Alcohol marketing and youth drinking: a rejoinder to the alcohol industry.

As societal concern has increased in relation to alcohol consumption and alcohol-related harm (Rehm et al., 2009; Nutt et al., 2010), attention on factors potentially influencing drinking behaviours has grown. One such factor that has been identified is alcohol marketing (Babor et al., 2010). Numerous studies assessing the impact of alcohol marketing on youth drinking have been carried out over the past 25 years. Early econometric studies investigating the associations between total alcohol advertising spend and drinking behaviour did not find any effect (Hastings et al., 2005). However, several recent sophisticated consumer studies, using a longitudinal cohort design, have found small but significant associations between exposure to, awareness of, and involvement with alcohol marketing, and youth drinking behaviours (Ellickson et al., 2005; Collins et al., 2007; McClure et al., 2009). This is unsurprising given the potential exposure to alcohol marketing experienced by young people. Although the majority of these studies have been conducted in the USA, a recent study from the UK published in Alcohol and Alcoholism also suggested that alcohol marketing influences youth drinking behaviour (Gordon et al., 2010). A briefing note published by the Portman Group (an agency linked to the alcohol beverage industry) presented the argument that alcohol marketing merely encourages brand switching, rather than encouraging consumption (Portman Group, 2010). The note commented that Gordon et al. (2010) found no association between awareness of alcohol marketing and either initiation of drinking or volume of alcohol consumed. This view was restated, and the topic debated, at the annual conference of Alcohol Concern in London, November 2010. The briefing note is selective in its reporting of the study of Gordon et al. (2010). It is true that no association was found between awareness of, or involvement with alcohol marketing at baseline, and ‘amount’ of alcohol in units consumed at follow-up. However, the briefing note fails to state that associations were found between ‘involvement’ with alcohol marketing and both ‘uptake’ of drinking and increased ‘frequency’ of drinking. Furthermore, ‘awareness’ of alcohol marketing at baseline was also associated with increased ‘frequency’ of drinking at follow-up. These findings offer support to the research hypotheses that awareness of, and involvement with, alcohol marketing would be positively associated with uptake and frequency of drinking. When measuring a complicated and sophisticated behaviour such as drinking, it is natural to expect that significant associations will not be found across all measures examined. The sample size in this study was limited, particularly as the analyses focused on sub-groups within the study (such as those who were drinkers at follow-up or those who were non-drinkers at baseline etc.) making it difficult to detect significant associations. Yet the fact that this study did find significant associations between alcohol marketing and youth drinking across several measures cannot be ignored. With reference to another point made in the briefing note, that confounders in associations found between marketing exposure and subsequent drinking were often not taken into account, our analysis showed associations that persisted after adjusting for a wide set of the likely social confounders. As with any scientific research, there are methodological and design limitations with consumer studies in this area. There will always be discussions over exactly how exposure to alcohol marketing is measured. It can be difficult to generate large enough random samples of participants, and interpretations of the analysis and findings can differ. However, although the perfect study remains out of reach, much of the consumer research in this area is of high academic quality, using recognizable, and tried and tested methods. Another view taken by alcohol producers is that the effects found by such studies are insignificant, and that other covariates such as parental and peer influence play a greater role in driving behaviours. Such factors do undoubtedly influence behaviours (Bobo and Husten, 2000). Indeed, it is unsurprising that a complicated social phenomenon such as drinking behaviour is influenced by a range of factors. Nevertheless, the evidence base has highlighted small but statistically significant associations between alcohol marketing and youth drinking behaviour, even after controlling for parental and peer influences, and this cannot be simply ignored (Anderson et al., 2009). Indeed, three recent systematic reviews have all suggested that alcohol marketing does influence youth drinking behaviour (Anderson et al., 2009; Smith and Foxcroft, 2009; Meier et al., 2008). This evidence base continues to develop with an Australian cross-sectional study (Jones and Magee, 2011) that found an association between exposure to alcohol advertising and increased alcohol consumption. The UK parliament and government policy seems to have responded, with the House of Commons Health Committee report identifying alcohol marketing as a matter for concern, and a policy response being considered in the coalition government’s public health consultation (Department of Health, 2010; House of Commons Health Committee, 2010). The alcohol industry seems to be focused upon continuously debating the evidence base; ‘a negative impact of alcohol marketing cannot be ruled out but it has not yet been proved by the research evidence’ (Portman Group, 2010, p2). However, with numerous Alcohol and Alcoholism Vol. 46, No. 4, pp. 369–370, 2011 doi: 10.1093/alcalc/agr043