Related Experiment Video
Updated: May 28, 2026

Using a Virtual Store As a Research Tool to Investigate Consumer In-store Behavior
Published on: July 24, 2017
A Randomized Intercept Survey Trial to Test the Effectiveness of Multiple Traffic Light Labels on Online Grocery
Soye Shin1,2, Ali Shubbar Jawad3, Buthaina Yusuf Ajlan4
1Health Services Research and Population Health, Duke-NUS Medical School, Singapore 169857, Singapore.
Abstract:
Background/Objectives: Multiple Traffic Light (MTL) front-of-pack (FOP) labels are being considered in Bahrain. We tested whether an adapted MTL label improves the nutritional quality of grocery purchases. Methods: In a two-arm randomized controlled intercept trial (January-May 2025), adults (≥21 years) responsible for household grocery shopping were recruited in high-footfall public venues and asked to complete a one-time shop on a tablet-based, purpose-built online grocery platform. The MTL label was adapted for Arabic reading direction and displayed per-serving nutrients and % recommended daily intake. Treatment effects were estimated using ordinary least squares regressions with robust standard errors and covariate adjustment. Results: Of 395 randomized participants, 360 were included in primary analyses (control n = 183; MTL n = 177). MTL exposure was not associated with a significant change in the primary outcome (basket weighted average MTL score per serving; β = 0.037; p = 0.64) or in per-serving calories and nutrients of concern (all p > 0.17). In the post-shop assessment, only 47.2% of participants correctly interpreted MTL labels, indicating modest objective label comprehension under the study conditions. Conclusions: These findings suggest that the impact of front-of-pack labels likely depends on both implementation features and consumer understanding, and that pairing labels with public communication and nutrition literacy initiatives may be necessary to maximize the effectiveness of labels in Bahrain and the wider Gulf region.
Related Concept Videos
Blinding
Bioequivalence Experimental Study Designs: Completely Randomized and Randomized Block Designs

