Restricting child-directed ads is effective, but adding a time-based ban is better: evaluating a multi-phase regulation to protect children from unhealthy food marketing on television

Abstract

Background: As childhood obesity rates continue to rise, health organizations have called for regulations that protect children from exposure to unhealthy food marketing. In this study, we evaluate the impact of child-based versus time-based restrictions of "high-in" food and beverage advertising in Chile, which first restricted the placement of "high-in" advertisements (ads) in television attracting children and the use of child-directed content in high-in ads and, second, banned high-in ads from 6am-10pm. "High-in" refers to products above regulation-defined thresholds in energy, saturated fats, sugars, and/or sodium. High-in advertising prevalence and children's exposure to high-in advertising are assessed.

Methods: We analyzed a random stratified sample of advertising from two constructed weeks of television at pre-regulation (2016), after Phase 1 child-based advertising restrictions (2017, 2018), and after the Phase 2 addition of a 6am-10pm high-in advertising ban (2019). High-in ad prevalence in post-regulation years were compared to prior years to assess changes in prevalence. We also analyzed television ratings data for the 4-12 year-old child audience to estimate children's ad exposure.

Results: Compared to pre-regulation, high-in ads decreased after Phase 1 (2017) by 42% across television (41% between 6am-10pm, 44% from 10pm-12am) and 29% in programs attracting children (P < 0.01). High-in ads further decreased after Phase 2, reaching a 64% drop from pre-regulation across television (66% between 6am-10pm, 56% from 10pm-12am) and a 77% drop in programs attracting children (P < 0.01). High-in ads with child-directed ad content also dropped across television in Phase 1 (by 41%) and Phase 2 (by 67%), compared to pre-regulation (P < 0.01). Except for high-in ads from 10pm-12am, decreases in high-in ads between Phase 1 (2018) and Phase 2 were significant (P < 0.01). Children's high-in ad exposure decreased by 57% after Phase 1 and by 73% after Phase 2 (P < 0.001), compared to pre-regulation.

Conclusions: Chile's regulation most effectively reduced children's exposure to unhealthy food marketing with combined child-based and time-based restrictions. Challenges remain with compliance and limits in the regulation, as high-in ads were not eliminated from television. Yet, having a 6am-10pm ban is clearly critical for maximizing the design and implementation of policies that protect children from unhealthy food marketing.

Keywords: Childhood obesity; Children; Food advertising; Food marketing; Health policy; Television.

Fig. 1

Timeline of regulatory phases and months of television sampling to examine pre-regulation (2016), two years of post-regulation during Phase 1 (2017, 2018), and one year after Phase 2 (2019). “High-in” refers to products above regulation-defined thresholds in energy, saturated fats, sugars, and/or sodium according to Chile’s Food Labeling and Advertising Law [21]

Fig. 2

Bars represent average daily gross rating points (GRP) for 4–12 year-old child audiences for food/beverage ads (matched with nutritional data) in total, high-in ads in total, and high-in ads placed in TV programs attracting child audiences for samples of television from April–May 2016 pre-regulation and April–May 2017, 2018, and 2019 post-regulation. GRP is a standard industry metric, in which 1 GRP indicates the given type of ad reached 1% of the target audience and 2 GRP may indicate reaching 1% of the target audience twice or 2% of the target audience once. Final (June 2019) regulation thresholds in energy, saturated fats, sugars, and sodium are used to define “high-in” products across all years. Differences in amounts of ads within each category are tested with Pearson chi-square tests evaluated at P < 0.01. Percentage change reflects change relative to either pre-regulation or to prior year and calculated as (Year 2 – Year 1)/Year 1. **P < 0.001; *P < 0.01; ^‡ = P < 0.05, approaches significance