A QUALITATIVE INVESTIGATION OF SOUTH AFRICAN CIGARETTE SMOKERS ’ PERCEPTIONS OF FEAR APPEAL MESSAGES IN ANTI-SMOKING ADVERTISING

Cigarette smoking continues to pose a global health risk, including in developing countries. Fear appeal messages have been widely employed in health communication to reduce cigarette smoking, but studies provide confl icting results on their effi cacy. The present qualitative study explores smokers’ perceptions of fear appeal messages used in anti-smoking advertising. Focus group discussions were conducted with male and female smokers from Gauteng. A thematic analysis found that participants negatively viewed advertisements that use unrealistic images and failed to relate to the message portrayed. Information about the risks associated with smoking was perceived as patronising and as positioning smokers as ignorant and unintelligent. In addition, fear appeal messages that only focus on long-term consequences of smoking were perceived as ineffective. Participants failed to identify with content that solely relied on factual information at the expense of an emotive appeal. The fi ndings suggest that anti-smoking communication could benefi t from content that evokes shock without sacrifi cing realism, that it should include information about shortterm and immediately visible consequences of smoking and that it should avoid negative depictions of smokers that alienate them from the message being portrayed.


introduction and BacKground to tHe ProBLeM
Smoking as a health-risk behaviour is preventable, yet tobacco smoking is positioned high on the list of health-risk behaviours, and the burden of disease attributable to smoking in developed countries is greater than that of all other health-risk behaviours combined (Kaptein & Weinman 2004:27).Globally, approximately 35% of men and 22% of women in developed countries smoke cigarettes (Mackay, Erikson & Shafey 2006:22-24).South Africa has similar prevalence rates to other developing countries, with approximately 31% of men and 8% of women in South Africa who smoke cigarettes (Department of Health 2003:3).Although the global prevalence of smoking has decreased over the past decades, this decrease has mostly been among higher socio-economic groups.Studies also show that certain populations such as young adults (johnston, O'Malley & Bachman 2001:197) and adolescent girls increasingly take up smoking (Killen 1998:228-229).South African research shows that smoking signifi cantly increases risk of death due to tuberculosis, chronic obstructive pulmonary disease, lung cancer, other upper aerodigestive cancer and ischaemic heart disease (Sitas et al. 2004:397).Smoking contributes in a very particular manner to health risks in South Africa, as the country has a high incidence of tuberculosis, which is further fuelled by the HIV epidemic (Sitas et al. 2004:399).In investigating tobacco-attributable mortality in South Africa, Sitas et al. (2004:399) conclude that smoking contributes to approximately 20% of deaths caused by tuberculosis.Furthermore, approximately 8% of all adult deaths in South Africa result from smoking (Sitas et al. 2004:399).
South Africa previously had no measures in place to control or reduce tobacco use, but since 1993 has introduced increasingly progressive legislative measures to control smoking.In 1995, health warnings were introduced for tobacco packaging and advertising (Swart & Panday 2003:3) and more progressive measures followed when the Tobacco Products Control Amendment Act of 1999 took effect.The Act restricts smoking in workplaces and other public places and prohibits tobacco advertising and promotion, including through sponsored events (Tobacco Products Amendment Act, No. 12 1999).Within this context, various initiatives are also continuously being implemented by government as well as non-governmental organisations in order to reduce the prevalence of smoking.These initiatives include health communication campaigns that communicate the risks involved in smoking and that aim to discourage smoking.The focus of this qualitative study was to explore how South African smokers perceive fear appeal messages in anti-smoking advertisements.

