Five-arm, open-label RCT.
The study setting was a virtual web-based supermarket, with participants from the French NutriNet-Santé study. The eligibility requirement was grocery shopping involvement.
The intervention was to simulate one shopping situation with front-of-pack nutrition labels affixed on food products (December 2014 to March 2015). Participants were randomly assigned to one of five exposure conditions using a central computer system: Guideline Daily Amounts, Multiple Traffic Lights, Five-Color Nutrition Label, Green Tick, or control (no front-of-pack exposure). Given the nature of the intervention, masking of participants was not performed.
The primary outcome was the overall nutritional quality of the contents of the shopping cart, estimated using the United Kingdom Food Standards Agency nutrient profiling system. Secondary outcomes included energy and nutrient content of the shopping cart. Impact of the front-of-pack labels was also evaluated across sociodemographic subgroups based on age, educational level, income, and nutrition knowledge.
A total of 11,981 participants were included in the analyses (April 2015). The Five-Color Nutrition Label significantly led to the highest overall nutritional quality of the shopping cart, as reflected by lower Food Standards Agency scores (M=8.72; SD=2.75), followed by Multiple Traffic Lights (M=8.97; SD=2.68) and Green Tick (M=8.99; SD=2.71), compared with the control (M=9.34; SD=2.57) (p<0.0001). The Five-Color Nutrition Label was the only front-of-pack format that led to a lower content in lipids, saturated fatty acids, and sodium of the shopping cart (all p<0.05). The impact of the different front-of-pack labels was similar across sociodemographic subgroups.
The Five-Color Nutrition Label based on a color-coded and graded scale indicating overall nutritional quality is effective in promoting overall healthier food choices in all population subgroups.
This study is registered at www.clinicaltrials.gov NCT02385838.