Warning labels on cigarettes, threatening words combined with scary pictures, are popular with politicians. The European Union is advising its member states to accompany the warning texts with graphic illustrations of the dangers of smoking. However, all questions about the effectiveness of these warnings meet with evasive answers or references to research publications that cannot be found. Only in 2004 did the first peer-reviewed publication on cigarette warning labels appear, in the American Journal of Public Health. Using a no control group, post-test only, design, Hammond et al.1 suggest that cigarette warning labels do not have aversive effects and that policy makers should not be reluctant to introduce these labels. Their evidence is based on self-reports in a longitudinal survey, after the introduction of the labels. Their main argument is that reported emotional reactions to the labels predict reported cessation attempts at follow-up.
We are not convinced. Asking a population of smokers about intentions to quit always results in substantial percentages of intenders, just as asking smokers in any follow-up if they have tried to quit.2 Smokers will often say they quit because of their health. Without an experimental design, there is no evidence that warning labels are responsible for these outcomes, as ‘third variables’ may be overlooked, nor is there any evidence that quitting percentages are higher than before the introduction of warning labels.
Hammond and colleagues further ignore the large research tradition on fear-arousing communications.3,4 The evidence in this area suggests that especially those who are most at risk react defensively to these messages. After watching fear-arousing and non-fear-arousing messages, smokers and non-smokers prefer the fear-arousing message because of its expected effectiveness. Smokers express a stronger intention to quit after a fear-arousing message compared with a non-fear-arousing message. However, when asked about their priorities, quitting had actually become a lower priority compared with other health behaviours.5 Even more, EEG/ERP analyses during exposure to smoking-related health commercials suggested that smokers allocated less attention resources to high as opposed to low threatening messages.6,7 Other authors find similar results: self-declared positive reactions, but observed negative reactions, including less behaviour change, among those that are most at risk.8
In summary, high personal relevance (smoking), in combination with low self-efficacy for the recommended action (quitting), leads to defensive reactions as a result of fear-arousing messages. This ‘psychological immune system’ helps in maintaining a positive self-image and may operate largely outside of awareness.9,10 Defensive reactions serve to get rid of the fear, not necessarily the threat. Policy makers should thus be reluctant to introduce cigarette warning labels and should instead focus on more effective interventions and policies.2