Literature review The efficacy of fear appeal in anti-smoking advertising
Anti-smoking communication has had limited success in deterring smoking behaviour.Some studies report positive results (Hafstad, Aaro & Langmark 1996:34;Smith & Stutts 2006:277) but others report that even sustained and highly visible anti-smoking messages have been demonstrated to be ineffective in changing attitudes toward smoking or deterring smoking behaviour (Bauman et al. 1991:600;Murray, Prokhorov & Harty 1994:54).There is also evidence for a 'boomerang' effect in health promotion campaigns, where messages that appeal to behaviour change result in responses opposite to what is called for (Ringold 2002:27).Wolburg (2006:298) reports that smokers who are recipients of anti-smoking messages often employ psychological responses such as denial, anger and defiance.
It is in this context that anti-smoking communication has increasingly incorporated fear appeal in messages aimed at reducing smoking behaviour.Fear appeal in advertising can be defined as 'persuasive messages that arouse fear' (Witte & Allen 2000:591).This strategy relies on fear as a powerful motivator in persuading an individual to change an attitude or belief (Witte 1998:424).Dion (2005:7) states that the condition of fear involves physiological arousal that results in more cognitive, affective and behavioural attention focused on the threat; resulting in an individual acting to lessen the threat and thereby reducing the fear.Research into the effect of fear appeal in advertising has considered the role of perceived susceptibility to the threat, which entails perceptions of the likelihood of experiencing a threat, as well as the role of perceived severity of the threat, which entails the extent of harm one expects from the threat (Witte 1992:330).Furthermore, research on fear appeal messages has also taken into account the role of perceived efficacy, which refers to the extent to which one perceives oneself as able to respond in a manner that serves to avert the threat (Witte 1992:330).
Studies exploring the efficacy of fear appeal in inducing behaviour change in relation to smoking provide equivocal results.Some studies have indicated that fear appeal is effective in persuading people to reduce health-risk behaviours such as smoking (Hale & Dillard 1995:73;Wakefield et al. 2003:242), yet Witte and Allen (2000:606) conclude from their meta-analysis that the use of fear appeal could in some cases have the opposite effect.The use of fear appeal may actually result in certain individuals adopting maladaptive responses, such as denial and avoidance.Witte and Allen (2000:604) state that fear appeal messages are most effective when combining a strong fear appeal with a message high in self-efficacy.Messages that combine strong fear appeals with low levels of self-efficacy are most ineffective in that they prompt psychological defensive responses such as avoidance or reactance.

Investigating perceptions of fear appeal using qualitative approaches
Research concerned with fear appeal messages in anti-smoking communication generally focuses on the efficacy of fear appeal in changing attitudes and behaviour and commonly employs a quantitative approach.A case in point is the extensive review conducted by Witte and Allen (2000:596), which focused on fear appeals in public health campaigns.Of the 98 studies that they identified for review, only 5 employed a non-experimental design.There is, however, a small body of international research that investigates smokers' perceptions of fear appeal messages in anti-smoking advertisements using qualitative approaches.
From the latter type of research, several themes can be identified that relate to smokers' perceptions of anti-smoking advertising.Montazeri and McEwen (1997:31) found that antismoking advertisements that use fear appeal are perceived more favourably if they are realistic.Participants in their study preferred content that captures reality and messages that are not artificial or highly stylised.Maguire and Love (2006:100) describe this as a distancing technique, in which participants find it easier to ignore fear appeal messages that they perceive to be unrealistic.Another theme reported in qualitative studies is smokers' perception that anti-smoking advertising presents information that is already known or is viewed as common sense.This perception includes the notion that smokers are aware of the risks involved in smoking and perceive messages that share information as condescending (Wolburg 2006:307).
A further theme is that of a defiant reaction by smokers to the limits posed on their freedom by anti-smoking advertisements (Gilbert 2005:239;Maguire & Love 2006:100;Wolburg 2006:307).Smokers perceive these advertisements as infringing on their right to choose their own actions and this can result in active resistance in order to assert their right to smoke (Gilbert 2005:241;Wolburg 2006:307).It has also been found that smokers react to anti-smoking messages by assessing the risks involved in smoking as part of a risk-benefit analysis (Denscombe 2001:168-169;Gilbert 2005:236;Wolburg 2006:308).The perceived benefits of smoking, such as stress relief or relaxation, are weighed up against the health risks involved.In such a risk-benefit analysis, the world is seen as inherently risky and the decision to smoke is perceived as justified as it is made in this uncertain context (Gilbert 2005:238).Related to this theme is the finding that smokers deny their risk in the face of anti-smoking messages.Smokers react to fear appeal messages by minimising the risk involved in smoking or equating it with other calculated risks taken every day.Participants in these studies stated that one could die from a variety of causes, so why not die from smoking (Denscombe 2001:169-170;Wolburg 2006:294).Gilbert (2005:236) found that smokers perceive anti-smoking messages as simplifying the diverse and complex reasons that they draw on when choosing to smoke.The clear, rational and factual information about the health-risks involved in smoking presented by anti-smoking messages deny the complexity of these reasons.Furthermore, smokers in Gilbert's (2005:241) study found fear appeal messages to be ineffective, negative and counterproductive.Participants stated that being shown graphic imagery that relies on fear appeal results in their dismissing the message, and that the 'predominant focus on the negative medical effects of smoking often encourages smoking, because it fails to offer the positive effects of not smoking' (Gilbert 2005:240).However, in contrast to this finding, Montazeri and McEwen (1997:31) found that when comparing an advertisement that uses fear appeal to another advertisement that utilises a positive image of a non-smoker, participants had more favourable perceptions of the advertisement that uses fear appeal.
Most of the qualitative studies exploring perceptions of antismoking communication and the use of fear appeal have focused on adolescents and young adults and have pointed to salient factors in these groups that contribute to negative perceptions of fear appeal among young people (Denscombe 2001:158;Gilbert 2005:228-229;Wolburg 2006:320).In the qualitative studies identified for this review, ages of participants ranged between 10 and 25 (Denscombe 2001:162;Gilbert 2005:232;Maguire & Love 2006:98;Montazeri & McEwen 1997:30;Wolburg 2006:296), with only one study conducted including adult smokers (aged 40 to 49) in its sample (Montazeri & McEwen 1997:30).Considering that smoking prevalence is highest for adult South African smokers in the age group 25 to 49 (Department of Health 2003:3;Groenewald et al. 2007:677), perceptions of older age groups is an area worthy of further study.
It appears that the contradictory findings arising from some quantitative studies are also present in qualitative studies exploring perceptions of fear appeal in anti-smoking messages.In the light of this, as well as the fact that there is a lack of research that explores this topic in a South African context, the present study has the following objectives: To explore South African smokers' perceptions of anti-• smoking advertising that uses fear appeal To provide more in-depth understandings of the perceptions • of fear appeal advertising by using a qualitative approach.

