Commentary on Zacher et al. (2014): Emerging behavioural impact of packaging changes in Australia: time for action elsewhere

Authors


As with all tobacco control policies, the burden of proof for effectiveness of ‘standardised tobacco packaging’ lies with the tobacco control community, in line with a desire for evidence-based public health. This desire could be questioned. Arguably, when a product kills one in two of its regular users, the burden of proof should lie elsewhere: policy makers should err on the side of protecting the health of the public when introducing preventive measures (in this case by removing attractive tobacco packaging) unless there is strong evidence that such measures could increase harm [1].

Still, in our current topsy-turvy ‘Alice in Wonderland’ world, Zacher and colleagues [2] provide important new evidence on the impact of standardised packaging. In the absence of other advertising, and given the frequency with which smokers smoke over the course of a day, cigarette packaging is the strongest vehicle tobacco companies have to promote their products [3], and standardised packaging takes away this vehicle. Zacher et al's findings are consistent with the growing body of evidence showing benefits of standardised packaging, and show for the first time how its implementation can impact behaviour.

Zacher and colleagues used an innovative but robust methodology [4] to examine the impact of changes in packaging on propensity to smoke and cigarette pack display. It has several important strengths: (a) it is one of the first reports from a country that has introduced standardised packaging; (b) it uses a strong pre-post design with objective measurement; and (c) several potential confounders were taken into account. There are limitations; in particular, Australia updated/enhanced and enlarged graphic health warnings at the same time as introducing standardised packaging, meaning that changes can only cautiously be interpreted as pertaining to standardised packaging. However, standardised packaging will likely always be introduced alongside enhanced warnings and indeed this would be favoured given known synergies between tobacco control strategies [5].

The findings are interesting and not always intuitive. There was a reduction in observed smoking of 23% and a reduction in pack display by 15%. The authors rightly argue that it is unlikely that packaging changes reduced the proportion of smokers by a quarter and that instead it is more likely that packaging influenced propensity to smoke, or desire to be observed smoking. The phased implementation of the packaging changes affected the outcomes: reductions in smoking were observed as soon as changes were introduced (perhaps also due to media discussions, known to be an important factor [6]), but changes in pack display occurred only after branded packs were completely phased out. Longer term studies will show whether these changes will be sustained or possibly enhanced.

Taken together, the behavioural impacts found in this study are very important. If, as well as reducing the visibility of branding, standardised packaging influences smoking behaviour then this is tremendously important. Reducing smoking will benefit the health of smokers and the health of those around them. Additionally, it will reduce the visibility of smoking, thereby reducing opportunities for social and role modelling [7, 8], as well as lowering brand awareness; these will all reduce adolescent uptake of smoking.

How and where packs were displayed also changed. Before packs changed, smokers most commonly put them face-up, concealing the then larger warnings on the back [4]. Putting the pack face-up declined after the packaging changes and packs were also more likely to be hidden (using phones, wallets etc). Hiding or avoiding the pack suggests some discomfort about their smoking behaviour or the sight of their cigarette packs. Evidence suggests that these behaviours do not diminish the overall effect of health warnings [9] and they may well not diminish the overall impact of standardised packaging either.

Reducing inequality in smoking is one of the biggest challenges in tobacco control. Here again the findings are of interest. Smoking and pack display did not decline significantly in low socioeconomic status (SES) areas whereas they did in higher SES areas. However, face-up display declined only in low SES areas and hiding the packs increased in low SES. What does this mean? Social norms may indeed override the influence on propensity to smoke as the authors argued, but the behavioural changes are suggestive again of increased discomfort with being seen smoking cigarettes from standardised packs. Whether such discomfort translates into changes in smoking behaviour over the longer term remains to be seen. Nevertheless, removal of branding reduced marketing opportunities in low SES areas which should ultimately impact youth smoking in those areas.

Standardised packaging is likely not to be the silver bullet of tobacco control but we recommend implementation given this study builds on the growing consistent body of evidence [10] pointing to positive impact, which should quench the thirst for evidence-based public health. Evidence helps to counter the continuing obfuscation of the tobacco industry whose wealth and power [11] must not be underestimated [12]. The authors are to be applauded for their courage in the face of this industry and their efforts in producing this and other exemplary studies [e.g. [13]] in building the evidence base for tobacco control policies.

Declaration of interests

The authors declare no competing interests. AM is a co-author of a systematic review of standardised packaging [10].

Ancillary