The Effect of Cigarette Packaging and Illness Sensitivity on Attention to Graphic Health Warnings: A Controlled Study

Eye tracking applied to tobacco smoking: current directions and future perspectives

Over the years the general awareness of the health costs associated with tobacco smoking has motivated scientists to apply the measurement of eye movements to this form of addiction. On one hand they have investigated whether smokers attend and look preferentially at smoking related scenes and objects. In parallel, on the other hand eye tracking has been used to test how smokers and nonsmokers interact with the different types of health warning that policymakers have mandated in tobacco advertisements and packages. Here we provide an overview of the main findings from the different lines of research, such as the evidence related to the attentional bias for smoking cues in smokers and the evidence that graphic warning labels and plain packages measurably increase the salience of the warning labels. We point to some open questions, such as the conditions that determine whether heavy smokers exhibit a tendency to actively avoid looking at graphic warning labels. Finally we argue that the research applied to gaze exploration of warning labels would benefit from a more widespread use of the more naturalistic testing conditions (e.g. mobile eye tracking or virtual reality) that have been introduced to study the smokers' attentional bias for tobacco-related objects when freely exploring the surrounding environment.

The short-term impact of standardised packaging on smoking and snus use in Norway

Introduction

Standardised packaging on tobacco products was required in Norway July 1, 2018. We report pre-registered analyses of the potential impact on daily smoking and on daily snus use among women and men.

Methods

Interrupted time series (segmented regression) on repeated cross-sectional surveys (2012–2019) from two sources: probability samples (Registry Sample, N = 46,957) and market research samples (Market Research Sample, N = 64,465) of Norwegian adults aged 16–79. Self-reported daily smoking and snus use were regressed on a step change impact variable, controlled for trend and demographics (sex, age, region, and education based on national registers in the Registry Sample, and self-reported in the Market Research Sample).

Results

There were tendencies of a decline in smoking (Odds Ratio [OR] = 0.94; 95% confidence interval [CI] = 0.87, 1.02; lower-tail p-value [P_lower] = 0.07), and women’s snus use (OR = 0.89; CI = 0.77, 1.03; P_lower = 0.06), but not men’s snus use (OR = 1.01; CI = 0.92, 1.11; P_lower = 0.59). Analyses using only the Registry Sample did not detect declines in smoking (OR = 0.99; CI = 0.88, 1.11; P_lower = 0.43) or women’s snus use (OR = 0.99; CI = 0.80, 1.24]; P_lower = 0.48), and indicated no decline in men’s snus use (OR = 1.18; CI = 1.03, 1.35; P_lower = 0.99). Exploratory analyses suggested potential acceleration of the declining trend in smoking (change in trends, OR = 0.97) and of the increasing trend in men’s snus use (OR = 1.03).

Conclusions

The analyses indicate that standardised packaging in Norway did not produce a decline in men’s snus use. Results are inconclusive regarding smoking and women’s snus use. Exploratory analyses indicated a decrease in smoking and an increase in men’s snus use.

Implications

We could not confirm or disconfirm whether standardised packaging is an effective tobacco control measure in a Norwegian context. According to our analyses, standardized packaging may have effects on smoking prevalence and women’s snus use, but is unlikely to reduce men’s snus use. The present results may reflect higher effectiveness of standardised packaging for products with stronger health warnings. As the results varied according to samples and outcomes, the study underlines the importance of pre-registering future analyses on this topic. Future confirmatory research should test models of gradual impact of standardised packaging.

Effects of Different Graphic Health Warning Types on the Intention to Quit Smoking

Anti-smoking advertisements are widely used to demonstrate to smokers the harm of smoking, and graphic health warnings (GHWs) are expected to have a positive effect on the intention to quit smoking. This study investigated which type of GHW (health-related threat (H-GHW) vs. social threat (S-GHW)) is more effective. Two types of GHWs for tobacco were shown to 28 daily smokers and 25 non-smokers while measuring their eye movements using an eye tracker. The time spent fixating on the GHWs was measured as an index of attentional bias. Participants were also asked to evaluate the unpleasantness of the images. They stated their intention to quit smoking in response to each image in a separate session. Multiple regression analysis was used to identify the effects of psychosocial factors on the intention to quit smoking in smokers and the intention to remain as non-smokers in the non-smokers. Both smokers and non-smokers reported greater unpleasantness and cessation intentions in response to H-GHWs than to S-GHWs. Non-smokers found both types of GHWs more unpleasant than smokers did. No differences were found in gaze fixation on GHWs between the two groups. When smokers viewed S-GHWs, the intention to quit smoking was greater as they felt more unpleasant. For non-smokers, the intention to remain non-smokers was greater when they felt more unpleasant and when the attention to H-GHWs was lower. Different psychological factors in anti-smoking advertisements are involved in the intention to quit smoking in smokers and to maintain a non-smoking status in non-smokers. Different approaches should be used according to the types of warning (e.g., warnings emphasizing a negative influence on others or on their own health) in anti-smoking campaigns.