Study deSign Sampling and data collection
In order to investigate smokers' perceptions of fear appeal messages in anti-smoking advertisements, focus group discussions were conducted with participants from Gauteng.Data were collected until saturation was achieved in that similar themes began to recur; this occurred when five focus group discussions had been conducted.A total of 30 participants were recruited by a professional recruitment company.In support of the exploratory nature of the study, purposive sampling was used to select a sample that would be diverse according to age, race and gender.Participants were further purposively sampled to include an equal spread of individuals identified as light, moderate and heavy smokers, based on the number of cigarettes smoked per day.Individuals who reported smoking one to five cigarettes per day were categorised as light smokers, six to ten cigarettes per day as moderate smokers and more than ten cigarettes per day as heavy smokers. 1 Once initial participants were identified, snowball sampling was used to recruit more participants.The motivation for recruiting participants in this manner was to ensure that a range of opinions was obtained.However, it is beyond the scope of the study to analyse differences in perceptions relating to race, gender or level of smoking and as such these constructs were not included in the analysis.
Participants' ages ranged between 25 and 49 years, with two group discussions being conducted with participants between 25 and 34 years (referred to as groups 1 and 2) and three groups with participants between 35 and 49 years (referred to as groups 3, 4 and 5).The focus groups were structured in this manner to allow for the emergence of potentially different perceptions held by different age groups -participants who share a similar life stage might be more open and comfortable in their interaction with each other.Considering the lack of research exploring the perceptions of adult smokers, as well as the higher prevalence rate of smoking for South African adults aged 25 to 49 (Groenewald et al. 2007:677), the current study excluded participants younger than 25 in order to focus on the perceptions of 'older' adult smokers.A further motivation for the chosen age group is that smoking is more likely to be established as a habit at this age than it would be among younger 'experimental' smokers such as adolescents and young adults (Gilbert 2005:232).
All of the focus group discussions followed the same procedure in which audio-visual equipment was used to screen a series of six anti-smoking advertisements.After viewing the advertisements, participants were guided in a discussion of their perceptions of the advertisements using a semi-structured interview guide.The same guide was used in all the focus group discussions and questions explored areas such as participants' general perceptions of the advertisements and their perceptions of the impact or efficacy of the advertisements.The focus group discussions were facilitated by the second author and were each approximately two hours in duration.The focus group discussions were tape-recorded.

Advertisements
The advertisements selected for the purpose of the study were sponsored by a non-governmental organisation, the South African National Council Against Smoking (NCAS), as part of the Council's Public Service Announcements.These advertisements were made available by the NCAS for use in 1. Wilson, Parsons and Wakefield (1999:140) use a classification of 25 or more cigarettes per day as heavy, 15 to 24 cigarettes per day as moderate, and less than 15 cigarettes per day as light.During recruitment for the present study it was found that very few smokers fall into the moderate to heavy categories and the range across categories was adapted to ensure that a spread across various levels of smoking was still obtained.
the study with the aim of generating research that can aid the Council in developing more effective anti-smoking messages.The series included advertisements that have been aired on South African television, as well as advertisements that have not been aired.The advertisements all have the objective of using fear appeal messages in creating greater awareness of the health risks associated with smoking, and are briefly described in Figure 1.

