Conclusions

This study identified ads that had the greatest resonance and potential effectiveness among African audiences, and supported adaptation as a viable approach for tobacco control mass media communication in Africa. Coughing Child was the strongest-performing radio advertisement across the broad range of measures for smokers and non-smokers. Suffering showed a comparably strong performance, although it was rated lower than Coughing Child on PE by non-smokers and Should Not Smoke Around Others by smokers. Baby Alive and Lung consistently performed strongly among the TV ads, although these were not rated higher than Industry Revealed on the measure of anti-industry sentiment and support for government actions.

This study replicates findings from other countries on the effectiveness of negative health harms-focused messaging for tobacco control behaviour change communication.6 All of the highest-performing ads featured hard-hitting and emotionally evocative content. Both of the highest-performing ads—Coughing Child and Baby Alive—focused on the effects of tobacco on vulnerable and blameless victims—children, in this case. This messaging approach encourages smokers to modify their behaviour and raises concern among non-smokers about SHS exposure, and has been used successfully previously.16 ,17

The high-performing ads were originally developed in high-income countries (HICs) yet performed strongly in Africa, particularly on the important measure of PE. Smokers’ ratings of personalised PE of television ads has been shown to predict change in quitting-related intentions and behaviours.18 Their high performance in this study—in three diverse African settings—suggests that such ads may be successfully adapted for use in Africa, thereby channelling resources from the development of advertisements to campaign implementation. The messages that performed most strongly in this study—and therefore lent themselves best to adaptation—relied on hard-hitting content that depicted bodily damage and evoked strong negative emotions. These findings underscore the universality of motivations for health across contexts: By graphically depicting the illness and burden caused by tobacco, these ads increase health concerns and the motivations to reduce risk by avoiding or quitting tobacco.

The findings highlight the importance of the clarity of messages and their ability to resonate with the target audience. The testimonial ads, Emphysema and Break the Chain, performed more modestly in this study. Previous message-testing research has also shown that while testimonials can be powerful in their countries of origin, they rely on cultural context and may not resonate with populations in other countries.15 Sponge, which has consistently been among the strongest-performing TV ads in message-testing in other LMICs,15 showed a positive performance overall, but the qualitative discussions in this study were found to be less easily understood. One possible explanation may have been its reliance on metaphorical imagery, albeit graphic in nature. Compared with the more literal and horrific sounds and imagery in Coughing Child and Baby Alive, the message in Sponge may not have been as accessible. This ad also relies heavily on an understanding of the concept of tar build-up in the lungs, but qualitative discussions revealed that many participants did not have prior knowledge of what tar was.

Though anti-industry messages tended to receive the lowest ratings in motivating behaviour change, they were the most likely of the ads to arouse anti-industry sentiment and support for government actions. This finding suggests that anti-industry messages are not optimal for communications that aim to directly prevent tobacco use or promote cessation. However, other studies have shown that anti-industry messages may be used successfully when the goal is to create anti-industry sentiment19 ,20 and advocacy around an issue, which may itself ultimately support longer term behaviour change at a population level.21

This study found that hard-hitting ads that portrayed the harms of SHS exposure were at least as highly rated as other visceral ads that focused on smoking cessation. As noted above, some features of the ads, including their graphic health messages and their focus on tobacco's impact on vulnerable children, may have been the main reasons for their higher performance. However, it is also conceivable that in a population where smoking rates are generally low, the health issues that affect vulnerable young non-smokers are more salient, urgent and universal.

There were minimal subgroup differences, with non-smokers tending to rate ads more positively than smokers; parents more than non-parents; older adults more than younger adults; and women more than men. These findings are consistent with others that have found these groups to be generally more sympathetic to tobacco control messages.15 However, the overall pattern of findings between groups remained consistent.

This study had several limitations. First, only participants with basic literacy could be included; the ads’ effectiveness on non-literate populations was not studied. However, this protocol has been similarly implemented in other LMICs, where ads that rated highly in the message-testing were later found to be effective in campaign impact evaluations, including among lower-literacy populations.16 Such findings increase the expectation of similar positive campaign outcomes in Africa. The sample was also limited to those in the age group 18–40 years, to ensure homogeneity within groups. Second, owing to cost considerations, only basic ad adaptations were tested, which could have interfered with their performance. However, the successful performance of the ads, despite minimal adaptation, suggests that they may perform more strongly with appropriate contextualisation. In addition, for efficiency and to optimise comparability, the same scales were used in all countries, ad types and smoking status groups. Third, the strong positive response bias across the Senegalese sample was mitigated by coding only ‘strongly agree’ ratings as positive responses, so response coding was not uniform across countries. Lastly, the study was conducted in African countries with developed media markets where rigorous message-testing was possible (ie, widespread consumption of mass media and presence of research agencies). This may limit the generalisability of the results to other African contexts, particularly lower-income settings and those with less pre-existing exposure to mass media. However, these same ads have evaluated well in urban and semirural settings of other low-income countries, such as China, India and Bangladesh.8

This study provides important guidance for the development of tobacco control campaigns in Africa. It is the first systematic study of tobacco control mass media advertisements and included television and radio materials. Hard-hitting, health harms-focused advertisements were perceived to be most effective and most likely to motivate positive behaviour change. In this respect, the findings from Africa were consistent with findings from other countries, which have led to greater pressure on governments to run antitobacco campaigns using graphic and hard-hitting imagery.22

This study also demonstrated the feasibility of adaptation as a successful approach in Africa. The upfront costs of developing and testing new campaign materials can be daunting and prohibitive. Ads from HICs that are found to be effective in LMICs through pretesting can be quickly and cost-efficiently adapted for local use as long as they are culturally appropriate. The identification of effective radio and television advertisements through this study provides programme managers with materials that can be expected to resonate with African audiences, create concern and motivate healthier tobacco-related attitudes and behaviours. Saving money through adaptation allows resources that would otherwise be expended on campaign development to be used to boost campaign reach, while taking the guesswork out of developing effective population-wide campaigns.23 Programme managers should consider adaptation as one measure among various strategies to save costs on developing and sustaining campaigns. Additional strategies include buying radio airtime,14 extending campaign reach through the press or social media, and advocating governments to pass legislation that allocates free prime time to tobacco control advertising.