Oral Presentation ANZOS Annual Scientific Meeting 2021

Identifying the most effective on-bottle warning labels to reduce consumption of sugar-sweetened beverages (#68)

Caroline Miller 1 2 , Kerry Ettridge 1 2 , Melanie Wakefield 3 , Simone Pettigrew 4 , John Coveney 5 , David Roder 6 , Sarah Durkin 3 , Gary Wittert 1 , Jane Martin 7 , Joanne Dono 1 2
  1. University of Adelaide, Adelaide, SA, Australia
  2. Health Policy Centre, SAHMRI, Adelaide, SA, Australia
  3. CBRC, Cancer Council Victoria, Melbourne, Victoria, Australia
  4. Food Policy, George Institute, Sydney, NSW, Australia
  5. Flinders University, Adelaide, SA, Australia
  6. University of South Australia, Adelaide, SA, Australia
  7. Obesity Policy Coalition, Melbourne, Victoria, Australia

Background: Australians are high consumers of sugary drinks. Interventions aimed at reducing population-level consumption would produce public health benefits. On-bottle warning labels that convey the risks of consumption without requiring the consumer to understand complex nutrition information is gaining momentum internationally. This study aimed to identify warning labels that resonated most with consumers.

Method: This online study involved random allocation of n=3500 sugary drink consumers aged 14-60 years, to view a beverage from one of 5 drink conditions: (a) 250ml Coke; (b) 600ml Coke; (c) 250ml lemon mineral water; (d) 500ml lemon mineral water; or (e) 500ml unsweetened orange juice. Respondents viewed this beverage a total of 6 times with randomly assigned label variants from the following categories: Health-text (5 label variants), Health-graphic (5), Sugar-text (4), Sugar-pictogram (5), exercise information (4), and energy information (4). Respondents rated each label on four perceived effectiveness scales: cognitive elaboration, persuasiveness, emotional response, and overall effectiveness.

Results: The difference in average overall effectiveness score (range 0-10) across the 27 labels was small, with four out of five graphic health labels scoring between 6.5 and 6.7, and three out of five sugar pictogram labels scoring between 6.3 and 6.6. The energy information labels received the lowest scores, ranging between 3.9 and 5.0. The results were similar across drink types, including fruit juice. The health effects labels had slightly lower scores on ‘taught me something new’ and ‘is relevant’ compared to sugar and exercise information labels.

Conclusions and implications: Labels conveying simple and easily understood information (e.g. teaspoons of sugar) were consistently rated highly. These results can be used in future studies to evaluate whether on-bottle warning labels contribute to behaviour change in real-world settings. On-bottle warning labels have the potential to improve the food environment through product reformulation and increasing consumers’ awareness of unhealthy drinks.