Message on a bottle: are alcohol warning labels about cancer appropriate?

• Main Authors: Emma R. Miller, Imogen J. Ramsey, Genevieve Y. Baratiny, Ian N. Olver
• Format: Article
• Language: English
• Published: BMC 2016-02-01
• Series: BMC Public Health
• Subjects: Alcohol; Cancer prevention; Warning labels
• Online Access: http://link.springer.com/article/10.1186/s12889-016-2812-8

Abstract Background Although most Australians are unaware of the risk, there is strong evidence for a direct link between alcohol consumption and many types of cancer. Warning labels on alcohol products have been proposed as a cost-effective strategy to inform the community of this health risk. We aimed to identify how Australians might respond to such an approach. Methods We conducted a national online survey canvassing responses to four separate cancer warning messages on labels. The graphically presented messages were informed by qualitative data from a series of focus groups among self-identified ‘light-to-moderate’ drinkers. For each label, participants were asked their level of agreement with impact statements about raising awareness, prompting conversation, influencing drinking behaviour and educating others about cancer risk. We analysed responses according to demographic and other factors, including self-reported drinking behaviour (using the 3-item Alcohol Use Disorder Test – AUDIT-C – scores). Results Approximately 1600 participants completed the survey, which was open to all Australian adults over a period of 1 month in 2014. Overall, the labels were well received, with the majority (>70 %) agreeing all labels could raise awareness and prompt conversations about the cancer risk associated with alcohol. Around 50 % or less agreed that the labels could influence drinking behaviour, but larger proportions agreed that the labels would prompt them to discuss the issue with family and friends. Although sex, AUDIT-C score and age were significantly associated with agreement on bivariate analysis, multivariate analyses demonstrated that being inclined to act upon warning label recommendations in general was the most important predictor of agreement with all of the impact statements. Having a low AUDIT-C score also predicted agreement that the labels might prompt behaviour change in friends. Conclusions The findings suggest that providing detailed warnings about cancer risk on alcohol products is a viable means of increasing public awareness of the health risks associated with alcohol consumption. Further research is needed to explore the ability of such warnings to influence behavioural intentions and actual drinking behaviour.