead this finding highlights the need to investigate how public health practitioners can work more effectively with media outlets to advance smoking and pregnancy public health goals. This includes capacity building in public health advocacy to support effective media reporting of smoking and pregnancy issues that improve rather than hinder pregnant women’s ability to quit smoking.

One limitation of this study was that the dataset comprised online news stories and blogs and did not include on-air coverage of the story by television and radio outlets. However, a substantial number of the online news items included an embedded audio or video link, a television clip or a transcript of radio coverage relating to the story, so it was unlikely key content was missed.

This study found the main themes within the online media coverage of this celebrity health story were primarily focused on celebrity hype and societal attitudes towards pregnant women who smoke. The study also found a lack of evidence of a measured public health response to the story. A readiness to respond to unplanned media coverage of controversial public health issues in order to participate in and influence the media’s reporting is an essential public health advocacy strategy. Identifying issues that may have influenced the public health response in this instance would be useful to inform future capacity building for initiating effective advocacy and communication activities that enable pregnant smokers to quit. The recruitment of relatable and accessible women, including celebrities, who have successfully quit smoking during pregnancy, could also be explored in formative research on future smoking and pregnancy campaigns as a potential communication strategy.