Background High sugar consumption contributes to the global obesity epidemic. Sugar-sweetened beverages (SSBs) constitute a significant proportion of sugar consumption and are independently associated with non-communicable diseases. Some equivocal evidence indicates the influence of marketing on sugary food and drink sales and consumption. However, little research has explored this relationship for SSBs, among adult samples, or using diverse marketing exposures. There are no cross-country comparisons of this relationship.

Aim We aim to answer:

1. Does SSB consumption vary according to self–reported exposure to sugary drink marketing?

2. Does the relationship between SSB consumption and self–reported exposure to sugary drink marketing vary by country?

Methods We analysed data from the 2017 International Food Policy Study (IFPS), a cross-sectional, online survey that sampled adults aged 18 to 64 years in Australia (n=3,767), UK (n=4,047), Mexico (n=4,057), USA (n=4,868), and Canada (n=2,139; aged 31 to 64 years). Surveys covered eating patterns, food environments, purchasing, knowledge of and attitudes to a range of policy measures. This included a seven-day Beverage Frequency Questionnaire. Dependent (SSB consumption) and independent variables (amount of self-reported sugary drinks marketing exposure) were computed, and potential confounders identified. A multinomial logistic regression model examined relative odds of non-, low- or high- weekly SSB consumption by marketing exposure, country, sex, age, and work status. The same model was reconstructed with interaction between marketing exposure and country, and a log likelihood test was used to compare the models.

Results SSB consumption varied according to self-reported exposure to sugary drinks marketing. 57% of respondents consumed SSBs, and 72% self-reported exposure to sugary drinks marketing. Compared to those reporting no sugary drinks marketing exposure, those reporting greater exposure were more likely to have low- than no-SSB consumption and more likely to have high- than no-SSB consumption (ps<0.05). Respondents in Mexico (relative to Canada) and men (vs women) also reported higher sugary drinks consumption, whereas those from the UK, USA (relative to Canada), and older respondents reported lower consumption (ps<0.05). The relationship between self-reported sugary drinks marketing exposure and SSB consumption varied by country (p=0.0286).

Discussion SSB consumption was positively associated self-reported exposure to sugary drinks marketing in all countries, but this relationship varied by country. These analyses are cross-sectional, but the IFPS panel will be re-contacted annually enabling future longitudinal analyses. Improving understanding of the influence of marketing on SSB consumption could help policymakers to augment current strategies to reduce SSB consumption.