Do Sugary Drink Policies Increase Purchases of Non-Calorically Sweetened Beverages? Evidence from Chile

Abstract

Abstract Objectives Chile's 2016 on food labeling and advertising reduced household (HH) sugary drink purchases by 24%. However, it is unclear if this regulation was also associated with a subsequent increase in non-calorically sweetened (NCS) beverage purchases, nor who is most likely to be a top consumer of NCS beverages. Methods We used longitudinal monthly-level data from Kantar WorldPanel Chile 2015–2017 (HH = 2383) linked to yearly updated nutrition facts panel data to examine changes in NCS beverage purchases from the pre-implementation period (January 2015-June 2016) to the post-implementation period (July 2016-December 2017). Beverages were categorized into 4 mutually exclusive groups: CS: sweetened with caloric sweeteners, NCS: sweetened with non-caloric sweeteners, CS + NCS: sweetened with both sweeteners, and unsweetened. For the pre-implementation period, we used multinomial logistic regression to model HH sociodemographic predictors (head of HH educational level, HH assets, HH composition and region) of being in tertiles of NCS beverages volume purchased. We estimated the changes in beverage volume purchases between pre- and post-implementation using linear regression models adjusted for HH sociodemographic characteristics and month of data collection to account for seasonality. Results HH with higher education (35%) and HH assets (38%) were more likely to be high NCS beverages purchasers, while HH with lower education (37%) and HH assets (42%) were more likely to be low NCS beverages purchasers. After policy implementation, volume of purchases of NCS beverages increased by 6.2 mL/capita/day (95% CI: 4.7, 7.6), while volume of CS, CS + NCS and unsweetened beverages purchases decreased by 24.6 mL (95% CI: −27.5, −21.7), 4.3 mL (95% CI: −6.4, −2.2) and 16.4 mL (95% CI: −20.8, −12.1), respectively. Conclusions In Chile, people of high SES purchase more NCS beverages. After policy implementation, the volume of CS, NCS + CS and unsweetened beverages decreased while the volume of NCS sweetened beverages increased. Funding Sources Bloomberg Philanthropies, International Development Research Centre, and Chilean National Agency of Research and Development. This project was also supported by the Carolina Population Center.