Data analysis
The focus group discussion data were transcribed and a thematic analysis was conducted.The thematic analysis entailed coding the transcribed data, grouping the codes into patterns or themes and then grouping related patterns into sub-themes (Aronson 1993:para.7).The coding process was primarily inductive in that predetermined categories were avoided and the codes were instead created as the data were examined (Charmaz 2008:155).
After coding, the transcriptions were re-read to identify patterns Advertisement 2: Suzy Q This advertisement is set at a house party with music playing in the background -a song called 'Suzy Q'.The people at the party are dancing, lazing around and chatting in a smoke-filled room.The camera closes in on a tall blonde girl.Two young men are shown to be watching her every move.They appear to be in awe of her beauty, but are puzzled by her next action: she begins picking her nose.At first, the two guys appear to be in disbelief.They look at one another and seem confused, but soon begin doing the same thing, until eventually the entire party of people are picking their noses and wiping their fingers off in the ash trays.The advertisement concludes with a message: 'What is so cool about a filthy habit?'

Advertisement 3: Children against smoking
This advertisement is set in a peaceful forest, featuring six-year-old twin sisters.
The music for the advertisement is in contrast to the setting and seems aggressive, attempting to reflect the sisters' rebellion against their parents' smoking.They speak about their parents and would like their parents to wash their hands after smoking a cigarette, to smoke outside instead of inside and want their parents to still be around when they are older.The message towards smokers is focused on adjusting their behaviour out of a responsibility towards their children.The final message of the advertisement is: 'As children, we hate the fact that our parents smoke!'

Advertisement 4: Chemical facts
This advertisement presents the voice of a male character talking about the various harmful chemicals in cigarettes.There are no people or music featured in the advertisement, and the tone of the male voice is factual and scientific.The final message of the advertisement is: 'Cigarettes contain deadly substances'.

Advertisement 5: Lung runs away
This advertisement begins with a young man sitting in a restaurant early in the morning.There are no other people in the restaurant and he orders breakfast from the waitress.As the waitress approaches the young man, he begins coughing profusely, until eventually a lung emerges from his mouth and falls onto the table in front of him.The waitress and the young man seem astounded and shocked.His lung literally jumps off the table and runs out of the door of the restaurant.The final message of the advertisement is: 'Get your lungs back'.

Advertisement 6: Inside-out
This advertisement is set in a forest.The camera pans in on a very attractive young woman with a cigarette in her hand.The woman slowly takes the cigarette towards her mouth and inhales, the camera then pans past her and a tree blocks out her face for a few seconds.When her face comes into view again, it is distorted and resembles the inside of a smoker's lungs.The final message in the advertisement is: 'If smoking did to your outside what it does to your inside, wouldn't you quit?'

Figure 1
Series of advertisements used in the study of perceptions, which were then listed by direct quotes or by paraphrasing common threads (Aronson 1993:para. 4).These patterns of perceptions were then sorted into themes.In the analysis, themes were defined as units derived from patterns such as 'conversation topics, vocabulary, recurring activities, meanings and feelings' (Taylor & Bogdan 1989:131).The themes that were identified from the focus group discussion data were pieced together to form a comprehensive matrix of the collective perceptions of anti-smoking advertising.
The final stage of the process of thematic analysis involved the development of an argument for choosing the selected themes to work with.At this point in the analysis the researchers referred back to literature on the topic, which allowed inferences about the information gathered.Internal validity was enhanced through matching the literature and the theoretical ideas developed about the data (Bryman 2004:273).Once the themes were collected and the literature studied, theme statements were ready to be formulated (Aronson 1994:para.9).
The second author conducted the primary analysis and an independent analysis was conducted by the first author.Where discrepancies occurred in the themes that were identified, they were reviewed by repeatedly returning to the data until a common understanding could be reached.This ensured the internal reliability of the data by subjecting it to more than one observer and reaching agreement on the analysis between two members of the research team (Bryman 2004:273).Direct quotes from the focus group discussions are presented where ever possible in order to allow the reader to make his or her own judgements about the conclusions reached during the analysis.This presents an opportunity for the credibility of the accounts to be established (Bryman 2004:274).

Advertisement 1: Don't jump
This advertisement features a 26-year-old male, Jonathan.He is standing on the edge of a very high building.As Jonathan is about to jump off the edge of the building, his friend is begging him not to jump by repeatedly pleading: 'Jonathan, don't do it, please don't do it …' The advertisement concludes when instead of seeing Jonathan jumping off the edge of the building, he lights up a cigarette, and the disappointment on Jonathan's friend's face is clear.The final message of the advertisement is: 'Smoking can kill you -literally'.